Abstract
In this paper, I present and defend a novel account of doubt. In Sect. 2, I make some preliminary observations about the nature of doubt. In Sect. 3, I introduce a new puzzle about the relationship between three psychological states: doubt, belief, and confidence. I present this puzzle because my account of doubt emerges as a possible solution to it. Lastly, in Sect. 4, I elaborate on and defend my account of doubt. Roughly, one has doubt if and only if one believes one might be wrong; I argue that this is superior to the account that says that one has doubt if and only if one has less than the highest degree of confidence.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This is not to say that no one has attempted to define ‘doubt’. For a helpful discussion and critique of the views of doubt of Russell, Pierce, and others, see Peels (ms). None of those philosophers employ the standard procedures in analytic philosophy that I mention above.
Not all types of doubt ascriptions will be explored. I will focus on those doubt ascriptions that will be necessary for the reader to grasp the sense of ‘doubt’ in which I am interested. For an exploration of more doubt ascriptions, see Peels (ms).
A minor criticism of Howard-Snyder’s suggestion is that the doubt need not incline one to disbelieve p (i.e., believe \(\sim \)p); it need only incline the person to not believe p. For example, some doubts might come in the form of undercutting defeaters. To use Pollock’s well-known example, suppose some widgets look red and I come to believe they are. I then learn that they have red lights shining on them. Then I do not gain a reason to disbelieve that they are red, but only to not believe that they are red. I would then have a doubt about whether they are red. This correction is accounted for in \(\hbox {Doubts}_{2}\).
Consider that internalists think that my demon world twin and I are evidentially identical; this supports the claim that the reasons one appears to have are the reasons one has, and vice versa.
Cf. Schellenberg (2005, p. 95).
I am thankful to a referee of this journal, whose comments led me to develop these points about how the strength of one’s doubts could come in degrees.
Thanks to John Schellenberg for drawing my attention to this sort of ascription. It is also Howard-Snyder’s (2013) focus.
Thanks to Brent Braga and Andrew Melnyk for helping me see the fruitfulness of distinguishing verb and noun doubt ascriptions.
Consider also, “It is doubtful that Sally will arrive on time,” which uses an adjectival form of ‘doubt’. It seems to express that there are reasons to doubt that Sally will arrive on time. This is the sense of ‘doubt’ that I am putting aside for this paper. Thanks to Adam Auch for helpful discussion.
Thanks to Philip Swenson for helpful conversation about these examples.
It is not always the case that if one can have much or little of X, then X comes in degrees. One can have much or little gold, but gold does not come in degrees. However, this inference is plausible in the case of mental states: if one can have much or little of mental state X, then X comes in degrees. For example, it is plausible to infer from the fact that we can have much or little desire, much or little hope, and much or little happiness, to the conclusion that these mental states come in degrees. So, this inference is plausible in the case of doubt.
It is interesting that, even with modifiers, verb forms of ‘doubt’ resist expressing low degrees of doubt. One can say, “I highly doubt that p”, but it is very awkward to say, “I lowly doubt that p” and still at least a little awkward to say, “I doubt, a little bit, that p.”
Degree of doubt will be discussed more in Sect. 4.2 and should be distinguished from the degree of strength of one’s doubts, which I discussed earlier in this section.
Contra Schellenberg’s (2005, p. 96).
Thanks to Daniel Howard-Snyder and John Turri for helpful discussion here.
Hawthorne et al. (2016, p. 1395) provide a similar argument to show that belief is a weak attitude (that is, the norms of belief are weaker than the norms of assertion).
Note that the infinitive form also indicates a high degree of doubt. If I say, “Fred has reason to doubt that Sally will arrive on time” or “Fred is inclined to doubt that Sally will arrive on time,” I convey that Fred has reason to believe that Sally will not arrive on time, or that Fred is inclined to believe that Sally won’t arrive on time. Thanks to Matt Duncan and Hayoung Shin for helpful conversation.
This follows Audi’s (1994) distinction between dispositional beliefs and dispositions to believe.
It is part of standard probability theory to assign the number 1 to the highest degree of confidence that p (when one is certain that p) and the number 0 to the lowest degree of confidence that p (when one is certain that \(\sim \) p) and the numbers in between 0 and 1 to the varying degrees of confidence in between. I am assuming here that there is a highest degree of confidence.
Thanks to Jon Matheson for help in formulating the following considerations.
Earl Conee has objected that ‘complete confidence’ does not indicate the highest degree of confidence. I disagree. However, notice that we could replace (13) with “I have less than the highest degree of confidence that p, but I have no doubt that p.” This also seems inconsistent. However, since we are less likely to talk this way in ordinary language, I will stick to “complete confidence” talk. Thanks to Jeremy Fantl for this helpful suggestion for how to respond to Earl Conee
Clarke (2013) and Greco (2015) have recently argued that, given some assumptions, belief just is credence 1 (in certain contexts). It is not obvious how relevant their work is to (C). Greco (2015, p. 180) says he is using ‘belief’ in a technically defined sense, and he seems to take Clarke to be doing the same. On the other hand, I intend to mean by ‘belief’ what it means in ordinary English. Furthermore, Clarke does not mean by “belief to degree 1” anything that entails certainty (2013, p. 11), whereas my discussion is about the highest degree of confidence (or certainty). Lastly, nothing they say counts against my argument that it seems that my degree of confidence could increase, and therefore, my initial confidence was not the highest degree of confidence.
Thanks to Neil Sinhababu and Matthew McGrath for pushing me to address this point.
Thanks to a referee of this journal for encouraging me to discuss my reply to the equivocation response more thoroughly.
I revisit the equivocation response near the end of Sect. 4.1.
This second objection and my response are motivated by Unger (1975, pp. 63–87). The first objection arose in conversations about the puzzle.
Thanks to Matthew McGrath for this sort of case. Scott Edgar has pointed out to me a similar case, according to which many scientists say that certain scientific theories—say, the theory of evolution—should be held without any doubt. However, they would not ascribe 100% confidence toward the theory. What I say about the baseball case will apply to these science cases.
Thanks to Daniel Howard-Snyder for conversation that helped clarify the reasoning in this paragraph.
One might be concerned by my use of first person ascriptions instead of the third person. But note that it does not always seem inconsistent to say of the person in the classroom scenario, “S/he is not completely confident that the class clown is there, but s/he has no doubt that he’s there.” Cases of unreflectively formed belief, such as the classroom scenario, are just those cases where such statements can be true.
One could stipulatively define ‘doubt’ to mean has a doxastic attitude but not the highest confidence, just as one could stipulatively define ‘knows’ to mean justified true belief. Then there would be a meaning of ‘doubt’ according to which (B) is true. But then we would no longer be talking about doubt, just as we would no longer be talking about knowledge.
These examples should make clear that epistemic possibility is distinct from other types of possibility, according to which possibility (or possible truth) is a property of propositions and not dependent on individuals’ information states. Most accounts of metaphysical possibility are like this, as are accounts of possibility depending on statistical, frequentist, or logical accounts of probability. On these accounts, truths like \(26 \times 23=598\) are all necessarily true, independent of any individual’s information state.
This evidence-based account is derived from Dougherty and Rysiew (2009, p. 127).
Fido also appears in a similar example in Yalcin (2007, p. 997).
Thanks to Hud Hudson and Daniel Howard-Snyder for helpful conversation. Nothing I say here requires that Yalcin’s own view, credal expressivism, is false, but only that this particular criticism that Yalcin makes fails. For a brief discussion of credal expressivism and its application to this paper, see the next footnote.
According to credal expressivism, to believe that \(\sim \) p is possible is to have a nonzero credence that \(\sim \) p. If this is true, then the additional complexity that is missing in the classroom scenario is as follows. You do believe that attendance is good. But you have not formed any degree of credence toward the proposition that attendance is not good. In other words, you will have formed a doxastic attitude toward a proposition without having formed an attitude toward its negation. That additional complexity is required for you to have doubt that attendance is good.
Note that \(\hbox {Doubt}_{1}\) does not require that for one to have doubt that p, one must believe that it is possible that p is false. The concepts of falsity and truth do seem to be a bit more difficult to grasp, and I do not want to say that a grasp of either of those concepts is required for doubt. Thanks to Chris Tweedt for the objection that led to this clarification.
Thanks to a referee of this journal for pressing this objection, which helped me to think more clearly about my account of degrees of doubt.
A referee has offered the following challenging objection. There are various properties associated with doubt: “hesitancy, inhibition, anxiety, curiosity, unwillingness to assert, unwillingness to stake values upon, and perhaps a handful of other phenomenal/dispositional properties.” Couldn’t a child exhibit varying degrees of these properties, and hence, have varying degrees of doubt, even if she doesn’t believe it’s likely that \(\sim \) p, it’s very likely that \(\sim \) p, and so on? In response, I say ‘no’. If the degree of the property is not had in virtue of believing something like it’s very likely that \(\sim \) p, then it is implausible that the child has a degree of doubt. For example, suppose a child is very unwilling to assert that p, but not in virtue of believing it’s very unlikely that \(\sim \) p; it is because the child simply doesn’t like asserting things. Then it is implausible that the child has a high degree of doubt that p, despite being very unwilling to assert that p. I would say similar things about the other properties in the referee’s list of properties.
But perhaps the referee holds to a dispositionalist account of doubt, according to which having a high degree of doubt just is having a high degree of enough of the properties in that list. Perhaps such an account could be developed along the lines of the dispositionalist account of belief held by Schwitzgebel (2002). But if the referee says that, then it would also be natural for the referee to accept a dispositionalist account of belief, where believing that it’s very likely that \(\sim \) p is also identical to having a high degree of enough of the properties in the list. Then it is still impossible to have a high degree of doubt that p without believing it’s very likely that \(\sim \) p. Thanks to Liz Jackson for helpful discussion about these points.
Following DeRose (2009, p. 112) again, perhaps we should “check the negations” of the relevant doubt ascription. While it seems inappropriate to attribute doubt to the strong atheist, it seems worse to say of her, “She lacks doubt that God exists,” “She doesn’t doubt that God exists,” and especially, “She has no doubt that God exists.” Is this good evidence that there is doubt? No. These sentences also have a false implicature; they imply that she believes that God does exist, which is certainly false! Therefore, the intuition of the falsity or inappropriateness of those negations might not be because doubt is present, but because they falsely imply that belief is present.
Thanks to Brad Rettler for this objection.
Thanks to Kenny Boyce and Daniel Howard-Snyder for helpful conversation about the previous objections and how to respond to them.
Thanks to Matt Duncan for helpful conversation.
I am thankful to a referee of this journal for the objection to that analysis of doubt, which in turn led to the current formulation of \(\hbox {Doubt}_{2}\) here.
For more defense of the value and philosophical respectability of this sort of analysis, see chapter 2 of Strawson (1992), especially pp. 18–20.
They are building on the work of Dougherty and Rysiew (2009, p. 130), who also endorse this claim, though more hesitantly.
For reasons to define fallibilism and infallibilism in something like this way, see Fantl and McGrath (2009, pp. 7–15). Fallibilists will differ on when the degree of probability is high enough to preclude knowledge. Proponents of pragmatic encroachment, like Fantl and McGrath (2009, pp. 25–26), think that the threshold is determined at least in part by pragmatic factors.
To repeat, this process requires reflection. Consider my earlier example. It might be that S knows that p if and only if S has ungettiered, justified, true belief that p. It does not follow that if you believe an instance of the analysandum, you will automatically believe the relevant instance of the analysans. That would take reflection.
References
Alston, W. (1996). Belief, acceptance, and religious faith. In J. Jordan & D. Howard-Snyder (Eds.), Faith, freedom, and rationality. Lanham, MD: Rowman & Littlefield.
Audi, R. (1994). Dispositional beliefs and dispositions to belief. Nous, 28, 419–434.
Carruthers, P. (2006). The architecture of the mind. Oxford: Clarenden Press.
Clarke, R. (2013). Belief is credence one (in context). Philosopher’s Imprint, 13, 1–18.
Cohen, L. J. (1992). An essay on belief and acceptance. Oxford: Clarendon Press.
DeRose, K. (1991). Epistemic possibilities. The Philosophical Review, 100, 581–605.
DeRose, K. (1995). Solving the skeptical problem. The Philosophical Review, 104, 1–52.
DeRose, K. (2009). The case for contextualism. Oxford: Oxford University Press.
Dougherty, T., & Rysiew, P. (2009). Fallibilism and concessive knowledge attributions. Philosophy and Phenomenological Research, 78, 123–132.
Egan, A., & Weatherson, B. (2011). Epistemic modality. Oxford: Oxford University Press.
Erikkson, L., & Hajek, A. (2007). What are degrees of belief? Studia Logica, 86, 183–213.
Fantl, J., & McGrath, M. (2009). Knowledge in an uncertain world. Oxford: Oxford University Press.
Foley, R. (1993). Working without a net: A study of egocentric epistemology. New York: Oxford University Press.
Gettier, E. (1963). Is justified true belief knowledge? Analysis, 23, 121–123.
Greco, D. (2015). How I learned to stop worrying and love probability 1. Philosophical Perspectives, 29, 179–201.
Hawthorne, J., Rothschild, D., & Spectre, L. (2016). Belief is weak. Philosophical Studies, 173, 1393–1404.
Howard-Snyder, D. (2013). Propositional faith: What it is and what it is not. American Philosophical Quarterly, 50, 357–372.
Huemer, M. (2007). Epistemic possibility. Synthese, 156, 119–142.
Jeffrey, R. (1992). Probability and the art of judgment. Cambridge: Cambridge University Press.
Kratzer, A. (1977). What ‘must’ and ‘can’ must and can mean. Linguistics and Philosophy, 1, 337–355.
Lewis, D. (1981). A subjectivist’s guide to objective chance. In W. R. S. Harper & G. Pearce (Eds.), Ifs. Dordrecht: D. Reidel.
Peels, R. (ms). Faith and doubt. Currently unpublished.
Plantinga, A. (1993). Warrant: The current debate. New York: Oxford University Press.
Schellenberg, J. (2005). Prolegomena to a philosophy of religion. Ithaca: Cornell University Press.
Schwitzgebel, E. (2002). A phenomenal dispositional account of belief. Nous, 36, 249–275.
Sennet, A. (2016). Ambiguity. In Stanford encyclopedia of philosophy. http://plato.stanford.edu/entries/ambiguity/#DeteAmbi. February 8, 2016 version.
Smithies, D. (2012). The mental lives of zombies. Philosophical Perspectives, 26, 343–372.
Stanley, J. (2005). Fallibilism and concessive knowledge attributions. Analysis, 65, 126–131.
Strawson, P. (1992). Analysis and metaphysics: An introduction to philosophy. Oxford: Oxford University Press.
Unger, P. (1975). Ignorance: A case for skepticism. Oxford: Oxford University Press.
Yalcin, S. (2007). Epistemic modals. Mind, 116, 983–1026.
Yalcin, S. (2011). Nonfactualism about epistemic modality. In A. Egan & B. Weatherson (Eds.), Epistemic modality (pp. 295–332). Oxford: Oxford University Press.
Acknowledgements
I am thankful for helpful feedback from the audience at Dalhousie University’s philosophy colloquium (10/2013), the philosophy faculty of Western Washington University (12/2013), the audience at the Central APA (2/2014) (especially my commentator, Earl Conee), the participants of the Notre Dame Belief and Credence reading group (9/2016) (including Hugh Burling, Brian Cutter, Liz Jackson, Ross Jensen, Ting Cho Lau, and James Nguyen), and the audience at a Saint Louis epistemology brown bag session (9/2016) (especially Jonathan Nebel, Jonathan Reibseman, and Julia Staffel). Thanks to Jon Kvanvig, Daniel Howard-Snyder, and Trent Dougherty, for allowing me to participate in the Summer Seminar on the Nature of Faith (6/2014–7/2014). The paper benefitted from discussions with the following participants of that seminar: Kenny Boyce, Daniel McKaughan, Rik Peels, and especially Daniel Howard-Snyder. Thanks to the Center for Philosophy of Religion at Rutgers University and the Center for Philosophy of Religion at University of Notre Dame for research time and funding for this project. Thanks to Peter Markie, Jon Matheson, Kevin McCain, Matthew McGrath, and Paul Weirich for helpful written comments on Sect. 3. In addition to those mentioned in the footnotes throughout the paper, I am also thankful for helpful conversations with Bob Beddor, Chris Gadsden, Richard Fumerton, Simon Goldstein, Peter Markie, Kevin McCain, Matthew McGrath, Andrew Melnyk, J.L. Schellenberg, Hayoung Shin, John Turri, Peter van Elswyk, Jonathan Vertanen, Paul Weirich, and Julie Zykan. Lastly, I am thankful to Ting Cho Lau for helpful editing that led to clarifications and improvements in the paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Moon, A. The nature of doubt and a new puzzle about belief, doubt, and confidence. Synthese 195, 1827–1848 (2018). https://doi.org/10.1007/s11229-016-1310-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-016-1310-y