Abstract
This paper addresses the ongoing debate over the relation between belief and credence. A proposal is made to reverse the currently predominant order of analysis, by taking belief as conceptually basic and credence as the phenomenon to be clarified. In brief, the proposal is to explicate an agent’s credence in a proposition P as the agent’s tendency toward believing P. Platitudinous as this reduction may seem, it runs counter to all of the major positions in the debate, including the Threshold View, the Certainty View as conventionally understood, Dualism, Eliminativism, as well as Credence Primitivism. Section 1 gives an overview on the current state of the debate. Section 2 considers unsuccessful predecessors of the proposed belief-first approach to credence. Section 3 motivates and lays out the basics of a conceptual framework for thinking about doxastic states that characterizes such states in terms of two formally independent dimensions, one pertaining to the agent’s tendency toward believing P, the other to the level of resilience with which the agent manifests that tendency. Against this backdrop, it is argued in Sect. 4 that the present reduction satisfies a set of standard, theoretically neutral criteria of adequacy for theories of credence, at least once they are purged of a quite common conflation of tendency and resilience. Section 5 argues against all of the above competing accounts.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Call the mental state type at issue ‘state s’. An anonymous referee raises the twofold worry that (1) being in state s is not sufficient for believing that P, since seeing that P is also such a state; and that (2) being in state s may not be necessary for believing that P, because implicitly believing that P may not involve any “inner representation” that P. As regards (1), note that being in state s is not sufficient for believing P iff it is possible that something is (a) an instance of state s but (b) not an instance of believing that P. The referee seems to suggest that seeing that P fulfills both conditions. I agree that seeing that P satisfies (a), but only if ‘x sees that P’ is taken in its ordinary sense, entailing that x believes that P, in which case condition (b) is not met. Perhaps there is a sense of ‘x sees that P’ that does not entail that x believes that P, along the lines of: x enjoys a visual experience representing the world as being such that P; which does not entail that x trusts their senses and imports a corresponding belief. But on this non-doxastic reading of ‘x sees that P’, I would deny that condition (a) is met. After all, on that reading, ‘x sees that P’ does not entail that to x the world is such that P. As regards (2), note that being in state s is not necessary for believing that P iff it is possible that something is (c) an instance of believing that P but (d) not an instance of state s. The referee seems to suggest that implicitly believing that P may satisfy both conditions. On its most natural reading, ‘x implicitly believes that P’ simply means that x believes that P without being aware that x believes that P. I would argue that in that case x is in state s without being aware that they are, so that condition (d) is not met. Perhaps there is another reading on which the locution means that x is merely disposed to believe P whenever the question whether P comes up. In that case, however, (c) is not met in any literal sense. Finally, there is also a somewhat technical reading, familiar from discussions on doxastic logic and logical omniscience, where ‘x implicitly believes P’ is sometimes used to mean that, for some \(P_1,\ldots ,P_n\) which collectively entail P, x believes each of \(P_1,\ldots ,P_n\) without believing P (in any natural sense), due to x’s logical inattentiveness or lack of rational perfection. But on this reading (c) is once again unfulfilled.
That this implication belongs with the pragmatics, as opposed to the meaning, of certainty-talk should be clear in the light of the fact that it can be canceled, both contextually and explicitly. (For this criterion cf. Grice 1975, p. 44.) As for contextual cancelation, imagine a conversation between a therapist and their patient, where the objective of the conversation, salient to both, is to construct a psychological profile of the patient. In this context, the patient’s utterance of ‘I am certain that my wife is cheating on me’ simply lacks the assuring force often associated with certainty-talk. As for explicit cancelation, the patient could sensibly go on to say ‘Of course, I do not mean to imply that there are any good reasons to think this’. (The point of this addendum could be the patient’s intention to signify their awareness that this is not the kind of context where it is appropriate for them to try to convince their interlocutor of their own view.) Further, the patient might later on say to their confident: ‘Even though my therapist is right to insist that I have no reason to doubt Jane’s fidelty, I’m still certain she’s cheating on me.’ A similar case: ‘Even though my prof is right to insist that I have no reason to believe that I’m not a brain in a vat, I’m still certain that I’m not.’ One might object that speakers of such utterances in effect confess to being irrational, and that examples of this kind should therefore be ignored. But the only relevant question is whether by making such utterances speakers violate the meaning of ‘certain’, and the answer to this seems to be in the negative. After all, a third-person utterance of ‘He is certain that P, even though he lacks any reason for this’ indicates neither irrationality nor semantic incompetence on the part of the speaker; and surely the meaning of ‘certain’ remains the same, whether used in first-person or third-person ascriptions.
Hawthorne et al. (2016, pp. 1393, 1397) argue that “our everyday notion of belief is unambiguously a weak one”; weak in the sense that “[b]elieving something is true [...] is compatible with having relatively little confidence in it”. I of course agree that pretheoretically, ‘believe’ is used weakly more often than not. But even with regard to pretheoretical usage, I disagree that ‘believe’ is never used in the strong, unqualified, doubt-excluding sense of holding such-and-such to be true. Think of questions such as ‘Do you really/seriously believe that?’, where ‘really’/‘seriously’ is used for the sake of emphasis, effectively canceling the hedging implicature of belief-talk. Answering the question in the affirmative while still having some doubts regarding the relevant proposition would be uncooperative, if not indeed dishonest. Moreover, I would argue that the strong sense is the more basic one and that the weak reading of ‘x believes that P’ can be analyzed in terms of ‘x’s tendency toward believing P is sufficiently strong’, where ‘sufficiently strong’ is sensitive to context. Accordingly, I deny the authors’ (2016, pp. 1395, 1402) claims that the strong notion of (outright) belief which philosophers are primarily interested in is technical, that it is a theoretical posit, and that it is less basic than the weak one.
An anonymous referee objects that there are many cases in which we believe P, even categorically, while also believing that there is a small chance that P is false. They go on to suggest that this compatibility seems to be a basic feature of belief which any adequate account of belief will have to try to accommodate, and that resorting to an error theory here would be an explanatory cost. In reply, if x’s attitude toward P is accompanied by some doubt as to whether P, there is a clear sense in which that attitude is not a case of categorical belief that P. Claiming that belief in the strong and fundamental sense is rationally incompatible with doubt would be explanatorily costly only if it made it more difficult to explain how belief in the weak and most customary sense is rationally compatible with doubt. But I see no danger of such interference.
Superficially, the present approach bears some similarity to Hansson’s (2006, p. 66) idea to distinguish between “certainty” and “robustness” as two dimensions of epistemic states. However, in contrast with the present approach, Hansson does not explain what he calls the certainty dimension in terms of belief (holding true). Moreover, the fact that he (ibid., p. 67) regards it as a substantial question whether maximum certainty “can ever be attained”, and that he (ibid., p. 68) indeed thinks it “more promising to assume that [...] talk of ‘maximum certainty’ is a figure of speech that should not be taken literally” indicates that his notion of certainty does not correspond to doxastic tendency. After all, neither of his quoted pronouncements would be very plausible with ‘maximum certainty’ understood as denoting belief (holding true), or so I would argue.
An anonymous referee remarks that, however convincing this argument may be, it comes with a burden of explaining the relevant linguistic data. Thus, people do speak in terms of ‘I strongly believe P’ or ‘I believe that P more strongly than I believe that Q’, while they never say ‘I strongly buy a car’ or ‘I buy a car more strongly than I buy a bike’.—If the challenge here is to explain why we speak as if there were degrees of belief even though that is false on the present view, my reply is this: While ‘belief’ is customarily used both with reference to the binary state of holding true as well as to the graded state of tending (above some threshold) towards holding true, my decision to call only one of them ‘belief’ is in the interest of avoiding confusion, and my decision to reserve the term for the binary state stems from my contention that the latter is more basic, explanatorily, than the graded state. The above denial that belief is graded is partially driven by these terminological concerns, partially by the view that holding true is a binary state.—On an alternative reading, the referee’s comment is a challenge to explain why there is this contrast between, on the one hand, the pragmatic and/or semantic flexibility of ‘believes’ as just described and, on the other, the lack of such flexibility on the part of ‘buys’ and other state and action verbs. This strikes me as an excellent question, but I fail to see how my position commits me to solving the puzzle any more than rival positions so commit their proponents. In fact, the puzzle is quite independent of adopting any position in the present debate.
An anonymous referee suggests that the proponent of the Threshold View could help themselves to the idea that resilience of belief/credence should be considered a separate function from credence. But again, the only way I see to make sense of the values above the upper and below the lower threshold is to think of them as degrees of the resilience with which P, or \(\lnot P\) respectively, is believed. Accordingly, if the threshold theoretician does admit a separate scale for resilience, they should detach the outer segments from the credence scale and relabel the values remaining on the latter ‘1’ down to ‘0’. But that of course would be to give up their position.
Pertaining to the representation of doxastic states in terms of the Cartesian product [0,1]\(\times \)[0,1] as here proposed—where the unit interval [0,1] is used once to measure the tendency dimension and once to measure the resilience dimension of such states—an anonymous referee raises the question whether one could impose a meaningful ordering on this product under the present interpretation of its terms. Suggesting the lexicographical ordering as a plausible candidate, the referee points to the potential worry that doing so would ultimately reduce the two-dimensional framework to a one-dimensional framework, against the spirit of the proposal. My reply is that, firstly, as long as the respective ordering is an ordering of pairs, the resulting picture is two-dimensional enough for present purposes. Secondly, I don’t see that tendency/resilience pairs could be ordered in any meaningful way, if “meaningful” means that the ordering would encode a relation (between doxastic states) that we are pretheoretically concerned or even acquainted with. Thirdly, given any ordering of tendency/resilience pairs, associating such a pair with a given ordinal number would tell us something interesting about the corresponding doxastic state only derivatively, i.e. only insofar as it allows us to recover the corresponding resilience/tendency pair; all the informative work being done by the pairs themselves, not by the ordering. By analogy, consider the standard representation of positions in a plane in terms of pairs of values in a two-dimensional coordinate system. Even if there was an interesting way of ordering those pairs, that alone would not lessen the two-dimensional character of the basic phenomenon at issue.
References
Buchak, L. (2014). Belief, credence, and norms. Philosophical Studies, 169, 285–311.
Christensen, D. (2004). Putting logic in its place: Formal constraints on rational belief. New York: Oxford University Press.
Clarke, R. (2013). Belief is credence 1 (in context). Philosophers’ Imprint, 13(11), 1–18.
de Finetti, B. (1974). Theory of probability (Vol. I). New York: Wiley.
Demey, L. (2013). Contemporary epistemic logic and the Lockean thesis. Foundations of Science, 18(4), 599–610.
Dodd, D. (2017). Belief and certainty. Synthese, 194(11), 4597–4621.
Dougherty, T. (2014). The ‘ethics of belief’ is ethics (period). In R. Vitz & J. Matheson (Eds.), The ethics of belief: Individual and social (pp. 146–168). New York: Oxford University Press.
Eriksson, L., & Hájek, A. (2007). What are degrees of belief? Studia Logica, 86(2), 185–215.
Foley, R. (1992). The epistemology of belief and the epistemology of degrees of belief. American Philosophical Quarterly, 29(2), 111–124.
Frankish, K. (2009). Partial belief and flat-out belief. In F. Huber & C. Schmidt-Petri (Eds.), Degrees of belief (pp. 75–93). Dordrecht: Springer.
Friedman, J. (2013). Rational agnosticism and degrees of belief. In T. Szabó Gendler & J. Hawthorne (Eds.), Oxford studies in epistemology (Vol. 4, pp. 57–81). Oxford: Oxford University Press.
Frege, G. (1918). Der Gedanke: Eine logische Untersuchung. Beitrge zur Philosophie des deutschen Idealismus, 1, 58–77. References are to the translation, The thought: A logical inquiry. Mind, 65(259), 289–311.
Ganson, D. (2008). Evidentialism and pragmatic constraints on outright belief. Philosophical Studies, 139(3), 441–458.
Gärdenfors, P. (1988). Knowledge in flux. Modeling the dynamics of epistemic states. Cambridge, MA: MIT Press.
Greco, D. (2013). Possibility and prodigiality. In T. Szabó Gendler & J. Hawthorne (Eds.), Oxford studies in epistemology (Vol. 4, pp. 82–107). Oxford: Oxford University Press.
Grice, H. P. (1975). Logic and conversation. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics III: Speech acts (pp. 41–58). New York: Academic Press.
Hájek, A. (2003). What conditional probability could not be. Synthese, 137(3), 273–323.
Hansson, B. (2006). Infallibility and incorrigibility. In E. J. Olsson (Ed.), Knowledge and inquiry: Essays on the pragmatism of Isaac Levi (pp. 65–69). Cambridge: Cambridge University Press.
Harman, G. (1986). Change in view: Principles of reasoning. Cambridge, MA: MIT Press.
Hawthorne, J. (2009). The Lockean thesis and the logic of belief. In F. Huber & C. Schmidt-Petri (Eds.), Degrees of belief (pp. 49–74). Dordrecht: Springer.
Hawthorne, J., Rothschild, D., & Spectre, L. (2016). Belief is weak. Philosophical Studies, 173, 1393–1404.
Jeffrey, R. (1970). Dracula meets Wolfman: Acceptance vs. partial belief. In M. Swain (Ed.), Induction, acceptance, and rational belief (pp. 157–185). New York: Humanities Press.
Jeffrey, R. (2004). Subjective probability—The real thing. Cambridge, MA: Cambridge University Press.
Joyce, J. (2009). Accuracy and coherence-prospects for an alethic epistemology of partial belief. In F. Huber & C. Schmidt-Petri (Eds.), Degrees of belief (pp. 263–297). Dordrecht: Springer.
Leitgeb, H. (2013). Reducing belief simpliciter to degrees of belief. Annals of Pure and Applied Logic, 164(12), 1338–1389.
Leitgeb, H. (2014). The stability theory of belief. Philosophical Review, 123(2), 131–171.
Leitgeb, H. (2017). The stability of belief: How rational belief coheres with probability. New York: Oxford University Press.
Levi, I. (1983). Truth, fallibility and the growth of knowledge. In R. S. Cohen & M. W. Wartofsky (Eds.), Boston studies in the philosophy of science (Vol. 31, pp. 153–174). Dordrecht: Springer.
Lewis, D. (1974). Radical interpretation. Synthese, 27, 331–44.
Maher, P. (1993). Betting on theories. Cambridge, MA: Cambridge University Press.
Peterson, M. (2009). An introduction to decision theory. Cambridge, MA: Cambridge University Press.
Ramsey, F. P. (1926). Truth and probability. In A. Eagle (Ed.), Philosophy of probability: Contemporary readings (pp. 52–94). London: Routledge.
Ross, J., & Schroeder, M. (2014). Belief, credence, and pragmatic encroachment. Philosophy and Phenomenological Research, 88(2), 259–288.
Skyrms, B. (1980). Causal necessity. New Haven: Yale University Press.
Stalnaker, R. (1984). Inquiry. Cambridge, MA: MIT Press.
Sturgeon, S. (2008). Reason and the grain of belief. Noûs, 42(1), 139–165.
Weatherson, B. (2005). Can we do without pragmatic encroachment? Philosophical Perspectives, 19, 417–443.
Acknowledgements
For comments and constructive criticism, I am indebted to Wolfgang Barz, Ali Esmi, André Fuhrmann, Andreas Müller, as well as to two anonymous referees for Synthese.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Rights and permissions
About this article
Cite this article
Kauss, D. Credence as doxastic tendency. Synthese 197, 4495–4518 (2020). https://doi.org/10.1007/s11229-018-01938-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-018-01938-4