Abstract
There has been a surge of interest over cases where a subject sincerely endorses P while displaying discordant strains of not-P in her behaviour and emotion. Cases like this are telling because they bear directly upon conditions under which belief should be ascribed. Are beliefs to be aligned with what we sincerely endorse or with what we do and feel? If belief doesn’t explain the discordant strains, what does? T.S. Gendler has recently attempted to explain all the discordances by introducing a controversial new cognitive category—associative clusters called ‘alief’. Others think that belief explains all the discordancy cases, while others argue that in-between belief does the trick, and so on. Most advocates of the different positions, indeed, assume that their favoured analysis will explain the whole range of discordancy cases. This paper defends what I call the ‘contextual view’, where I argue that overturning this assumption of uniformity leads to more nuanced account of belief-ascription. On the contextual view, which analysis applies to which case depends on the discordancy case at hand. Perhaps a height-phobic stepping on a glass platform deserves different treatment to a hesitant stepper. I ground the contextual view in a biologically functional account of the alief/belief distinction, which construes alief as a real cognitive category but without the explanatory reach Gendler gives it. This functional distinction yields a principled strategy for determining the correct application of analysis to discordancy case.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
A point made eloquently by Zimmerman (2007, pp. 61–67).
E.g. see Mandelbaum (2012), Kung (2012), and Kwong (2012). Kwong explains the discordances in terms of inferences made from a number of beliefs held by the subject. Although I do not directly address Kwong’s view, the position developed in my paper for the functional unity of alief (in both humans and animals) would imply that any application of Kwong’s analysis must be limited.
It is plausible that whenever a subject is disposed to sincerely affirm that P, the subject is disposed to judge that P. If so, we can define endorsement in terms of dispositions to judge alone.
I have heard it suggested that the notion of judgment presupposes the notion of belief, and that subjects count as judging P only if they believe P. This claim seems too strong, as belief is typically regarded as a dispositional state. It seems possible that a subject who does not believe that P and who lacks a disposition to judge that P could nevertheless judge that P on one occasion. However, if one thinks that the connection between judgment and belief is trivial, one can also adopt an understanding on which a subject endorses P iff they seem to themselves to believe that P. Much of what I say in this paper will still apply. (Thanks to Wayne Davis for bringing this point to my attention).
The emphasis here is on belief-ascription rather than the nature of the belief ascribed (e.g. dispositional, representational, functional, etc.). One’s stance on belief-ascription may nevertheless constrain the sort of accounts about the nature of belief that are compatible, and vice versa. For a recent taxonomy on what it is to believe, see Schwitzgebel (2010b).
I owe this terminology to David Chalmers. Zimmerman (2007, pp. 72–73) outlines a similar taxonomy, but with different terminology, and without reference to the disjunctive view. Each of these views about criteria for belief has analogues (outlined in subsequent notes) in Schwitzgebel’s taxonomy of analyses of discordancy cases. To avoid confusion with Schwitzgebel’s taxonomy, I will from now on refer to the positions within his taxonomy as ‘analyses’ and those in my taxonomy as ‘views’. In keeping with this switch, the ‘contextual view’ becomes the ‘contextual analysis’.
So long as judgment-based criteria are met, discordancy cases are to be described as fitting the pro-judgment analysis. If the judgment-based criteria are not met, then advocates of the view should describe the discordancy case as fitting an in-between analysis insofar as S neither believes P nor not-P (although S need not have an in-between belief). When pushing for a universal application of the pro-judgment view, however, its advocates don’t always clearly do this.
From Engel (2004, p. 83), in relation to Bernard Williams's argument against belief-voluntarism.
Should an endorsement of P persist in the face of contrary evidence it can still suffice, on this criteria, for a belief that P. In normally rational subjects, an isolated failure to follow evidential norms would make it appropriate to criticise the belief in that instance as false or unreasonable, but it would not make the belief defective in my sense. (An example might be where one takes oneself to be a better driver than the evidence would suggest). In cases that I regard as genuinely defective (which may include various pathologies of clinical delusion or extreme self-deception), a disposition to judge that P is not part of a broader or multi-track disposition to follow norms of rationality. Here, an endorsement of P would not suffice for a belief that P.
Insofar as they "privilege evidence accessible from the third-person perspective", Zimmerman (2007, pp. 71–73) attributes variants of an action-based position to Williamson (2000), Smith (1994), Stalnaker (1984), Davidson (1984) and Dennett (1987). We should not assume that they all subscribe to the action-based view discussed here.
On other variants, S may simply fail to believe P. On any action-based position, if the discordant strands do meet the relevant action-based criteria, the case merits an anti-judgment analysis. If the discordant strands do not meet the action-based criteria, but the case as a whole does, then a pro-judgment analysis should be taken. If the discordancy case altogether fails to meet action-based criteria, then the case should be described as fitting an in-between analysis (see note 8).
So if S believes she is in danger, and has an overriding desire to be safe, then she will be disposed to behave in ways that extricate her from the danger. Velleman does not himself endorse this "purely motivational" conception of belief ascription.
It would be wrong to assume that a lack of explicit pairing of belief with norms of rationality (associated with the judgment-based view) makes belief, when ascribed on action-based criteria, a non-rational cognition. As Schwitzgebel points out, many of our unreflective (verbal and non-verbal) behaviors and emotions are sensitive to evidence in a way that can be judged rational or irrational (2010a, pp. 538–541). Nevertheless, those ascribing beliefs on the basis of action-based criteria (including proponents of the disjunctive view) have more leeway to ascribe beliefs to non-rational subjects such as infants and animals. This becomes relevant later in the paper.
It will be argued later that despite appearances, aliefs do not properly fit these standard behavioral and
affective patterns, and so do not conform to action-based criteria.
The argument also rules out variants of an action-based position upon which S simply fails to believe P (see note 12).
At the end of the paper, I indicate how these analyses, on a contextual approach, could apply to various discordancy cases.
For an account of this hypothesis, see Damasio (1999, pp. 40–42).
This feature of alief is defended further in the next section.
Note that Velleman’s claim that belief aims at truth doesn’t require that no other mechanisms are also at work. Hence, he acknowledges that evolution “may have given us dispositions to err on the side of caution in perceiving predators” (2000, p. 254).
This directly conflicts with how Kriegel views alief. At the heart of his position is the idea that aliefs, unlike beliefs, are intrinsically motivational and hence, “architecturally connected to action” such that they “govern behaviour as it freely unfolds in the normal go of things, where explicit deliberation is not called for” (2012, pp. 479, 478). His mistake, on my view, is to conflate an architectural connection to behavioral proclivities (which aliefs have) with an architectural connection to actionguiding behavior.
I do not discuss how aliefs might interact with other cognitive states such as imagination.
Are aliefs restricted to sensory representation, or can they also attach to concepts? Mandelbaum (2012, online) has recently pointed to evidence suggesting that concepts are associated with subtle affective resonances (‘microvalences’) and motor responses. It would indeed be strange if, in creatures capable of abstract thought, aliefs did not attach to concepts.
In the case of animals, aliefs and belief will usually work together in a concordant way, except without the ‘all-things-considered’ circuitry that permits a contrary response to countervailing evidence. As with the case of Sally and the apple, the images will immediately trigger an alief, which is then readily actioned by the accuracy-aiming belief-system.
By this, I don’t suggest that alief is not present in all the discordances—it may well be.
What about cases where one really dislikes the feelings of fear and exits the Skywalk or movie-theatre? Would such emotion and behavior be caused by a belief that the structure is unsafe or that one is in danger from the shark? Not necessarily. The emotions (and accompanying thought-patterns), even if unpleasant, may not be as they typically would be if one really believed oneself to be in danger.
Gendler discusses racist alief at length in 2008b, pp. 574–578.
References
Damasio, A. (1999). The feeling of what happens: Body and emotion in the making of consciousness. New York: Harcourt Brace.
Davidson, D. (1984). Inquiries into truth and interpretation. Oxford: Clarendon.
Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press.
Engel, P. (2004). Truth and the aim of belief. In D. Gillies (Ed.), Laws and models in science. London: King’s College.
Frankish, Keith. (2009). Delusions: A two-level framework. In M. Broome & L. Bortolotti (Eds.), Psychiatry as cognitive neuroscience: Philosophical perspectives (pp. 269–284). Oxford: Oxford University Press.
Gendler, T. S. (2008a). Alief and belief. Journal of Philosophy, 105, 634–663.
Gendler, T. S. (2008b). Alief in action (and reaction). Mind & Language, 23, 552–585.
Gertler, B. (2011). Self-knowledge and the transparency of belief. In A. Hatzimoysis (Ed.), Self-knowledge. Oxford: Oxford University Press.
Kriegel, U. (2012). Moral motivation, moral phenomenology, and the alief/belief distinction. Australasian Journal of Philosophy, 90(3), 469–486.
Kung, P. (2012). Review of ‘Gendler, Tamar Szabo, intuition, imagination and philosophical methodology’. Australasian Journal of Philosophy, 90, 806–809.
Kwong, J. M. C. (2012). Resisting aliefs: Gendler on belief-discordant behaviours. Philosophical Psychology, 25(1), 77–91.
Mandelbaum, E. (2012). Against alief. Philosophical Studies (in press).
Schwitzgebel, E. (2010a). Acting contrary to our professed beliefs or the Gulf between occurrent judgment and dispositional belief. Pacific Philosophical Quarterly, 91, 531–553.
Schwitzgebel, E. (2010b). Belief. In E. N. Zalta (Ed.), Stanford Encyclopedia of Philosophy (Winter 2011 ed.). http://plato.stanford.edu/entries/belief/.
Smith, M. (1994). The moral problem. Oxford: Blackwell.
Sommers, F. (2009). Dissonant beliefs. Analysis, 69, 267–274.
Stalnaker, R. (1984). Inquiry. Cambridge, MA: MIT Press.
Van Inwagen, P. (1983). An essay on freewill. Oxford: Oxford University Press.
Velleman, D. (2000). On the aim of belief. In D. Velleman (Ed.), The possibility of practical reason (pp. 244–281). New York: Oxford University Press.
Williamson, T. (2000). Knowledge and its limits. Oxford: Oxford University Press.
Zimmerman, A. (2007). The nature of belief. Journal of Consciousness Studies, 14(11), 61–82.
Acknowledgments
My biggest debt of gratitude is to David Chalmers and the anonymous referee whose suggestions helped significantly shape the paper. I would also like to thank Wayne Davis, Nic Damnjanovic, Keith Frankish, J. David Velleman, Christian Lee, and especially John Maier for their feedback on various drafts of the paper. Thanks also to Rob Rupert, Adam Pautz, Patrick Greenough, Zoe Drayson, Georgie Statham, Hartley Slater, Barry Maund, Michael Rubin, Douglas Campbell, David Ward, and Uriah Kriegel for their helpful comments on versions of this paper that I gave at conferences.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Albahari, M. Alief or belief? A contextual approach to belief ascription. Philos Stud 167, 701–720 (2014). https://doi.org/10.1007/s11098-013-0122-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-013-0122-x