The Nature of Belief: Evaluating Schwitzgebel’s In-Between Cases Argument
Dispositionalism and representationalism are both theories that attempt to explain the nature of belief. In “A Dispositional Approach to Attitudes: Thinking Outside of the Belief Box,” Eric Schwitzgebel makes the case that dispositionalism provides a more accurate depiction for what it means to have a belief than the alternate view of representationalism. One source of evidence Schwitzgebel appeals to as a reason to prefer dispositionalism to representationalism is so-called “in-between cases”; however, I do not believe that his discussion of in-between cases provides sufficiently convincing evidence for preferring dispositionalism to representationalism.
In Schwitzgebel’s words, representationalists view having a belief as “a matter of possessing some particular internally stored representational content, a content perhaps poised to play some specific set of cognitive roles depending on the attitude type…to believe that P (e.g. to believe that snow is white is to have a representation with the content ‘P’ (‘snow is white’) stored in a metaphorical ‘Belief Box’ (75). Essentially, the idea behind the representationalist view is that having a belief is a matter of having literal symbols that are stored in the brain that represent a certain proposition and that these beliefs are stored in a “belief box” that is functionally defined by how things get into the box and how the beliefs affect behavior.
Alternatively, Schwitzgebel presents the dispositionalist view, the one he makes a case for, writing, “to have an attitude is mainly a matter of being apt to interact with the world in patterns that ordinary people would regard as characteristic of having that attitude” (75). In other words, dispositionalism conceives of having a belief as a matter of having a certain set of dispositions or tendencies: tendencies to think, feel, and act in a way that is stereotypical of someone who has that belief (under relevant, normal conditions). For example, according to dispositionalism, to believe that giraffes are born six feet tall is to have the tendency to think, feel, and act in a way that sufficiently matches the thoughts, feelings, and actions one would expect of someone who has this belief; this might include stating that giraffes are six feet tall when born or not thinking that a six-foot-tall baby giraffe is exceptionally tall for a baby giraffe or not feeling surprised when someone tells you that giraffes or six feet tall, etc. A dispositionalist would consider anyone who has a set of dispositions that sufficiently matches the set of dispositions aligned with the stereotype for having a given belief as having that belief.
Bearing these conceptualizations of representationalism and dispositionalism in mind, Schwitzgebel claims that dispositionalism, unlike representationalism, can explain (what he dubs) “in-between cases” in a satisfying way. An “in-between case” is a situation in which someone matches some of the tendencies stereotypical of someone with that belief, while also having some tendencies that do not match the stereotype of someone with that belief.
One such in-between case Schwitzgebel appeals to is Juliet the implicit racist. (Note that this particular in-between case will be referenced throughout the rest of this paper, but that analogous conclusions can be drawn for other in-between cases.) Some of the Juliet’s tendencies include “warmly espous[ing]” that all races are equally beautiful, “spontaneously comment[ing]” on the beauty of black people, and sincerely professing to be “awed by the beauty of certain black celebrities”; however, her tendencies also include being more “viscerally engaged” by white celebrities, and more swiftly and spontaneously “favor[ing] the white bathers at the beach” (Schwitzgebel 85-86). Thus, Juliet has some of the tendencies or dispositions of someone with the belief that all races are equally beautiful, but also some dispositions that do not align with someone with that belief.
The naturally following question is whether Juliet believes that all races are equally beautiful; Schwitzgebel maintains that the correct answer is “kind-of” – Juliet kind-of believes that all races are equally beautiful – and that dispositionalism can satisfyingly explain this answer (86). Schwitzgebel asserts that dispositionalism can satisfyingly explain that Juliet kind-of believes that all races are equally beautiful because she has some dispositions that align with someone with the belief that people of all races are equally beautiful and some that do not. He writes, “The dispositional stereotype approach, partly because of its superficiality, handles in-between cases…with a flexible minimalism: Display the dispositional structure and you’re done; nothing more to report” (86). In other words, Schwitzgebel claims that the reason dispositionalism can satisfyingly explain in-between cases is that it is flexible. That is flexible in the sense that having a belief is a matter of how closely one matches the stereotypical dispositional profile and therefore, if one kind-of matches the stereotypical dispositional profile, they kind-of have that belief.
However, according to Schwitzgebel, unlike the dispositionalist, the representationalist cannot answer correctly that Juliet “kind-of” believes that all races are equally beautiful because representationalism does not allow for the same level of flexibility that dispositionalism does. Schwitzgebel writes, “Juliet might (or must?) really have ‘all races are equally beautiful’ in her Belief Box, or ‘white people are more beautiful’ there, or maybe both, and until we have figured this out, we don’t know what her attitude really is, even if we know every inch of her superficial dispositional structure” (86). Essentially, Schwitzgebel says that a representationalist cannot correctly answer with “kind-of” because according to Schwitzgebel’s conception of the representationalist view, Juliet can only have the belief that people of all races are equally beautiful or not have that belief based on whether or the belief is or is not in her “belief box” (as there is no way for the belief to “kind-of” be in her belief box).
However, I believe that Schwitzgebel’s discussion of in-between cases does not provide sufficiently convincing reasons for preferring dispositionalism to representationalism because a more convincing claim that suggests a different interpretation of the belief box can be made against his second assumption that representationalism cannot satisfyingly explain the answer “kind-of.” In their article “Against Dispositionalism: Belief in Cognitive Science” Jake Quilty-Dunn and Eric Mandlebaum make a case that representationalism can satisfyingly achieve the “kind-of” answer by asserting that belief storage in the brain is fragmented. Quilty-Dunn and Mandlebaum summarize/characterize Schwitzgebel’s reasoning for why representationalism cannot explain in-between cases writing, “Schwitzgebel’s criticisms are all part of the same critique: representationalism is inflexible insofar as it supposes there is a single place where a mental representation is stored and thereby becomes a belief” (2357). In the next line, they explicate why they believe these criticisms are not valid: “We think these criticisms fail because they rest on an inaccurate depiction of representationalism, and of the belief box metaphor in particular” (2357). Quilty-Dunn and Mandlebaum’s point here is that Schwitzgebel is mistaken in believing that representationalism can’t explain in-between cases because he has a misconception of belief storage in the brain. Namely, he misconceives that representations of belief are all stored in one location.
Quilty-Dunn and Mandlebaum instead propose that belief storage in the brain is “fragmented” (2358). They introduce the idea of belief fragmentation writing, “rather being stored in a single box – or, to borrow another metaphor, in one consistent web – our beliefs are stored in disparate, perhaps mutually inconsistent fragments” (2358). The example they give to illustrate this idea of belief fragmentation is Lewis who believes that “Nassau St. ran north-south, that it was parallel to a certain railroad, and that the railroad ran east-west” (2358). This demonstrates the idea of belief fragmentation because Lewis has clear inconsistencies in his beliefs (as a street cannot be parallel to a railroad that runs in a perpendicular direction) and therefore means these beliefs are independently stored.
The idea that belief is fragmented can be applied to in-between cases like Juliet and therefore allows for the representationalist to provide a satisfying explanation for how Juliet “kind-of” believes that all races are equally beautiful. With belief fragmentation, the representationalist can explain that Juliet kind-of believes that all races are equally beautiful because she has one belief fragmentation that is egalitarian and represents the proposition that all races are equally beautiful and one separate belief fragmentation that is racist and represents the proposition that all races are not equally beautiful. Furthermore, Quilty-Dunn and Mandlebaum assert that fragmented belief architecture not only allows for in-between cases but can also “explain them in causal-mechanistic terms” (2358). In the case of Juliet, they explain that whereas the egalitarian belief fragment is accessed for conscious actions, the racist belief fragment is accessed for actions and thoughts below her conscious control (2358). Therefore, the representationalist can satisfyingly achieve the answer “kind-of” in response to in-between cases, and by demonstrating why Schwitzgebel’s second assumption is ill-informed, Quilty-Dunn and Mandlebaum severely weaken Schwitzgebel’s argument that in-between cases are a reason to prefer representationalism to dispositionalism.
One reason Quilty-Dunn and Mandlebaum’s objection is successful and thus, presents a strong case against preferring dispositionalism to representationalism due to in-between cases is because it is not ad hoc. As the Lewis example (or any other example of inconsistent beliefs) demonstrates, the idea of belief fragmentation is not ad hoc, that is not created to “save” representationalism, but rather is a satisfying theory behind inconsistent beliefs. The appeal of belief fragmentation is further proven by that fact there are theorists who accept fragmentation while rejecting representationalism (Quilty-Dunn and Mandlebaum 2358).
Quilty-Dunn and Mandlebaum’s objection is also successful because they also provide experimental evidence that supports fragmented representationalist architectures and poses problems for dispositionalism. One example of evidence is sorting experiments in which participants are given a deck of cards and asked to make a pile of the clubs and spades or make a pile of non-diamonds and non-hearts, which is essentially the same task but given in different terms; however, those who asked to make the pile of clubs and spades are much faster and less error-prone than those who are asked to make a pile of non-diamonds and non-hearts (Quilty-Dunn and Mandlebaum 2362). Quilty-Dunn and Mandlebaum believe that representationalism can explain these findings because the representational structures result in different behavior as it easier to process the “positive” representation of clubs and spades than the “negative” representation of non-diamonds and non-hearts that requires more translation steps (2362-2363).
They also explain that these experimental results pose a problem for dispositionalism. The dispositionalist cannot refer to the notion of representations in the brain. Instead, Quilty-Dunn and Mandlebaum imagine that a dispositionalist might instead say “to believe that one is to sort the spades and clubs is just to be disposed to sort the spades and clubs in a quick and error-free manner, while to believe that one is to sort the non-diamonds from the non-hearts is to be disposed to sort them slowly and poorly,” (2363) essentially explaining the results by including them in the set of stereotypical dispositions. Without having a representational structure to refer to though, there appears to be no reasons to include quick and error-free sorting in the dispositional set of believing the task is to sort spades and clubs but not non-hearts and non-diamonds. This leaves the dispositionalist in a difficult situation and because Quilty-Dunn and Mandlebaum’s objection is not ad hoc and has experimental evidence supporting it that poses problems for dispositionalism, Schwitzgebel’s in-between cases argument does not hold up.
In conclusion, Schwitzgebel’s discussion of in-between cases does not provide sufficiently convincing evidence for preferring dispositionalism to representationalism because Quilty-Dunn and Mandlebaum demonstrate that representationalism can satisfyingly explain in-between cases using a fragmented belief architecture; moreover, Quilty-Dunn and Mandlebaum’s argument is successful because it is not ad hoc and presents experimental evidence that further supports the notion of a fragmented belief structure and poses problems for dispositionalism.
Works Cited
Quilty-Dunn, Jake, Mandelbaum, Eric “Against dispositionalism: belief in cognitive science.” Philos Stud 175, 2353–2372 (2018). https://doi.org/10.1007/s11098-017-0962-x
Schwitzgebel, Eric. 2013. “A Dispositional Approach to Attitudes: Thinking Outside of the Belief Box.” In New Essays on Belief: Constitution, Content, and Structure, edited by Nikolaj Nottelman, 75–99. New York: Palgrave Macmillan.