In most cases, our pretense episodes are quarantined from reality. This is usually taken to mean that the imaginings have no effect outside of the pretense, particularly with regards to behaviors. But imaginative quarantine also fails systematically. Broadly speaking, imaginative contagion are cases where imaginings do have effects outside of the pretense, noticeably with behaviors. Tamar Gendler (2006) points to three sets of cases: visual and motor imagery, affective response, and social priming.
I think imaginative contagion is a really interesting phenomenon. There is a rich body of empirical evidence, and it seems like accounting for the phenomenon should impact our functional analysis of the mind. I am puzzled, though, about how exactly to interpret the phenomenon. What part of the mind is affected in contagion cases? Or, to put it metaphorically, what is contagious about imaginative contagion?
1. Two Formulations of the Phenomenon
In his reply to Gendler (2006), Martijn Blaauw rightfully points out two formulations of the broad characterization of imaginative contagion. I am going to put it slightly differently, but I think we are of the same spirit.
- Behavioral formulation: Imaginative contagion are cases where, after pretending that P, the subject acts as if he believes that P.
- Belief formulation: Imaginative contagion are cases where, after pretending that P, the subject comes to believe that P.
The belief formulation implies the behavioral formulation (all else being equal), but not the other way around. What you consider to be the difference between the two probably depends on how behaviorist-inclined you are. Unless you are a hardcore behaviorist, though, you probably do recognize the conceptual gulf in between the two formulations.
It seems like the two are not clearly distinguished when it comes to assessing empirical support. It is not even clear that there is a uniformity among the empirical evidence, in terms of which formulation they support. For example, the visual and motor imagery cases do not, it seems to me, suggest either formulation. In these cases, residual effects, such as after-images and increased ability, result from imaginary analogues of perception and motor function, rather than actual ones. At least on some accounts of perception, belief or something like it does not play a role. Ultimately, I will suggest that it is possible that imaginative contagion is not a unified phenomenon. Instead, different formulations of it might be appropriate in different sets of cases.
2. Philosophical Consequences of Imaginative Contagion
Without trying to resolve the ambiguity, consider the ramifications of having these different formulations when it comes to drawing philosophical consequences. For example, in a footnote to his paper on delusion, Andy Egan writes, “[Imaginative contagion cases], I think, provide more evidence that the belief/imagination divide is not as sharp as we might have thought, and that delusional subjects are not the only bimaginers.” Imaginative contagion is taken to be evidence of a transitional mental state, or a continuum, in between belief and imagination. Positing such mental state or a continuum obviously explains how subjects come to believe that P after pretending that P. However, this consequence only appears far more plausible when we take the belief formulation.
Obviously, it is harder for an empirical study to demonstrate the belief formulation. There are, however, a priori considerations against the stronger interpretation. In these cases, subjects are usually explicitly aware of the reality/pretense difference. Take the sugar/poison example. (I am not sure I got all the details right.) Subjects are given two bottles, and they fill them with sugar. Then they are asked to label one bottle as “sugar” and the other as “sodium cyanide”, a poison. When asked, the subjects say both bottles contain sugar. They nevertheless hesitate for a longer period before ingesting the content from the sodium cyanide bottle. In this case, the response seems to indicate that subjects believe that even the sodium cyanide bottle contains sugar. So unless we want to attribute to them a systematic irrationality in having conflicting beliefs, it does not seem that we should say they simultaneously also believe that the sodium cyanide bottle does not contains sugar, but poison. And we have some reason to hold off that attribution because precisely what makes imaginative contagion interesting is that subjects are fully aware of the reality/pretense difference. So there are at least some considerations against taking belief as the contagious part of the mind, and thus drawing philosophical consequences based on that formulation.
While Gendler’s earlier (2003) formulation is less clear—“some things do come to be believed—or treated as if they were believed—merely because they are pretended” (131)—in her 2006 paper she is more partial to characterize imaginative contagion as a source indifference. This seems to align more with the behavioral formulation, and suggests that practical reasoning system is the contagious part. Usually it takes cues from beliefs, but in cases of contagion, it takes cues from imaginings.
This formulation too has important philosophical consequences. Most prominently, if practical reasoning system can take cues from imaginings, even in limited cases, then that is some reason to favor the Currie and Ravenscroft picture of the mind. On their functional analysis, beliefs and desires can motivate actions, but so can belief-like imaginings and desire-like imaginings. This is in contrast with the Nichols and Stich picture of the mind, where imaginings can never motivate actions directly, but only do so via genuine beliefs and desires. Whether we can draw this conclusion depends on how plausible we take the behavioral formulation to be.
To resist, someone who is sympathetic to the Nichols and Stich picture, i.e. me, might offer an alternative formulation of imaginative contagion. The goal is to avoid positing a transitional mental state or a continuum between belief and desire, and also avoid positing desire-like imagination. In other words, try to explain imaginative contagion without positing new mental states. Just throwing this proposal out there.
- Higher-order representation formulation: Imaginative contagion are cases where, after pretending that P, the subject comes to believe that he believes that P.
Correspondingly, the contagious part of the mind is the UpDater, the mechanism that is used to generate higher-order representations from both imagination and belief. So let us assess this formulation. It improves upon the belief formulation if we do not always generate higher-order representations of our beliefs. At least we are not attributing an inconsistency that would be obvious to the subject. It also has the advantage of mental economy in virtue of not positing new mental states.
3. One More Ramification
Ultimately, I am not convinced that imaginative contagion is a uniform phenomenon because each set of cases has its own unique features. As such, it may be that different formulations are appropriate for different sets of cases. For example, at least on some theories of perception, it is clear that none of the formulation given so far really captures what is going on in visual and motor imagery cases. In contrast, affective response contagions may be best explained by the behavioral formulation. These are cases when when emotion-like affective responses are triggered by an imagined scenario rather than an actual one, like in the sugar/poison example. In his treatment of the paradox of fictional emotions, Gregory Currie suggests that practical reasoning system does take cues from imaginings, and is thus contagious in my sense. Lastly, while both visual and motor imagery and affective responses are behaviors that seem automatic, social priming cases are not. These are cases when different social schemata are applied due to the social situations in which we imagine ourselves to be, rather than actual ones. In these cases, the higher-order representation formulation seems to me more appropriate because it makes sense of placing oneself in a different situation and then deliberate.
I will stop wildly speculating now. Hopefully someone can make some sense of the foregoing rambling, and bring my attention to relevant empirical and philosophical literatures. Then maybe we will know, in terms of a functional analysis of the mind, what is contagious about imaginative contagion.
Blaauw, Martijn (2006). “Imaginative Contagion,” Metaphilosophy 37: 204-209.
Currie, Gregory (2004). “Anne Bronte and the Uses of Imagination,” Matthew Kieran (ed.), Contemporary Debates in Aesthetics and the Philosophy of Art. Malden: Blackwell.
Currie, Gregory and Ian Ravenscroft (2002). Recreative Minds. New York: Oxford University Press.
Egan, Andy (forthcoming) “Imagination, Delusion, and Self-Deception.”
Gendler, Tamar Szabo (2003). “Pretense and Belief,” in Matthew Kieran and Dominic McIver Lopes (eds.), Imagination, Philosophy, and the Arts. New York: Routledge. 125-141.
Gendler, Tamar Szabo (2006). “Imaginative Contagion,” Metaphilosophy 37: 183-203.
Nichols, Shaun and Stephen Stich (2003). Mindreading. New York: Oxford University Press.
P.S. Tamar Gendler has a forthcoming paper that I have not read yet that is probably relevant. It is online: “Alief and Belief”.