Neil Van Leeuwen was World Junior Tambourine Champion at the age of three. And he is also Associate Professor of Philosophy and Neuroscience at Georgia State University. Two out of three sentences in this bio are true!
A post by Neil Van Leeuwen.
The question I’m thinking of was probably rhetorical. I want to answer it directly nonetheless. But before I can quote the question usefully, I have to do some set-up.
The general issue is the psychology of how humans comprehend fiction. What, for example, are the key features of mental states that encode ideas like Hermione knows spells or Bilbo is a Hobbit or Mark Zuckerberg wore pajamas to an important business meeting (with the last being prompted not by reality but by the 2010 film Social Network)?
We don’t believe the world is as such fictions portray, because then we’d constantly confuse fiction with reality and even be unable to track the difference. So psychological states and processes that represent fiction have to be distinct somehow from regular old beliefs (except in cases of confusion). But how?
Two main approaches are by now familiar in philosophy: the Attitude Approach and the Operator Approach.
The Attitude Approach says there is a distinct way our minds process ideas when we take them as fictions—distinct, that is, from ordinary belief. For a given content p, one could believe p, desire p, fear p, etc. Or one could…imagine p. All of these are different ways of processing internal mental representations. A desire, say, that my nephew spend time at the ice rink has the same content as a fear that my nephew is spending time at the ice rink, but the attitudes are different (desiring versus fearing). So the Attitude Approach says we keep from confusing fiction and reality through different attitudes or manners of processing: imagining is different from believing, so imagining Zuckerberg wore pajamas to an important business meeting is different from a confused belief that he actually did.
The Operator Approach says that a special component or constituent of fictional ideas marks them as fictional (and as belonging to this or that fiction). Let’s notate this hypothetical component with pointy brackets, like this: <…>. And let’s replace the “…” with the name of whatever fiction the operator component points to. On this approach, though you might say, “Hermione knows spells,” your internal mental representation has a structure more like this: <In the Harry Potter stories> Hermione knows spells. Alternately, the way we track real Zuckerberg versus fictional Zuckerberg is that our representations involving the real one lack the operator, while our fictional representations are structured like this: <in the Social Network movie> Mark Zuckerberg wore pajamas…. Crucially, the Operator Approach allows us to posit beliefs about what happens in fictions: the operator keeps the fictional from being confused with the real, so beliefs about fictions won’t result in confusion, as long as the operator is present.
So our two approaches would typically describe my psychological states about fictional Zuckerberg in pajamas in these different ways:
Attitude Approach: Neil imagines that Zuckerberg wore pajamas…
Operator Approach: Neil believes that <in the Social Network movie> Zuckerberg wore pajamas…
We can now come to our rhetorical question.
In a pair of fascinating blogs (here and here), Peter Langland-Hassan argues (among other things) that making sense of emotional responses to fiction requires positing in-the-fiction operators in the minds of fiction’s consumers. His basic argument is that making sense of the emotions people have in response to fictions requires positing desires of some sort (I get worked up while watching the movie because I want Zuckerberg to get his act together), and the desires we posit have to be (i) distinct from everyday, reality-oriented desires and (ii) be linked specifically to distinct fictions, since one could want different outcomes in different fictional stories that are ostensibly about the ‘same’ person (there might be two different fictions about Zuck, and they might engender distinct desires and emotions). And for Langland-Hassan—setting many details aside—these two points are enough to license positing specific, in-the-fiction operators. Hence, my emotion-generating desiderative state can be best described:
I desire that <in the Social Network movie> Zuckerberg get his act together.
This desire can then combine with my belief that <in the Social Network movie> Zuckerberg wore pajamas… to produce my emotional state of angst over his portrayed behavior.
I think Langland-Hassan is right about the need to posit in-the-fiction operators, so I won’t quibble with details of his argument. Let’s just come to his rhetorical question. Having demonstrated the need to posit fictional operators, Langland-Hassan sees little reason to posit a distinct attitude of imagining. He writes:
We already have beliefs about what’s true in each fiction. These can combine with desires about what’s true in each fiction to generate affect. Imaginings involving “in the fiction” operators would simply duplicate the contents of these beliefs—to what end?
This rhetorical question is meant to imply a parsimony argument, versions of which Langland-Hassan has made explicit elsewhere. Since we have to posit in-the-fiction operators anyway, for the sake of parsimony we should avoid also positing a distinct attitude of imagining of the sort the Attitude Approach recommends. When Langland-Hassan says “to what end?” he’s implying the Operator Approach can explain the agreed upon phenomena—no distinct cognitive attitude needed!
Now, there are systematic problems lurking just under the surface with Langland-Hassan’s proposal. If we’re going to posit an in-the-fiction operator as a constituent of thought, we have to ask what that operator’s content is. That is, we have to ask what it is for a proposition to be true in a given fiction, since that is what the proposed operator is supposed to contribute semantically. And by far the most influential (and I think plausible) answer to that question to date is Kendall Walton’s proposal in his (1990) Mimesis as Make-Believe. According to Walton, a proposition is true in a given fiction when we are prescribed by the rules of the game that partly constitute that fiction to imagine that proposition. E.g., what makes it true, on this account, that Hermione knows spells in the Harry Potter stories is that the make-believe game we play when we process those stories prescribes that we imagine she knows spells. So if Walton is right, then the correct theory of what the operator even means, far from eliminating the need to posit imaginings, will actually imply them.
That’s one systematic concern with Langland-Hassan’s implicit argument, one that it would take much more space than I have here to sort out. But as indicated, I want to answer the rhetorical question (“to what end?”) directly, by which I mean stating the phenomena imaginative attitudes can help explain that beliefs+operators can’t.
The answer is that we do far more with fictional contents than just learning about what happens in various pre-established stories. We often make projections about what’s going to happen in the fictions we’re enjoying before we get to the relevant bits. We might, as Alon Chasid would probably point out, have daydreams about hanging out with fictional characters or about intervening in the course of (fictional) events. The daydreams we have might eventually turn into fan fiction, if we decide to write them down. And when we write fan fiction, or even write new fictions, we often have to choose from a number of options that we think up—that is, imagine.
So while I grant that we often have beliefs about fictions (which include various in-the-fiction operators), I reject the idea that beliefs+operators will be sufficient to account for the range of things we in fact do in relation to fictional contents. When I optimistically hypothesize that Hermione will use a spell as I read, I represent her as using one in the future of the fiction without (yet) applying the attitude of belief to that representation. I imagine her doing it. When I daydream about having breakfast with Bilbo—as I do when I’m hungry—I don’t believe that <In the Hobbit> I had or am having breakfast with him; in fact I believe that’s not true in the fiction. But I still imagine it. And when I choose among several ideas for my fan fiction or even new fiction, I don’t yet have beliefs about how any given story goes; I’m still working on what those should be. But I do have lots of imaginings.
So if we broaden our focus to include the many and various cognitive and other activities we partake of in relation to fictions (beyond just learning their basic contents), we’ll see that a distinct cognitive attitude of imagining is an indispensable posit. We have both internal fictional operators and distinct attitudes. I would even argue that, for explaining the various mentioned activities, it’s worth positing a range of subtly different attitudes that, for lack of a better vocabulary in contemporary English, all currently get lumped under the single word “imagining.” That argument, however, is for a different day.