Luke Roelofs is an assistant professor of philosophy at the University of Texas at Arlington. Their research focuses on the metaphysics of consciousness, the moral role of empathy, and whatever else is on their mind recently.
A post by Luke Roelofs
There’s a common idea that sometimes the best aesthetic choice is to “leave something to the imagination”, where that primarily means not presenting it explicitly.
Sometimes this is a claim about horror — that the monster is scarier if it stays mostly hidden, and showing it clearly is often a let-down (Lovecraft 1927 famously said that “the oldest and strongest kind of fear is fear of the unknown”). Sometimes it’s a claim about sexiness – that the most arousing outfit isn’t necessarily the most revealing one, that eroticism can be better served by suggestive hints than graphic depiction. And I think similar dynamics can come up with other emotions: someone’s tragic backstory might seem more tragic if given only through vague suggestions, a perfect day might seem more perfect if we’re not fully shown what happened. And so on.
But what are we actually doing — what is the “imagination” that things are being left to? Setting aside the question of when and why it’s a good technique, I want to focus on the more basic question of what it even means to leave something to the imagination. We can’t just take the faculty or process that we call “imagination” in other contexts and plug that in here. That won’t work, for at least two reasons.
First, whenever we engage with fiction, we’re plausibly using the imagination (see, famously, Walton 1990). But “leaving it to the imagination” often happens within fiction – e.g. a horror writer leaving to the imagination exactly how their monster is killing its victims offscreen But if the imagination is being employed whether the killing is shown or not, then leaving to the imagination must go beyond simply using the imagination.
We might respond by invoking some standard distinction, like voluntary vs. involuntary imagination, or imagination governed by rules and props vs. not. Maybe when the horror director leaves things to “the imagination”, that means leaving it to the viewer’s unconstrained imagination, rather than directing them to imagine something specific.
But (second) I don’t think this works either. If we do focus on the thing left to the imagination and make a point of imagining it, there’s a decent chance that we’ll destroy whatever special allure it had. Trying to picture it, fully and concretely and explicitly, means no longer leaving it to the imagination, but rather spelling it out – even if we are spelling it out in our imagination.
And sure, sometimes we’re very satisfied with how we explicitly imagine something. Sometimes it becomes scarier, sexier, sadder, or whatever when we actually spell it out. But sometimes it doesn’t: often, no explicit image that we can come up with can match the intensity or power of what is “left to the imagination.”
(When the person leaving things to our imagination is a professional author, director, or other artistic creator, I think we can reasonably expect that this should often be true. If we could easily come up with a better specific thing to imagine than the creator can, what are we paying them for?)
For these two reasons, it seems like we need a more specific characterization of the sort of imagination to which things are sometimes best left, to differentiate it from other sorts.
We might consider the literature on mind-wandering, which has the right sort of undirected, inexplicit, free-ranging character (see e.g. Irving and Thompson 2018, Irving et al. 2020). But I don’t think mind-wandering is quite what we’re leaving things to. For one thing, mind-wandering tends to take time – the mind has to be allowed to wander, and given free space to do so. Moreover, the common result of letting one’s mind wander is that at least some more-or-less fleshed out visions or ideas do actually come up and pass before the minds’ eye.
It seems like things that are left to the imagination can sometimes prompt this kind of mental exploration, but not always. Often our attention stays on what is shown (the half-hidden shape of the monster, the sexy outfit, or whatever), and the further stuff that’s left to the imagination serves to imbue that with greater significance without competing with it for time or attention
We also might consider the literature on perceptual co-presentation, which some analyze in terms of imagination (see e.g. Nanay 2010, Roelofs 2018). Everyday perceptual experience shows us complete objects, but presents them as having both revealed aspects, which we directly perceive, and concealed aspects, like their rear surfaces and occluded parts. This lets the unperceived be present in perception, and one account is that the mind does this “filling in” of unperceived elements through unconscious mental imagery or low-level imagining.
Might this be what’s going on when we leave things to the imagination? Well, clearly it’s sometimes involved. If only one part of a creature extends into the light, or only half of a body part is exposed by a garment, the mind tries to complete the shape as best it can. But it seems neither necessary nor sufficient. It’s not sufficient because perceptual co-presentation is so pervasive: every object we see or imagine is likely to have both revealed and concealed aspects. And it’s not necessary because some cases of leaving things to the imagination don’t involve anything as concrete as the unity of a perceived object (e.g. remarks that gesture at a tragic backstory, or hint suggestively at unspecified sexual prospects).
I’m attracted to the thought that leaving things to the imagination involves a sort of combination of co-presentation and mind-wandering. The thought runs: in perceptual co-presentation the mind makes something like a probabilisitc inference, completing an object in the most likely way based on past perceptions, and then incorporates that inference into the phenomenology of perception. It predicts likely future perceptions and infuses them into the actual perception.
If this sort of infusing is possible, perhaps something like mind-wandering – the free, unconstrained, play of associations – could also be infused into a percept. The one tentacle that reaches out from the shadows might be experienced as imbued not only with the fairly mechanical completion of the shapes that likely continue into the dark, but also the open-ended and basically speculative range of bizarre and sinister possibilities it brings to mind.
Note that we could express this idea without linking it to either mind-wandering or perceptual co-presentation. When we leave something to the imagination, that means that something actually perceived or imagined somehow “half-activates” multiple associated ideas and images. By “half-activate” I just mean bringing them to mind enough to influence our pheneomenology but not enough to experience them distinctly.
Observe three things about this idea of simultaneously half-activating many associations:
1. It explains why leaving things to the imagination is sometimes so powerful. By activating all of these different ideas and images at once, we can build up a greater intensity of feeling than they would yield individually, without putting undue demands on our time and attention.
(This is compatible with thinking there might be other mechanisms at play in particular cases, e.g. something being unknown might itself be a scary-making feature, compounding with this more general effect.)
2. It can explain why spelling things out is often disappointing. Half-activated ideas don’t need to be consistent with each other, so we can have lots of disparate or actively conflicting associations all working together. But committing to one of them banishes this happy camaraderie, and suddenly most of the ideas are being ruled out and, very often, the one idea we commit to is less potent than all the half-activated ideas were together.
3. It provides a data point on the folk concept of the imagination. Someone might accept the sort of psychological architecture sketched above, but deny that it has anything to do with the imagination. We can talk about ‘automatic half-activation of associated content that infuses a particular experience’ without the word “imagine”, and if we think of imagination as essentially an active or intentional or conscious process, we would actively deny that this half-activated penumbra involves imagining.
But in fact this everyday phrase does use the specific word “imagination”. That seems to suggest that, insofar as there is a folk concept of imagination, it can cover automatic and unconscious, or semi-conscious, processes, if they have the right characteristics. That doesn’t necessarily show that such a concept must correspond to a real or natural kind, but it might lend some support to such a supposition.
Of course it may be that I’m reinventing the wheel here, and this is something that’s already been spelled out somewhere. If so, I hope someone tells me! And I also hope that it’s valuable to draw out the significance of this topic for thinking about “the imagination” in other contexts.
References
Irving, Zachary, Glasser, Aaron, Gopnik, Alison, and Sripada, Chandra Sripada (2020) "What Does ‘Mind-Wandering’ Mean to the Folk? An Empirical Investigation" Cognitive Science. https://doi.org/10.1111/cogs.12908
Irving, Zachary, and Thompson, Evan. (2018) "The Philosophy of Mind-Wandering" in Christoff and Fox (Eds.) Oxford Volume on Spontaneous Thought And Creativity. Oxford University Press.
Lovecraft, Howard Philips. (1927). “Supernatural Horror in Literature.” The Recluse 1: 23–59.
Nanay, Bence. 2010. “Perception and Imagination: Amodal Perception as Mental Imagery.” Philosophical Studies 50: 239-254.
Roelofs, Luke. (2018). “Seeing the Invisible: How to Perceive, Infer, and Imagine Other Minds.” Erkenntnis 83 (2): 205–229.
Walton, Kendall. (1990). Mimesis as Make-Believe. Harvard University Press.