In the Early Modern period many philosophers took ideas to be mental images of the objects they stood for.
Close your eyes, and imagine a bedroom you lived in as a child. Take a moment in order to do a thorough job. Try to visualize the spatial layout of the room, the colors and textures, details of any clutter around your floor and desk. All but a few of us experience mental images when we try to do this.
But what is a mental image? If you can summon one to mind, you probably already have a sense of what it is. If you don’t, it’s hard to define in any other terms. You might say it’s like a picture that belongs to the “mind’s eye.” Or you could compare what it’s like to have a mental image with what it’s like to have a visual experience of something. It’s like a visual experience, you might say, only less vivid.
Here’s where I want to pause. This claim gets made all the time in philosophical discussions of the imagination: mental images are less vivid than genuine visual experience of the world. Something like this claim is almost undeniably true. But what, exactly, does it mean? What is it for mental images to be less vivid than genuine visual experiences of the world?
The idea of “vividness” is sometimes traced back to the great British empiricist David Hume’s claim that our mental images have less “force and vivacity” than the sense impressions we have in firsthand experience; they are “faint and languid” where perception is “lively and strong.”
Psychological tests designed to elicit valid self-reports aren’t much more specific. One metric asks subjects to rate their mental imagery on a scale from “perfectly clear and lively as real seeing” to “no image at all, you only ‘know’ that you are thinking of the object.” Between those extremes lies another point labeled “dim and vague; flat.” These questions introduce some new terms—clarity, dimness, vagueness, flatness—but aren’t much help in analyzing what we mean when we talk about vividness of mental imagery.
Beyond unhelpful, there’s counterproductive. It’s easy to jump to conclusions about the vividness of mental imagery based on errors about what we mean. Recently, the following image circulated on the internet asking readers to imagine a red star, then choose which of the following six images best matched what they ‘saw’ in their mind’s eye:
This is meant to be a proxy test for the vividness of mental imagery. Perhaps there is some feature of people’s mental imagery that would make imagined red stars approximate a pink star seen on a screen. But lack of vividness is not just lack of color saturation. It’s a kind of degeneracy of a mental image, as compared to the phenomenology of ordinary visual experience.
Presumably any mental image could suffer from degeneration. It’s important to keep this in mind to avoid making other errors as well. For example, Elaine Scarry says in Dreaming By the Book that the transparency of gauzy, vague, or ghostly things—consider a see-through curtain, for example—are easier to imagine than solid things because they exploit the gauzy vagueness of the imagination. But this point conflates a feature of a physical object—its transparency—with a more abstract, all-over feature of a mental image: its lack of vividness. Surely a mental image of a ghost can fail to be vivid in the same way as a mental image of a person can fail to be vivid.
Recently, philosopher Amy Kind argued that the very idea of vividness is “so problematic as to be philosophically untenable.” Kind considers various definitions of vivacity in terms of clarity, detail, determinacy, brightness, color, and a combination of all of the above. These are all concepts that are most literally applied to physical images—drawings and photographs—and perhaps only metaphorically applied to the mental images that constitute our episodes of imagination. Kind argues that none of these features capture the difference in feeling between mental images and real visual experiences that we call “vividness.”
Kind also rejects the suggestion that vividness is a primitive notion—that is, an idea so basic it cannot be substantively defined in other terms. She recommends we give up the idea of vividness as (ironically) it lacks clarity and determinacy. Is Kind right that we should simply give up on the idea? I myself feel a lot of resistance to the suggestion, in part because the concept seems so natural. I feel that I have a grip on what I’m talking about when I say that my mental image of my rug at home is more vivid than my mental image of Hoover Tower. But that doesn’t sound very philosophically responsible.
Fortunately, there’s a further way to save the idea of vividness. Vividness is itself a phenomenal property: a feature of your experience in virtue of which it feels like something to you. And our concept of that property, correspondingly, might be a phenomenal concept. These are concepts of feelings that allow us to recognize and imagine those feelings.
Notably, though, you can’t analyze phenomenal concepts in substantively different terms—and you can’t define phenomenal properties for those who don’t already have experience of them. You need to have felt something yourself in order to have a phenomenal concept of it. For instance, if you haven’t ever felt pain, you might have a descriptive concept of it (something that people don’t like) but not the kind of concept that would let you imagine it for yourself. You certainly wouldn’t be able to tell what we mean when we talk about the difference between dull aches and sharp stabs of pain.
The simple fact that you cannot analyze a concept into other concepts, and that you can’t come to understand what property it captures by way of definition, does not immediately imply that the concept itself is theoretically suspect, or that we must do without it. If that were the case, we should have to do without all sorts of phenomenal concepts that we all use and understand quite well—including pain concepts, and concepts of itches, tastes, sounds, looks, and other feelings. If the concept of vividness is like these concepts, we can hold onto it even though we cannot analyze it at all. What would be remarkable about that would simply be the fact that this phenomenal concept tracks a somewhat more abstract property than many other phenomenal concepts: the most salient property by which a mental image of something, and an actual experience of the same thing, differ.