Margherita Arcangeli is a member of the Institut Jean Nicod as maîtresse de conférence (associate professor) at the École des Hautes Études en Sciences Sociales. Imagination is the more specific object of her research, but she is also interested in other topics in philosophy of mind, philosophy of science and aesthetics.
This week at The Junkyard, we’re hosting a symposium on Bence Nanay’s recent book Mental Imagery (Oxford University Press, 2023). See here for an introduction from Bence. Commentaries and replies will follow Tuesday through Thursday.
* * *
Commentary from Margherita Arcangeli
In 1969 Alan Richardson wrote, in the foreword of his Mental Imagery, that the time was ripe for a synthesis of the work done in psychology and philosophy on mental imagery and the aim of his book was “to serve as a guide to research in this field until a more comprehensive treatment becomes available”. Bence Nanay’s Mental Imagery offers us a new compass for navigating a prolific research topic that, despite its fluctuating fortunes, is still engaging the concurrent efforts of philosophers, psychologists, and neuroscientists.
The book shows how mental imagery connects to almost all major mental capacities (perception, action, imagination, memory, desire, emotion). I necessarily have to narrow down the scope of my commentary by focusing on the relationship between mental imagery and imagination. I take it that although Bence frees mental imagery from imagination, the latter is, in his view, still dependent on the former. While I agree with the first claim, I am more sceptical about the second.
Mental imagery is a notion somehow caught between two others: perception and imagination. To the former, mental imagery owes its sensory nature, but it is legitimate to ask to what extent mental imagery is like perception. Precisely what makes mental imagery different from perception (in primis the fact that it is not dependent on direct sensory stimulation) seems to be explained by it being a form of imagination.
Bence resists this imaginativist view of mental imagery and advances a perceptualist one, according to which “mental imagery is (representational) perceptual processing that is not directly triggered by sensory input” (p. 251). The notion of mental imagery is thus detached from that of imagination: perceptual processing that is not directly triggered by sensory input can be recruited by imagination, but also by other capacities, without necessarily involving the latter.
A conceptual and terminological distinction between mental imagery and imagination seems to be essential in order to better sort out problems and to avoid the risk of talking past each other in philosophical debates in which mental imagery and imagination are invoked (I myself have pleaded for such a distinction, on other – though compatible – grounds than those of Bence). However, mental imagery and imagination are not completely detached in Bence’s view. As anticipated before, when he comes to imagination, Bence seems to make it somehow dependent on the notion of mental imagery.
A widespread view is that mental imagery is an essential component of sensory imagination and a mere accessory element of propositional imagination, which would mainly come without it. This view has pushed propositional imagination closer to supposition, thus casting doubt on its genuinely imaginative nature, the underlying idea being that supposition is not a form of imagination.
In contrast to this view, Bence tries to strengthen the case for the essential role of mental imagery in propositional imagination, while remaining an adjunct to supposition. More specifically, he claims that mental imagery can explain why, contrary to supposition, propositional imagination is subject to the phenomenon of imaginative resistance and has a temporal unfolding involving elaboration on the given proposition.
Many philosophers have emphasised that, to use White’s words, “things are not unsupposable as they are unimaginable” (The Language of Imagination: 145). In line with other analyses, Bence holds that the reason that not all propositions that can be supposed can also be imagined lies in their different dynamics: while supposition requires only the grasping of a content, imagination calls for further exploration of that content. To take an example offered by Bence, if I am asked to imagine that Paris is the capital of Italy, I am pushed, by the very fact of imagining, to elaborate on what it might mean that Paris is the capital of Italy (e.g., the Colosseum stands not far from the Tour Eiffel and a sign says ‘Seine’ where the Tiber should flow; on a political map we can read that the territory known as ‘France’ is called ‘Italy’ and its capital is ‘Paris’).
The question is: where does the difference in dynamics between propositional imagination and supposition come from? Bence’s answer is: from mental imagery, which is substantially involved in both sensory and propositional imagination, but not in supposition. Bence is careful in pointing out that in the case of propositional imagination the elaboration cannot be explained only in terms of further propositions (e.g., that the Colosseum is not far from the Tour Eiffel), otherwise the difference between propositional imagination and supposition would collapse, insofar as the latter can also be developed by means of additional propositions, as happens in reductio ad absurdum arguments (note that, thus, supposition has a dynamics and it is not a mere grasp of a thought).
On this view, imagination constitutively depends on mental imagery and supposition can hardly be seen as a type of imagination. I would like to resist this view.
Bence’s proposal seems to put the emphasis on the content level of our mental states: it is the representational format of their contents, the fact that supposition does not require perceptual-imagistic content, that explains the dynamic difference between imagination and supposition. It seems to me that it is what we do with a certain content, rather than how that content is conveyed, that can shed light on how supposition differs from other forms of imagination.
In supposing a certain content (e.g., that Paris is the capital of Italy), we are in a sense committed to it, we have to take it as if it were true and stick to its logical consequences (e.g., that Rome is no longer the capital of Italy). This is not the case with imagination in its perception-like and belief-like forms (shortly I explain why I prefer this terminology to propositional imagination), which ask me to substantiate the suggested possibility by exploring it in different directions. Put this way, we can see a dynamic difference between supposing that p and (e.g., belief-like) imagining that p even if it happens that the latter is elaborated only by means of propositions, or even if both involve mental imagery on the grounds that language processing engages mental imagery (a suggestion made by Bence in the book). What counts is not the type of content, but the psychological attitude via which we grasp it. On this view, imagination is freed from mental imagery.
More emphasis on the attitudinal aspect also has the merit of highlighting the extent to which propositional imagination is a poorly defined notion. To contrast propositional imagination with sensory imagination is to convey the idea that types of content individuate types of imagination, but it is not clear which type of imagination ‘propositional imagination’ is meant to pick out. Sensory imagination is intuitively seen as similar to perception, and this similarity is expected to show up at the content level, but the content of perception is not unanimously taken to be non-propositional. If there is propositional perception (e.g., Dretske’s epistemic perception) and also its imaginative homologue, then some sensory imaginings would be propositional. Under the umbrella of ‘propositional imagination’ very different types of imagining can be lumped together: belief-like, desire-like, hope-like, and even supposition. So the fact that “there is not much logical space remaining for propositional imagination” (p. 167) is rather due to a lack of clarity about what that space should consist of than to a lack of space.
Response to Arcangeli
I am extremely grateful to Margherita Arcangeli for her commentary. Margherita and I very much agree on probably the most crucial point about mental imagery, namely, that it is very different from imagination. We also agree that it’s something of a category mistake to confuse the two.
I think Margherita and I also agree on what follows from this for imagination and supposition and the relation between the two, but I suspect that Margherita may disagree with this general kumbaya sentiment. Let me take a step back and cover some grounds that are presumably familiar terrain for readers of a blog about imagination.
Imagination is different from supposition. I’ve been avoiding using an example that is on everyone’s mind now (I’m writing this on November 4, 2024), but here it is: I can suppose for the sake of argument what would happen if the Republican presidential candidate wins the election. But that is different from imagining this candidate winning the election. One grand question in the philosophy of imagination is about what the difference is between imagination and supposition.
Margherita has a very original and convincing story to tell about this imagination-supposition divide, which she wrote an excellent book about five years ago. I have a different story, which I take to be at least partially compatible with hers. Imagining the Republican presidential candidate winning the election is, for me at least, an emotionally extremely draining, painful mental episode. Supposing that the Republican presidential candidate wins the election is not that bad. What’s the difference?
Margherita’s line is – I’m simplifying considerably here – that when we are supposing, we do something extra: we are flagging what this mental episode is about as something we are committed to. It’s flagged that we are taking the proposition – in this example, the proposition that the Republican presidential candidate wins the election – as if it were true to explore its logical consequences. And that does not happen in imagination.
My own line is that it is when we are imagining that we do something extra: we elaborate the imagined content with emotionally laden mental imagery. It is not just that I am entertaining the proposition that the Republican presidential candidate wins the election. Entertaining this proposition is not itself imagination. In addition, I also need to elaborate this proposition, and do so with emotionally laden mental imagery. Thus the strong emotional reaction. And this emotionally laden mental imagery is missing when I merely suppose that the Republican presidential candidate wins the election. That is why I don’t have the same emotional reaction then. (What I hope to be my next big project is exactly on this concept of elaboration, which I take to be an attempt at making the content of a mental state more determinate, and which I take to be an important feature of most of our mental states.)
I’m not trying to adjudicate between Margherita’s story and mine. Rather, I want to see how the two could fit together. I am committed to saying that in imagination we do elaborate the content of the imaginative episode with emotionally charged mental imagery. But this view is very much consistent with saying that there is also the kind of flagging of supposition Margherita talks about.
In short, Margherita and I may not disagree as much as she suspects. Having said that, there are some bigger picture disagreements that may drive both her and my thoughts about the difference between imagination and supposition. And that has to do with another notoriously difficult distinction that should also be familiar to readers of this blog: the difference between sensory and propositional imagination. Imagining that the Republican presidential candidate wins the election is propositional imagination. Imagining watching the victory speech of the Republican presidential candidate live is sensory imagination.
Margherita is not a fan of the concept of propositional imagination. Nor am I, but for very different reasons. Margherita doesn’t like it because it seems to suggest that sensory imagination is proposition-free. I don’t like it because it seems to suggest that belief-like imagination is propositional. Margherita takes both kinds of imagination to be at least partly propositional. I take both to be at least partly imagistic. While there is some agreement here (and some possibility of a consensus), I suspect that we are arriving at it from very different directions.