Jonathan Mitchell earned his Ph.D. in Philosophy from the University of Warwick and subsequently held a British Academy Postdoctoral Fellowship at the University of Manchester. He later joined Cardiff University, where he is now Senior Lecturer in Philosophy. His research focuses on Philosophy of Mind and Phenomenology, with particular attention to spatial perception, value, and emotion.
A post by Jonathan Mitchell
One puzzle which has animated philosophers of perception from the classical phenomenologists to the present day sometimes goes under the name of the ‘puzzle of perceptual presence’. Consider the following situation. Sitting in the park, looking at the oak tree opposite, what I seem to see is a complete three-dimensional thing, that is a bounded voluminous object. Yet from where I am sitting, I can only see the side of the tree facing me. Now, one side of a tree does not make a whole tree, and so what I am presented with from where I sit is somehow less than the whole tree. But at the same time, I don’t seem to be seeing a mere tree-façade, as something akin to a stage prop of a tree; much less a mere tree-like surface. What I am visually aware of as I sit there and look at the tree is nothing less than a complete three-dimensional thing, the ‘whole tree’. We can articulate the resulting puzzle more generally as follows: how is it the case that we can enjoy visual experiences as of complete three-dimensional entities given the fact that we are limited to seeing the side or sides facing us from a specific spatial perspective? How are these phenomenological facts to be reconciled?
In my recent book – A Sense of the Possible: The Horizons of Visual Experience (Oxford University Press, 2026) – I contend that we can solve this puzzle through a correct understanding of the role that a ‘sense of the possible’ plays in visual experiences. More specifically, what is required is a detailed and plausible account of the kind of awareness we enjoy of the sides and aspects of the relevant entities that are currently ‘out of view’, or otherwise occluded, but that nonetheless contribute to ‘making up’ the relevant complete three-dimensional object. The classical phenomenologists had a special term for this aspect of perceptual experience that ‘refers beyond’ what is currently presented of an object to its occluded sides, namely an intentional horizon.
Various accounts of what these intentional horizons consist in have been proposed, and so various purported solutions to the puzzle of perceptual presence are on the market. Here I want to mention one tempting, but in my view incorrect, account, namely one that draws on the role of imagination – I’ll call it the ‘imagination view’. Here is how the imagination view goes: a visual experience of an object as a complete three-dimensional entity consists in a visual component and a visualised component, that is, an imagistic representation of the occluded sides of the object. Jennifer Church articulates the view as follows:
When we see a tree, we imagine distinct perspectives on distinct parts of that tree, and it is the combination of those distinctive perspectives and parts that gives us the experience of the tree as a well-rounded object in space. (Church 2011: 76)
Amy Kind suggests something similar:
Working in tandem with our perceptual capacities, our imaginative capacities contribute to our perceptual experience by making unseen features of objects seem present. As I’m looking at the Diet Coke can on my desk, it’s via a conjunctive effort of vision and imagination that I have the perceptual sense of the can as a voluminous whole. The front side of the can is seen; the back side of the can is imagined. (Kind 2018: 176)
Further details notwithstanding, this view runs into significant problems, two of which I explore here.
The first I’ll call the horizonal structural problem. Consider the following line of thought, separately from our puzzle of perceptual presence. When we visualise a material thing, say a cube, we visualise it perspectivally, in the sense that we don’t (and cannot) visualise all sides of the cube simultaneously. Put otherwise, in visualisation we still only enjoy, as we do in visual experience, a limited presentation of the facing sides of the material thing we are visualising. Nonetheless, we visualise a whole cube, not a cube-part, and so we seem to be drawn into thinking that visualisation, separately from visual experience, has its own problem of ‘presence’ in terms of how, in visualising a complete three-dimensional object, the ‘non-facing’ sides are given to us, and so a question arises concerning its putative ‘intentional horizons’.
If the above is right, then this causes problems for the imagination view, as a solution to the problem of perceptual presence. Let me explain why: a visualisation of the backsides of a cube, the front sides of which I can currently see, is not just some unstructured visualisation of mere cube-parts, but a perspectival visualisation as of a complete object, and so one which must have its own intentional horizon. But surely, we don’t want to say that when we see a cube, are perspectivally presented with its front sides, and putatively visualise its back sides, that internal to that visualisation is another visualisation of front sides relative to back sides – we don’t want to give a ‘more visualisation’ answer to the question concerning the putative intentional horizons of visualisation, including those visualisations that are supposedly constitutive of our visual experiences as of complete three-dimensional objects. But note that if we give a non-visualisation answer to what the intentional horizon of visualisation consists in, then the following problem arises: why can’t we just appeal to whatever is said there in answer to our original problem concerning the intentional horizons of visual experience, and so do away with appeal to visualisation entirely? The issues here concern the structure of visualisation itself and are motivated by the thought that if visualisation ‘mirrors’ the horizonal structure of visual experience then appealing to visualisation or imagination to solve the puzzle of perceptual presence pushes this puzzle one rung back to questions concerning the horizonal structure of visualisation.
The second problem I’ll call the unity-redundancy problem. Let’s grant for the sake of argument that when we visually experience an object as a complete three-dimensional thing, that this is, as Kind puts it, ‘a conjunctive effort of vision and imagination’; simply, we strictly see the facing sides and visualise the non-facing sides. We can now ask the following question: what supplies the ‘unity’ between vision and visualisation, in the sense that we enjoy a well-motivated rather than arbitrary ‘synthesis’ of different kinds of representations of the object (such that the representations are ‘joined up’ in the right way). At a minimum the following should hold: I have to be aware of the non-facing sides so represented in visualisation as making up for what is missing – what is not seen – in the visual presentation of the facing sides, and vice versa; I have to take the visual presentation of the facing sides as making up for what is missing – what is not visualised – in the visualised representation of the non-facing sides. It is only on this basis that I can justifiably take my perceptual representation of the object to be a unified representation of one and the same object, albeit from different perspectives.
However, if this is right, we get the following problem. If a unified representation of perception and imagination can only be ‘achieved’ as described above, then both presentations – vision and visualisation – must already include their own ‘horizonal references’ before any bringing together of vision and visualisation. The central line of thought is as follows: to make sense of the imagination view as yielding a plausible answer to how vision and visualisation can form a unified representation, it has to be the case that the different presentations of the object make up for what is missing in their counterparts. But if that is right, then prior to any ‘synthesising’ of vision and visualisation – to the forming of a unified representation – the original presentations already must involve what Husserl would call ‘empty horizonal references’ to parts of the object that could be ‘fulfilled’, that is could be given to us in a sensory representation of those parts of the object, but critically are not yet so given. However, if we accept this, then the imagination view looks to be redundant. Here is how Husserl puts the point:
But that is already to say that every such imagination is a full phantasy, which could also exist for itself as a mere phantasy and which binds the presentation to the components that refer beyond. Likewise, it is to say that perception, even without phantasy, delivers a full representation, namely as presentation of the front side along with components that refer beyond. (Husserl 1907 [1907] §18: 48)
Critically, we then have a situation in which the ‘horizonal references’ of visual experience cannot be explained by appeal to visualising the occluded sides of the objects we can see.
The two problems I have highlighted are by no means knock-down arguments against the ‘imagination view’ as a solution to the puzzle of perceptual presence, and this is not least because they depend on accepting some more or less controversial premises (e.g., concerning the ‘horizonal structure’ of visualisation, conditions on unified representations, etc). Nonetheless, they are difficult problems for the view to grapple with, and ones which, as far as I’m aware, haven’t been considered in detail by advocates of the view.
I think that we should look elsewhere than imagination for a solution to the puzzle of perceptual presence. We should look more towards distinctively visual forms of modal awareness characteristic of our experiences of complete objects; indeed, I explore such alternatives in my book, along with a range of other possible solutions.
References
Church, J. (2011) Possibilities of Perception. Oxford University Press.
Husserl, E. (1997 [1907]) Thing and Space: Lectures of 1907. R. Rojcewicz, trans. Dordrecht: Kluwer Academic Publishers.
Kind, A. (2018) ‘Imaginative presence’, in F. Macpherson & F. Dorsch (eds.), Phenomenal Presence. Oxford: Oxford University Press.
Mitchell, J. (2026) A Sense of the Possible: The Horizons of Visual Experience. Oxford: Oxford University Press.