Max Jones is a Lecturer in Philosophy at the University of Bristol. His research is primarily motivated by the conviction that recent developments in the sciences of the mind have significant implications for traditional philosophical debates in metaphysics and epistemology. He is particularly interested in the implications of embodied cognition, predictive processing, ecological psychology, and active perception, and has recently been working on embodied and enculturated approaches to the imagination.
This week at The Junkyard, we’re hosting a symposium on Epistemic Uses of Imagination, a recently published volume edited by Chris Badura and Amy Kind (Routledge 2021). See here for an introduction from Chris and Amy. Commentaries will appear Tuesday through Friday.
* * *
When does (and when doesn’t) imagination support reasoning?
Philosophical debates about the epistemic usefulness of the imagination often seem to proceed as if nobody has ever bought an item of furniture and then failed to fit it through their front door. We like to think that we’re good at seemingly simple perceptual-motor planning tasks like this, but our assessments of our own abilities can be deceiving. I may be a foolish outlier, but I was once given a garden table, which I imagined easily getting into my house, but sadly I was mistaken.
I mention the now classic example of using the imagination to assess the furniture-door challenge, as it nicely points to a problem I see in the way in which philosophers tend to address the epistemic usefulness of imagination. In short, we tend to focus on whether imagination can be epistemically useful in general, while neglecting to pay attention to the specific instances in which it may be less useful and perhaps even misleading, as well as why these instances might arise.
It’s clear, as emphasised in, for example, Kind (2016) and Balcerak Jackson (2016, 2018), that if imagination can be relevantly constrained by reality then it can contribute to epistemic justification. However, this gives rise to the question of whether and when imagination can be relevantly constrained by reality. Most work on the role of imagination in reasoning, including three contributions to Section II of the collection (Chapters 5-7), assumes that the relevant reality-orienting constraints on “offline” imagination can be inherited from more straightforwardly reality-oriented “online” faculties such as those for believing, thinking, or perceiving.
The usual assumption is to see these faculties as generally reality-oriented, such that imagination can (when put to epistemic uses) inherit a domain-general capacity to be relevantly reality-oriented. This assumption is at least implicit in Myers’ (Chapter 5) focus on general rules of inference as the relevant operations in reasoning with the imagination (107) and the way in which both Myers (Chapter 5) and Badura (Chapter 7) use formalised abstract generalisations of patterns of (imaginative) inference. This assumption is particularly clear in Berto’s (Chapter 6) attempts to formulate a logic of the imagination, with domain-generality arguably being a defining feature of what makes something a “logic”.
Yet, despite these assumptions, it is clear that we are naturally better at thinking about some domains rather than others and that our perceptual faculties are geared towards representing or picking up on some aspects of reality but not others.[1] As a result, one should expect our ability to apply relevantly reality-oriented constraints to vary from domain to domain depending on how good we are at tracking reality within a given domain. Importantly, this domain-specific variation may not be obvious “from the inside”. We may need to look to empirical work in psychology and neuroscience to determine where our domain-specific strengths and, importantly, our weaknesses lie.
This can be made more concrete by considering perceptual illusions that operate within specific domains. Two illusions that are particularly pertinent to the now infamous furniture-door example are:
i. The size–weight illusion (SWI): In which, when a subject lifts objects of equal weight, but different sizes, they perceive the smaller object as heavier (Murray et al. 1999).
ii. The weight–size illusion (WSI) In which subjects perceive lighter objects as being smaller and heavier objects as being larger (Hirsiger et al. 2012)
It turns out that we’re not as great as we might expect at judging the size or weight of objects, counterintuitively, particularly when we’ve been able to access more information by integrating information from visual and haptic perceptual channels. More empirical work needs to be done, but it seems plausible that our errors in online interaction might also be inherited by offline imagination-supported judgements of size and weight. It’s usually a better idea to bring a tape-measure to the furniture store rather than relying on imagination alone.
Insofar as imagination inherits reality-oriented constraints from supposedly reality-tracking faculties, it is also susceptible to inheriting reality-distorting illusions and biases. Therefore, one cannot assume the justificatory capacity of imagination within a given domain unless one can be sure that a domain-specific illusion or bias hasn’t been brought along for the ride.
Returning to the specifics of the furniture-door case, one option is to doubt the imagination’s general epistemic usefulness in providing justification for our size or weight judgements, based on the the weaknesses in our everyday size and weight perceptions. Moreover, this case needn’t be an outlier. There is plentiful evidence of systematic illusions in other sensory domains, and one might expect similar inherited reality-distortions to arise for more cognitive-oriented theories of imagination, given the extensive evidence for a whole range of systematic domain-specific cognitive biases and common mistakes. The negative story that could be gleaned from this is that we shouldn’t expect our imaginative capacities to solely inherit reality-oriented constraints from our often reality-distorting cognitive or perceptual capacities (thereby adding further support to Mallozzi’s (Chapter 8) challenge to the epistemic usefulness of imagination).
Thankfully, a more positive approach is, however, available. Rather than focusing on aspects of reality for which our imaginatively-aided judgements are potentially subject to distortion, we can look for those aspects of reality that our natural constraint-providing faculties are more reliably able to track, and the we can therefore hope more reality-oriented constraints to emerge from. Determining which domains our perceptual or cognitive faculties are reliable in is a highly contentious issue, with one’s answer likely to vary depending on one’s broader theoretical commitments. Here, I’ll focus on the ecological approach to perception as one case of a theory that can provide a more positive assessment of the seemingly problematic implications of the SWI and WSI for the classic furniture-door challenge.
Proponents of the ecological approach to perception argue that in the case of apparent size-weight illusion, the perceptions only seem illusory if one assumes that the size or weight are the properties of the objects in question that our perceptual systems are sensitive to. If one instead takes subjects’ assessments of “weight” to really be assessments of pick-up-ability or moveability, then these subjects’ assessments can be seen as largely correct and sufficiently reality-oriented (Amazeen & Turvey 1996; Shockley et al. 2004) (albeit not objective in the sense of being independent of the subject’s own object-picking-up abilities). There is, in effect, no illusion. We are just sensitive to different, more action-oriented, and narcissistic aspects of reality than a more traditional account of perception assumes.
If this kind of non-illusory interpretation generalises to other cases of apparent perceptual illusions (as proponents of ecological psychology would argue it does (Michaels & Carello 1981; Chemero 2009)) then we should expect our action-oriented perceptual abilities to give rise to an aptitude for action-oriented imagining. We may be more justified in our imagination-supported judgements about what we (ourselves, in all our specifically embodied glory) can do with things than we are in similarly imaginatively-supported judgements about the objective properties of objects (see Hanrahan’s contribution to the target collection).
I happen to favour an ecological account of perception. Yet, in the bigger picture, nothing hangs on this. The key lesson is that our cognitive and perceptual capacities vary from domain to domain and whether we take such capacities to be successful may depend on how the domain in question is specified. Moreover, we shouldn’t expect homogeneity in the variation among domains. People’s skill at imagining should vary depending on their other perceptual and cognitive skills. Kind (2020) may be right, in one sense, that imagination is a skill, but it may be best understood as a secondary skill that is primarily dependent on other domain-specific skills. We’re better at thinking about or perceiving some kinds of things and not others, and we’re consequently likely to be better at imagining some things than we are at imagining others. Thus, insofar as imagination’s justificatory power is derivative from our other reality-tracking faculties, we should go beyond solely trying to explain how imagination can in general provide justification, and start seeking the particular cases where imagination is set up to be reality-oriented, as well as being careful to determine which aspects of reality our cognitive/perceptual, and thus imaginative, capacities are oriented towards.
All this needn’t be a problem for the wider project of providing a general case for the epistemic usefulness of using imagination in reasoning. After all, nobody is positing infallibility, and one can simply hold that imagination is as useful as the faculty from which it derives its constraints. However, it may make one more sceptical of the possibility of developing a one-size-fits-all account of imagination’s epistemic uses independently of the content of the imagining and the modality or medium of the imaginative episode in question. At the very least, it should make us suspicious of trusting our own internal judgements about which cases of the use of imagination in reasoning are likely to be useful, at least without first consulting the relevant empirical evidence (which may not yet have been sought or found). A piecemeal approach to the epistemic usefulness of imagination may be far less satisfying, as well as being far harder to formulate, with the bulk of the relevant empirical investigation still to be undertaken. Yet, it may be the most realistic way to approach the question of when the imagination can usefully aid our reasoning.
[1] One need not be committed to a strong Evolutionary Psychology view that postulates domain-specific mechanisms to endorse the weaker claim here that we have some domain-specific capacities or just some domain-relative variation in natural epistemic abilities.
References
Amazeen, E. L., & Turvey, M. T. (1996). Weight perception and the haptic size–weight illusion are functions of the inertia tensor. Journal of Experimental Psychology: Human perception and performance, 22(1), 213.
Balcerak Jackson, M. (2016). On the epistemic value of imagining, supposing, and conceiving. In Kind & Kung (eds.) Knowledge through imagination, 41-60. OUP.
Balcerak Jackson, M. (2018). Justification by imagination. In MacPherson & Dorsch (eds.) Perceptual imagination and perceptual memory, 209-26. OUP.
Chemero, A. (2009). Radical embodied cognitive science. MIT press.
Hirsiger, S., Pickett, K., & Konczak, J. (2012). The integration of size and weight cues for perception and action: evidence for a weight–size illusion. Experimental brain research, 223(1), 137-147.
Kind, A. (2016). Imagining under constraints. In Kind & Kung (eds.) Knowledge through imagination, 145-59. OUP.
Kind, A. (2020). The Skill of Imagination. In Fridland & Pavese (eds.) The Routledge Handbook of Philosophy of Skill And Expertise (pp. 335-346). Routledge.
Michaels, C. F., & Carello, C. (1981). Direct perception. Prentice-Hall.
Murray, D. J., Ellis, R. R., Bandomir, C. A., & Ross, H. E. (1999). Charpentier (1891) on the size—weight illusion. Perception & Psychophysics, 61(8), 1681-1685.
Shockley, K., Carello, C., & Turvey, M. T. (2004). Metamers in the haptic perception of heaviness and moveableness. Perception & Psychophysics, 66(5), 731-742