This week at The Junkyard we’re hosting a symposium on Anja Berninger and Íngrid Vendrell Ferran’s (eds.) recent book: Philosophical Perspectives on Memory and Imagination. On Monday, we began with an introduction from Anja Berninger and Íngrid Vendrell Ferran. Today we have a commentary from Seth Goldwasser. Additional commentaries will appear the rest of the week.

It’s not uncommon to hear that someone has a good memory or is particularly imaginative. At first glance, such attributions appear to pick out some innate quality or disposition. However, philosophers have begun investigating whether memory and imagination might be cognitive skills or abilities (see Hopkins 2014, 2018, 2022, n.d., especially chapters 1 and 4-6; Kind 2020, 2022a,b,c; Michaelian 2021; and my 2022). In that case, to have a good memory or be imaginative might mean being a skilled rememberer or imaginer. Or it might mean that one is able to accurately recall some detail or vividly picture some far-off, alien possibility (more or less) at will.

The chapters in Part 3 of Anja Berninger and Íngrid Vendrell Ferran’s edited volume, Philosophical Perspectives on Memory and Imagination, develop this line of thought. All three draw lessons about the possibility of continuity between memory and imagination. Particular attention is paid to whether remembering is in some respect the same process or attitude as imagining, as the metaphysical continuist maintains.[1] All three chapters express measured doubt. Here, I highlight a claim from each chapter and raise concerns with a view to providing some minimal defense of continuism on the grounds that remembering is a type of imaginative project.

Starting with Kind (chapter 9), she argues that the training regimens by which we build and improve our mnemic and imaginative skills can tell us something about their nature as dispositions for action. Specifically, if similarities between training regimens for memory and imagination are due solely to features that skills in general exhibit then there’s some defeasible reason against thinking that memory and imagination are metaphysically of a piece. Divergence between regimes gives us more reason against continuism.

Kind highlights two similarities and three differences (203-206). First, training regimens for memory and imagination involve concatenating series of units, say, imagined movements or digits, into chunked sequences. Second, both greatly facilitate improvement if they’re personalized. That is, training regimens are more effective the more they tap into a domain that the agent already knows well or are tailored to her pre-trained abilities or dispositions. However, as Kind points out, skill development in general involves concatenation (see Verwey 2001, Wymbs et al. 2012, and Fridland 2019) and is better facilitated with personalized training.

By contrast, the differences between training regimens are telling. First, while imagination training can be facilitated by publicly accessible scripts, skilled rememberers appear to be (mostly) self-trained. Second, in part because accuracy in recall is arguably the principal aim of its exercise, it’s possible to get immediate feedback while training one’s memory. By contrast, since accuracy isn’t the principal aim of its exercise, it’s difficult to see how one could get (direct) feedback while training one’s imagination. Third, for similar reasons, judging that memory has improved is straightforward—just look for increases in accuracy or in the amount that’s accurately recalled. By contrast, judging that imagination has improved proves difficult.

Do these differences in training regimens point to deeper differences in the respective skills? There are at least two reasons to think they don’t. First, the self-training Kind discusses might be an artifact of how we often study memory (195-196, 203-205). In the lab, the items subjects are meant to recall are usually arbitrary. This helps keep encoding and consolidation of those items within the lab setting. However, it makes guided or joint training difficult, since what gives some otherwise arbitrary item significance sufficient for successful recall likely differs between individuals. By contrast, much of what we remember outside the lab is already significant for us and, in addition, ripe for joint acts of recollection with others.[2] The point extends to memory competition. Because they’re standardized, these tasks acquire or end up drawing on shared significance for skilled rememberers, otherwise known as “mnemonists”. Mnemonists regularly share and discuss their training regimens and techniques, train together, give each other advice, etc. (see Foer 2012, especially chapter 8).

The second reason to think that differences in training regimens don’t run (much) deeper is that both memory and imagination training aim at getting the agent to construct the right content(s). While accuracy might not be the principal aim of imagining, both memory and imagination training in the context of Kind’s discussion aim to develop dispositions for producing contents that are otherwise reality tracking or instructive. That remembering involves an additional constraint of accuracy is by itself insufficient to distinguish remembering from other imaginative projects with epistemic or practical aims, for instance, (accurately) imagining the correct golf swing (cf. 199-200).

Moving on to Hopkins (chapter 10), he focuses on memory for experiences from the personal past—usually called “episodic memory”—and sensory/experiential imagining. He argues that we make headway explaining a number of relations between memory and imagination by shifting focus away from occurrent states of remembering and imagining and towards the agent’s powers to remember and imagine. Due to space, I focus on one relation: the identity of powers.

Hopkins (215-219) argues that we should resist thinking of the ability to intentionally imagine as distinct from the capacity for unbidden imagery. Ditto for the ability to intentionally remember and the capacity for unbidden memories. Hopkins claims that there’s a defeasible inference to identity of powers from “shape” (see Locke 1974, Small 2017). Shape picks out features of an individual’s ability or capacity across multiple dimensions which together at least partly constitute that ability or capacity. For instance, the shape of Michael Phelps’s ability to swim includes his speed, the strength of his kicks, etc. If two powers have the same shape then, other things equal, they’re the same power (cf. 218-219). Hopkins maintains that because the ability to intentionally imagine has the same shape as the capacity for unbidden imagery, “the ability incorporates the capacity” (217). Ditto for the ability to intentionally remember (219).[3]  

The argument for identity of powers raises the question whether an individual’s ability to remember and their ability to imagine share a shape, as metaphysical continuism might suggest. Hopkins claims they don’t. The ability to episodically remember is constrained by its target, namely, representing particular past episodes to oneself (220-222). By contrast, the ability to imagine isn’t constrained by such specificity. Even when one fails to imagine a particular episode, say, a shipwreck, the failure is likely “general: an inability to imagine anything, or colours, or ships” (221). This is a difference of shape. An essential part of the shape of memory is how well one represents particular episodes to oneself, whereas this isn’t part of the shape of imagination.[4]

I’m not quite convinced. It might turn out that the shapes of abilities to engage in certain imaginative projects include how well one represents particular episodes. Say one experientially imagines how their spouse might react to a specific anniversary gift to determine whether to buy it. It won’t do to imagine something other than a particular episode—giving their spouse that gift on that occasion. Imagining particular episodes for epistemic or practical purposes might thus require specificity of the sort Hopkins thinks distinguishes the ability to remember. And Hopkins allows that specificity might not be a defining feature of memory in general (222, fn.18). If all that’s different between episodic memory and other experiential imaginative projects are their aims then the powers that underwrite them might share a shape. In which case, we have reason to think they’re the same power.

Finally, Robins (chapter 11) argues that the method of loci (MoL) is a way of remembering that involves the agent’s building novel mental representations through what Van Leeuwen (2013) calls “constructive imagining.” Agents using the method of loci memorize the layout of a familiar place and then intentionally construct novel imagery associated with to-be-remembered items and positioned along a path through that place. At retrieval, agents take a “mental walk” along the path, recalling the imagery and, with it, the to-be-remembered items. Robins argues that comparing MoL with episodic memory provides insight into the nature of memory and imagination, including how they’re related. First, MoL and episodic memory both involve construction of novel contents (241). Second, both involve autonoetic consciousness, viz., awareness of what’s represented as being of the past and what it’s like representing it (see Tulving 2002).

However, MoL is deployed deliberately such that the agent is aware of her constructive activities, whereas episodic memory involves no such awareness. Otherwise, this awareness would undermine the agent’s autonoetic consciousness of the constructed contents. Alternatively, if she were aware of the constructed nature of some content in remembering, she’d eliminated it as extraneous or inaccurate (242). Similarly, the direct control the agent has over the content of her construction in MoL is absent in episodic memory (ibid).[5] This provides some defeasible reason to think that episodic memory and imagining are discontinuous along the dimensions of awareness and control (see also Robins 2020, 2022).

Again, I’m not quite convinced. At least in cases of active remembering, awareness that she’s constructing the contents of her episodic memory isn’t inconsistent with the agent’s having autonoetic consciousness of those contents nor with her eliminating extraneous content or inaccuracies as they crop up. Exerting control over construction is likewise consistent with experiencing the content as recalled. Autonoetic consciousness might be part of the phenomenology of success along with (other) metacognitive feelings of, for instance, fluency. Moreover, it’s consistent with there being strict constraints on the content constructed. Similar constraints are in force when the agent is engaged in other epistemic imaginative projects. For instance, imagining her couch as having its actual spatial dimensions is a strict constraint on content that’s in force if the agent decides to try to figure out whether her couch fits through a door using only imagination (see Dorsch 2012).

Zooming out, a route the metaphysical continuist can take is to insist that there’s nothing (much) more to imagining as an agent-involving process than the collection of projects that call on us to construct contents appropriate for their completion. If that’s right then some (minimal) constraint(s) on the content constructed is always in force. Episodic remembering, then, might just be one type of imaginative project among others, distinguished primarily by its aim, say, representing particular past episodes. The epistemic and phenomenological hallmarks of (veridical) episodic memory would then fall out of the agent’s making good on this aim.    

Notes

[1] On the metaphysical (dis)continuism debate, see Perrin (2016) and Michaelian, Perrin, and Sant’Anna (2020). Hereafter, I focus on the part of the debate concerned with whether episodic recall and sensory/experiential imagination are the same kind of process.

[2] As I’ve argued elsewhere (2022), memory development in early childhood is heavily influenced by joint acts of narratively structured recollection (see also Reese 2002, Ornstein et al. 2004, and Fivush 2019).

[3] Nonetheless, Hopkins also maintains that instances of unbidden memory or imagery are not actions (216-219).

[4] Nonetheless, Hopkins also maintains that the powers are intimately related (224-225).

[5] Similarly, Robins argues that MoL and episodic memory differ with respect to what’s episodic, that is, what’s experienced by the agent as a particular event. The construction and subsequent mental walk in MoL are both experienced as events. By contrast, only the output of retrieval in episodic memory is properly episodic (242-243).

Reference

Aronowitz, S. (forthcoming) Semanticization Challenges the Episodic-Semantic Distinction. British Journal for the Philosophy of Science.

Dorsh, F. (2012) The Unity of Imagining. De Gruyter.

Fivush, R. (2019) Family Narratives and the Development of an Autobiographical Self: Social and Cultural Perspectives on Autobiographical Memory. New York: Routledge.

Fridland, E. (2019) Longer, Smaller, Faster, Stronger: On Skills and Intelligence. Philosophical Psychology 32 (5), 759–783.

Goldwasser, S. (2022) Memory as Skill. Review of Philosophy and Psychology, 1–24.

Greenberg, D. and M. Verfaellie (2010) Interdependence of Episodic and Semantic memory: Evidence from Neuropsychology. Journal of the International Neuropsychological Society 16 (5): 748–753.

Hopkins, R. (2014) Imagining the Past: On the Nature of Episodic Memory. In Memory and Imagination, eds. F. Dorsch, and F. Macpherson, 46–71. Oxford: Oxford University Press.

——— (2018) Episodic Memory as Representing the Past to Oneself. Review of Philosophy and Psychology 5(3): 313–31.

——— (2022) Architecture and Cultural Memory. In Heritage and War, eds. H. Frowe, and D. Matravers, Abingdon: Routledge.

——— (n.d.) The Profile of Imagining [Unpublished Manuscript]

Kind, A. (2020) The Skill of Imagination. In The Routledge Handbook of Skill and Expertise, eds. E. Fridland, and C. Pavese, 335–46. Abingdon: Routledge.

——— (2022a) Learning to Imagine. British Journal of Aesthetics 62(1), 33–48.

——— (2022b) Fiction and the Cultivation of Imagination. In The Philosophy of Fiction: Imagination and Cognition, eds. J. Langkau and P. Engisch, 262-281. Routledge.

——— (2022c) Imagination and Creative Thinking. Cambridge University Press.

Locke, D. (1974) Natural Powers and Human Abilities. Proceedings of the Aristotelian Society 74(1): 171–84.

Michaelian, K. (2021) Imagining the Past Reliably and Unreliably: Towards a Virtue Theory of Memory. Synthese.

Michaelian, K., D. Perrin, and A. Sant’Anna (2020) Continuities and Discontinuities between Imagination and Memory: The View from Philosophy. In The Cambridge Handbook of the Imagination, the Cambridge Handbook of Imagination, ed. A. Abraham, 293–310. New York: Cambridge University Press.

Ornstein, P., C. Haden, and A. M. Hedrick (2004). Learning to Remember: Social-communicative Exchanges and the Development of Children’s Memory Skills. Developmental Review 24(4), 374–395.

Perrin, D. (2016) Asymmetries in Subjective Time. In Seeing the future: Theoretical perspectives on future-oriented mental time travel, eds. K. Michaelian, S. B. Klein, and K. K. Szpunar 39–61. New York, NY, US: Oxford University Press.

Reese, E. (2002). Social Factors in the Development of Autobiographical Memory: The State of the Art. Social Development 11(1), 124–142.

Robins, S. (2020) Defending Discontinuism, Naturally. Review of Philosophy and Psychology 11(2), 469–486.

——— (2022) Episodic Memory Is Not for the Future. In Current Controversies in the Philosophy of Memory, eds. A. Sant’Anna, C. McCarroll, and K. Michaelian. London: Routledge.

Small, W. (2017) Agency and Practical Abilities. Royal Institute of Philosophy Supplement 80: 235–64.

Tulving, E. (2002) Episodic Memory: From Mind to Brain. Annual Review of Psychology 53: 1–25.

Van Leeuwen, N. (2013) The meanings of "imagine" part i: Constructive imagination. Philosophy Compass 8(3), 220–230.

Verwey, W. (2001) Concatenating Familiar Movement Sequences: The Versatile Cognitive Processor. Acta Psychologica, 106, 69–95

Wymbs, N. D. Bassett, P. Mucha, M. Porter, and S. Grafton (2012). Differential Recruitment of the Sensorimotor Putamen and Frontoparietal Cortex During Motor Chunking in Humans. Neuron, 74(5), 936-946.