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 Eva Backhaus. Additional commentaries will appear the rest of the week.

If perception is a good source for knowledge of what is in front of our nose, memory is a good source of knowledge of what was in front of our nose yesterday, last year or some 30 years ago. Even though we know by experience and a whole bulk of psychological research that our memory might not be as reliable as we think, the capacity to look into our past is an important part of our self-understanding and crucial for many practices.

The reason why memory can tell us something about the past, in a way that imagination cannot, is that we assume a special causal connection between what we saw or heard or felt yesterday and what we remember now. What I experienced at the time necessarily came from a first person perspective, or what is sometimes called a field view. Hence, if I remember an episode from the past, it should be in field view too. Yet, introspection and testimony by others reveal that many memories do not follow this principle. In remembering we may see ourselves from the outside, from an observer perspective, or even from a sort or disembodied perspective like a bird’s-eye view. These memories are not only from a different perspective than in the original situation but include additions and subtractions, e.g. ourselves in the observer perspective. Even though these memories seem distorted and hence problematic regarding their veracity, they are pretty common, which makes it obvious why philosophers try to get a theoretical grip on them (Peeters et al, 2023, 168; Arcangeli, 2023, 146; McCarroll, 2023, 35).

The chapters in section two of Berninger’s and Vendrell’s Philosophical Perspectives on Memory and Imagination each deal with the philosophical problems around memory, perspective and imagination. This commentary focuses specifically on two of those papers.

In “Constructing a Wider View on Memory,” Peeters, Cosentino and Werning put pressure on the idea that there are only two or three perspectives to consider in memory (and imagination). They claim that field and observer perspective are only two points in a continuous spectrum that is, at the same time, more complex. An important observation that serves as a kind of obvious starting point is that observer memories still feel like our memories and not as if we were merely bystanders. We could, for example, see ourselves fall on our knees from the outside but still remember our intense pain and helplessness from the inside. Peeters et al claim that the embodied sense of ownership stems in that case from the agential and emotional aspects of memory. All our memories vary in at least these three aspects (and a fourth, the social aspect) and this explains their different phenomenology and makes them suitable for different purposes. Peeters et al draw on psychological research identifying rules of thumb about the correlation between perspective and purpose of memories, e.g. Nigro and Neisser (1983) have shown that situations where one is highly self-conscious tend to be remembered from an observer perspective.

Peeters’ et al scenario construction account moves in the direction of McCarroll’s simulationist account since they believe that memories are constructed – and in that sense continuous with imagination. Contra McCarroll (2023) they do not believe that perspective is encoded in memory which is supposed to be in line with recent psychological research. Their scenario construction view holds, instead, that perspective is a result of mechanisms in retrieval of memory and depends crucially on the aforementioned social, agential, emotional and visuospatial dimensions. Leaving perspective out of the picture at the stage of encoding puts all perspectives in memory on a par regarding their claim on veracity (Peeters et al, 2023, 179).

This puts them in sharp opposition to preservationists like Bernecker (2010) who claim that memory is the preservation of representational content. Yet, they agree with the preservationists in that memory is not independent of causal traces, since the causal chain is an important part in the truth-approximating process of memory. Memory, in short, is imagination that comes with the right causal tag.

What, then, is the trace? The relevant trace is called minimal trace and this is not a misnomer for it is only the bare skeleton of a memory, like the succession of events, that needs to be connected to the original experience via a direct or indirect causal link. This, however, seems a little overoptimistic with respect to how much work a causal chain can do. It is well known from action theory and theory of perception that causal chains are tricky, the argument in a nutshell being that there are no independent criteria as to what counts as the right causal chain. Obviously, the right causal chain results in memory, action or perception. But this begs the question. If we cannot give an independent account of what exactly the right causal connection is and does, to assume just any right connection seems like a desperate move in order to anchor memories firmly in the past.

In her paper “Imagining in Remembering from the Outside,” Arcangeli thinks about similar questions surrounding perspective but with a focus on imagination. Her point of departure is the following argument (Arcangeli, 2023, 147) which suggests that the problems around observer memories stem from the mental imagery involved episodic memory:

h1.  Points of view in visual mental images are always occupied.

h2.  Episodic memories cannot involve visual points of view occupied by selves other than the rememberer’s self.

h3.  Observer memories involve the rememberer’s self as an imagined object and cannot involve it also as the origin of the relevant visual point of view.

c. Observer memories must be false episodic memories.

If recreativism is right about the claim that episodic memories recreate non-imaginative states like hearing or seeing it may be forced to accept the conclusion that observer memories are not genuine memories. Since Arcangeli would like to grant observer memories the status of memories and thinks that recreativism is basically right, she develops two lines of argument against the aforementioned conclusion. She basically challenges the idea that (1) mental imagery necessarily involves the experience of seeing, hearing, etc. and suggests (2) that even if we occupy a point of view in mental imagery (like in the observer perspective) we can do so in a less than “full-blooded” manner (Arcangeli, 2023, 147).

No one doubts that sensory imagination is a very common type of imagination. Still, Arcangeli warns us against using mental imagery and sensory imagination as synonyms. She applies her well-established difference between mental imagery as an attitude and mental imagery as a type of content (Arcangeli, 2020) to the problem of perspective. Mental imagery in the attitudinal sense is equivalent to sensory, i.e. perception-like, imagination. Mental imagery in the content sense is a sensory content that is sufficiently rich, and that can be grasped via different mental attitudes (e.g. memory, imagination, desire). Very often these two types of mental imagery work together, e.g. if we imagine the smell of a lilac or the taste of a risotto. Her main point is that only visual imagination in the attitudinal sense necessarily involves imagining sensory experiences, e.g. seeing or hearing. If we think, instead, of mental imagery as a certain kind of content, there is no reason to grasp this content in a sensory- or perception-like way. Keeping the two senses of mental imagery nicely separated allows us to understand how episodic memories involving mental imagery can be unlike perception. And this makes way for understanding how observer memories do not necessarily involve an occupied point of view.

However, even if it is possible that observer memories do not involve the experience of seeing, many do. This is why Arcangeli develops a second argument that proposes the idea that a point of view in imagination (and hence in memory) can be less than fully occupied. Even though there are cases where we imagine that we ourselves see or hear something Arcangeli denies that this is always the case. In imagination or memory, but not in perception, it may be that it is not us experiencing something but a thin self instead, as opposed to the thick selves we are. She writes: “such an imagined viewer can be interpreted in a loose way without any commitment to the idea that it should coincide with either the imaginer’s self or with any other thick self. The visual perspective involved in the imaginative projection can remain thin. The ensuing account is perfectly compatible with positive views on observer memories, which do not see them as distorted memories.” (Arcangeli, 2023 155) What, then, is this so called thin self? Arcangeli argues that the relevant thin self is unspecified, which means that its traits, personality, history, body are not specified. The thin self may even lack all traits and thus becoming a virtual perspective. A virtual perspective is a mere counterfactual or hypothetical viewpoint, a viewpoint only a thin self can occupy but certainly not the imaginer’s self or any other thick self.

Both accounts argue for a continuity of memory and imagination – now the fact that memory and imagination are so closely related also means that they share a common problem: perspective. Arcangeli’s account helps us understand how imagination does not inherit problems of perspective from perception since imagination is freer than other mental capacities, especially insofar as we are not committed to a single perspective and especially not to our own thick perspective. And Peeters et al refer to the wide variety of memory and imagination that need careful theoretical treatment in order to avoid unwanted conclusions regarding observer memories. It is notable that both Peeters et al and Arcangeli solve the respective problems around perspective by mentioning the freedom of and variety in imagination. On a more personal note, it is exactly the heterogeneity of imagination and its many theoretical surprises that keep me fascinated.

References

Arcangeli, M. 2023. Imagining in Remembering from the Outside. In Philosophical Perspectives on Memory and Imagination, ed. A. Berninger and I. Vendrell Ferran, 146-164. New York: Routledge.

Arcangeli, M. 2020. The Two Faces of Mental Imagery. Philosophy and Phenomenological Research 101(2): 304–22.

Bernecker, S. 2010. Memory: A Philosophical Study. Oxford: Oxford University Press.

McCarroll, M. 2023. Memory and Imagination, Minds and Worlds. In Philosophical Perspectives on Memory and Imagination, ed. A. Berninger and I. Vendrell Ferran, 35-53. New York: Routledge.

Nigro, G., and U. Neisser. 1983. Point of View in Personal Memory. Cognitive Psychology 15(4): 467–82.

Peeters, A.; Cosentino, E.; Werning, M. 2023. Constructing a Wider View on Memory: Beyond the Dichotomy of Field and Observer Persepectives. In Philosophical Perspectives on Memory and Imagination, ed. A. Berninger and I. Vendrell Ferran, 165-190. New York: Routledge.