Still Life, a Mirror: Phasic memory and re-encounters with artworks

Re-encountering certain kinds of artworks in the present (re-listening to music, re-reading novels) can often occasion a kind of recollection akin to episodic recollection, but which may be better cast as ‘phasic’, at least insofar as one can be said to remember ‘what it was like’ to be oneself at some earlier stage or phase in one’s personal history. The kinds of works that prompt such recollection, I call ‘still lives’ - they are limited wholes whose formal properties are stable over time. In the first part of the paper, I spell out a way of making sense of the peculiar power that certain artworks have to occasion such recollection – it is, as I explain, a power or ductus that derives from the form of the artwork, though possession of such a power is not limited to art. I then detail three dimensions along which episodic recollection and phasic recollection as occasioned by re-encountering ‘still lives’ differ: metaphysical, phenomenological, and descriptive. In the second half, I explore a challenge for my account of phasic recollection, which in turn helps make more vivid my proposal as well as the spectral analogy at the heart of it: Just as one can see regions behind one by looking in the direction of a mirror located in the same space in which one is, sometimes by re-encountering certain kinds of artworks now, past intervals or phases ‘behind one’ can be ‘made present’ in a way that the paper aims to make plain. I also explain to what extent phasic recollection might be understood as a form of mental time travel, and what the attendant phenomenology of ‘transportation’ involves.


The Datum
I defend two thoughts in relation to this phenomenonone negative and one positive.
Negatively, I suggest that some memories occasioned by re-encountering artworks are not best cast as episodic. Nelson (1993, p.7) tells us that "an episodic memory has the phenomenal characteristic of referring to something that happened once at a specific time and place". For Conway (2008, p.19), episodic memories are "specific momentfragments". Robins (2017, p.77) has it that they are "for one-off experiences". But Odette loved Swann for weeks; what is recalled is a season of love. I will suggest that sometimes re-encounters with artworks occasion what I will call phasic memorythat is, they involve the recollection of past phases in one's personal history.
Positively, I propose that phasic recollection as occasioned by re-encounters with artworks depends on the perceiver's being perceptually related to a formal object, the form and nature of which is partly constitutive of the recollection, as well as partly explanatory of our capacity to notice our undergoing any such episodes. What is meant by "constitution" is something to be spelt out later on (and incompletely). I get at the idea roughly by way of my peculiar title which is intended to invite a loose analogy with perception with a mirror: When looking at a mirror in the same space in which one is located, regions behind one can be seen. In this respect, location and direction are cleaved apart; things located behind one can be seen by looking in the direction of the mirror and one's spatial perspective is augmented. 1 Analogously, sometimes on reencountering artworks, past phases in one's personal history can become 'present' to one again; there is a sense in which they overlay and thus come to augment the present. However, as I propose, this mode of augmentation is not best cast as a form of mental time travel, at least if by that it is assumed that the same kinds of mental events and occurrences are involved in such recollection as are involved in imagining the future or counterfactual possibility, a thought that has been recently cast as metaphysical continuism, imagination and memory being supposed continuous in mental kind (See Michaelian et al. forthcoming, Bernecker andMichaelian 2017, andMichaelian 2016). Rather, because episodes of phasic recollection as occasioned by re-encounters with artworks depend on the presence of the artwork, the experience that such recollection involves could not be of the same kind as the experience that underlies mere imagining of the future (say). At the same time, insofar as what is sometimes occasioned when one undergoes phasic recollection is, as I will show, a way of occupying and relating to the space one currently occupies -albeit in a way that is characteristic of a past phase of one's personal history -such recollection can nonetheless serve to orient future action and movement in actual, non-imagined space and across time. To this extent, phasic recollection can involve a certain sort of Janus-headed temporal orientation, though it needn't always.
The paper unfolds as follows. First, I set a constraint on which artworks can be reencountered and in virtue of what they are re-encounterable. I introduce a concept now mostly lost in contemporary discussion: the medieval notion of ductus. I explain in what sense the ductus of a work is or can be conceived of as a kind of power to conduct patterns of perception, thought, imagination, and feeling ( §2). In §3 I defend the negative claim by distinguishing episodic and phasic memory along three dimensions: roughly, metaphysical, phenomenological, and descriptive. I develop the positive claim in §4 by canvassing a candidate objection and in §5 I sketch the peculiar ontology of experience that the title of the paper hints at. I close by saying something about the phenomenology of transportation that such experiences involve.

Still Life
I will assume that at least some artworks can be re-encountered -that one can re-listen to a song, re-read the same novel. This is not to endorse a so-called 'lump' view of the ontology of artworks however, one that identifies artworks with material objects. Instead I pick up, as something like a constraint, a thought from Iris Murdoch. Murdoch writes that if I stand in front of a painting, and "nothing happens" then "something has gone wrong" (Murdoch 1992, p.3). I think we can acknowledge both this obscure reality and the thought that, in the main, most works of art, irrespective of medium or genre, require material objects and processes to keep them, as Murdoch puts it, 'continuously available'. That we can re-encounter such material objects and processes ensures that we can also re-encounter the artworks that they, by various means, as she puts it, 'body forth' (Murdoch 1992, p.2).
In broad strokes, I will assume that what is re-encountered is some kind of stable structure or form which is such that it can be 'bodied forth' through space and time. Many artworks are special in this sense. For instance, their size or shape (pictures), their ontology (musical works, novels) and, importantly, their normative status (we don't typically tamper with artworksfor example, we don't as a matter of course reuse the material that makes them available) -ensures that they can easily move through space and time, or be realised in different spaces and times, mostly without loss. While the world around them changes, they do not. Accordingly, I will call such works 'still lives'. Still lives have a stable identity.
Yet that still lives have a stable identity does not mean that they are essentially unchanging. Some 'still lives' maintain their identity over time despite change; if, for instance, time is part of the content of the work, or if the material that 'bodies forth the work' is not merely material but vital, as in a garden. It follows that works that are timespecific are not 'still lives' in the sense I am envisaging -for example, highly ephemeral landart or works such as Roman Singer's time-structures are not 'still lives'. Such works cannot be re-encountered at a distant later time because although their form may have been stable at a time, it did not persist over and through time. 2 I focus entirely on artworks that are 'still lives', though I consider that phasic memory on re-encountering stable form or structure is also occasioned by many other stable forms that were repetitively engaged with at a distant interval in one's past personal historyfor instance, the childhood home, places that one used to frequent as a teenager, perhaps even a well-known forest at a certain time of year. One reason for this limited focus is that artworks, because of their normative status, are more likely to survive with their form in-tact (literally: not-touched). Likewise, those that are not sitespecific can be moved through space, which makes re-encountering them not only at a different time but place possible.
A second reason for this limitation arises from the way in which we tend to approach art. We approach it, as the Kantian aesthetician might put it, disinterestedly. 3 This doesn't mean that we ought not to approach art inquisitively or even without interest or worse inattentively, the word 'disinterested' notwithstanding. It means only that when one contemplates an artwork as an artwork this partly involves treating it and one's experience of it non-instrumentally. This might be easily disputed, of course. Perhaps a teenager might feign musical tastes that are apt to make him seem 'cool' to his peers. However the idea I am after is closer to one nicely expressed by Lucy O'Brien. Artworks afford what she describes as "practical rest" (O'Brien 2017, p.149). We don't typically act on them or with them. And that we approach them thus disinterestedly, not using them instrumentally, not acting on or with them, means that we are more apt to notice the effect they have on us. And that artworks are designed to be contemplated and to reward attention means that the possibility of our becoming aware of our being thus affected is amplifiedand can sometimes even be savoured.
My focus on artwork is heuristic then. I think the phenomenon of phasic memory is more easily described by appeal to art and perhaps, for the same reasons, is most familiar in the context of such re-encounters. However, I also think that peculiar ontology and metaphysics of phasic memory is best brought into view by these cases and I hope the paper is proof of that. It strikes me that only once this is in sight can questions concerning the normative status of the occasioning structure be fruitfully framed. For instance, if episodes of phasic recollection are valuableand they may be so for many reasons that I don't even broacha perceiver that does not, or cannot, reencounter stable forms, including 'still lives', is missing out in various respects. Questions concerning elitism, aesthetic injustice, cultural appropriation and memorialisation are nearby here, but I won't bring my discussion into conversation with any of these themes.
What then is special about form and, given my heuristic, artistic form in particular? I will assume that 'still lives' possess a certain power in virtue of their form, which I illuminate or at least articulate by appeal to a medieval conception of form, now lost or forgotten: the notion of a work's having a ductus. I don't defend this notionin fact, it is only a placeholder for the kind of conception of form that the argument requires. Nonetheless, appeal to ductus will allow me to capture why works that have a ductus (or some such quality) are apt to support phasic recollection when re-encountered. Here I assume that the relevant artworks were previously repetitively engaged with over some remote interval before being retired, how remote being, I take it, an empirical matter.
So, what is the ductus of a work? Revisit the epigraph at the start of this essay, which comes from a French courtly treatise on rhetoric called Les Douze Dames de Rhétorique. Twelve rhetorical procedures are represented by distinct handmaids. The words quoted are spoken the eleventh, Deduccion loable or 'praiseworthy planning', who represents arrangement, or the ductus of a work. Medievalist Mary Carruthers (2010) glosses this as follows. The ductus of a work is: "the way by which a work leads someone through itself: that quality in a work's formal patterns which engages an audience and then sets a viewer or auditor or performer in motion within its structures" (my emphasis, 2009, p.190).
Etymologically, ductus was used as the past participle of ducere -'led'. In its nominal form, it means way or path, a directed movement or course, and names "a principle of movement not stasis, of process rather than product, the conduct of a thinking, listening and feeling mind on its way through a composition" (ibid., p.206).
One might wonder what the role of the artist is on this conception. Carruthers (2010) explains: For Cicero, the consilium of the work is "the aggregate of the rational decisions and selections of the composer as s/he works" (De invention I.25.36, see Carruthers p.200). Consultus Fortunatianus has it that while "the consilium is a matter of [the author's] choice, ductus is part of the work itself. Likewise, in that the ductus develops from the consilium, not the consilium from the ductus" (Ars rhet. I. 8 (75.19-21). The ductus then is intrinsic to the work itself, even while it is relational insofar as it is a capacity to engage the viewer or auditor, to "set them in motion". It is intentionally produced.
I have implicitly cast the ductus of a work as a power. Powers can be characterized in terms of dispositions; for instance, it might be thought to be in the essence of a power to be disposed to bring about a certain effect without being identified with those dispositions (Groff 2017). When I use the notion of a power I mean it to encompass the looser and more intuitive notion of a disposition as a capacity or ability (Mumford 2009). Dispositions can be conceived as grounded or ungrounded. If certain disposition kinds are necessarily grounded, they are non-fundamental. I take it that the power of an artwork to 'lead' is uncontroversially non-fundamental. I suppose that the ductus of a work is at least partially grounded in its form and that this form must be at least perceptually accessible (Petit 1983). For instance, while novels are not perceived, their form is perceptually accessible. They need to be read or listened to. Many artforms involve sensuous perceptible form, but the ductus that a work possesses in virtue of its form need not be exhausted by perceptible form.
Naturally, a great deal more needs to be said about what ductus is or could be. For the purposes of this paper however I have introduced the idea with only a high level of generality. What matters for now is the simple idea that certain artworks possess an ability to conduct patterns of perception, imagination, feeling and thought, partly in virtue of their form, form which must be perceptually accessible (if not constituted by sensible form as many artworks are). This helps articulate less opaquely I think what Murdoch's Constraint is getting at. If one stands in front of a painting (say), and nothing happensthat is, you are not led anywhere, perceptually, emotionally, cognitively, or imaginatively by the workthen it is intelligible to say that something has gone wrong. Of course, something can be wrong with youyou may be distracted, tired, perhaps you can't make out what the work is about because your hermeneutic resources are not up to the task, or perhaps you are confronting the object at the wrong moment in history. But likely too something may be wrong with the work. Put simple, it may lack the relevant power or capacity to lead.
Importantly, this does not mean that works that do lead, that do possess a ductus, typically reproduce those patterns when re-encounteredviz. that they lead the perceiver in the same way on each occasion. Rather, that certain artworks possess a power or ductus in virtue of perceptually accessible form is compatible with the idea that there are countless ways to 'move through' a given work or to be 'set in motion' by its structures. Indeed, the ductus of the good artwork will typically have a kind of inexhaustibility which explains, in part, why one returns to it again and again. But not only that. Many artworks are composed not only to repay aesthetic attention, and so to sustain interest over some present interval but also to be re-encountered at a later time and so to be re-encounterable. And this too often shows up in the ductus and is a consideration which surely shapes the consilium, likely often tacitly. For instance, the great modernist composer Schoenberg, though supposing repetition crucial to the intelligibility of music, nonetheless reduced the amount of repetition in his work in light of the anticipated innovation that would mean that his work could be recorded and re-listened to. It is a familiar phenomenon that some music, when listened to 'on repeat', will soon, to borrow Hume's delicate phrasing, "pall the taste ". 4 In what follows I propose that only works that have a ductus can sustain phasic recollection. This explains why works that are designed to be re-encountered or repetitively engaged with are apt to induce it. Earlier I designated such works as 'still lives', mostly to highlight the fact that they tend to travel through space and time intact, partly due to their normative status. I can now make that notion more precise: A work of art, the formal features of which are stable over time, such that the same work can be re-encountered at a later time, where those stable featureswhich compromise its form -partly ground the ductus of the work and are at least perceptually accessible.
Two Caveats. Though I distinguish episodic memory from the kind of phasic recollection that I claim re-encountering artworks can sometimes occasion, I do not explore the relation between episodes and phases. It is likely that certain events do serve to demarcate phases and can stand as temporal landmarks in one's personal historycultural rites of passage and other personally temporally significant events like one's first kiss. 5 Second, that re-encounters with such artefacts can occasion phasic recollection does not yet mean that other structures do not -they can, though I think we are much less apt to notice them. 6 Nor does it mean that phasic memory does not also involve canonically episodic memories; it often does. Proust tells us that when Swann hears the violin rise to a series of high notes, he 'sees' again "the snowy, curled petals of the chrysanthemum which [Odette] had tossed after him into his carriage…."this is a one-off temporal fragment, a recollected episode. But Swann remembers too "all the network of mental habits, of seasonable impressions, of sensory reactions, which had extended over a series of weeks its uniform meshes, by which his body now found itself inextricably held" (Proust (1913(Proust ( /1956), Swann's Way, p. 496). Swann recalls a season of love which Vinteuil's sonata reflects. Later, I will explain why this latter italicized term is itself strangely reflective of the ontology and metaphysics of those phasic recollections that are occasioned by re-encountering 'still lives'.

Episodic versus Phasic Memory
Here I set out three considerations in favour of the negative claim -the idea that the memories occasioned by re-encountering artworks are often not standardly episodic but phasic. My method is to draw on fragments of testimony and descriptive phenomenology to buoy up my analysis, which I later draw together through the spectral analogy I make explicit in closing. I also make plain the respects in which phasic memory as occasioned by re-encounters with artworks both does and does not involve 'mental time travel'.

The Artwork Is Present
To begin, we might remark that in phasic recollection as occasioned by reencounters with artworks, the artwork is present; what is salient is that the work previously engaged with is here now. In her wonderful The Garden as an Art, Mara Miller uses the striking idiom of being 'face-to-face' with something: 5 For an account of the boundaries of episodic memories, see Williams et al (2008). 6 A reviewer wonders if phasic recollection can occur without it being triggered or occasioned by some currently-perceived structure. My analysis does not preclude this. Rather it identifies a core case (reencounters with artworks) and offers a way of spelling out the ontology of those phasic memories that are so occasioned.
"The work of art reveals itself to me….in encountering such a structure, which remains substantially identical with itself over time, I encounter myself. I come face to face with memories of myself as I was, and as I felt, in previous encounters. The sameness of the structure prompts recognition of the difference in the viewer. This is inevitably part of the experience of a work of art" (Miller 1993, p.112) The descriptive phenomenology that Miller offers in this passage might seem to invite an episodic reading. Even so, a key feature of her description marks off the mode of recollection that she describes from standard forms of episodic recollection. Her recollective experience depends on the temporal presence of the artworkshe is in the garden, now. Yet, as Matthew Soteriou notes: "episodic recollection does not strike one as being the kind of event that depends on the temporal presence of the object. Indeed, it is arguable that an aspect of the phenomenology of such an episode is the current absence of the event recollected" (Soteriou 2013, p.90) In the cases that we are considering, it is the temporal presence of the artwork, here and now, that is a salient aspect of the phenomenology of the experience. Call this Presence.
As it happens, Presence is a feature of the phenomenology of perception -it seems to us that we are co-located in space and time with the objects and events that we perceptually experience. Soteriou links this feature of our perceptual phenomenology to a further claim. Perceptual experience is temporally transparent to its objects. Here the idea is that the temporal location of the act of perceiving is the same as the temporal location of the objects of one's experience and this, moreover, is phenomenologically salient to us. When we introspectively reflect on our perceptual experience, the inception and duration of any episode so reflected on seems to depend on, and be determined by, the inception and duration of the events that one is aware of in having the experience. In the case of awareness of a 'still life' however, perhaps we can say more. It seems that the course of our experience, the way in which we are 'set in motion' by the work, seems to depend, partly, on the form of the work itself. Again, there is no implication that the same work reproduces the same experience now as at an earlier time, despite the perceiver being temporally transparent to it. Indeed, it is a common experience that works re-encountered after a remote interval though temporally transparent to the perceiver are often not experienced as historically transparent. As Patricia Meyer Spacks notes in her semi-autobiographical Rereading: "Reading a novel at the time of its publication, we may hardly register how it has been shaped by its specific location in time…because we share the author's historical moment, that location may remain largely invisible, as it does in daily life. Reading the same book fifteen or twenty years later, we sometimes perceive often with a sense of shockthat it belongs to a bygone era" (Meyer Spacks 2011, p.85).
To modify Spacks' way of framing things, sometimes re-encounters with artworks can make past temporal locations visible to one. It takes an imaginative effort to become aware of contingencies of one's present historical situation, though not impossible, and sometimes these become clearer as one recalls them in later times and re-encounters with artworks can facilitate this -you can see the atmosphere of an era at work in the piece. 7,8 This brings into view the second respect in which the form of recollection I want to capture differs from episodic memory, standardly cast.

Time-Travel and the Transportative Phenomenology of Phasic Memory
Presence foregrounds the thought that part of the phenomenology of the recollective experience undergone is its dependence on the artwork now re-encountered. Critically, however, what is recalled is often not or only described as episodic, at least on the assumption that episodic memory is for recall of one-off episodes (more on that assumption later). For instance, consider the passage below: "[Re-reading] is time travel, a reliable way to reawaken feelings sparked by a book at first encounter. George MacDonald Fraser's series of Flashman novels summons for me an early stretch at university, when I picked up one in a stranger's room, skimmed a paragraph, and realised with excitement and dread that my set-text reading plan would now implode. Nineteen Eighty-Four brings back a thrilling first sense of professional life and the daily commute, Orwell's novel finished while travelling across town for work experience at 15. Salinger's slim book of stories will forever be a ski-trip coach that smelled not unpleasantly of Chewits; Laughter in the Dark a summer spent dumped and misanthropic and grateful for Nabokov's mean wit." (Tom Lamont, writing in the Guardian, April 2012).
While the writer refers to one-off episodesskimming a paragraph, a coach-journeyhe also refers to phases or intervals: a stretch at university, the daily commute, a summer. But the idiom of 'time travel' seems to apply equally to these and it also appears that the recollective experience has some kind of transportative phenomenology that applies both to episodes and phases. Yet here what is relevant is not that a work to which one is not now historically transparent is re-encountered -Nabokov in particular might anyway be thought to have a temporally transcendent quality -but rather that past episodes and phases in one's personal history are now recalled. 7 Though I do not explore cultural, collective or shared remembering in this paper, there is a clear sense in which the concepts of ductus and orientation towards a past phase as I develop it later can be applied to groups. Given that artworks are public artefacts designed to be contemplated this should not be surprising. In particular, as a reviewer has noted, a clear application to Bradd Shore's fascinating work on family memory and its mediation is suggested. See especially Shore and Kauko (2017). 8 Thanks to Andy Hamilton for suggesting a way of framing this experience.
Of course, for some philosophers and psychologists of memory, the use of the phrase 'time travel' may be taken up as more than idiomatic. Notoriously, Endel Tulving has it that episodic memory is that form of memory that makes possible "mental time travel through subjective time, from the present to the past, thus allowing one to re-experience, through autonoetic awareness, one's own previous experiences" (Tulving 2002, p.5). Here 'autonoetic awareness' refers to conscious awareness that one personally experienced the event now recollected. 9 For Tulving then, episodic memory involves a kind of mental time travel insofar as it allows for the re-experiencing of past events personally witnessed. It hence allows for the reliving of those events. Now, perhaps as far as skimming a paragraph goes, such an analysis might well be granted. But with respect to phases, plainly described as part of the phenomenology, it might be wondered: how can phases be relived -how can one relive an entire summer, or an early stretch at university? The current paper is a part answer to that question but for now, I want only to highlight the respect in which Presence interacts with the important fact of the experience being occasioned by a re-encounter with an artwork. To do so, I consider the way Soteriou (2013) makes sense of the characterisation of recollective experience as involving 'reliving'.
Soteriou points out that the metaphor of 'reliving' might suggest that what is retained in memoryand is relived -is an occurrence, something with temporal parts. But this seems problematic. Something with duration, something with temporal extension, cannot be retained. So what must be retained, he proposes, is a stateknowledge of a past apprehension or "what it was like to apprehend the particular event now recollected" (p.173). This 'know-what' is not a kind of propositional knowledge however, but a special kind of 'know how', namely the ability to put oneself in an occurrent state of knowledge, the propositional knowledge that this is what it was like to undergo that eventa state that obtains in virtue of the phenomenally conscious mental episode of recollecting the event which manifests the recaller's knowledge of what it was like. Put more simply, the idea is that episodic memory involves a kind of capacity or ability to put oneself into a conscious sensuous state that manifests one's knowledge of what it was like to undergo a past event.
Yet suppose this is granted and suppose that such know-how can be flexibly expressed, as I think is plausible, 10 this might seem to place a constraint on the possibility of re-encountering an artwork in episodic memory. These two claims need to be spelt out.
There are two dimensions along which we can make sense of flexible expression. 11 First, the ability itself might be something that could be expressed in different ways at different times. For instance, sometimes an external performance 9 Malcolm (1963) and Shoemaker (1967) emphasise a distinction between 'personal' and 'factual' memories. 'Personal' memories, so understood, need not be solely episodic. It is worth noting that for Shoemaker (1976, p.266) personal memories are memories of events experienced, but this leaves it open whether one can have personal memory of a repeated event. I say a little more about spectrum accounts in fn.13 10 Soteriou does not argue for this claim, but I think it is compatible with his ontology and epistemology of mind. 11 For discussion of how occurrent episodic recollection may be conceived as an expression of knowledge rather than a source of knowledge see Hopkins 2018). might be necessary, as when the production of a physical picture is essential to the remembering (for instance, when a police artist assists a witness in remembering what a perpetrator looked like). Or we might think of the vehiclethe phenomenally conscious episode that manifests one's knowledge of what it was like -as capable of flexible expression. This is because such knowledge is essentially firstpersonal. So, the vehicle need only manifest one's knowledge of how things truly struck one then -viz. as 'like' this; different vehicles might manifest and thus be expressive of the same knowledge of 'what it was like'.
Imagine, for instance, that you remember what it was like on some occasion to see your friend wearing a striped, colourful skirt. The vehicle that manifests your knowledge of what it was like to see your friend then might express that knowledge by representing a skirt of some determinate stripyness and colour. But the same knowledge could be expressed at a later time by a different phenomenally conscious vehicle. It only struck you that her skirt was 'colourful' and 'stripy'.
However, suppose now it were possible to re-encounter an artwork through the re-living of past personal experience of the work. Surely it would have to be granted that the form of the phenomenal vehicle that manifests one's knowledge of what it was like to encounter that artwork ought not to be flexibly expressed. After all, certain aspects of what it is like to listen to Debussy's Clair de Lune ought not to be flexibly expressed if I am to be credited with remembering what the piece itself sounds like. 12 Now, prima facie, encounters with visual art may seem different in this respect. Debus (2007) considers a recollected experience of the seeing of a correctlypositioned Mondrian on a gallery wall, and a hypothesised memory of a seeing the same work up-side down. Both experiences have as their object the same stripy, colourful formal object, but they have different phenomenal vehicles. Might this suggest that recollections with different phenomenal vehicles can, after all, manifest knowledge of what a past encounter with an artwork was like? 13 Perhaps. But in the course of her exposition, Debus draws our attention to a visual analogue of the point made above with respect to Debussy and which is, I think, relevant to the difference I am trying to explain.
Try to remember an encounter with a painting. Notice that in remembering the painting, typically perspectival detail is elided. You might remember what the painting looked like, 14 but not what it was like to see the paintingwhere you were standing while looking at it, from which angle it was viewed and so on. Debus calls such memory experiences unmediatedthey are typically not consciously mediated by the memory of seeing the artwork, though the experience might still be autonoetic in Tulving's sense. But if this description is phenomenologically apposite, as I think it 12 As it happens, I think it is implausible to suppose that successively remembering what (aspects of) the symphony sounds like could ever be sufficient for a re-encounter with the artwork, leaving aside complicated matters concerning the ontology and essential nature of such kinds of musical works, as well as limitations on representation in working memory etc. 13 It is widely accepted that episodes of recollection can take an 'observer' rather than a 'field' perspective. Sutton (2010) characterises such differences in perspective as differences "of form rather than of content:", suggesting that in both cases "the same underlying (complex and distributed) representations can animate occurrent memories involving either perspective" (p.33). 14 For an account of what our knowledge of looks consists in see Martin (2010). is, it seems a dilemma is suggested. Why so? For if the recollection of a past personal encounter with an artwork can be flexibly expressed, it is not form-preserving. On the other hand, if it is unmediated and so is suitably form-preserving, the personal dimension is elided and it is arguably no longer apt to be characterised as episodicviz. as involving a phenomenology that can be adequately captured as the reliving of a past personal experience and hence, by Tulving's lights, as constituting a form of mental 'time travel '. 15 To remove the dilemma (if it is real), it seems Presence must be foregrounded. Presence teaches that in cases of conscious re-encounter with an artwork in the present, the artworkand its formal featuresis there, temporally transparent to the act of perceiving it. There is hence no need to reconstruct those features in memory. Since the experience is also one of conscious re-encounter, the firstpersonal dimension remains salient; you are perceptually related to the temporally transparent work. But if so -and the work is there, here and now -in what sense can the re-encounter engender a sense of 'living anew' or reliving at all? And how might the re-encounter support a transportative phenomenology that pertains not only to episodes but to phases?
Soteriou's proposal, only sketched above, suggests an angle. Part of Soteriou's argument for retained knowledge of acquaintance recognises, in reports of recollection, occasional ellipsis. Remembering 'Mary falling asleep' can be read propositionally: one remembers that Mary fell asleep, or episodically; one remembers seeing Mary falling asleep, where this latter, recall, can be glossed as remembering what it was like to see Mary falling asleep.
But, intuitively, a similar line of thinking might be applied to phases. For consider: What the writer remembers is what it used to be like to commute in those early days (it felt thrilling) or to read Nabokov alone that summer (he was heartbroken). Here the use of a habitual progressive aspect might be thought to gesture rather at an existential sense of reliving. That is, the journalist remembers what it was like to be him at those earlier phases in his personal history. Call the phenomenology of time-travel involved Existential Reliving: one experiences in some renewed way what it was like to be oneself at the earlier stage in one's personal history. What is re-lived is not a past experienceor even set of experiencesbut what it was like to live at that past phase, now. I assume (hope) that this characterisation has some phenomenological traction. 15 Music psychologist Elizabeth Margulis links this circle of concepts in a different way by. In her 2014 On Repeat, she compares remembering hearing a concert performance with the experience of being beset by a socalled 'earworm', 'the obstinate unfolding of an imagined line of music' (p.75): "If I remember a hearing a concert performance of Brahm's Second Symphony, the memory might include something about the hall… the perfume of someone in the row behind me…the expressive resonance of the piece. These recollections emerge jumbled together, without a clear temporal order…and they emerge in a flash, the memory occupying a duration far shorter than the duration occupied by the actual event. But if the second theme gets stuck in my head it is a completely different experience -I seem not to remember but rather to rehear the entire thing, note by note, in a clear temporal sequence, and over an amount of time that roughly matches the duration its actual performance might have had." Here reliving is associated with precisely the kind of memory that Soteriou found to be ontologically problematic. But since occurrences cannot be retained, this suggests that we ought not to think of such reliving as involving the retention of a past occurrence after all. It is, rather, as Marguilis urges, a form of sensory imagining.
It might at once be pointed out however that the ontology of many artworks and the conditions of their cultural production and consumption just invites a straightforward appeal to intervals or phases and nothing more needs to be said. Novels take time to read; the same overproduced 'hit' may be ubiquitous for a whole summer before disappearing. But this being so, it is no surprise that re-reading and re-listening will then be evocative of or prompt recollection of a phase, so called.
Of course this is true. Even so, it strikes me that simply noting this fact does not do justice either to Presence or to the phenomenology of Existential Reliving, which often seems to be a deeply resonant part of the phenomenology. As Bergland comments: "We all know from life experience that nostalgic songs from your past are like time capsules that have the power to take every sense in your body right back to a specific time in your life, as if it was yesterday. That rare one hit wonder or random song that you haven't heard in decades will instantly unlock a hermetically-sealed memory box and transport your mind, body, and brain back in time. All the smells, feelings, friends, romantic partners, etc. come rushing back in a tidal wave of vivid memories, as if you were there again in the flesh." The experience is as if there again 'in the flesh'.
The third consideration I want to raise helps home in on the significance of the descriptive appeal to phases, especially with respect to trying to make sense of what the relevant species of time travel might involve.

Phasic Memory and Orientation in Space
We can grant that to undergo episodic memory is to be oriented in time. Specifically, it involves being oriented with respect to times in time, times which are ordered in time and are hence particular. In his 1994 Past, Space, Self, John Campbell sorts this kind of orientation in time from a more primitive form of orientation: orientation with respect to phase. A hibernating animal is said to be oriented in this way; it knows when it is late spring, but it may be incapable of differentiating one autumn from another -"it simply has no use for the conception of a particular autumn, as opposed to the general idea of a season" (p.38).
In his exposition, Campbell can be fairly read I think as opposing orientation with respect to particular times to orientation with respect to phase. This latter is a kind of orientation in time characteristic of the forms of life of non-human animals while mature human temporal orientation is characteristically to particular past times. But might this apparent opposition mask a way of being oriented with respect to phase in the human case? I think it does. To see why, consider what orientation with respect to phase involves.
A creature that is oriented with respect to a certain phase may be disposed to do certain things during that phase. For instance, 'in Autumn, the red squirrel stores surplus food in gaps in tree trunks'. This sentence is an example of what Thompson (2004) calls a natural historical judgment. The natural historical judgement articulates what the red squirrel does or the form of life peculiar to that species. Such judgments are atemporal and non-empirical; they have a so-called life-form term in the grammatical subject -the red squirrel. Now, there are plainly ways of applying this idea to descriptions of the 'look' of the human form of life, though I can't do that in any substantive way in this paper. Still, the rough idea can be sketched. Such descriptions would also have life-form terms in the grammatical subject, though qualified now with respect to developmental phasee.g. 'the human infant tends to put whatever it finds into its mouth', 'the human adolescent sleeps for 8-10 hours daily' etc.. And we can imagine qualifications of manifold other sorts and of ramifying degrees of specificitysocial, cultural, historical and epochal. For example, we can conjure descriptions of the ways of filling out space of 'the Victorian householder', 'the Sixities Liverpudlian youth', 'the Left-Bank flaneur' etc. But on this understanding, notice, orientation in time with respect to phase is in fact a way of being oriented in and filling out space. I pick up the relevance of this in §5.
So far, we have sorted orientation with respect to past times and with respect to phases, noting that the latter involves a kind of spatial orientation or way of filling out space over time. But might there not be a hybrid form of temporal orientation which is a composite of both these other forms: namely orientation with respect to particular past phases? If so, then such an orientation would seem to involve both the ordering of the past phase in time (like orientation with respect to past particular times) while also being such that being oriented with respect to that past phase would also involve being oriented in space now but in a way that was characteristic of that past phase. It is striking, I think, that much descriptive phenomenology associated with re-encountering artworks evinces not only a sense of movement through time (backwards) but concomitantly of imaginative re-situation in a distant space. Consider this passage from Marya Schechtman. Of a reformed party-girl, now middle-aged, she writes that: "listening to music she once loved may momentarily transport her back to her favourite clubs, and she may even feel a certain wistful nostalgia for those morning commutes to work after a particularly compelling week-night party" (Schechtman 2001, p.101) The listener finds herself transported while listening to music she once loved now. She is not transported to one specific site however, but to the 'lay of the land' as it was for her then as we might put it. Figuratively, we might say that the present moment becomes a moment in a remote phase in her life, itself characterised by a way of filling out space (those morning commutes). In this way, the present sub-region of space that she now occupiesthere is music in her perceptual ambit and she is listening to itmay itself be experienced, again speaking figuratively, as a subregion of that distant space, which is thereby somehow made re-present to her in some attenuated sense. 16 Granted, this passage does not yet fully paint the idea of one's being oriented in space now with respect to a past phase but it does, I think, at least draw attention to the spatiality of the relevant memory. And this, I hope, is sufficiently convincing for now. It also helps introduce a final point which makes way for the positive thesis I detail in §4.
It is intuitive to think that the reformed party-girl's nostalgia is part of the experience of her re-encounter. But if the experience involved the mere reproduction of an earlier experience or earlier set of experiences, the experience of nostalgia would have to 'fall outwith' whatever it is we would choose to thereby circumscribe as the experience. Nostalgia was not part of the original experience presumably, on this view, now reproduced. 17 I pick up this thread of thinking a little later.

The Positive Thesis
I have now gestured at three respects in which episodic and phasic memory as occasioned by re-encountering artworks differ. As a caveat, it is worth saying that I am marking this distinction for dialectical purposes. It may turn out (as I suspect) that the episodic/phasic distinction is less relevant than the difference that the presence of a formal object makes to the occurrence of particular phasic memories. 18 To recap, these differences were broadly as follows: In episodic memory, the events recollected are absent; what is recalled is typically a one-off event that occurred in the past, an event which one personally witnessed, and which it might be supposed is in some sense relived through the conscious sensuous phenomenology that characterises the episode of recall; the event is known to be past and is experienced as ordered in time. In contrast, in recall as occasioned by re-encounters with artworks, the artwork is present; what is recalled is often a past phase in one's personal history and there is a phenomenological sense in which one relives what it was like to be oneself during that phase. Finally, in being oriented with respect to a past particular phase, the recall may also orient one with respect to a past space now (though I do not claim this phenomenology is pervasivewhether it is so, seems to be an unexplored empirical matter).
In the rest of the paper I develop my positive claimthe idea that episodes of phasic memory constitutively depend on the formal character of the temporally transparent artwork. In §2, I said something about the nature of this form. I suggested that 'still lives' are those that can ground phasic recollection and I proposed that such works have a peculiar power or ductus that is grounded in their form. Such works are both reencounterable and have the power to direct thought, feeling, imagination and perception. All going well then, encountering such works should satisfy Murdoch's 17 While this might be considered just as true for episodic as for phasic memories, it is tempting to surmise that nostalgia pertains largely to phasic and not episodic recollections, though this is an empirical supposition that I cannot explore here. At any rate, as I hope to show, it strikes me that the question as to the scope of the experience helps motivate reflection on its metaphysical structure, the task of the remaining part of the paper. 18 That this is so suggests that the episodic/phasic distinction could well be treated as something of a spectrum with the crucial difference in kind pertaining to whether Presence obtains or not in either case. Such an approach could be fruitfully brought into contact with number of studies that seek to locate continuities rather than cleavages between putative memory kindsfor instance, Rowlands (2017) explores treating the episodic/ semantic distinction as continuous; Neisser's 1981 notion of repisodic memory is memory not for one-off events but a repeated series; Renoult et al's recent account of personal semantic memory also allows for a continuum reading on which personal semantic memory is intermediate between episodic and general semantic memory. For dialectical reasons, I think it is useful to keep them apart in this paper. The question of whether what we have here are genuine psychological kinds strikes me as a bit premature. At the same time, the descriptive phenomenology does invite such a distinction, though there are important individuative questions nearbyfor instance, if the two are to be distinguished what is the relation between them? As noted earlier, I leave this question aside.
Constraint. But first I raise a candidate objection to the idea sketched so far by considering a context, already hinted at, in which the notion of phasic recollection already has a natural home -Marya Schechtman's notion of 'empathic access'. Attendantly, I highlight ways in which the phenomenon I isolate is richer than an alternative explanation might allow: the suggestion that what we have here remains a case of episodic memory, but that what is recollected is only a generic episode with particularity, as I will put it, bleached out, or at least made diffuse.
As noted, Schechtman's conception of 'empathic access' chimes with the datum I have so far been assembling through snippets of testimony. In cases of recollection as occasioned by re-encounters with artworks, the artwork is there, and the experience is temporally transparent to the artwork (Presence). Often what is recollected are not oneoff episodes, but seasons or past intervals. Accordingly, it is not (merely) the case that specific episodes are recalled and sensuously relived. Rather, in some sense one re-lives what it used to be like to be oneself at the earlier phase (Existential Reliving), where this appears to involve a spatial dimension, something which reflection on Schechtman's notion of 'empathic access' will, I hope, help bring out. So, what does 'empathic access' involve?
According to Schechtman (2001), 'empathic access'or some complex psychological construct that answers to that descriptionis a psychological feature that is overlooked in neo-Lockean psychological continuity accounts of personal identity. 19 On those accounts so long as a there is a sufficient number of appropriately caused psychological connections between some set of psychological states that a person or 'time-slice' of a person could be 'in' from one moment to the nextoften states of recollection -the person is the same. 20 In contrast, Schechtman emphasises a 'point of view', where what is at stake in disavowals of the person being the same is "the loss of a particular kind of access to one's past point of view" (p.19); and where, when such access is lost, an individual can in some sense "no longer remember what it was like to be as they were before" (my emphasis p.20). What kind of 'access' does she have in mind?

"[I]individuals
[that lack such access] may be able to tell you well enough what they did and said and thought and felt in their 'previous lives', but they are no longer capable of inhabiting the first-person point of view that they once inhabited. They cannot feel as they felt before or look at the world through the same eyes…[there is] an inability to inhabit the first-person perspective of the earlier person in a gestalt phenomenological sense. It is no longer possible to see the world as one once saw it or to feel about things the way one once felt." (pp.20-21). 19 She also criticises narrative accounts for a like neglect. Note her concern is not with metaphysical or literal identity and survival but figurative identity, though, as she notes, for the neo-Lockean this amounts to roughly the same thing. For a distinctive criticism see Hamilton (2013), particularly Chs 2 and 3. 20 Of course, it can happen that we 'change our mind', modify behaviours, acquire new beliefs and so on. In such cases, we might be inclined to say that we have changed, and even say things like 'I no longer recognise that person'. For the neo-Lockean, despite such testimony, so long was the psychological binds from one moment to the next are appropriately yoked together, the same person survives.
For my purposes and, intriguingly, two central descriptive features of Schechtman's account line up with the position so far sketched in this paper. First, the species of remembering that Schechtman appeals to is plainly phasic. She writes that empathic access requires "that a person retain some sympathy for the psychological features of the life phase to which she retains access" (p.109), and that "remembering a past lifephase seems essential to being the same person who experienced that phase" (pp.102). Second, though not a theoretical feature of her account, she recognises that such access is often occasioned by re-encountering 'still lives', something I flagged up earlier with the case of the reformed party girl. Keeping these points in mind, I now want to consider a candidate challenge which I extrapolate from a reading of a 2014 paper by Christoph Hoerl. Shortly, I will bring Hoerl into conversation with Schechtman.
Hoerl takes up the datum that Debus (2007) also isolates, and which I mentioned earlier with her conception of unmediated recollection. Unlike Debus, however, Hoerl proposes that such experience should count as a distinct species of recollectionmemory for what x looks like. Here is the case he sketches (notably, again, the art status of the recollected object is not a conceptual feature of his proposal).

MONA LISA
During a spell living in Paris, I used to visit the Louvre to admire the Mona Lisa.
In casting my mind back to those occasions "I have a genuine episodic memory, even though my memory may not single out one of these visits amongst others by some unique feature. The phenomenology…..is [nonetheless] the same as that involved in recollecting a one-off event" (p.358). Now, while for Hoerl the phenomenology of the experience conjured up in so casting his mind back is said to be "the same" as recollecting a one-off event, he also acknowledges that there is in fact very little to distinguish his seeing the Mona Lisa from remembering what the Mona Lisa looks like. What marks the latter off from the former is autobiographical knowledge that he had spent the relevant spell living in Paris followed by periods of living elsewhere. But this knowledge, he says, "can provide for a way of making concrete to myself a sense in which what I remember when I recall those visits is something that will not occur again" (p.38). That is, it can be drawn upon in making concrete that the current experience is past insofar as it "mobilizes causal reasoning" about which events superseded which others in one's life. Accordingly, while he acknowledges a causal connection between, in this case, past episodes of apprehension and current experience, he denies that a mere causal connection between such states could ever be sufficient either to individuate the experience as one of recollection or to single out one particular past event. Besides, as he explains, visual imaginings are also casually related to past encounters 21 ; our imaginings draw on our past perceptions and are thereby also suitably causally related to those past encounters. What makes the experience a memory, however, is the fact that it is constellated in a wider pattern of autobiographical knowledge and what makes it episodic is the firstpersonal recognition of its pastness, due to this wider pattern, and this despite its generic character.
Hoerl's case is demanding, I think, since it suggests that appeal to phasic recollection, indeed that very designation, may just be superfluousepisodic memory can be construed as broad enough to make sense of phasic recollection, provided episodic memories are generic memories that are appropriately bound up with autobiographical knowledge. However, my concernsomething that my title reflects -is with such recollection as occasioned by re-encountering artworks and it should be plain that such experiences do differ. 22 The phenomenology is not the same as that of recollecting a one-off event or repeated series of events that is past; this is since the re-encounter is taking place now, and so does not involve appreciation that the experience is not something that will not occur again. Yet nor is it the case that what is remembered is what the artwork looked like or sounded like, a form of recollection which might seem to elide the first-personal dimension entirely. Instead, the first-personal dimension of the experience remains intactindeed it is highly salient, as the wistful nostalgia of Schechtman's reformed party-goer illustrates and reminds us. At the same time, for Schechtman, the first-personal dimension is far richer than that could be afforded by autobiographical knowledge, even when such knowledge is suitably integrated with generic sensory-perceptual episodes. And broadly, this is since the first-personal dimension is understood in terms of a point of view and not merely in terms of subjective sensory episodes which, though they may have perspectival content, only acquire first-personal content through their being integrated with autobiographical knowledge. I develop the import of this conception of a point of view, as I see it, in the next section. But with this end in mind, two features of Schechtman's treatment are worth foregrounding now.
Schechtman emphasises the plenitudinous character of the recollective experiences that she thinks empathic access involves. Such experiences are plenitidinous since they involve not just perceptual access to a generic past episode, but access to past feelings, patterns of thought and affect. What 'empathic access' enables is that such past feelings -and objectives inherent in past patterns of motivation over an interval -can be given weight in present deliberation now about future action. 23 The relevant 'access' then is not a wholly backward-looking phenomenon. How plenitudinous must the experience be? Well, submits Schechtman, we can hardly demand of the reformed party-goer that she recollect -"the thoughts, feelings and emotions of the entire decade of the party girl's wild daysthis is too many inner states to maintain even dispositionally. What is needed instead is a more modest connection to that era, together with enough empathic access to enable sympathetic phenomenological and behavioural representation of that era in the present" (my emphasis, Schechtman 2001, p.110) Call what Schechtman requires a Present Connection. There needs to be a connection to the past phase in the present.
The second feature of her account worth highlighting is as follows: a failure of empathic access does not appear to involve a failure of autobiographical memory at all, for such individuals "may be able to tell you well enough what they did and said and thought and felt in their "previous lives". Rather it involves an inability to occupy or inhabit "the first-person point of view that they once inhabited". What does this mean?
Schechtman doesn't say explicitly, but let us pick up a previous line of thinking. Earlier, I characterised the notion of a phase by appeal to different ways of filling out space and time. On this view, we might count among past personal phases those delimitated by the duration of, for instance, a romantic relationship, a job, a particular school, living in a certain neighbourhood and so on. And we might suppose all of these phases to be associated with (and partly individuated by) distinct nexi of relations of places and people of interest and concern. On this understanding, an ability to occupy or inhabit 'the first-person point of view once inhabited' might be an ability to orient oneself with respect to those distinct nexiand not merely in thought, and imagination, but in the space now currently occupies. But this was just the notion introduced earlier of orientation with respect to particular past phase. Now, it might well be wondered how this is possible, an empirical question surely. At the same time, it strikes me that the art status of the artefact re-encountered is not at all irrelevant, though it remains untheorized. And this is because the nexi of relations of places and people of interest and concern that are characteristic of phases typically involve artworks -works or types of work with which one repetitively engaged with over the phase, either by happenstance or deliberatively; many such works may even have been self-defining, particularly at an early age. 24 Such artefacts are what I have called 'still lives'while being associated with a phase they may transcend it, if not historically, then in the sense that they can move through space and time, in tact. And this confers on them a particular explanatory and mnemonic status. It could be objected that madeleines might well have been repetitively encountered insofar as they were frequently eaten. But they do not have the same status. Madeleines are not selfdefining. Generally, they are not baked to be contemplated, or to reward aesthetic attention, even if they could be. There is something special about 'still lives'. In the next section I make plain what.

Form and the Present Connection
In the last section, I suggested that the kind of recollection occasioned by reencountering artworks is not best thought of as memory for what the work of art sounded or looked like. This would anyway imply the current absence of the work, pace Presence. Further, the fact that the descriptive phenomenology points to Existential Reliving suggests that the first-personal dimension cannot be accommodated by autobiographical memory alonethe experience is, as it were, 'in the flesh'. At the same time, the phenomenon typically involves affective temporal emotions like nostalgia, something I have not said much about. I can now be brief: On an account where what is remembered is a generic episodic memory, feelings like nostalgia are not strictly part of the experience since they were not part of the 24 I think there is good reason to assume that Schechtman is implicitly thinking in these spatial and material terms, though I cannot give full elaboration or defence of this interpretation here. See instead Schechtman's development of the PLV (person life view) in her 2014 Staying Alive: Personal Identity, Practical Concerns and the Unity of a Life. particular experiences now bundled together and generically recalled. In contrast, on the positive account I now assemble there is no requirement at all that such feelings be conceived as 'outwith' the recollective experience as a whole. This is because rather than involving a generic memory of a set of past particular encounters, the particular present re-encounter, which involves a historical individual standing in a perceptual relation to a temporally transparent 'still life', is generic to a particular past phase or interval, namely the interval or phase over which the work was previously repetitively engaged with. But if so, one's current perceptual relation to the temporally transparent artwork can play the role of the Present Connection Schechtman envisages. I say a bit more about the nature of this Present Connection in the case of 'still lives' below. First a clarificatory point.
I introduced Schechtman's idiom of 'inhabitation' above to help elucidate what a point of view is. It is "a way of being in, and interacting with, the world" (ibid. p.23) where having access to a past point of view -being able to inhabit that past point of view -is to have access to the patterns of feeling, thought and objectives contained in motivations characteristic of that past phase. This is Schechtman's way of putting things and it chimes with Proust's characterisation of Swann's suite of recollections on re-encountering Vinteuil's sonata. He recalls "all the network of mental habits, of seasonable impressions, of sensory reactions, which had extended over a series of weeks its uniform meshes, by which his body now found itself inextricably held" (Proust (1913(Proust ( /1956), Swann's Way, p. 496). But I have also gestured towards more literal reading, specifically through the notion of being oriented with respect to phase, and specifically to a past particular phase. Since inhabiting a past point of view is not merely experiential but is tied to potential modes of acting, inhabiting a past point of view in the present can be said to be a way of being oriented in space now, albeit in a way that was characteristic of that past phase. I take it that this is a familiar, if undertheorised phenomenon (at least in analytic philosophy)crossing the threshold of a childhood home or revisiting an old haunt can sometimes occasion, even tempt, old ways of filling out space now, old patterns of behaviour. But homes and haunts are places that can be filled or moved through. Artworks typically can't, or are not supposed to be (there are exceptions). So, in what respect might past ways of filling out space be made available simply on re-encountering an artwork in the present? Hopefully by now I have assembled enough conceptual material to begin spelling out why.
The artwork and its perceptually accessible form is therethis is Presence. I have stipulated that works that induce phasic recollection are those that have a ductus, a power to conduct patterns of perception, thought, imagination and feeling. In reencountering an artwork, one is confronted anew with this power. To this extent, then, one comes to be in the same condition as one was then now, though it is not the case that the same experience is reproduced. Rather, in virtue of their reencounterable form, such works are apt to partially determine ways in which a perceptual relation to them is maintained and sustained over time (this among other things).
However, because the artwork now re-encountered once had a place in a nexus of places and people of interest and concern -'the lie of the land' as it was for one thenin being related to the artwork now, it seems one can spontaneously find oneself oriented with respect to that past space. And this seems particularly so of works of personal valuethe song the reformed party-girl once loved, the consolation of Nabokovalthough it seems that works that appeared in one's life at a phase by happenstance, simply by virtue of one's living through time at a certain place, can also have this effect. In either case, I suppose it is an empirical matter as to why and such exploration would likely be subpersonal, 25 but I am concerned rather with the phenomenology (which must surely be elucidated in step with, if not prior to, any such subpersonal enquiry) as well as with what I take to be a plausible metaphysical treatment of the experience. And, as I hope is now clear, the fact of the re-encounter's being with a formal object that was intentionally designed to be contemplated is crucial. I can now be explicit.
Phasic recollection seems to be a form of memory which Debus (2007) characterizes, though without theorizing, as constitutive. This is memory that is "composed of memorial and perceptual experiential elements" (p.193), where certain spatial aspects of the perceptual experience are constitutively part of the memory. It strikes me that the artwork and its formal features play this constitutive role in the case of phasic recollection, at least on re-encountering still lives. 26 How so?
If what I have argued is on the right track, in being perceptually related to the workin being 'face-to-face with it' (or the equivalent in the auditory case) -one is in the same condition now as one was at any earlier phase, namely the phase over which one previously repetitively engaged with the work, though there is no assumption that the same experience is reproduced. Rather the current particular encounter is generic to a past phase. But to this extent, there is at least potential for one's inhabitation of a past point of view; the particular artwork and one's perceptual relation to it to provides the Present Connection. Importantly, this connection is not causal. It involves the subject's conscious active awareness of a work of art, the ductus of which, grounded in perceptually accessible form, partly determines the way in which attention to it is sustained and modified over time. And this explains the existential dimension of Existential Reliving,. One's current experiencethe experience one is currently living throughis generic to past phase. Naturally, the intentional status of the 25 The hippocampus is known to play a role in route learning and spatial or topographical memory as well as in episodic memory, where episodic memory is understood to involve the encoding of where and when a particular event, say, (what) occurred (Hartley et al. 2014). For Eichenbaum (2017) however, though both space and time are processed in the hippocampus, they are represented by distinct neural network patterns. What and where need not, it seems, be wedded to a particular when. This invites a suggestion: supposing that the same what can occur in the same broad spatial region though at many different places (wheres) in that nexus and/or at many different specific whens over a remote interval, it is tempting to suppose that the specificity of the particular wheres constituting the nexus, as well as the particular whens of past engagement, can become transparent to the overall way that places are constellated over some interval. Tantalisingly, Tavares et al. (2015) argues that hippocampal neuronal populations can also code 'social space'networks of affiliation. In the case we are considering, the what is the perceptual event of an encounter with an artwork. If that event takes place repetitively over an interval and perhaps too at different places over that interval, a later re-encounter may prompt memories not specifically of particular past encounters, or particular places where encounters took place, though these may, of course, be recollected, but the region itself over an intervala past constellation of 'wheres', linked partly perhaps by filaments of personal value. 26 Colombetti (2017) develops a position on moods which has parallels with the current proposal. Moods are not merely influenced by events and situations but are 'scaffolded' and interrelated with the world in complex ways, even to the extent that they can 'experientially incorporate parts of the world'. She also explores ways in which moods can be regulated by artefacts that can be sought out to rekindle memories or stabilise selfidentity. Listening to music is identified as one such world-involving, mood-scaffolding regulatory activity. Of course such self-regulatory activity in the case of music involves re-listening. While often phasic memory occurs unbidden, it can also be sought out through active relistening to work that has long been retired. Indeed, this activity formed a happy part of the research for this paper! See also Krueger (2014). artwork plays a role here which standard madeleines cannot. 'Still lives' are designed to be contemplated and to reward attention. Their normative status further facilitates disinterested engagement, which is sustained over time. But this makes possible too a certain reflective awareness of the recollection occasioned. Importantly however, such reflective awareness and any associated feelingsnostalgia, for instanceneed not be conceived as falling outwith the experience. This is since, once a relationalist metaphysics of perception is in place, the experience can be understood to involve a historical subject standing in a perceptual relation to an artwork.
Granted, spelling all of this out adequately requires another paper, perhaps more than one, as does, relatedly, making fully explicit the explanatory role that a relationist metaphysics of perception plays in the current account (as well as the relation of ductus and form). Since I can't do that here, I close by setting out the spectral analogy that I think partly reflects this peculiar metaphysics and by saying something about the significance of this metaphysics for the mental time travel debate.
When one looks in the direction of a mirror in the same space in which one is, one can see regions behind one. 27 In the course of the paper, I have tried sketching an analogous position with respect to re-encountering 'still lives'. In re-encountering certain kinds of artworks nowat the same time at which one is -past intervals or phases in one's personal history can seem to be made present. In Proust's phrase, Vinteuil's sonata reflects a season of love; the past is reflected through one's current perceptual engagement with the powerful still-life, a work which once had a place in a past nexus of places and people of interest and concern; the 'lay of the land' as it was for one then. At the same time, since one's engagement with artwork is typically disinterested one can reflect, and even dwell upon, the feelings and thoughts engendered in one by and through one's engagement. But it is for this reason that our manner of relating to the work can sometimes also become the matter of our reflectiona mirror not only of a past phase, but of ourselves.
So, what lessons does the phasic recollection as occasioned by reencountering artworks have for theoristsphilosophers and psychologistsof mental time travel? I think it is plain that the phenomenon I am theorizing cross-cuts the debate between continuists and discontinuists with respect to imagination and memory. Imagery is part of phasic recollection (and their episodic components)both sensuous (the smell of chewits, the curled petals of Odette's chrysanthemums) and motor (the network of habits by which Swann's body is held)but imagery, it seems, is not fundamental. Strikingly, if phasic recollection is partly constituted by spatial or formal structure that is perceptually experienced, and the presence of that structure is necessary to its being occasioned and perhaps even being sustained, then perhaps phasic memory cannot occur in the absence of the subject's being perceptually related to that temporally transparent structure. 28 I have explored a phenomenon which without doubt does involve the phenomenology of temporal transportation, figuratively cast by one writer as 'time travel'. But, as I hope to have shown, this phenomenology is not best characterized in terms of recollection of a past event personally witnessed and now relived. My use of the word 'existential' is supposed to gesture rather at the respect in which the phenomenology of transportation ramifies beyond past time and place. It is not just a matter of the experiencer being 'taken back' to a distant time and place, as though the historical subject now undergoing the recall is the same as the previous witness. Instead, through the present encounter, the historical subject experiences or recalls, momentarily, what it was like to be herself at an earlier stage in her personal history. If anything, then, a difference between past and present selveshere I am speaking intuitively not theoretically -is made salient. But where remembering is effectively only imagining the past, it is not clear where this difference in the historical subject can be put.
Finally, as is perhaps dimly visible by now, if phasic recollection as I have cast it involves a mode of temporal orientation that has a spatial dimension that is configured by networks of earlier interests and concerns, however those nexi are to be empirically understood, there is a sense in which, as Schechtman wants to urge, it can be Janusheaded and have a future orientation too -and her reformed party girl can see the world, briefly, through unreformed eyes, and act on it too, unreformed. This suggests not only the transportative but the transformative power of art, and in this case of the temporally transparent still-life, when re-encountered. 29,30 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.