Situated Affects and Place Memory

Abstract

Traces of many past events are often layered or superposed, in brain, body, and world alike. This often poses challenges for individuals and groups, both in accessing specific past events and in regulating or managing coexisting emotions or attitudes. We sometimes struggle, for example, to find appropriate modes of engagement with places with complex and difficult pasts. More generally, there can appear to be a tension between what we know about the highly constructive nature of remembering, whether it is drawing on neural or worldly resources or both, and the ways that we need and use memory to make claims on the past, and to maintain some appropriate causal connections to past events. I assess the current state of work on situated affect and distributed memory, and the recent criticisms of the ‘dogma of harmony’ in these fields. I then deploy these frameworks to examine some affective dimensions of place memory, sketching a strongly distributed conception of places as sometimes partly constituting the processes and activities of feeling and remembering. These approaches also offer useful perspectives on the problems of how to engage – politically and aesthetically – with difficult pasts and historically burdened heritage. In assessing artistic interventions in troubled places, we can seek responsibly to do justice to the past while fully embracing the dynamic and contested constructedness of our present emotions, memories, and activities.

We understand history … through both evidence and affect. Memory is born from that most subjective of places: desire. Memory is love and hate; fire as warmth and fire as death. Memory is suffering and innocence; memory is the moan and weeping, and the sudden laughter.

Luke Stegemann, Amnesia Road: landscape, violence and memory (2021), p.185.

1 Resources for Constructing the Past

For creatures like us, the asymmetry of experienced time grounds key features of our cognitive and affective lives. We are not merely influenced by past events, but also sometimes take them as objects of thought, feeling, and social negotiation (Campbell 1994, 1997; Hoerl 1999; Sutton 2009a). History animates our minds and our activities in distinctive ways, at many timescales. As they come to mind in the dynamics of everyday cognitive and social life, we assign certain episodes to a source in past worldly experience, others to imagination (Mahr 2023), and we often try to make some partial sense of our present feelings, moods, or attitudes by comparing or connecting them with past emotions, by way of even minimal narratives or small stories (Goldie 2012; Hydén 2017; Trakas 2022; Fabry 2023). As has long been recognised in research on remembering in many fields, these are constructive processes – selection, abstraction, condensation, interference, consolidation, pattern-transformation, reconsolidation, generalization, and more. We do not store replicas or canonical versions of past events, and so must deploy resources in the present to create more or less stable, more or less adequate memories, often on the fly and iteratively in the ongoing whirl of bodily action and social interaction (Sutton 2009b; Michaelian 2011; Wagoner 2017; Wagoner et al. 2019).

The situated or distributed approaches to mind, memory, and emotion which have transformed the cognitive sciences from within over the last 30–40 years flesh out this picture by pointing to the rich, sometimes partly constitutive roles of heterogeneous non-neural resources in these processes of construction. Well-practised actions, trusted people, or familiar artifacts, for example, may transform or sculpt our cognitive and affective lives – what we think and remember and decide and do and feel, individually and together – both at a time and over time. This paper’s first focus or point of departure is the fascinating and productive recent shift in philosophical explorations of situated affectivity in which potentially negative or harmful effects of this cognitive and emotional openness come under intensified scrutiny, as we reject ‘the dogma of harmony’ to address ‘the dark side of niche construction’ and the dangers of ‘mind invasion’ (Aagaard 2021; Coninx 2023; Slaby 2016). Welcoming the new perspectives emerging within this line of work, I complicate them by applying them to domains of situated affectivity that are of independent interest and significance and that form the paper’s second point of departure. I examine place memory, in which (in some cases) the parts of the world that may be integrated into expanded processes and systems of remembering or feeling are neighbourhoods or regions, and the challenging problems of evaluating modes of engagement with difficult histories.

The present resources which participate in the ongoing construction of memories and feelings, often actively and iteratively at various timescales and levels, notably include ‘traces’ of past events. I use this familiar term very broadly. The primary focus is on traces as ‘representational’ rather than ‘non-representational’ resources, to use Heersmink’s terms for distinctive categories of cognitive artifact (Heersmink 2015, 2021), in that they carry some sort of information, and in some key cases information about the past events that produced or affected them. The paper’s third point of departure is the fact that traces of many past events are often layered together or ‘superposed’. This can occur, in distinctive ways, in the brain, in the body, or in the world¹.

In a moment I explain the notion of superposition in play here, and discuss its source in the neurocognitive sciences². But an initial grasp on how it also applies in the domains of place and environmentally situated affects comes from Kukla’s study of ‘how urban dwellers and urban spaces make one another’. Celebrating Berlin as a ‘repurposed city’ which does not freeze and fix its (troubled) past, in which vestiges of the past remain visible and open to reuse, Kukla writes (2021: 144):

Each layer of the landscape is polysemic; it is both a trace of multiple pasts and a structuring feature of the present. Berlin never gives the illusion of being able to show you its authentic or univocal history. Rather than putting its history on display, Berlin goes out of its way to let the ghosts of its past remain alive and visible.

Although Kukla is not engaging explicitly with cognitive theory or ideas about situated affect, I can use this striking description to introduce some key claims I explore here across domains, and which animate my treatment below of responses to places with difficult pasts. Firstly, the polysemy Kukla mentions is, in my terms, superposition in ecological action: many meanings are layered in the same material trace, and any one present place or memory may connect us to many past events³. Secondly, the idea of preserving or accessing an original or canonical version of the past is challenged here. As a result, thirdly, constructing something in the present is an ongoing activity: remembering or narrating or commemorating the past is – for individuals and communities alike – an achievement, and can be done well or badly, more or less responsibly.

In the connectionist networks employing distributed representation which significantly loosened the grip of classical cognitivism in the 1980s, traces are ‘stored’ only superpositionally, many across the same physical vehicles (McClelland and Rumelhart 1986: 193; Clark 1993: 17), with many ‘representations’ in or across one ‘representing’ system (Haugeland 1991; van Gelder 1991). Every enduring trace is a composite, out of which some specific memory, for example, is constructed at the time of recall (O’Brien 1991; Sutton 1998). Until that present process of construction, distinct traces have in a sense disappeared, persisting only implicitly (Elman 1993: 89). This notion of superposition provided the first of an expanding series of forms of ‘distribution’: for Andy Clark, memory and cognition were distributed firstly thus within individual neural networks; then doubly, across distinct but dynamically interacting brain systems; and then multiply, across brain, body, and world operating together (Clark 1997; Sutton 2015a). Here my attention is not primarily on the neural wings of the larger distributed systems that came to be the focus of the burgeoning alternative mainstream movements in cognitive science (Hutchins 1995; Sutton 2010; Newen et al. 2018), because I am examining exported or transported forms of superposition that operate beyond the brain. But I quickly note two points about this neurocognitive form of superposition which inform my treatment of situated affectivity.

Firstly, these notions of superposition and distributed representation are not uniquely tied to these particular connectionist and post-connectionist neural network models of cognitive processes. They operate at an abstract level, and can be identified in very different scientific and historical contexts, in which we can spot familiar concerns about the implications of the radically constructive visions of memory and mind that they suggest (Sutton 1998, 2020a). Secondly, looking forward from the ‘80s and ‘90s rather than back, recent innovations in deep learning are still animated by technical attempts to work with or work around the radically constructive nature of systems heavily reliant on superposition (McClelland et al. 2020; Shea 2023). They thereby gel in as yet under-recognized ways with pressing problems in contemporary cognitive neurophilosophy about how we can ever remember specific past events if memory is so thoroughly constructive. If the neural systems and mechanisms of remembering turn out to be identical with those of imagining, counterfactual thinking, and other forms of event simulation, perhaps we have to drop the idea that memory has some particular causal connection with the past events on which it makes a claim (Michaelian 2016; Addis 2018). This would be surprising and, perhaps, troubling, because we want to be realists about the past, no matter how fragile our access to it may be: it’s because we know we can be wrong about past events that we struggle so hard to be right, contesting and revising as we go (Craver 2020). There are a number of fronts in the debate about whether remembering is just one form of imagining or simulating events (Addis 2020; Michaelian 2016, 2022). It’s not yet clear that there really is a single, undifferentiated internal ‘constructive episodic simulation’ system (Andonovski, Sutton, & McCarroll forthcoming). And we can lean on liberalized conceptions of causal processes, which allow for the layered multiplicity of causal connections over time, without requiring linear and singular links between a single past event at encoding and a single current act of retrieval (Schechtman 1994; Andonovski 2020; Mac Cumhaill 2020; Sutton and O’Brien 2022). Realism about the past does not mean that truth is either simple or singular. But it’s hard to think and feel a way into such complicated relations between past and present, in which multiple coexisting traces may be entangled, in which we can acknowledge loss and selectivity and change over time without losing grip on the need to make, evaluate, contest, and remake claims on the past. These challenges in the philosophy of memory are not centre stage here, as I move on to consider situated affectivity and place: but in examining places and practices, artifacts and artworks that can viscerally unsettle or move us, touching our embodied and emotional being deeply, the intention is not to bypass or supplement existing cognitive theory about how we might construct the past well, but to provoke better (cf Anderson 2022).

To prepare the ground to return to these puzzles about superposition and the resources we use in constructing the past, I first address directly the current state of play in broadly situated and 4E approaches to affective technologies and ecologies. While embracing the complexities and critical perspectives highlighted by recent critics, I argue that attention to forms of disruption and affective imbalance among the components of such dynamic ecologies is natural within these approaches, when correctly understood, rather than a dramatic innovation. To put flesh on how this point applies in a relatively under-studied domain, I then discuss cognitive and affective ecologies of place. In a briefer final section, I consider the narrower topic of artistic engagements with difficult pasts, sketching a provisional aesthetics of superposition which falls out of the consideration of place as affective technology: I point to case studies that might feed directly back into cognitive theory, where artistic practices are already deeply informed by sustained reflection on memory and emotion, places and the past, on attempts to shape or wrangle or rework troubled pasts. The turn to art is not to apply cognitive theory, but actively to expand, disrupt, or refine it.

Before pushing ahead, I consolidate my concerns here and their connections. Established work on cognitive ecologies within the distributed cognition framework can encompass and effectively apply the critical turn in 4E cognitive theory and studies of affective technology. The case of place memory directly illustrates how the right kind of 4E theory naturally addresses negative and violent forms of cognitive and affective distribution, especially when we apply it to problems of historically burdened heritage that are of independent importance. Focus on the multiplicity and heterogeneity of resources across which mind and feeling may be distributed also highlights the significance of superposition, when many meanings or traces of past events are overlaid, in brain and world alike. Examination of the challenges posed by superposition and of how we do and should construct the past from multiple traces, as individuals and as communities, connects neurophilosophical concerns about memory’s faithfulness to the past with political and aesthetic concerns about apt modes of engagement with difficult pasts. The paper covers a lot of ground, and throughout I provide a range of references to help connect currently disconnected projects and fields.

2 Affective Ecologies and Distributed Disruption

As work on distributed and extended cognition matured, focussing less on revolutions in metaphysics, more on identifying neglected phenomena of rich and sustained interaction between brains, bodies, and worlds, and on transforming methods for the study of issues of independent interest, emotion became a natural target topic (Griffiths and Scarantino 2009; Greenwood 2013; Varga and Krueger 2013). I follow leading theorists in these fields in using ‘affectivity’ as a label to signal the broad scope of the approach, intended to apply to many affective phenomena, both occurrent and dispositional, from momentary emotion episodes and appraisals to sentiments, temperaments, character traits, and moods (Colombetti and Roberts 2015; Candiotto 2016; Slaby 2016). In the explosion of outstanding work on situated affectivity that has followed, making this perhaps the most productive application of a broadly distributed approach, we have seen increasingly precise and refined differentiation in treating the disparate integrated resources in question – a great variety of artifacts and technologies, other people, practices, environments, collectives – and in pinning down ways to study them (Krueger and Osler 2019; León et al. 2019; Chung et al. 2024). In stressing the heterogeneity of the resources in question, each of which has its own history, format, and dynamics, much of this work highlights the complementary relations between disparate but meshing components of expanded systems which have, as a result, new or transformed affective qualities (Krueger 2014: 538; Colombetti and Krueger 2015). This broadly ‘second-wave’ approach to distributed affectivity encourages us to pick out particular dimensions of interaction between elements of wider affective ecologies as frameworks within which to locate specific case studies (cf. Sutton et al. 2010; Heersmink 2021). It also ensures that study of distributed affectivity cannot become encapsulated or cut off from other cognitive and socio-emotional domains: attention to the development and entrenchment of practices or resources within a larger affective ecosystem also confirms the essential roles of embodied memory, active self-scaffolding, social relations, and communication in establishing and maintaining reliable emotion-regulation systems over time (Sutton 2018).

Alongside such case studies in situated affectivity, we can now enjoy, evaluate, and apply more precise evaluation of key descriptive concepts such as affective artifact, affective scaffold, affective niche, affective milieu, atmosphere, affective incorporation, and affective ecology. Work on ‘affective artifacts’, for example, effectively connects ideas about situated emotions to questions about personal identity and the ‘distributed self’ (Heersmink 2018; Piredda 2020)⁴. The fact that it’s becoming impossible to keep track of all the literature on these concepts even within philosophy alone confirms that we have entered a wonderfully fruitful phase in which ideas from quite different areas, from philosophy of biology to phenomenology and beyond, are feeding research on distributed cognition and affect, which is in turn offering a new lens on these other fields. This is exactly what early theorists dreamed of when struggling to establish credibility for the idea that cognitive and affective processes were not entirely ‘brain-bound’. Nearly 25 years ago, Clark looked forward to a ‘spectacular’ future pay-off from the study of ‘the complex and iterated interactions’ between brains and technologies: ‘nothing less than a new kind of cognitive scientific collaboration involving neuroscience, physiology, and social, cultural, and technological studies in about equal measure’ (Clark 2001: 154). That Clark did not explicitly include politics in this wish list is notable now, after a decade in which a more critical and political philosophy of mind centrally focused on norms and normativity has gradually emerged and been effectively implemented (Protevi 2009; Slaby 2016; Maiese and Hanna 2019). I am delighted at and engaged with the variety and quality of critiques and especially case studies of ‘callous design’, ‘oppressive things’, ‘hostile scaffolding’, and more, currently arising from a broadly situated perspective, and especially at their increasingly sophisticated incorporation of evidence from ethnographies and social theory (Rosenberger 2017; Liao and Huebner 2021; Meissner and Huebner 2022; Spurrett and Brancazio 2023). The couple of complicating notes that follow should be read in this light.

In suggesting that distributed cognition was always attuned to the possibility of oppressive disruption at the site of human-technology interfaces, and that attention to more or less smooth coupling between agents and artifacts never settled into such a blinkered form of boosterism as to deserve to be called a ‘dogma’ (Aagaard 2021), I can briefly address historical and thematic issues in turn. In terms of recent research history, first, as well as excavating the exchanges between Clark and Protevi which underlined that the same socio-historically constructed scaffolding processes could at least as easily have politically significant negative effects (e.g. Clark 2005: 257; Protevi 2009: 29), it would also be worth going further into the prehistory of situated and distributed cognition in education, anthropology, science studies, media, and sociology, where I suspect we’d find recognizable distributed-cognition themes driving ethnographies and case studies of highly contested or conflictual agent-artifact interactions (Michaelian and Sutton 2013). But I also direct philosophers to flourishing research programs in the history of distributed cognition, which have come to fully infiltrate mainstream scholarship in fields from ancient philosophy of technology to early modern literary culture, as demonstrated in the remarkable four-volume collection A History of Distributed Cognition (2019), edited by Miranda Anderson, Douglas Cairns, Mark Sprevak and Mike Wheeler. In the case of early modern affective technologies, the work of Evelyn Tribble in testing, revising, and expanding distributed approaches to emotion and memory across a range of early modern cultural-cognitive practices including theatre, education, religion, design, and embodied skills (Tribble 2005a, 2017a; Tribble and Keene 2011) has also been criticized for privileging smoothness and integration across disparate components of cognitive ecosystems while neglecting moments of failure, contingency, noise, or friction, for failing ‘to contemplate disharmony or resistance’ (Mazzola 2023: 12). In fact Tribble has arguably always highlighted tough affective effort and intense emotional interaction in concrete settings then and now (Tribble 2005b, 2017b, 2022; on affective technologies in early modern theatre see also Mullaney 2007; Rzepka 2015).

To return to the thematic or conceptual issue, I accept that some leading advocates for an extended or distributed approach to mind, memory, or emotion also sometimes revealed an optimistic technophilia, thus eliciting criticism of the ‘many fancy formulations’ that served sometimes to keep theoretical projects clear of troubling political issues (Slaby 2016: 5). But I deny that any such blindness, or any motivated neglect of disruption, contradiction, and conflict was (or is) intrinsic to distributed cognition as a framework.

As often, it depends on where you were looking or seeking your (constructive or critical) inspirations. For me, the happy benefits of pre-formed individual users’ smooth coping in manipulating artifacts created for their possession and use, their profit and pleasure were never at the heart of distributed cognition, as critics allege (Slaby 2016; Williams 2016; Aagaard 2021). Firstly, minds soak in, rather than developing as autonomous users with clear instrumental needs: the label ‘extended’ cognition has become less helpful because it encourages the misconception that mind is first in the head, and only then spreads outwards to colonize things. On the actual distributed approach, in contrast, a broadly Vygotskyan take on cognitive and affective development treats remembering and feeling as capacities we learn over the course of gradual enculturation within specific socio-affective contexts (Miller et al. 1990, 2014; Menary 2007; Wang 2013), such that (relational) autonomy is an ongoing achievement, fallibly forged in (rather than before or behind) the weave of our worldly lives (Sutton 2010: 213). And more broadly, once we start thinking in terms of the cognitive ecologies within which memory and affect are situated, it is obvious that we are dealing with vast and uneven arrays of disparate resources, many or most of which are beyond any individual’s immediate control, no matter how much they have been shaped by human culture and history. In that the ecological approach directs us to notice and track shifting balances among the heterogeneous resources in any interactive cognitive ecosystem (Hutchins 2010), it offers a new perspective on our cognitive and affective vulnerability. On the one hand, we are by nature interdependent and thus intrinsically vulnerable to changes in other parts of our wider affective ecologies (Mackenzie et al. 2014). On the other hand, differential vulnerability in the form of unequal access to the resources that permit flexible cognitive and affective expansion or adjustment is one of the subtler mechanisms of power. As Krueger and Salice put it (2021), it is the relational nature of mind and action that means ‘we are deeply vulnerable to manipulation by ecological constraints’, sometimes even with our awareness and consent. A distributed approach to affectivity as anchored or layered in to particular practices, places, or objects, perhaps in conjunction with cognitive-ethnographic observation of the ways that participants in any affective regime manage and repair glitches, trouble, or breakdown, can potentially reveal flexible and often communal distributed resilience in people-place ecosystems (Throop and Duranti 2015; Gillett 2022; Tribble 2022).

So disruption of various kinds and with various sources is structural rather than optional, given the uneven and dynamic nature of distributed cognitive and affective ecologies. Slaby does acknowledge that ‘the adult human mind is structurally invaded’ (2016:11). But in some of the critical literature there is perhaps a danger of again treating reliance on particular systems, environments, or other people as in tension with either autonomy or an active democratic moral-political consciousness. In talking of ‘invasion’ (as indeed of ‘extension’), we need to explicitly reject the implication that there is something unformed just there waiting to be influenced, whether for good or ill (Sutton 2011). We also need to remember that while the natural search for the bad actors who are doing the invading or disrupting may sometimes locate particular (individual or corporate) agents who are deliberately manipulating our open and porous mental lives for their own ends (Timms & Spurrett 2023), we are also perfectly capable of disrupting ourselves or choosing our own nasty niches, both through uneven and idiosyncratic processes of moral enculturation and internalization, and simply as a result of inevitable conflicts or shifting balances among the various disparate resources which partly constitute our capacities to remember, navigate, or feel. Likewise, a problem with the language of cognitive and affective ‘scaffolding’ is the implication that it can or should be dismantled once the edifice inside is constructed, a misleading implication that takes us back to the internalist positions we wanted to reject (Sutton 2015b; Larvor 2020).

It is important to recall that the individualist forms of classical cognitivism against which distributed frameworks were developed had no conceptual resources at all for incorporating an interest in normativity within cognitive theory. It is only once we have replaced the individualist conception or ideal of mind or mature agency as self-sufficiency with the alternative vision of cognitive and affective interdependence at the heart of human nature that we can even raise ethical and political concerns about the dangers of specific cognitive or affective artifacts and institutions, and begin to draw on and contribute to a genuinely interdisciplinary political philosophy of mind. Despite the welcome rise of the distributed and 4E alternatives, internalism is still a default assumption in many contemporary domains and debates, from mainstream neuroscience and cognitive psychology to mainstream politics and popular culture. Alignment between those powerful forces is apparent, for example, in some treatments – both scientific and popular – of our increasing and diversifying reliance on technologies in everyday cognitive life. While there is outstanding research on the different ways that different people lean or rely on particular technical systems in different contexts and for different tasks (Finley et al. 2018), there is also a vein of work on the impairment or degradation of our cognitive capacities over time by (over-)reliance on technologies like photography and GPS, or by ‘offloading’ and ‘outsourcing’ to Google and the internet what our brains should be doing: some such work persists in treating the relevant unit of analysis or comparison as the unaided biological brain operating ‘naked’ or in isolation. I can’t here address or respond to these new versions of very old moral panics (but see Heersmink 2016; Heersmink and Sutton 2020; Orben 2020): in the current context I want merely to remind us that outside cognitive philosophy it is even harder to find anything like a ‘dogma of harmony’ about human-technology relations, and easier to find technical systems being treated as wholly external sources of disruption or invasion. So while it is apt to caution those of us who do treat cognition and affect as open and distributed against excessive focus on and confidence in smooth coping and problem-solving, leading us towards more nuanced treatments of the complexity of particular ecologies, I also underline just how difficult it is, in modern Western culture especially, really to hold on to a sustained acknowledgement of our constitutive interdependence, right across the lifespan, on other people and other resources beyond skull and skin (Clark 2003; Harcourt 2016).

The new critical turn is offering fruitful attention to a range of particular negative cases, developing useful taxonomies of the sources of vulnerability to affective and cognitive harms, and criteria for the normative evaluation of affective niches (Nagatsu and Salmela 2022). Loaded normative terms for describing particular distributed systems – like ‘virtuous’, ‘empowering’, ‘vicious’, ‘harmful’, ‘problematic’ or ‘unhealthy’ (Slaby 2016; Williams 2016) – may sometimes have clear and uncontroversial applications when we consider obviously oppressive ecologies. But normative judgements won’t always be so easy just to read off descriptive accounts of the operations of more ambiguous forms of interaction and distribution, and the same affective ecology may have radically different temporary or enduring effects in different contexts or for different people and groups. In some cases it may not be appropriate for the theorist to make such open and contested evaluations, which may be better left to those directly engaged with the artifacts or equipment in question. These points, indeed, are familiar in the adjacent philosophical discussion of historically burdened heritage such as monuments, statues, and memorials, which is now a prime site for integration of political as well as aesthetic considerations into theories of cognitive and affective technology (Archer 2024). It is in this spirit that I move on to consider one particular strand of the interconnected sets of resources that can in certain circumstances partly constitute our individual and shared processes of thinking, feeling, and remembering – place.

3 Affective Place Memory

‘Place memory’ is used as a useful umbrella label for a variety of phenomena, not as naming a putative kind

Places can have visceral power to evoke past experiences. Certain memories and emotions do not arise if we are not back in a particular place, such that it can feel as if place partly holds our pasts for us or with us. It is just because memory and emotion can attach so deeply to accustomed terrain or neighbourhoods that displacement is such violent disruption, affective and cognitive as much as economic and political. The negative side of distributed cognition and situated affect is easy to see on this point, and this is the most straightforward connection between these themes here (cf. Nine 2018; Piredda 2020: 556). To take this natural link further, I suggest that a strongly distributed approach to place memory is needed to make sense of the deep emotional significance of place in the lives and projects of individuals and groups.

Places can play a great variety of roles in our cognitive and affective lives⁶. To simplify, and to pick out memory and emotion as interconnected test domains, we can consider three kinds of relation. First, places can be the objects or contents of memory or emotion: I remember Lisbon (from my visits in 1992 and 2023), and I have a range of feelings about it and specific places in it. Second, places can be stimuli or cues to memory or emotion, notoriously often triggering intense or surprising feelings and forgotten episodes, experiences that may be affectively mediated in the present by associations between the place and other people, songs, or things.

It may initially seem as if these two significant modes or forms – places as objects of and as cues to memory and emotion – are the primary or even the only significant relations between place and the mind. If this is all, then places – like, perhaps, bodies, other people, and artifacts – are always and inevitably external to the mind, merely providing stimuli and fields for action while the real psychological processes occur in the brain. Below I address some possible implications for political and environmental action from this thought that there is no constitutive cognitive or affective interdependence between places and minds. First, I move on to the third, stronger and more controversial, way of thinking about relations between place, memory, and emotion, directly driven by a distributed framework.

Perhaps in certain circumstances places can partly constitute the processes and activities of feeling and remembering. Historically and culturally unique landscapes, architectures, technologies and ecologies will then not always be external to mental life. Places can sometimes be parts of (distributed) vehicles of memory and emotion that span brain, body, and world, complementing our biological resources in place-people cognitive-affective ecosystems. Although places have all kinds of properties of their own, and may be active components of such heterogeneous ecosystems, accumulating their own histories at various timescales (Basso 1996; Turkel 2007), this is not a turn to panpsychism (Candiotto 2022). The idea is not that places remember or feel on their own, no more than that the black tie I wear to a funeral is ‘doing my grieving for me’ (Harris 2004: 729), or that the disconnected or naked brain remembers or grieves on its own. As Haugeland put it, it’s not that the road to San Jose knows the way on its own, but that the road and I collaborate: not all the structure of intelligence is ‘external’, but some or much of it may be, ‘in a way that is integral to the rest’ (1998: 233-5; Sutton 2020b)⁷.

One immediate benefit of this picture of places as potentially partly constitutive of remembering and feeling is that it gives us initial grip on how some people develop a deep embodied familiarity with particular places. When we inhabit neighbourhoods or territories in a more or less stable way, in cycles of activities, tasks, or routines, the sense of belonging or attachment need not be an object of reflection but can be taken for granted in seamless embodied interactions in well-trodden locations. This does not require that individuals or groups be permanently settled – deep place knowledge need not be fixed, and can also arise in patterns of movement over time, sometimes being transferred or transformed in inhabiting new places. It is affective through and through, but it is not only affective. It involves many interacting forms of memory and knowledge as well as highly attuned perceptual, spatial, and social capacities that are not easy to study in controlled laboratory settings. We must draw on disparate research resources such as history, ethnography, and philosophy to catch it in action. Wood’s cultural history of ‘topographies of remembrance’ in early modern England reveals ‘an inhabited, known landscape, one walked across, worked on, ploughed over, dug into … a taskscape’ that invited and supported rhythms of embodied and affective interactions at multiple timescales (Wood 2013: 188, 198; Sutton 2020b). Brown and Laurier (2005) track the extraordinarily flexible decision-making of a lifelong inhabitant of London as she negotiates the daily challenges of work travel across the city in changing contexts. Stirling and colleagues (2022) confirm the remarkable accuracy of directional and place knowledge revealed in casual conversation among long-term workers in the vast Kimberley regions of north-western Australia. Kukla points us to the exquisitely refined attention characteristic of spatially and socioculturally skilled urban agents, profoundly attuned to the affective tone of embodied micro-negotiations on the streets and to any salient novelties in the built or lived urban setting (2021: 13–82), just as Wood’s early modern workers were constantly ‘reading, monitoring and remembering change in the local world down to its most precise details’ (2013: 229). In contrast, with the notable exception of one research program on memory in London taxi drivers (Maguire et al. 2000; Griesbauer et al. 2022), experimental psychologists are only beginning to find ways to operationalize the effects on memory of deep familiarity with particular neighbourhoods (Penaud et al. 2022). Deep place knowledge is challenging to study not just because of the need to find appropriate participants, but because it seems to involve rich integrations of what are usually treated as distinct forms of knowledge or memory. It is neither like the case of someone who knows a lot about a city but has never been there; nor the case of someone with hippocampal deficits who retains a capacity to follow overlearned routes (Jeffery 2019: 858). This kind of knowledge of places is vitally intelligent, and may have distinctive psychological and epistemic features as perceptual and cognitive expertise operate alongside affective engagement (Kukla 2023).

This distributed perspective on affect and cognition in people-place ecosystems offers an extra dimension to studies of forced mobility, displacement, diaspora, and ‘solastalgia’⁸ in geography, memory studies, and cultural theory (Read 1996; Creet and Kitzmann 2014; Albrecht 2019; Erll 2020; Hage 2021; Mayblin and Turner 2021). These are fields in which cognitive theory and indeed psychology more broadly has often, understandably but unfortunately, been seen as irrelevant at best, actively imperialist and reductionist at worst, just because of the individualism that was characteristic of classical cognitivism, which left no easy roads between the disciplines. But it is clear that the distributed cognition approach to place actively encourages attention to normative issues in understanding cognitive and affective ecologies. Differential access to and control over places, spaces, and routes is not simply a political concern, but can be directly and simultaneously a means of cognitive and affective manipulation. At the level of bodily-affective styles and interactions, ‘the politics of space’ plays out constantly in the ways some buildings and public spaces operate actively to disorient or thwart or threaten some potential occupants, reflecting or enforcing or celebrating certain values and capacities, and not others (Krueger 2021; Crippen 2022). On a larger scale, it is just because land and buildings can directly scaffold and partly constitute joy, comfort, embodied skills, and shared memories as well as economic well-being that colonial removals of people from their land were such brutal and all-encompassing violence, obliterating values and crafts and situated memories as well as livelihoods (Campbell 2014a, b). This aligns also with the inclusion of ‘deportation’ – the forcible transfer of a group from one territory to another – among the ‘crimes against humanity’ established at Nuremberg and confirmed in the Geneva Convention and at the International Criminal Court, as mobilized for example in repeated international legal proceedings to force the United Kingdom to allow the Chagos Islanders right of return to their stolen Indian Ocean homelands (Jeffery 2013; Sands 2022).

Indeed, again, given the vast and uneven array of place-related resources that can partly constitute emotion and memory, there will never be invulnerable, entirely secure or stable relations between places and minds. No genuine or thoroughgoing distributed approach to place memory could fall for a dogma of harmony. There is always risk, openness, change, actual or potential trouble in the ways that we think, feel, and remember the past individually and together in and with and through our familiar environments. One brilliant analysis of a place-based cognitive and affective ecology is the Homeric scholar Aldo Paolo Bottino’s (2020) work on ‘space, time, and remembering in the orchard of Laertes’ in book 24 of The Odyssey⁹. An episode that has puzzled ancient and modern readers alike – Odysseus’s final reunion with his father, after the slaughter of the suitors and his night with Penelope – comes to life in Bottino’s ecological reading. Trying to drop his disguise and have his father recognize and acknowledge him, Odysseus takes him walking among the trees, naming them by type as Laertes had years ago when Odysseus was a child. The two men reconnect with their shared history, overcome the grief of absence, and find joy and common purpose only in thus moving among and noticing the trees with their stability and differently-paced existence, themselves re-instantiating that past event. Remembering here is a joint and site-specific performance, with the men’s embodied activity as they retrace their own past footsteps realigning the disparate components of this orchard ecology. Bottino’s analysis does justice to the non-human elements in this complex scene which play their part in holding and reactivating the shared past. While the analysis ends, like the passage of the poem, with the father and son united and heading together into the future, it shows clearly how intense were the troubling emotions like despair that were activated along the way.

A range of other enquiries follow naturally from adoption of a situated or distributed conception of affective place memory. We can investigate substantial individual differences in ways of inhabiting, navigating, and describing routes and places. We can deploy ideas about socially distributed cognition and emotion to address collaborative wayfinding, a topic barely studied in the cognitive sciences until recently (Dalton et al. 2019; Velasco 2022; Curtin and Montello 2023): this is strange given its ubiquity and importance, most of us being familiar with conflict and failure in trying to find our way together, even when using GPS. And we can expand the unit of analysis in studies of technology use, acknowledging possible changes in spatial learning but attending to augmented GPS designs that may promote active engagement or ‘cognitive diligence’ (Hebblewhite and Gillett 2021; Wunderlich and Gramann 2021). In each case, whether examining cognitive and affective aspects of digital spatial technologies, or treating places and spaces themselves as forms of technology which can exploit or empower particular populations, we are putting the situated perspective actively to work to better understand the interactive unfolding of our affective lives. To finish, I zoom in to the more specific arena of political and artistic engagements with places with multiple or difficult pasts, which also brings us back to superposition.

4 Modes of Engagement with Difficult Pasts: A First Sketch of an Aesthetics of Superposition

The Rivesaltes Camp Memorial near Perpignan in south-western France is an extraordinary site. Across thirty years of the mid-20th century, this windswept plain of makeshift huts saw the detention of four very different groups: Spanish republicans fleeing Franco; French Jews rounded up under Vichy to be taken East to die; captured German prisoners of war; and Algerian refugees, the ‘Harkis’, seeking asylum in France after independence (Peschanski 2002). The dark histories of these distinct groups are all superposed in the very same huts where they were interned, now standing in ruined formation in the bleak bright dusty landscape. A stunning memorial building by Rudy Ricciotti opened in 2015, embedded in the earth at the heart of the Camp, bewilderingly hard to access but in its powerful multimedia exhibitions vividly evoking not just the specific histories of those who lived and died there, but also our ongoing inability to find humane space for today’s vast refugee populations.

This application of the concept of superposition, to catch places with traces of multiple pasts, is clearly related to the scientific one with which I began, and also picks up directly Walter Benjamin’s references to ‘co-spatial’ layerings of different temporal strata of the same city, and the related notion of the ‘palimpsest’ in media and literary theory: indeed Rivesaltes has been called ‘France’s concentrationary palimpsest’ (Cantoni 2022: 253)¹⁰. Compared to other war internment sites like the Camp des Milles near Aix-en-Provence (Sumartojo and Graves 2018), and against the wider category of historically burdened heritage, superposition at Rivesaltes is in a sense straightforwardly enacted on the ground, with it falling on the visitor to construct responses and reactions in tracing a path now through the site and the rich and heartbreaking archival materials. But there is a sense in which Rivesaltes thus only literalizes or magnifies a feature of many monuments or memorials (for the distinction, as drawn by Danto, see Shapshay 2021: 149), and indeed of many places with complex or difficult pasts, where we in the present (whoever we are, positioned in whatever relations to the various past events in question) can only deploy the disparate resources and traces available in the present to construct our best cognitive and affective responses.

Archer’s paper (2024) in this special issue takes a great step forward in philosophical discussion of public monuments and artifacts precisely by bringing ideas of situated affect to bear. Compared to other recent philosophical literature on commemoration (see Lim and Lai 2024 for a rich and economical survey), Archer’s approach draws on ideas about distributed affect to widen our sense of the channels by which statues in particular have their visceral impact as affective technologies. As Archer recognizes, like Lim and Lai, the philosophical debate could also be helpfully further expanded by setting studies of particular toxic monuments in the broader contexts of the interconnected social practices and institutions within which objectionable artifacts have for a time held a fixed place (Lim and Lai 2024: 6). One fruitful source here is work in memory studies like Rigney’s (2022) application of a more holistic conception of mnemonic regime change to the Colston case in Bristol, where the dynamics of cultural memory had already been in increasingly tense motion for some time, and alternative narratives were in place, where decommissioning Colston went together with a broader ‘un-forgetting’ of slavery, violence, and colonial injustice (Rigney 2022: 22–32; Araujo 2020). My last speculations here look to memory studies and cultural theory to sketch a preliminary aesthetics of superposition as one among many possible modes of engagement with difficult places and multiple traces.

Typically, we no longer accept denial and suppression of difficult past events; and increasingly we are also dissatisfied with either the straightforward destruction or obliteration of objectionable artifacts or pasts, or the mere juxtaposition of counter-narratives, both which strategies fail actively to face and unsettle, dissolve, or replace the takes on the past that we now question or reject (Marschall 2019). This leaves us with a range of alternative political and aesthetic modes of engagement, that may more actively confront the past by in some way re-enacting or performing it, or may intervene and alter present traces of the past so as to reimagine it afresh¹¹. Or, we can find or settle into quieter modes of engagement which might celebrate, rearrange, and flaunt the superposed traces we find around us.

Gibson (2015) sees much effective memory art as a ‘forensic activity’, where artworks are ‘built from traces … left lying around in archives, in landscapes, in objects, in people’s bodies, in biographies and in family histories’. Such ‘memoryscopes’ detect ‘some lurking change’ in a remnant or trace, and channel it now in ‘a surge that zings some vivacy or aggravation’, helping us to sense how ‘the past is abroad in our present-day experience’. Developing Gibson’s ‘aesthetics of seepage and submergence’ (2015: 60) against both his own case studies like the history of Jamaican dub music (Veal 2007; Gibson 2010), and further case studies in contemporary memory art practice, including that of Janet Cardiff, William Kentridge, and Norman Klein, I suggest the following provisional principles for an aesthetics of superposition.

First, as foreshadowed above in Kukla’s account of Berlin’s counter-preservationiist impulse, it challenges the idea that we can recover an original or canonical version, whether of a song or of the past – all we will construct now are more versions, which we can do more or less responsibly and more or less effectively, in ongoing process. Second, and relatedly, it ‘flaunts its constructedness’ (Gurr 2023: 75), accepting or celebrating the creative refashioning that occurs in putting traces together now: as the dub pioneers said, ‘every spoil is a style’ (Veal 2007: 45–46). This is, again, not to give up on the past or on the responsibility to make claims on it now: rather, problems arise when we try to disguise the filters and processes that we are using to access or narrate it now. If we retain the disjunctures between past and present, double-coding our work or our memories or our actions as both ‘this-was’ and ‘this-is’, we do not conjure an illusion of presence or transparency, or of some literal form of ‘mental time travel’, but rather allow the seams of the presented past to fray, the interface to reappear, and the medium between us and the past to stay visible (Wheeler 2019; Lucas 2020). Finally, what is productive in this mode of engaging the superposed traces of the past is transformation, not preservation. The changes across versions, the erasures and accumulations, the seepage and the sedimentations and the selections, are the engine of the work (Rothberg 2012) or of the active processes of feeling and remembering, not seeking to reproduce one single past event, but evoking the tangling or overlaying of multiple traces.

These are merely one set of responses to our current crises of commemoration, among other ways of engaging with the ‘skewing dynamism of the past’ (Gibson 2015). The hope is that, as our initial concerns with the challenges of constructing the past from multiple present resources find urgent and specific instantiation in these place-based projects, we can enrich memory theory and cognitive theory too: we see that constructing something now need not be in tension with the attempt to be faithful to the past, but may precisely be a way of being responsible to that past. Constructing a usable past is a paradigmatic exercise and component of individual and community agency (Campbell 2006; Brown and Reavey 2015; Kukla 2022), and it can be done well or badly, better or worse, in any context and given any sets of goals and constraints. Managing to do so well enough is always an achievement, settling on who and where and when and why in ways that we hope will guide action well in the present. It is something we do, even if in our world of violence and distress ‘this is often the work of broken hearts or exhausted souls’ (Stegemann 2021: 31). I hope to have shown that examining place as a key dimension of ecologies of memory and affect opens up a range of independently interesting topics for insistently interdisciplinary enquiry in cognitive theory and beyond.