Memory

views updated

MEMORY

Remembering is one of the most characteristic and most puzzling of human activities. In particular, personal memory—the ability mentally to travel back into the past, as leading psychologist Endel Tulving puts it—often has intense emotional or moral significance: It is perhaps the most striking manifestation of the peculiar way human beings are embedded in time, and of humans' limited but genuine freedom from their present environment and immediate needs. Memory has been significant in the history of philosophy as much in relation to ethics and to epistemology as in theories of psyche, mind, and self.

The philosophy of memory is a fascinating, diverse, and underdeveloped area of study, which offers difficult but rewarding connections not only with psychology and the cognitive sciences, but also with the social sciences and political theory, and with literature and the arts. Outside philosophy, interest in memory increased massively and disproportionately in the late twentieth century in both the neurocognitive sciences and the humanities, driven both by internal developments within disparate disciplines and by wider social and cultural concerns about trauma and recovered memories, about the politics of forgetting and collective responsibility, about memory loss in an aging population, and about the manipulation, control, ownership, and protection of individual memory. The widespread and troubled fascination in Western culture with this last set of concerns in particular, and with challenging associated questions about moral psychology and personal identity, is suggested by the success of films like Bladerunner (1982), Memento (2000), and Eternal Sunshine of the Spotless Mind (2004).

As a result, just as in other areas of the philosophy of mind, it has become increasingly difficult to cordon off a set of questions about memory, or methods for its study, which are uniquely or primarily philosophical. Some philosophers treat memory as a case study in philosophy of science, asking for example whether the psychology of memory might be reducible to the neuroscience of memory. Others begin with the phenomenology of memory, the ordinary experiences and practices of remembering; others still inquire into cross-cultural or historical differences in these practices. It seems likely, further, that psychopharmacological influences on memory, and their potential misuse, will make memory a central topic in the emerging fields of neuroethics and philosophy of psychiatry. This entry covers more traditional philosophical issues about the nature of memory, but includes some consideration of the need for a broader framework that can encompass the neural, embodied, psychological, and social aspects of remembering.

Forms of Remembering

When a person is remembering, there are many different activities he or she may be engaged in, and the expression of the individual's memory can take many different forms. One reminisces with old allies about shared experiences; one finally calls to mind that obscure fact; one mindlessly cycles off down the lane, despite not having been on a bike for years; one sits alone and ruminates on one joyful or agonizing moment long ago; one gathers with others to commemorate a significant occasion; one writes or fashions something in memory of a person or an event; a photo, an odd memento, or a long-forgotten melody suddenly immerses a person in the emotions of another time.

It is not easy to pinpoint just what is common across this range of activities, and some philosophers have argued that not all of them involve true memory. But the present-day consensus in both philosophy and psychology is that there are at least three distinct forms of remembering that can helpfully be detected in the variety of ordinary experience.

First, in remembering specific events or episodes from an individual's personal past he or she draws on personal memory (also known as experiential or event memory ): For example, one remembers walking down by the river with a friend that spring afternoon. Psychologists often call this episodic memory, or sometimes autobiographical memory.

A different form of memory is naturally expressed with a "that" complement: One remembers that Aristotle was Alexander's tutor. This factual or semantic memory is akin to simple belief, and the remembered facts can be about events in the remote past, or indeed the future, as well as personally experienced events. One can factually remember details one has been told about one's early life, for example, for which one has no personal memory, no sense at all of what the past experiences were like.

In English, and many other languages, people sometimes contrast things that they "just know" from what they genuinely (personally) remember, thus treating personal memory as the basic or essential kind of memory. But in other contexts people are happy to talk also of remembering facts, and to attribute their general beliefs about the world to "memory" in a broader sense.

Personal/episodic remembering and factual/semantic remembering are both forms of declarative memory, in which individuals seek to hook up to reality, to represent the world or the past. Although remembering activities often have quite different functions as well, under normal circumstances such memories aim at truth. This is so even though, as both scientific and common-sense psychology increasingly suggest, people do not always get there. The point is not that memory necessarily or even reliably achieves this aim, but that one's ordinary practices include a general commitment to its reliability in doing so. For example, an individual may or may not in fact have walked by the river with a friend that spring afternoon, and Aristotle may or may not actually have been Alexander's tutor. But if one is sincerely expressing that personal memory, or that factual memory, one is (among other things) making a claim about what happened.

In these declarative forms of memory, the content of one's memory can in principle —at least in central cases—be articulated. But when a person wonders if a friend remembers how to play the flute, or how to drive a car, the person is asking not about the friend's personal or factual memories, but about his or her skills or embodied memories. Philosophers have often talked here of habit memory, while psychologists identify these cases as types of procedural memory, where this category is also taken to include more basic/primitive forms of conditioning and associative learning.

Procedural memory has been sharply divided from declarative memory for a number of reasons: Perhaps most important is the case of H.M., an epileptic patient who suffered terrible amnesia after brain surgery in the 1950s. H.M., who had lost his hippocampus and other brain structures now known to be central to declarative memory, was no longer able to lay down event memories, so that he would forget everything minutes after its occurrence, and lose any clear sense of time passing. Yet H.M. was still able to learn new games, and to improve his performance at new perceptual-motor skills, despite having no idea each time that he had ever tried them before.

Procedural memory is philosophically important for a number of reasons, although habits and skilled activities have been little studied. For example, neither philosophers nor psychologists have a clear grip on the various ways that personal memory and other high-level cognitive processes interact with remembered embodied skills. Competition and coordination between the different memory systems can both occur. On the one hand, skilled performers in dance or sport know that their motor habits often run best in a groove, when not consciously or verbally controlled: yet the skills involved are robust and flexible, unlike more primitive forms of procedural memory, and can sometimes be directly shaped by mood, context, verbal instruction, and conscious decision.

These conceptual, grammatical, and experiential distinctions between personal, factual, and habit memory have in contemporary cognitive psychology been developed into theories of distinct memory systems. There is considerable disagreement about the psychological status of these systems, and about whether the distinction between episodic and semantic memory, in particular, should be characterized by reference merely to the kind of information in question, or by an essential phenomenological difference. Since there is little agreement more generally about what a psychological system or module is, or about the nature of any putative natural kinds in psychology, these debates about memory systems are likely to be resolved only in conjunction with progress on broader questions in philosophy of psychology.

Personal Memory

An individual's capacity to conjure up experiences, emotions, and events from long in the past involves the same kind of memory as the mundane ability to keep track of just what he or she has been doing, feeling, and thinking in the last day or week. Personal remembering does not seem to be distinguished from other related activities—imagining, dreaming, factual remembering, for example—by the level of sensory detail or vividness which it involves: some memories, after all, are both faint and fragmentary, while some scenes of fantasy can be richly imagined. Memory capacities, even in their normal and reliable functioning, are both fallible and selective: human beings don't need either total or precise recall to maintain sufficient coherence and continuity of self over time, for personal memory works in part through an ongoing condensing, editing, and summarizing of life experiences, on which people draw in specific autobiographical narratives. One's narratives or other memory expressions can be public or private, and they can be more or less under one's control, either smoothly tailored to specific audiences or emerging in involuntary fragments.

Personal remembering is a context-sensitive activity from the start. As young children build on their earlier abilities to understand typical sequences of events, their capacity to remember particular past experiences is supported and shaped by adults. Joint attention to the shared past emerges in an interactive social environment, as children come to see that there can be different perspectives on the same past time. Spontaneous self-conscious thought about the personal past is a gradual development out of these memory-sharing practices, which can vary considerably in nature, frequency, and significance across contexts and cultures. One condition for the full emergence of such self-conscious thought about his or her own past experiences, which may be surprisingly late, is that the child picks up the causal connections between events in time, and within the child's own history. Some grasp of the temporal asymmetry of experience is needed to understand that, in principle at least, remembered events can be integrated on a connected temporal dimension. Children's personal memory, then, is a highly sophisticated achievement closely linked not only to their emerging self-awareness and understanding of other minds, but also to their recognition that they cannot change the past, and that their current and future actions are unique and irrevocable.

Because early personal remembering is socially situated in this way, it is also tightly meshed with emotional and social/moral development. Key social practices, such as promising and forgiving, and some central complex emotions, such as grief, love, and regret, depend essentially on personal memory and on one's grasp of temporal relations. The point here is not just that the fallible but more-or-less reliable operation of memory in two or more people is needed to give those people current informational access to the past times at which their paths have crossed. Memory's affective tone and influence means that, in addition to its role in retaining the past, it also has a forward-looking function, as Richard Wollheim argued in his Thread of Life (1984): Remembering can keep what happened in the past alive, giving it significance for one's ongoing relationships and projects. According to this view, memory is not just a means for checking on the continuity of the self over time, but also itself partly produces or creates personal identity: As Wollheim puts it, the past affects people in such a way that they become creatures with a past.

The particular ways in which, through memory, individuals deal with events and experiences that are no longer present varies according to context and aim. Most dramatically, for example, legal contexts impose demands and standards on the memory narratives witnesses must produce that differ greatly from the norms operating in other remembering activities. But questions about the reliability of memory and about its mechanisms arise in many different circumstances just because memory, with its orientation to truth, is in these ways intimately involved in both personal identity and significant social practices. Two connected lines of thought have raised the most serious concerns about people's access to the past in remembering: philosophical views about representations and memory traces, and psychological accounts of the constructive nature of remembering.

Theories of Memory

People can, sometimes, remember past events and experiences in the absence of immediate external cues or prompts to memory. It is natural, then, to think that somehow individuals carry around with them what they will need in order to remember when circumstances are right. Even one's ordinary conception of memory, C. B. Martin and Max Deutscher argued in their influential causal analysis Remembering (1966), requires the existence of an appropriate causal link between one's past experience and one's present remembering. Although the notion of the "memory trace" has appeared in many strange metaphors and theories in the history of philosophy and the history of science, it need be no richer than this idea of a state that causally connects experience and remembering in a certain way. This causal analysis embeds the theory of memory in the broader representational theory of mind which has come to characterize mainstream philosophy of cognitive science; however the bare invocation of memory traces is compatible with many quite different views about their nature and operation.

However, even this basic view about memory traces, in the eyes of its critics, engenders serious problems about the nature of a person's access to the past. If the past is thus truly lost, so that a person can only make contact with it by examining certain representations in the present, critics complain, there is a real danger of scepticism, to be countered by affirming that the person is in fact aware of the past directly in memory. The ensuing, long-running debate between representative realists or indirect realists, who accept memory representations, and direct realists or phenomenologists who reject them, is exactly parallel to that found in theories of perception. Although the dichotomous nature of this debate no longer fits the range of positions available, and many quite different views are often condensed by critics into a monolithic target, there is some common ground.

Contemporary trace theorists tend to work in a broadly materialist framework, and do not in general think of traces as direct objects of awareness from which the nature of the past is consciously inferred at the personal level of psychological analysis. If complex nonconscious processes, operating subpersonally on representations which may themselves be partial and context-sensitive, are involved in the shaping and constructing of the contents of memory, this does not mean that the experience of remembering is indirect. On this point, the positive direct realist contribution is convincing: Remembering, under normal circumstances, is a kind of immersion in which one has a pre-reflective confidence.

But this idea that an individual typically inhabits the memory, rather than judging and assessing it for plausibility and coherence, is in fact entirely compatible with the existence and involvement of subpersonal mechanisms operating on enduring but modifiable traces. Such mechanisms can be typically reliable even if they are fallible in particular instances. To raise a general skeptical worry again at this point against the invocation of memory representations would be unrealistically to demand incorrigible access to the past, to seek a blanket guarantee of accuracy in memory. Such blunt certainty about memory was expressed, for example, by the eighteenth-century Scots philosopher Thomas Reid, the most ardent critic of philosophies of "ideas" or "traces," who wrote that "those things really did happen which I distinctly remember" (Reid, Essays, 1849, p. 444). But this renders the indisputable evidence—both everyday and scientific—of errors in memory quite mysterious, and thereby threatens to erode commonsense realism about the past.

Theorists who posit memory traces are also criticized for adherence to what is seen as an arbitrary metaphor of "storage," unfortunately entrenched in the philosophy of memory since Plato's Theaetetus. The bare retention of capacities or dispositions to act or respond in certain ways, the critics complain, implies nothing about the means by which such capacities are retained: Storage is a mistakenly concrete way of thinking, as if each memory had to be stashed away separately, like sacks of grain in a storehouse or fixed entries in an archive. Ludwig Wittgenstein, for example, mocked the static but inaccessible inner records he identified in the psychological theories of his time: In notes of 1935–1936, he wondered "whether the things stored up may not constantly change their nature" (Stern 1991, p. 204).

In some invocations of memory representations, each trace has indeed been treated as distinct, with each single remembered item mapped on to one storage element. Such atomist or localist representational schemes make control over the contents of memory easier to imagine or achieve: The remembered items are passive, and must be manipulated or altered by an external executive. In this separation of data from process, ordinary digital computers exemplify the localist memory scheme: But what is "stored" in human memory displays more intrinsic dynamics than this, tending in some contexts naturally to interfere, blend, and generalize without deliberate or voluntary control. But just as such computers do not exhaust possible computational devices, so localist representational schemes are not essential to the general framework of memory traces. Both historical theories of memories as patterned flows of "animal spirits" through the pores of the brain, and contemporary connectionist models in cognitive science employ distributed (rather than localist) representation: What can be distinctly remembered need not be held distinctly or independently, since each item is spread or "superposed" across many elements in a system or network. This entry examines the implications of these distributed models of memory after setting them in the context of recent developments in cognitive psychology and the cognitive sciences.

Remembering and the Cognitive Sciences

The recent history of the sciences of memory offers a sharp contrast and corrective to the stereotyped image of cognitive science as a scientistic quest to reduce the human mind to the dull mechanism of digital computers. Memory research was one of the first areas to be taken out of the lab in the 1980s and 1990s, as psychologists sought to address the kinds of memory that matter in everyday life (such as autobiographical memory), and to find ecologically valid methods of studying such memories outside artificial isolated situations. The difficulty facing philosophers or scientists with an urge toward synthesis is not that psychological results are irrelevant to wider concerns about memory, but that the daunting diversity of methods and traditions even within cognitive psychology makes it hard to see how different levels of explanation might relate to one another. There are issues of considerable interest for the philosophy of science in understanding the connections between neuroscientific and cognitive-psychological descriptions and methods; and, equally, robust and philosophically intriguing research traditions on autobiographical memory in developmental, personality, and social psychology. This entry briefly examines ideas about the constructive nature of remembering that seem to have direct relevance to concerns about truth in memory.

Remembering is a multifaceted activity that takes place in the present, and so the best explanatory frameworks for understanding it will attend closely to the context of recall, rather than simply investigating the nature of encoded traces. Memories are often compiled or constructed for particular purposes when needed, not held fully formed. There is room for considerable internal plasticity in memory traces, which are (on the connectionist model) always composites shaped by the entire history of their network. The most dramatic work on construction in memory has come not from connectionist modeling, however, but from the research on suggestibility and false memory by Elizabeth Loftus and her colleagues (2003). Misleading information from external sources can be incorporated into personal recollection. Confident, entrenched childhood "memories" of spilling a bowl of punch at a wedding, for example, or of gazing long at an exceptionally colorful mobile in the days after birth, can be elicited artificially in certain circumstances. This work is partly motivated by a wish to confirm the possibility of false confessions, in which individuals may come sincerely and passionately to believe that they have committed horrible crimes in the past; but the mechanisms in play are just the ordinary and normally robust processes of shaping and generalizing memories to make them fit. Although Loftus has adopted the high moral tone of a crusade, ongoing careful investigation of individual differences and integration of these results with social and personality psychology promises a much richer picture of the conditions which make different kinds of distortion more likely.

Again, the point of this research is not to show, implausibly, that reliability in memory is impossible or unlikely. Psychologists assume that understanding the mechanisms of distortion will also throw light on the processes involved in veridical remembering. Reliability and accuracy are not transparent notions here. Pre-reflective confidence in personal memories can, and in certain contexts should, coexist with attention to the other evidence about the past which is often available, and care for the defeasible but subtle and robust capacities to winnow evidence that individuals have developed in the rich and complex social context of early memory-sharing and memory-using practices.

Social Memory and Shared Memory

The general constructive picture of remembering can be accepted while acknowledging that external influences—particularly social influences—on memory need not inevitably lead to error. As Sue Campbell argued in her powerful philosophical responses to the "memory wars," there are vital features of relational interaction with others that contribute positively to practices of good remembering, both in development and in adult social life: To treat the true unit of memory as the isolated individual, free from the distorting influence of other people, is to miss the value we often appropriately place on negotiating the past—both the personal past and the shared past—in company.

Indeed a need for attention to shared remembering and social remembering in both psychology and philosophy can be motivated from within the broadly constructivist framework itself. It is because one's internal memory is partial and context-sensitive, and does not naturally retain information in distinct and unchanging form between experience and recollection, that one relies so pervasively and—in the main—successfully on external social and technological scaffolding. A challenge for psychologists is to find ways to study shared memories that do not focus solely on the conformity induced or sought by powerful external authorities; and a challenge for philosophers is to construct a social ontology of memory by which to understand the diverse ways in which people manage to hook their incomplete inner systems of traces with the vast social and cultural resources in which cognition is situated.

Mark Rowlands (1999) and Rob Wilson (2004), for example, have suggested specific ways in which external symbol systems—in their many distinct historical and cultural forms—allow individuals to leave information and skills out in the world, saving on the resources and capacity required for biological memory. Drawing on the more precise invocations of terms like social memory and collective memory in the contemporary social sciences, this distributed cognition framework suggests that researchers can study the transmission of particular representations across different individuals and media, and the specific forms of interplay between group dynamics and individual recall. It also promises to throw better light on the influential work on memory by the French sociologist Maurice Halbwachs (1980).

Halbwachs's notion of the collective memory is often cited by contemporary social scientists and historians as deeply anti-psychological, or as sociologically determinist: but in fact his work focuses on the incomplete or shrouded nature of the individual's memory, which (outside of dreams) must be sculpted and completed within a social framework, which provides the context and the means for the construction of a specific recollection. Philosophical analysis can potentially be of immense service to empirical disciplines like cognitive anthropology and historical theory in the study of memory by showing how case studies of remembering activities in particular times and places might be embedded in robust broader theories of memory. So in addition to the long-standing philosophical concerns about truth and the self previously outlined, it is likely that philosophical attention will increasingly engage, through topics like memory, with the urgent challenge of connecting the cognitive sciences and the social sciences.

Bibliography

Bartlett, Frederic. Remembering: A Study in Experimental and Social Psychology. Cambridge, U.K.: Cambridge University Press, 1932.

Bursen, Howard. Dismantling the Memory Machine. Dordrecht: Reidel, 1978.

Campbell, John. "The Structure of Time in Autobiographical Memory." European Journal of Philosophy 5 (1997): 105–118.

Campbell, Sue. Relational Remembering: Rethinking the Memory Wars. Lanham, MD: Rowman and Littlefield, 2003.

Campbell, Sue. "Models of Mind and Memory Activities." In Moral Psychology: Feminist Ethics and Social Theory, edited by Peggy DesAutels and Margaret Urban Walker. Lanham, MD: Rowman & Littlefield, 2004.

Carruthers, Mary. The Book of Memory. Cambridge, U.K.: Cambridge University Press, 1990.

Casey, Edward S. Remembering: A Phenomenological Study. Bloomington: Indiana University Press, 1987.

Connerton, Paul. How Societies Remember. Cambridge, U.K.: Cambridge University Press, 1989.

Deutscher, Max. "Remembering 'Remembering.'" In Cause, Mind, and Reality, edited by J. Heil, 53–72. Dordrecht: Kluwer, 1989.

Draaisma, Douwe. Metaphors of Memory: A History of Ideas about the Mind. Translated by Paul Vincent. Cambridge, U.K.: Cambridge University Press, 2000.

Draaisma, Douwe. Why Life Speeds Up as You Get Older: How Memory Shapes Our Past. Translated by Arnold and Erica Pomerans. Cambridge, U.K.: Cambridge University Press, 2004.

Engel, Susan. Context Is Everything: The Nature of Memory. New York: W. H. Freeman, 1999.

Foster, Jonathan K., and Marko Jelicic, eds. Memory: Systems, Process, or Function? New York: Oxford University Press, 1999.

Hacking, Ian. Rewriting the Soul: Multiple Personality and the Sciences of Memory. Princeton, NJ: Princeton University Press, 1995.

Halbwachs, Maurice. On Collective Memory. Translated by F. J. and V. Y. Ditter. New York: Harper & Row, 1980.

Hamilton, Andy. "False Memory Syndrome and the Authority of Personal Memory-Claims: A Philosophical Perspective." Philosophy, Psychiatry, and Psychology 5 (1998): 283–297.

Hoerl, Christoph. "Memory, Amnesia, and the Past." Mind and Language 14 (1999): 227–251.

Hoerl, Christoph, and Teresa McCormack, eds. Time and Memory: Philosophical and Psychological Perspectives. New York: Oxford University Press, 2001.

Hyman, Ira E., and Elizabeth F. Loftus. "Errors in Autobiographical Memory." Clinical Psychology Review 18 (1998): 933–947.

Krell, David Farrell. Of Memory, Reminiscence, and Writing: On the Verge. Bloomington: Indiana University Press, 1990.

Loftus, Elizabeth F. "Our Changeable Memories: Legal and Practical Implications." Nature Reviews Neuroscience 4 (2003): 231–234.

Malcolm, Norman. Memory and Mind. Ithaca, NY: Cornell University Press, 1977.

Margalit, Avishai. The Ethics of Memory. Cambridge, MA: Harvard University Press, 2002.

Martin, C. B., and Max Deutscher. "Remembering." Philosophical Review 75 (1966): 161–196.

Martin, M. G. F. "Perception, Concepts and Memory." Philosophical Review 101 (1992): 745–763.

McClelland, James L., and David E. Rumelhart. "A Distributed Model of Human Learning and Memory." In Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Vol. 2, edited by J. L. McClelland and D. E. Rumelhart, 170–215. Cambridge, MA: MIT Press, 1986,

Misztal, Barbara. Theories of Social Remembering. Maidenhead: Open University Press, 2003.

Neisser, Ulric. "The Ecological Study of Memory." Philosophical Transactions of the Royal Society B: Biological Sciences 352 (1997): 1697–1701.

Nelson, Katherine. "Self and Social Functions: Individual Autobiographical Memory and Collective Narrative." Memory 11 (2003): 125–136.

Owens, David. "The Authority of Memory." European Journal of Philosophy 7 (1999): 312–329.

Owens, David. "A Lockean Theory of Memory Experience." Philosophy and Phenomenological Research 56 (1996): 319–332.

Reid, Thomas. "Essays on the Intellectual Powers of Man." In The Works of Thomas Reid, edited by W. Hamilton. Edinburgh: McLachlan, Stewart, & Co., 1849.

Ricoeur, Paul. Memory, History, Forgetting. Translated by Kathleen Blamey and David Pellauer. Chicago: Chicago University Press, 2004.

Rowlands, Mark. The Body in Mind: Understanding Cognitive Processes. Cambridge, U.K.: Cambridge University Press, 1999.

Rubin, David C. Memory in Oral Traditions: The Cognitive Psychology of Epic, Ballads, and Counting-Out Rhymes. New York: Oxford University Press, 1995.

Schacter, Daniel. Searching for Memory: The Brain, the Mind, and the Past. New York: Basic Books, 1996.

Schacter, Daniel. The Stranger Behind the Engram: Theories of Memory and the Psychology of Science. Hillsdale, NJ: Lawrence Erlbaum, 1982.

Schechtman, Marya. "The Truth about Memory." Philosophical Psychology 7 (1994): 3–18.

Sorabji, Richard. Aristotle on Memory. London: Duckworth, 1972.

Stern, David G. "Models of Memory: Wittgenstein and Cognitive Science." Philosophical Psychology 4 (1991): 203–218.

Sutton, John. Philosophy and Memory Traces: Descartes to Connectionism. Cambridge, U.K.: Cambridge University Press, 1998.

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

Tulving, Endel, and F. I. M. Craik, eds. The Oxford Handbook of Memory. New York: Oxford University Press, 2000.

Warnock, Mary. Memory. London: Faber, 1987.

Wilcox, Stephen, and Stuart Katz. "A Direct Realist Alternative to the Traditional Conception of Memory." Behaviorism 9 (1981): 227–239.

Wilson, Robert A. Boundaries of the Mind. Cambridge, U.K.: Cambridge University Press, 2004.

Wollheim, Richard. The Thread of Life. Cambridge, U.K.: Cambridge University Press, 1984.