Matter and Memory by the French philosopher Henri Bergson is a seminal text in memory studies, the philosophy of mind, and metaphysics more broadly. It has exercised immense influence on the likes of Emmanuel Levinas, Maurice Merleau-Ponty, Jean-Paul Sartre, Paul Ricoeur, Walter Benjamin, and Gilles Deleuze. William James even said its publication amounted to a Copernican revolution in philosophy comparable to the critical turn made in the Critique of Pure Reason by Immanuel Kant. The acclaim heaped upon Matter and Memory is attributable to the innovative solutions Bergson proposes to classical philosophical problems such as the relationship between spirit and matter, past and present, and perception and memory. Bergson articulates his thesis as follows: “This book affirms the reality of spirit and the reality of matter, and tries to determine the relation of one to the other by the study of a definite example, that of memory” (9). Memory, then, provides the clue to the problem of how spirit relates to matter and vice-versa; it allows us to see that while spirit and matter are distinct, they are nevertheless related, albeit not in the problematic ways that dualistic metaphysical theories most often claim. While Bergson admits that his account is dualistic, even radically so, he insists that to whatever extent he accentuates this classical dichotomy, he does so in order to lessen, if not overcome, “the theoretical difficulties which have always beset dualism” (ibid.). To this end, Bergson situates his theory in between, or even beyond, realism and idealism, both of which he rejects, and each of which he thinks inevitably leads to the other in a kind of vicious circle.
While the principal thesis of Matter and Memory is that memory provides the clue to the relationship between spirit and matter, a secondary thesis that helps bolster this claim is that there is a crucial difference between perception and memory. Memory is not, in other words, an attenuated perception that recedes from the present moment and thereby loses its vitality; it is, rather, an autonomous and dynamic process in which the past that is preserved in itself can come to present consciousness in an infinite number of different and creative ways. Memory is not, then, so much a faculty of reproduction or repetition as one of synthesis; in fact, for Bergson, it is what makes consciousness possible in a duration that links the past and present. Related to this secondary thesis is yet another central claim: while matter, specifically the brain and the wider nervous system, may be necessary in order to actualize memories in consciousness, memories are not stored in the brain as if the brain were a container. In this sense, memories are not “in” the brain, but rather “in” time, i.e. in the past as part of the unconscious. Bergson therefore defends an anti-reductionist position that maintains a clear distinction between spirit and matter which parallels the distinction between memory and perception: in both cases, the former is not reducible to the latter.
To make his case about the difference between memory and perception and, more broadly, about the centrality of memory to the problem of spirit and matter, Bergson articulates a theory of pure perception in contrast to pure memory. Both pure perception and pure memory represent ideal types, extremes between which consciousness typically operates. As Bergson stresses, “there is no perception which is not full of memories” (33). Hence the ideal of pure perception intentionally brackets memory from its concept so as to render salient what perception, in and of itself, really is. For Bergson, pure perception relies on the concept of the image, which is neither an entity as it exists in itself as presupposed by realism nor a mere representation as presupposed by idealism. Images, Bergson claims, are all that we sense: “images [are] perceived when my senses are opened to them, unperceived when they are closed.” Yet there is one image that “is distinct from all the others, in that I do not know it only from without by perceptions, but from within by affections: it is my body” (17). On the one hand, then, there is the totality of images that comprise the material universe or matter, while on the other hand, there is the specific image that is my body. For Bergson, perception is simply the referral of some portion of the images that comprise matter to the eventual action of one particular image, my body (22). In this sense, my body is but one image amidst other images, albeit a unique one that marks the center of my possible action. As Bergson explains this point, “as my body moves in space, all the other images vary, while that image, my body, remains invariable. I must, therefore, make it a center, to which I refer all the other images” (46).
My body is therefore the locus and source of perception—or, more precisely, perception just is the referral of images that are not my body to the image that is my body. Yet because my body is merely another image that demarcates the center of my possible action in relation to other images, perception is “part of [the objects perceived] themselves; it is in them rather than they in it” (228-29). In other words, just as the brain is not a container for memories, the mind is not a container for perceptions; perception simply relates some set of images to the image that is my body, on account of which it has unmediated access to the images (i.e. the objects perceived) themselves. This means that the traditional distinction between objects that are outside the mind and representations of those objects inside the mind breaks down; or rather, this distinction is more appropriately conceived as “a distinction between the part and the whole,” where the part refers to the image that is my body, the locus of my perception, and the whole refers to the totality of all the other images that comprise the material universe (47). Bergson’s theory of pure perception is therefore not reducible to either realism or idealism: contra realism, matter “outside” the mind does not have the mysterious ability to produce representations “inside” the mind; and contra idealism, our perception of matter is not relative, mediated by mind-dependent structures universal to all humans. On Bergson’s theory of pure perception, there is no distinction between the phenomenon and the noumenon, between appearance and reality, but only, as we have seen, between the part and the whole (230).
Perception therefore has access to material objects as they exist “in themselves,” as Kant would put it, but this does not entail that it has access to the whole of the material universe. As Bergson frequently maintains, perception is limited in its scope, and this is because it is “entirely directed toward action” (31). That is, perception for Bergson is fundamentally vital and not speculative, a crucial point that both idealism and realism miss and that leads both theories into errors and contradictions. Because perception is ordered toward action in accordance with what interests bodily functions, it not only does not add to the image it perceives, but in fact subtracts from it, since it detaches “from the totality of objects the possible action of my body upon them.” Put differently, its function “is to eliminate from the totality of images all those on which I can have no hold, and then, from each of those which I retain, all that does not concern the needs of the image which I call my body” (229). For this reason, Bergson claims that perception exhibits a “necessary poverty” (38).
Pure perception is impoverished in that it only concerns itself with the needs of the image that is my body. Yet it is also impoverished in that it is an ideal type with which we have no actual familiarity. As noted earlier, pure perception never exists on its own in lived experience; it is always already complemented by memory, which when combined with perception constitutes what Bergson calls “my concrete and complex perception” (34). Unlike concrete perception, pure perception is entirely absorbed in the present and with matter—that is, it exclusively concerns my body in relation to the images upon which my body can act. Conversely, concrete perception is informed by memory, on account of which it “offers always a certain breadth of duration,” a plurality of moments that extend into the past (ibid.). Parallel to and concomitant with the distinction between pure perception and concrete perception is one between the present simply (another ideal type, a kind of analytical fiction) and the “real, concrete, live present,” the present that “necessarily occupies a duration” and hence “must be both a perception of the immediate past and a determination of the immediate future” (137-38). It is at this point, with the influential idea of Bergsonian duration in view, that we can turn to memory, the lynchpin concept of the entire book.
One of the several ways that Matter and Memory is so influential has to do with its treatment of memory and its many forms. Bergson identifies at least four, perhaps even five, forms of memory (there is some debate as to whether pure memory is equivalent to recollection memory, as the latter is contrasted with habit memory in the second chapter). In the first chapter, Bergson differentiates between what Trevor Perri calls “contraction memory” and “perception memory,” where the former refers to the contraction of a number of independent moments into one, internal moment or duration that constitutes the lived present, and the latter refers to “a cloak of recollections” that covers over “a core of immediate perception” (34). One could perhaps say that contraction memory provides the form of consciousness whereas perception memory provides consciousness with its content (combined with the content provided by perception). In the second chapter of Matter and Memory, Bergson makes yet another distinction between two types of memory, one which Gilles Deleuze stresses we should not conflate with the distinction between contraction and perception memory. On the one hand there is “habit memory,” or what contemporary theorists would call procedural memory, while on the other hand there is “recollection memory,” now known as episodic memory. Habit memory is a non-representational “motor memory” of the body that, as Perri observes, “manifests itself as a disposition to react in a more or less fixed way” to one’s environment (Perri, 511). Recollection memory is the representation of some past event in one’s life; it has a place and date and by nature cannot be repeated, unlike habit memory (81). Bergson considers recollection memory as “memory par excellence,” the true form of memory, whereas habit memory is “habit interpreted by memory rather than memory itself” (84). This seems to be because Bergson thinks that habit memory is closely related to perception; while it has to do with the acquisition of sensorimotor habits over time, and in this sense has some kind of relation to the past, it is actualized in relation to the present needs or interests of the body.
This leaves “pure memory” as the final form or aspect of memory, which several commentators associate with recollection memory. The two concepts are, I think, ultimately different, if nevertheless closely related. As far as I can tell, recollection memory is to pure memory as the part is to the whole; pure memory, in other words, is comprised of recollection memories, each of which refers to some past event or episode. Pure memory therefore refers to the totality of one’s past experience virtually preserved in the unconscious; it is the source of what is actualized in perception memory and the recollection of what Bergson calls “memory-images,” from which it is nevertheless distinct. “From the moment that it becomes image, the past leaves the state of pure memory and coincides with a certain part of my present,” Bergson explains. “Memory actualized in an image differs, then, profoundly from pure memory” (140). For Bergson, the relationship between pure memory and the unconscious is especially important. While memories are not “stored” in the brain, they are in some sense “housed” in the unconscious, with the crucial caveat that this spatial metaphor is not exactly appropriate to how Bergson understands the unconscious. This is because the unconscious for Bergson is not a spatial entity or container, but more like the past itself, the whole of which influences, if not in a deterministic way, our present experience in consciousness. As Bergson puts it, “the whole of our past psychical life conditions our present state; . . . whole, also, it reveals itself in our character” (148). Of course, the vast majority of our past psychical life is inaccessible to us; we only ever perceive a small part of it in the present, and it sometimes seems that our past recollections have either disappeared or only reappear at random. “But this semblance of complete destruction or of capricious revival,” Bergson contends, “is due merely to the fact that actual consciousness accepts at each moment the useful and rejects in the same breath the superfluous” (146).
Pure memory, then, includes all one’s past memories preserved in an unconscious state out of which certain memories come to the fore of consciousness based on the demands imposed by the present in concrete perception. To demonstrate this relationship between pure memory and perception, and also the relationship between recollection and habit memory, Bergson utilizes the famous image of the “cone of memory.” He describes an inverted cone whose apex intersects with plane P at point S and whose base is marked with points A and B, one on each side of its diameter. AB represents pure memory, the entirety of my past experience as preserved in the unconscious; S represents the sum of the present and the habit memory inscribed upon and within my body; and P represents the totality of the material universe within my field of present perception. On the one hand, this “cone of memory” indicates that pure memory and perception, recollection and habit memory, are distinct and of separate natures: habit memory is represented by point S that intersects with the field of perception, whereas pure memory, comprised of recollection memories, is represented by the base of the cone on a separate plane from plane P. On the other hand, the cone SAB underscores that the two forms of memory in conjunction with pure memory and perception are unified and “lend each other a mutual support” (152). As Bergson explains, pure memory offers to the sensorimotor mechanisms of habit memory the recollections needed to help them with their task, while the sensorimotor apparatus in turn provides the unconscious memories of pure memory the means to actualize themselves in present action (152-53). Moreover, the continuity of the cone, the fact that it unifies its base with its apex, accentuates that between the plane of perception and that of pure memory there are an infinite number of planes that correspond to what Bergson describes as the different “tones of mental life” (14). As Bergson makes this point, which he claims is one of the central claims of the book, “between the plane of action—the plane in which our body has condensed its past into motor habits—and the plane of pure memory, where our mind retains in all its details the picture of our past life, we believe that we can discover thousands of different planes of consciousness,” each of which constitutes a unique repetition of the whole of our lived experience (241).
From the above, we see how body and mind, matter and spirit are related. In fact, the cone of memory represents how these two sets of ostensibly opposed terms relate to each other as part of one continuous whole. Pure memory, which Bergson associates with the mind and spirit insofar as it is not “housed” in the brain and, as constitutive of the unconscious, is distinct from the images that comprise the material universe, conditions present perception and manifests in material “memory-images” conducive to action. Conversely, habit memory and perception, which Bergson associates with the body and matter insofar as the former refers to the sensorimotor apparatus of the body and the latter apprehends images that comprise matter, draw on the content of pure memory in order to perform actions that satisfy the needs and interests of the body. In ordinary life, we are almost always in between the base and apex of the cone, between the “plane of action” and the “plane of pure memory.” Typically, we are neither entirely impulsive, utterly absorbed with present needs, nor oblivious dreamers, hopelessly lost in past recollection, and this is because the two planes consistently interpenetrate, “so that each has to abandon some part of its original purity” (155). For Bergson, then, we are not so much composed of mind and body (as many of the ancients maintained) as we live between these two extremes, which are mixed in different proportions that correspond to different tones of mental life. Likewise, to the extent that mind and body are distinct, it is not in terms of space, with extended matter (like the body) outside the mind and unextended consciousness inside the mind, but in terms of time, with a past preserved in memory that bears upon the lived present of concrete perception (220). Hence time, just as it provides a principle of distinction between mind and body, also accounts for their intimate relation: the mind (i.e. pure memory, spirit) relates to the body (i.e. perception, matter) as the past relates to the present, and vice-versa.
Perri, Trevor. “Henri Bergson.” In The Routledge Handbook of Philosophy of Memory. Edited by Sven Bernecker and Kourken Michaelian. New York: Routledge, 2017: 510-18.