Mneme, Anamnesis and Mimesis: The Function of Narrative in Paul Ricœur’s Theory of Memory

Ridvan Askin,

University of Freiburg

Im Blick zurück entstehen die Dinge
Tocotronic – “In höchsten Höhen”


Paul Ricœur develops his phenomenological-hermeneutical theory of memory in his seminal Memory, History, Forgetting, and several preliminary studies to his monumental book.[1] As its title indicates, the monograph treats memory in conjunction with forgetting and history, placed within a wider horizon of what could be termed an ethics of forgiving. For the purpose of this article I will focus on the problems of memory and forgetting, ignoring history for the most part. Similarly, I do not explicitly deal with the more specific issue of collective memory, which marks a transitional step from a proper phenomenology of memory to the epistemological problems of history and historiography. Even though the three main objects of Ricœur’s study are firmly interlinked, leaving aside history does not impinge on its general line of thought since the most fundamental problems of Ricœur’s discussion on history are imported from his discussion on memory, most notably that of the Greek eikon – representation. Rather than to engage in a wide-ranging ethical discourse, my aim is to focus on a more basic critique of Ricœur’s understanding of narrative, and the role he assigns to it in his theory of memory.

I argue that Ricœur’s Aristotelean-structuralist notion of narrative runs into problems when confronted with narratives which do not comply with its notions of order and synthesis. In addition, despite its pervasiveness in his theory, narrative is not adequately scrutinized by Ricœur. That my critique implies severe consequences for his ethics should become obvious, even though an analysis of what exactly these consequences entail must remain outside of this study. For the purpose of this article, it is enough to know these two points: first, that Ricœur’s theory of memory and forgetting is constitutive for his theory of history, which in turn functions as a critical remedy to our often unreliable memories; and second, that this dialectical relation between memory and history is situated within an ethical horizon.

Mneme and Anamnesis

As a first step in the development of his theory of memory, Ricœur distinguishes between two dimensions of memory: a cognitive and a pragmatic one. These two dimensions correspond to the Greek notions of mneme and anamnesis and designate memory, in the first instance, as affection (pathos) claiming a truthful relation to the past, memory as a kind of knowledge, and the active search of recollection, memory as praxis, in the second. Regarding the first, cognitive, aspect of mneme, the most pressing problems revolve around the notion of representation, which includes the problem of the relation between presence and absence, plus that concerning the temporal dimension of memory.[2] The problem in the first case is that a representation itself is present but stands in for something absent. This is exemplified in Plato’s Theaetetus with the seal’s imprint left in a piece of wax. Plato hereby ties the notion of representation (eikon) to that of the imprint (tupos) and thus originates a prolific discourse centered on the concept of the trace. Having thus introduced the concept of the trace, Ricœur proceeds by distinguishing between three different types of traces: the cerebral imprints and traces that neuroscience is concerned with, the affection-impressions caused by events, and the archival document (Memory, History, Forgetting 13-5).[3] What makes these traces problematic, however, is the fact that even though the traces are present, whatever caused them is not – not present meaning both not here and not now. This temporal dimension, remaining implicit in Plato and only introduced explicitly by Aristotle, is of major importance, since otherwise we would not be able to distinguish between memory and imagination, which is similarly characterized by absence. Indeed, according to Ricœur, Plato advocates “enclosing the problematic of memory within that of imagination” whereas Aristotle “argues for including the problematic of the image within that of remembering” (7). Following Aristotle, Ricœur asserts that memory thus marks our access to the past – and not just one access among others, but the only one we have (21). This, of course, is highly problematic. Our memories, the present traces of something absent, are not necessarily reliable. Traces can be erased, for example. Such erasure amounts to one kind of forgetting – one kind, because Ricœur distinguishes between different types of forgetting which correspond to the different aspects of memory. However, before turning to the specific difficulties of forgetting, the pragmatic dimension of memory needs to be considered.

Anamnesis, the praxis of recollection, presents us with a whole new array of problems: as praxis, recollection constitutes an action and thus denotes an exercise, the use of memory. Yet, Ricœur adds, the use of something always allows for the possibility of its abuse. Thus, according to Ricœur, we can observe three different levels of abuse in the context of memory: on the pathological-therapeutic level there is the phenomenon of blocked memory, on the practical level that of manipulated memory and on the ethical-political level that of obligated memory (57). These possibilities of abuse pose yet another threat to memory’s truth claim, to its faithfulness towards the past, the praxis of anamnesis thus directly affecting memory’s cognitive dimension. In turn, the practical operation of recollection is driven by a cognitive aim, its objective being re-cognition: it is precisely the reliability of re-cognition which is at stake here. Since the two dimensions of memory are thus firmly tied to each other (this is why they are not two distinct phenomena but two dimensions of one phenomenon) one should keep in mind that Ricœur’s distinction between them does not denote a substantial difference but only serves heuristic purposes.[4]

As to the three levels of abuse, in the first case of blocked memory Ricœur refers to Freud and his work on remembering and repression. Freud’s notion of the work of remembering is central to Ricœur in this context as it opposes the dynamics of repression consisting in the compulsion to repeat. Giving leeway to the compulsion to repeat instead of opposing it with the work of remembering constitutes the first kind of abuse. In the second case, abuse consists of the manipulation of memories through narrative. Narrative is fundamentally selective, allowing for variation: one can always tell the same story differently. Thus, the abuse in this case amounts to the manipulation of memory through master narratives. Finally, the third level of abuse concerns the duty to remember. The injunction “You must remember!” can also be expressed in terms of forgetting: “You must not forget!” As the ability to forget is central to Ricœur’s ethics of forgiving, such a duty to remember marks the third possible abuse of memory. Ricœur suggests replacing such obligated memory with the aforementioned Freudian work of remembering. However, because such a work of remembering would still be prone to manipulations, it needs to be accompanied by a work of mourning, which accepts loss and thus paves the way for reconciliation.[5]

These are the three kinds of abuse memory can fall prey to. One can easily see how these abuses threaten the truth claim of memory: in the first case, indulging in the compulsion to repeat means denying true recollection, in the second case, master narratives reinforce particular versions of stories and thus manipulate memories according to their purposes, and in the third case, the clamor of the “frenzy of commemoration” (90) threatens to drown out any work of remembering and mourning, thus facilitating the workings of repression and the reinforcement of master narratives. As the proper aim of anamnesis, characterized as zetesis – search – by Aristotle, is truth, or rather the establishing of a truthful relation between the representation and the represented past, it is this faithfulness towards the past which is threatened. The abuses of memory do not constitute the only danger, however. There is yet another threat to memory; this threat concerns forgetting.


What is forgetting? In accordance with the distinction of the two dimensions of memory, mneme and anamnesis, Ricœur says of the first one: forgetting amounts to a partial or complete erasure of traces. This seems fairly reasonable when considering two of the aforementioned three types of traces, namely cerebral traces and the document. The third kind, however, poses a serious problem: if the affections caused by events were really destroyed, could we still speak of something like memory? Would there still be a memory if the first impression did not persist? For his answer to this problem, Ricœur turns to Bergson and Heidegger. With Bergson he speaks of an unconscious duration of memory which, as it is unconscious, amounts to a kind of forgetting. In this case, however, we can no longer speak of an erasure of traces. Quite on the contrary, with Heidegger Ricœur interprets forgetting as immemorial resource in this case and thus as constitutive for memory (440-443).[6] Seen in this way, forgetting provides what the work of remembering tries to recover. Echoing Nietzsche, Ricœur thus distinguishes between forgetting as destructive and forgetting as creative.[7]

In a similar move, forgetting becomes constitutive for the search of anamnesis. It is exactly the loss of memory which propels the search and calls for the work of remembering. The forms of forgetting corresponding to this pragmatic dimension of memory correlate with the aforementioned types of abuse: in blocked memory, it is repetition as the dynamic of repression which amounts to forgetting; in manipulated memory, the selective nature of narrative causes forgetting – forgetting amounting to that what has been left out of the narratives; and finally, obligated memory finds its counterpart in commanded forgetting. The most striking example of such commanded forgetting is the institution of amnesty. If amnesty as commanded forgetting is not to lapse into pure amnesia, Ricœur insists, it must be treated in conjunction with the question of forgiving. Thus, Ricœur suggests substituting the command to forget with what he terms the “wish for a happy forgetting” (500).[8]

As we can see, forgetting as it is conceived in its complexity by Ricœur does not constitute the opposite of memory. Rather, memory and forgetting constitute each other in an intricate relationship with forgiving. Memory – forgetting – forgiving: this is Ricœur’s ethical trajectory.


There is still one element missing in this trajectory, however: narrative*. Already in the second sentence of the preface to Memory, History, Forgetting, narrative is explicitly mentioned when memory and forgetting are introduced as the “median levels” between “temporal experience and the narrative operation” (xv). In Time and Narrative, temporal experience had been directly coupled with the narrative operation: “time becomes human time to the extent that it is organized after the manner of a narrative” (Time and Narrative I 3)*. It is here that the discussion on memory and forgetting needs to be inserted, as memory denotes our only recourse to the past and “time indeed remains the factor common to memory as passion [mneme] and to recollection as action [anamnesis]” (Memory, History, Forgetting 18). One would thus have to rephrase the central assertion of Time and Narrative as follows: time becomes human time to the extent that memory and forgetting are organized after the manner of a narrative. Thus, it becomes clear that narrative has to occupy a central function in Ricœur’s discussion of memory and forgetting. And indeed, when he reminds his readers that anamnesis is, for Aristotle, characterized by the notion of search, Ricœur makes clear that this search aiming at the coherence of re-cognition has to be interpreted in the light of narrative:

But most importantly, it is in narrative that the search for coherence which motivates the effort of recollection finds its first articulation. In this way a connection is established between recollection as search and recollection as narrative. (“Erinnerung und Vergessen” 15)[9]

Recollection thus culminates in a fundamentally narrative reconstruction. It is important to note here that narrative for Ricœur essentially entails a mimetic relation. As mimesis, narrative is inextricably entangled with the whole problematic of representation. This is made explicit early on in Memory, History, Forgetting when Ricœur discusses Plato’s distinction between “eikastic” art and “fantastic” art, eikon and phantasma, ‘good’ and ‘bad’ mimetics in the context of memory’s truth claim and its relation to the notion of the image in general and that of imagination in particular. Ricœur’s own understanding of mimesis can be retraced in detail in his conceptualization of what he terms threefold mimesis in Time and Narrative. As narrative occupies such a central role in Ricœur’s theory of memory, it is worth recapitulating the essentials of his concept of narrative.

In the chapter on the threefold mimesis in Time and Narrative I (52-87), Ricœur conjoins his preceding separate analyses of the aporias of temporal experience in Book 11 of Augustine’s Confessions, and of Aristotle’s Poetics. In short, the circular threefold mimesis consists of the quasi-narrative prefiguration of our heterogeneous life-world experiences, the properly narrative configuration of these experiences consisting in a synthesis of the heterogeneous providing an intelligible whole, and finally the refiguration of this synthesis in its actualization within our life-world experiences. In what follows, I will have a closer look at each of these figurations in their interconnection, beginning with the prefiguration of mimesis1.

For Ricœur, in the wake of Aristotle, the “minimal narrative sentence is an action sentence” meaning that “[i]n the final analysis, narratives have acting and suffering as their theme” (Time and Narrative I 56). This presupposes a preliminary understanding of action in general, which is provided by what Ricœur terms practical understanding consisting of the capacity to identify and master the structural features, the symbolic mediations and the temporal elements of action. This understanding constitutes a whole “conceptual network” including goals, motives, agents, etc. (54-5). Regarding the structural features, narrative in general takes recourse to this conceptual network of action, but it does so in such a way as to transform the network: in narrative, the concepts of the network, which initially are related paradigmatically and synchronically, and thus basically remain unstructured, are inscribed into a syntagmatic and diachronic order.[10] But this narrativization of action is only possible because action is “always already articulated by signs, rules, and norms. It is always already symbolically mediated” (57). Drawing on Clifford Geertz, Ricœur in this context speaks of a texture of symbolic mediation. It is this texture which provides the grounds on which specific actions can be interpreted: “In this way, symbolism confers an initial readability on action” (58). To transform this “quasi-text” (58) into a text proper is the task of narrative. Regarding the third feature of temporality, it is exactly “the way in which everyday praxis orders the present of the future, the present of the past, and the present of the present […] that constitutes the most elementary inductor of narrative” (60).[11] Thus, according to Ricœur, it is primarily temporal experience which shows a prenarrative structure (59, 82ff.). The three factors practical understanding draws on – (un)structure, symbolism, temporality – with a clear emphasis on temporality, constitute the prefiguration of mimesis1, the prefiguration being grounded in the fundamentally prenarrative structure provided by these factors. This structure can be qualified as prenarrative since, although being essentially heterogeneous, it is not entirely chaotic but lends itself to narrativization.[12]Narrativization in turn means essentially ordering and structuring. The narrativization of the prefiguration constitutes the configuration of mimesis2, whereby practical understanding is transformed through narrative understanding. We thus turn to the configuration of mimesis2.

Otherwise remaining amorphous, everyday-life is emplotted and thereby ordered and structured in the configuration of mimesis2, the ensuing structure consisting in an “intelligible whole” (65). Even though this intelligible whole consists of individual events, these events do not remain separate successive occurrences since they are situated within the totality of a story. This is exactly what configuration means – emplotment resulting in an intelligible whole: “In short, emplotment is the operation that draws a configuration out of a simple succession” (65). This operation of emplotment is characterized by bringing together the heterogeneous factors which make up the aforementioned conceptual network of action, the inscription of these factors into a syntagmatic order, and the ensuing plot’s temporal characteristics. It is due to these temporal characteristics that Ricœur calls plot a “synthesis of the heterogeneous” (66). But on which grounds? As Aristotle does not explicitly treat temporality in this context, Ricœur resorts to Augustine and asserts that one can say of the “operation of emplotment both that it reflects the Augustinian paradox of time and that it resolves it, not in a speculative but rather a poetic mode” (66). This resolution of the paradox is attained through the combination of a chronological with a non-chronological dimension of time: the manifold episodic events (chronological dimension), by being emplotted, result in the configuration of a story and therefore in one temporal totality (non-chronological dimension). Thus, the act of emplotment denotes the solution to the Augustinian temporal paradox. This act of poiesis is revealed to the reader in what Ricœur terms the ‘followability’ of the story:

To follow a story is to move forward in the midst of contingencies and peripeteia under the guidance of an expectation that finds its fulfillment in the “conclusion” of the story. This conclusion is not logically implied by some previous premises. It gives the story an “end point,” which, in turn, furnishes the point of view from which the story can be perceived as forming a whole. To understand the story is to understand how and why the successive episodes led to this conclusion, which, far from being foreseeable, must finally be acceptable, as congruent with the episodes brought together by the story. (66)

Thus, it is from the perspective of the “end point” of the story, which gives it its intelligibility, that the paradox of Augustinian intentio and distentio can be said to be solved and transformed into a “living dialectic” (67). This means that a sense of temporal unity is only established because of the temporal characteristics of narrative. The paradoxical time of everyday-life becomes a temporal unity only insofar as narrative configures a beginning, middle, and ending. The configurational act of mimesis2 thus synthesizes and renders intelligible the heterogeneous and amorphous “quasi text” of the prefiguration of mimesis1. This configuration of mimesis2 must be in turn actualized in the refiguration of mimesis3. I will now sketch out this last step, which closes the circle of Ricœurdian mimesis.

With the refiguration of mimesis3, Ricœur stresses that it is not, in fact, until the act of reading that the act of configuration is completed. Referring to Wolfgang Iser’s theory of reading and Hans-Robert Jauss’ theory of reception, Ricœur emphasizes that the act of configuration is not to be located within the text only but rather in the interaction of text and reader. This is why Ricœur in this context speaks of a Gadamerian fusion of horizons between the world of the text and the world of the reader (77). Thus, one could say that the refiguration of mimesis3 takes place in this fusion of horizons. In all this Ricœur assumes a referent outside of language, namely a world: “What a reader receives is not just the sense of the work, but, through its sense, its reference, that is, the experience it brings to language and, in the last analysis, the world and the temporality it unfolds in the face of this experience” (78). In the act of receiving a narrative and the corresponding fusion of horizons, the reader experiences an enlarging of their horizon of existence, which in turn has its impact on actions in the practical world of everyday-life. It is exactly this enlarging of one’s horizon of existence and its ensuing impact which designates the refiguration of mimesis3. Since in all these figurations the main focus is on temporal experience, one can say in conclusion that “making a narrative [le faire narratif] resignifies the world in its temporal dimension, to the extent that narrating, telling, reciting is to remake action following the poem’s invitation” (81).[13] Thus, this is the trajectory of threefold mimesis: from the prefiguration of heterogeneous-paradoxical temporal experience via the configuration of this experience in the synthesis of the heterogeneous by means of emplotment, to the actualization of this synthesis in the fusion of horizons between the world of the text and the world of the reader in refiguration. Correspondingly, refiguration always also means prefiguration, thus completing the hermeneutical circle or rather spiral of Ricœurdian mimesis.

What Ricœur emphasizes in his theory of mimesis is basically the structuring capacity of emplotment regarding temporal experience. Thus, coming back to the problematic of representation concerning memory, it is no surprise that Ricœur treats narrative proper under the second set of problems, namely that of the temporal dimension. Accordingly, it is in the search of anamnesis where the workings of emplotment are to be located. The re-cognition of mneme then provides the “end point” from which to understand (cognize!) the story: “The end of the story is what equates the present with the past, the actual with the potential” (“Narrative Time” 186). One can say that the narrative character of recollection configures our memories into an intelligible whole. In other words: the search of recollection emplots our memories, rendering them coherent and intelligible. In light of narrative’s essentially selective character, such a constitutive function is highly problematic for the truth claim of memory, which Ricœur wants to retain at all costs. This is why he deals with the problem of selectivity in the framework of his discussion on the abuses of memory and the corresponding type of forgetting: it denotes one of the sources of memory’s potential unreliability. Selectivity does not per se preclude a true account or an account of truth, however. It may become difficult to recognize and to determine whether or not an account is true, but as long as selectivity does not directly impinge on the cognitive dimension, it remains possible in principle. One can still speak of truth and speak truthfully in relation to the past; the distinction between eikastic art and fantastic art, ‘good’ and ‘bad’ mimetics, remains fundamentally intact and is “vigorously affirmed” by Ricœur (Memory, History, Forgetting 12).

It is here that I want to pause and ask a few questions. The first concerns Ricœur’s concept of narrative: is it really appropriate to conceive of narrative as a synthesis of the heterogeneous and as a configuration resulting in an intelligible whole? The second aims at the function narrative occupies in Ricœur’s theory of memory: dealing with it explicitly only under the heading of anamnesis and in the context of the uses and abuses of memory and the corresponding type of forgetting, does Ricœur not in fact ascribe a much more pervasive role to narrative than he is inclined to admit?[14] In other words: if narrative qua recollection provides the grounds for recognition, if we re-cognize narratively, are we not in need of a thorough and explicit examination of narrative’s role within memory’s cognitive dimension of mneme? And finally: if I answer the first question with no, and the second with yes – what does that mean for Ricœur’s theory? In the following I will delineate some possible answers to these questions.

Mneme, Anamnesis, Mimesis

For heuristic reasons, I will approach the question of narrative as a more pervasive factor for Ricœur’s theory of memory from the perspective of forgetting. As to anamnesis, it should not be too difficult to see that all three types of forgetting can be explained in terms of narrative, namely that forgetting basically amounts to a silence created by the absence of narrative. This is already implicit in Ricœur’s understanding of recollection as narrative reconstruction: recollection provides a narrative where there has not been one before, where silence reigned supreme. The three different kinds of forgetting illustrate this point. Ricœur points out himself that an absence of alternative narratives in conjunction with what has been left out of the master narrative constitutes forgetting in the case of manipulated memory. However, what about the case of blocked memory? In his reading of Freud, Ricœur identifies repetition with forgetting since repetition as the dynamic of repression constitutes the main obstacle “along the path of recalling traumatic experiences” (Memory, History, Forgetting 70). Given that the Freudian work of remembering consists of the “search of a truthful relation to [one’s] past” (73), and given that this search is the search of recollection and that recollection finds its articulation first and foremost in narrative – that recalling fundamentally means (re)telling – what does repetition as forgetting denote if nothing other than a fundamental silence, an absence of narrative in relation to the traumatic experience? Otherwise, why should the psychoanalytic operation be a matter of “allowing the representatives of the unconscious to speak and, in this way, as far as is possible, to ‘tell all’” (88), as Ricœur himself asserts?[15] If recollecting and thus verbalizing, (re)telling the traumatic experience, is the remedy for compulsory repetition, it is so because repetition here denotes a fundamental silence in relation to the traumatic experience. Turning to the third type of forgetting, that which corresponds to the practical dimension of anamnesis and commanded forgetting, one similarly encounters an absence of narrative as constitutive. “Do not recall!”: this is the injunction of commanded forgetting such as in the institution of amnesty. That this injunction implies the injunction ‘not to tell’ should be clear by now. This is exactly the reason why Ricœur wants to see the appeal “narrare aude!” joined to enlightenment’s “sapere aude!” (449).

Directing our attention now to the dimension of mneme, one must remember that Ricœur describes forgetting as the erasure of traces. This causes a major problem concerning the affection-impressions and Ricœur therefore, turning to Bergson, suggests speaking of an unconscious duration of memory. However, is this explanation not similar to the point of being identical in terms of structure with the notion of forgetting in the discussion on the workings of repression in blocked memory? Even if unconscious duration does not always amount to repression, repression is never anything other than unconscious duration. Ricœur himself explicitly establishes the link between Freud’s and Bergson’s respective notions of the unconscious (445). Thus, if forgetting concerning the workings of repression can be explained in terms of the silence created by an absence of narrative, then I cannot see why forgetting as the Bergsonian unconscious duration of memory should not be explained in the same manner. The unconscious survival of affection-impressions in forgetting as the unconscious duration of memory then requires narrative in order to attain the level of consciousness, to escape the realm of silence. Otherwise these affection-impressions remain forgotten. Lyotard, in a reading of Freud’s case of Emma, seems to argue in a similar vein when stating the following:[16]

With the idea of original repression, I return to Freud’s question to Husserl. An event (an excitation) occurs at T2. There is no representative trace of this event in the vertical series of T´2, T´´2, etc. The psyche (of Emma at T2) does not then have representations of the event. These images are not merely too confused or too pale, they are not at all. In the place of the vertical line, then, a blank – T2 is forgotten straight away. It is not inscribed in the representative order. The same can also be said in mechanical language: the energy introduced by the excitation at T2 is not and has not been tied up in representative formations, neither consciously nor unconsciously. […] It [the psychic appearance] has been affected without the power to imagine this affectation, that is to say, in good Freudian doctrine, without the power to control and “liquidate” it. […] And, finally, according to the problematic of the Repräsentanz: this affect is how the excitation is present, i.e., as a cloud of energy not entirely fixed in psychic appearance but also not “free” either. The affect is present but not represented. (“Emma” 32)

If this presence does not indicate the presence of representation, what presence does it indicate? How is the affect present? Speaking with Ricœur and Bergson, if the survival of affection-impressions, the survival of images, is not a survival in representation, how can it be explained? It can be explained with absence. It is absence which is present, a “blank,” a gap. Accordingly, Lyotard says* a few lines further: “The phrase of affect ‘says’ it is something, as Da, here and now, inasmuch as this something is nothing, not meaning, not referent, not address” (33). In the Bergsonian terms Ricœur adopts, this means that the survival of images paradoxically depends on their absence. They are not even present unconsciously. What does loom in the unconscious, however, is the “blank,” the gap indicating their absence. Such a reading can also be drawn from Ricœur’s own text, as the following passage dealing explicitly with Bergsons’s notions of virtuality and pure memory indicates:

It is also then in the work of recollection that this operation of putting the “pure memory” into images can be grasped in its origin. We can speak of this operation only as a movement from the virtual to the actual, or again as the condensation of a cloud or as the materialization of an ethereal phenomenon. Other metaphors suggest themselves: movement from the depths to the surface, from shadows to the light, from tension to relaxation, from the heights to the lower levels of psychical life. Such is the “movement of memory at work” (171 [Ricœur quot. Bergson]). It carries memory back so to speak into a region of presence similar to that of perception. (Memory, History, Forgetting 52).

Pure memory needs to be actualized, carried back into presence, re-presented. To speak with Lyotard: although there, Da, it is nothing. It can thus be read as the presence of an absence of representation. It is exactly this presence of an absence of representation, this something as nothing, which points towards an understanding of forgetting in Lyotard: “To have forgotten is to say that she [Emma] does not possess, through what follows and up to T0, any representation of scene 2 or of the excitation of which it consisted (the temptation)” (“Emma” 35). Thus one can say that forgetting amounts to the silence of affects. This is why Lyotard, too, finally speaks of the “silent ‘presence’ of the affect.” The whole passage reads: “In truth, the silent ‘presence’ of the affect, a sigh, demands of articulated language an endless series of stagings, novels, tragedies, epics, an accumulation and linking of articulated phrases which are contradictory, undecidable, very numerous, or, at least, very ‘fair’” (43). Two things are remarkable here: first, remembering for Lyotard apparently means the accumulation of articulated phrases where there has been silence, in short the filling of the blank of forgetting. This is decidedly different from Ricœur’s understanding of remembering as the bringing to light of unconscious images present in a state of latency. Lyotard’s notion of remembering can thus not be discussed under the aegis of representation: what should these articulations represent if all they can build on is “nothing, not meaning, not referent, not address”? Second, the accumulation of articulated phrases demanded by the blank of forgetting is nothing other than the production of “stagings, novels, tragedies,” in short – narratives. The silent presence of the absence of representation in the survival of affection-impressions which is forgetting demands its articulation in the telling of stories. Again Ricœur’s text, against his general line of argumentation, seems to suggest a similar reading. This is how the passage quoted above continues:

But – and here we reach the other side of the difficulty – it is not just any sort of imagination that is mobilized. In contrast to the function of derealization, culminating in a fiction exiled to the margins of reality considered in its totality, what is celebrated here is instead the visualizing function of imagination, its manner of giving something to be seen. On this point what unavoidably comes to mind is the final component of the muthos that, according to Aristotle’s Poetics, structures the configuration of tragedy and epic, namely, the opsis, held to consist of “placing before the eyes,” showing, making visible. This is also the case when “pure memory” is put into images. (Memory, History, Forgetting 52).

Ricœur is so intent on disentangling the problem of memory from that of imagination – especially imagination as the ‘bad’ mimetics of fiction – that this prevents him from a proper and more exhaustive consideration of narrative and narrativity, even though everything is already there in his text: considering the narrative structure of recollection, the “placing before the eyes,” the showing effect of the opsis of Aristotle’s muthos [plot], suggests incorporating the problematic of the image with respect to memory within a treatment of the problematic of narrative and narrativity as the images provided by the work of recollection, rather than being self-sufficient, are always already inscribed within a narrative structure. With Freud, one could thus say that Ricœur’s almost obsessive preoccupation with the notions of image and imagination denotes nothing more than an acting out indicating the repression of the much more pervasive role of narrative for his theory of memory. That Lyotard in turn even speaks of an endless series of articulations and that these are contradictory and undecidable, already points towards an understanding of narrative which I will elaborate on shortly. For the moment I want to emphasize that with the reading I propose here, forgetting as the unconscious duration of memory, rather than signifying an immemorial resource, amounts to the silence of absent narrative.[17]

The same can be said of the erasure of cerebral and documentary traces: in these cases, there is simply nothing left to tell about. At best, one can report their destruction and absence: the medical case of dementia, the scrolls of the library of Alexandria… It is in this way that narrative is also constitutive for memory’s cognitive dimension of mneme. This raises significant problems considering the selective character of narrative. These problems are further accentuated to the degree of impasse if one denies the synthesizing powers of narrative. That Ricœur conceptualizes narrative as synthesis and as configuration in the first place has to do with the desire to find a remedy for Augustine’s disturbing temporal paradoxes. Indeed, Ricœur finds a powerful ally in Aristotelian narrative theory: besides his extensive references to Aristotle’s Poetics, he relies heavily on Czech and Russian formalists as well as on French structuralists.[18] This is evident in the importance of plot for Ricœur. Besides highlighting the synthesizing powers of emplotment time and again, Ricœur also relies on the notion of plot for his understanding of the literary tradition, for according to him, it is a typology of plots that allows us to write a history of literary genres and forms. Innovation amounts to nothing more than deviation from these forms and genres (cf. Time and Narrative I 68-70). However, Ricœur is bound to face unsurpassable problems in the face of so many literary narratives, which so obviously do not correspond to his conception of narrative (or literature), as these works deny both synthesis and the notion of an intelligible whole. One only needs to think of the literary experiments of the 1960s and 1970s. Ronald Sukenick, for instance, in “Death of the Novel” succeeds in simultaneously disseminating several narratives without subordinating one to the other, therefore escaping any notion of wholeness. Similarly, hypertext fiction from its early modernist manifestations in works such as Cortázar’s Hopscotch, Coover’s postmodern “Elevator,” and further refinements in works such as Ana Castillo’s Mixquiahuala Letters, defies synthesis and wholeness incorporating what Ricœur calls selectivity into a ‘single’ narrative causing its irreducible multiplication. Accordingly, these narratives do not present a definite beginning and a definite ending. Or consider genre-bending work such as Michael Ondaatje’s The Collected Works of Billy the Kid, which won a poetry award when it was first published and is sold as a novel today. Thomas Pynchon’s The Crying of Lot 49 is a last and somewhat more conventional example in terms of plot, which defies the narrative closure so dear to Paul Ricœur. Indeed, Ricœur fears these narratives and sees in them nothing less than the advent of the “death of narrative form itself” (Time and Narrative I 70).[19] This is necessarily so, as for Ricœur narrative denotes our capacity to synthesize our essentially paradoxical temporal experience so that we can obtain a certain orderliness, whereas the examples mentioned fundamentally stress the aporetic and the paradoxical. Thus, Ricœur tries to incorporate these ‘dangerous’ narratives into his theory:

By means of the frustrations engendered by their ironic mistrust of any paradigm, and thanks to the more or less perverse pleasure the reader takes in being excited and gulled by them, these works satisfy both the tradition they leave behind and the disorganized experiences they finally end up imitating by dint of not imitating the received paradigms. (73)

It is telling that Ricœur, who otherwise proceeds in a very careful and thorough manner, resorts to nothing more than polemics here: with expressions such as “frustrations,” “perverse pleasure,” “excited and gulled,” and “disorganized experiences” Ricœur discredits in a sweeping movement everything which does not conform to his notions of literary tradition and narrative. The works themselves, however, do not cease to defy intelligible wholeness and to stress the aporetic-paradoxic and thus constitute Ricœur’s fiercest nightmare turned real, namely that nothing “excludes the possibility that the metamorphosis of the plot will encounter somewhere a boundary beyond which we can no longer recognize the formal principle of temporal configuration that makes a story a whole and complete story” (Time and Narrative II 28). The Ricœurdian account of temporal configuration cannot be extracted from this kind of works as these works are fundamentally aporetic in nature. In a last sortie, Ricœur says about them:

[I]t is the “discordance” engendered in discourse by the ironic distance in regard to any paradigm that undermines from within the view of “concordance” sustaining our temporal experience and that overthrows the intentio without which there would be no distentio animi. We can then legitimately suspect the alleged discordance of our temporal experience as being only a literary artifice. (Time and Narrative I 73)

Against Ricœur, one could say that above all he presupposes a general wish for concordance, which the aforementioned works cannot and do not want to fulfill as their concern is not concordance, but the paradoxical and aporetic.[20] Ricœur’s general presupposition of a wish for concordance, of course, contradicts such a concern. That emphasizing discordance subverts such a wish for concordance is more than congruent with the concern for the paradoxical and aporetic. Rather than advocating the rule of pure discordance, or chaos, the emphasis placed on this rule is directed against a primacy of concordance and synthesis. Emphasis on discordance invigorates the paradoxical and aporetic. This is the whole point. The ironic distance towards any paradigm, in turn, is a strategic tool used to escape established orders. One could thus suspect that the alleged synthesis of the heterogeneous is a mere philosophical-literary artifice. The works which defy intelligible wholeness and highlight the paradoxical and aporetic call for a fundamentally different concept of narrative, a concept of narrative which does not synthesize but does justice to the paradoxical and aporetic as such.

With such a concept of narrative the truth claim of memory, understood as faithfulness towards the past as Ricœur envisions it, would be thwarted.[21] The reason for this, however, is not so much to be found in the “specter of the bad ‘mimetics’” (Memory, History, Forgetting 57) and the phantastic, but rather in the internal workings of the operation of narrative, whether fictional or factual in intention. What all of the aforementioned literary works highlight, and here we are coming back to Lyotard’s remarks from “Emma,” is an essential “contradiction” and “undecidability” within narrative. In short: narrative itself is aporetic. This is why there needs to be an endless series of “stagings, novels, tragedies” and this is why memory in the singular and memories in the plural will remain unreliable in essence. Not because they are too close to imagination, but because they are constituted by narrative, which places a differend, to borrow Lyotard’s central term, between mnemonic intention and mnemonic operation. My critique of Ricœur is thus not directed at the fundamentally narrative character of memory; quite on the contrary, I hope to have shown that narrative is even more pervasive in and more fundamental for Ricœur’s theory than he himself is prone to admit.

My critique instead aims at the character and function of this narrativity. Where Ricœur tries to overcome the aporias of temporal experience with his notion of narrative, comprising synthesis, configuration and intelligible wholeness, I think it is necessary to draw attention to and emphasize the aporetic. Where Ricœur asks us to take recourse to our literary tradition and its rules of composition in order to create stories that find their fulfillment in the conclusion, I see the necessity to emphasize the ‘unconcluded’. To quote Lyotard:

A postmodern artist or writer is in the position of a philosopher: the text he writes, the work he produces are not in principle governed by preestablished rules, and they cannot be judged according to a determining judgment, by applying familiar categories to the text or to the work. Those rules and categories are what the work of art itself is looking for. The artist and the writer, then, are working without rules in order to formulate the rules of what will have been done. (“What is Postmodernism?” 81)

In this sense, memory constitutes what will have been done by moving back through time.


1. As Adolf Polti points out, dealing with Ricœur can easily lead to an unintended Ricœurdian philology since Ricœur continuously reworks his topics and thus establishes an immense network of related texts which cover the same topics from different angles (cf. 234). In order to avoid too much of a philology, I only resort to these other texts when they make an additional point or contribute in a particularly interesting or different way.

2. This problem of representation provides a consistent thread running through the whole book since the problems concerning the truth claim of memory hinge on the particular relation of presence and absence within representation. Ricœur discusses this most extensively in his chapters on Plato and Aristotle (cf. Memory, History, Forgetting 7-21), and the specific relation of memories to images (cf. 44-55).

3. The third type of trace, the archival document, is of major importance in Ricœur’s discussion on history and historiography. In the context of this article only the first two types of traces will be of interest.

4. Already in “Erinnerung und Vergessen” Ricœur states that he distinguishes the two dimensions for “the purposes of analysis” [“zum Zweck der Analyse”] (11).

5. Ricœur links the work of mourning to the work of remembering: the work of remembering actually amounts to a work of mourning, and the work of mourning always entails the work of remembering (cf. Memory, History, Forgetting 72; 77).

6. Ricœur’s reference is to Heidegger’s notion of “das Unvordenkliche” prior to whose terms there is no thinking. Immemorial resource thus denotes that prior to whose terms there is no remembering.

7. The reference is to the beginning of Nietzsche’s “On the Uses and Disadvantages of History for Life.”

8. Happy forgetting thus denotes the corresponding term to the “happy memory” of the conjoined works of remembering and mourning (Memory, History, Forgetting 77).

9. My translation. This text is the publication of a lecture Ricœur gave at the University of Ulm when holding the Humboldt professorship during the winter term 1998/99. In German the passage reads as follows:

Aber vor allem findet die Suche nach Kohärenz, welche die Anstrengung der Wiedererinnerung motiviert, in der Erzählung ihren ersten Ausdruck. So entsteht eine Verbindung zwischen der Wiedererinnerung als Suche und der Wiedererinnerung als Erzählung.

In Memory, History, Forgetting a similar statement interestingly appears in the context of a discussion on Augustine, which in turn is embedded in an examination of the “tradition of inwardness” which Ricœur undertakes before turning to collective memory. Here too, it appears when broaching the issue of memory’s capacity “to move back through time”: “It is primarily in narrative that memories in the plural and memory in the singular are articulated, and differentiation joined to continuity” (96).

10.. Of course, it is only with the configurational act of mimesis2 that this transformation takes place. As we will see shortly, the notion of order is of major importance here.

11. Ricœur refers to Augustine’s distentio animi here.

12. Compare Chung’s notions of prefiguration as regular disorder (gesetzmäßige Unordnung) and disordered structure (ungeordnete Struktur) (51).

13. I would rather suggest translating “le faire narratif” with “the makings of narrative.” That Ricœur speaks of poem here, has to do with the fact that this statement appears in the context of a reference to his discussion of poetry in The Rule of the Metaphor. Still, all works of narrative character are being meant here.

14. In “Memory – Forgetfulness – History” Ricœur seems to assign narrativity to memory in general when he speaks of the “[n]arrative structure, which is common to memory and to history.” However, the whole passage reads:

Speaking about memory necessarily means speaking about forgetfulness, because one cannot remember everything. A memory with no gaps would be an unbearable burden; it is a cliché to say that memory is selective. Narrative structure, which is common to memory and to history, confirms this law of the necessity of forgetfulness. (“Memory – Forgetfulness – History” 9)

Thus, Ricœur here too treats narrative in the context of selectivity and the respective type of forgetting.

15. My italics.

16. The designations T2, T0 and scene 2 in the extracts from Lyotard’s text refer to the following chain of occurrences: the initial event of eight year old Emma being sexually harassed in a shop (scene 2 at T2); that of finding herself in a shop alone with the clerk at the age of thirteen which triggered her phobia of entering shops alone (scene 1 at T1); and that of Emma on Freud’s couch (scene 0 at T0).

17. In answer to the possible objection that my short discussion of Lyotard is only based on a few remarks and that these remarks are considered completely out of context, I ask in turn, whether this objection would be legitimate? The context is a discussion placed in-between philosophy and psychoanalysis (the English subtitle) and is concerned with emphasizing “the role of the differend that lies between adulthood and childhood” (Silverman 3). This would seem to have quite a lot to do with Ricœur’s Bergsonian elaborations on forgetting as an immemorial resource. Ricœur, too, places his examination in-between psychoanalysis and philosophy bringing together Bergson and Freud. In fact, when establishing the notion of forgetting as immemorial resource, he gives a philosophical reading of the psychoanalytic concept of the repression of traumatic events. Lyotard, in turn, when elaborating on the differend between adulthood and childhood does nothing other than examining the workings of Freudian repression in the light of philosophy.

18. This is apparent in his elaboration of the threefold mimesis in Time and Narrative and made more than explicit in “Life in Quest of Narrative”:

The narrative theory I shall now be discussing is at once very recent, since in its developed form it dates from the Russian and Czech formalists in the twenties and thirties and from the French structuralists of the sixties and seventies. But it is also quite ancient, in that it can be seen to be prefigured in Aristotle’s Poetics. (20)

19. It is more than telling that Ricœur speaks of schism in this context, suggesting that these narratives apostatize from the right belief.

20. That Ricœur’s ultimate wish is to establish a primacy of concordance over discordance is indisputable. Of course, he should also not be mistaken for wanting to establish a tyranny of pure order. “Emplotment is never the simple triumph of ‘order’” (Time and Narrative I 73). After all, emplotment must incorporate the contingencies of the peripeteia. However, in incorporating and/or bracketing them, it inscribes the contingencies within an intelligible whole. Concordance thus gains the upper hand over discordance: not by doing away, so to speak, but by subduing.

21. I am aware that a disorderly form of representation might be regarded as the only adequate form for a world perceived as disorderly: thus, one would still remain faithful. This, however, besides the fact that the world is not to be equated with pure disorder for Ricœur, as I have shown in my consideration of his threefold mimesis. This is not what Ricœur is aiming at when he speaks about faithfulness to the past. In a disorderly representation the aporias of time, which narrative is called upon to disentangle, would essentially remain unresolved. A disorderly representation of disorder offers no remedy. In addition to an adequate representation, faithfulness to the past fundamentally denotes an ordering and thus orderly representation. This presupposition, of course, also provides the grounds for Ricœur’s formulation of what I have termed his ethics of forgiving. Since narrative is constitutive for his notions of memory and history and subsequently for his ethics, which in turn requires an openness to the narratives of others and thus involves a negotiation of narratives, my repudiation of his notion of narrative renders his ethical program certainly more difficult, to say the least. If the narratives in negotiation cannot be faithful narratives in the Ricœurdian sense, the whole issue of negotiation must be reconsidered.

Works Cited

Castillo, Ana. The Mixquiahuala Letters. New York: Doubleday, 1992 [1986].

Chung, Ki Cherl. Der hermeneutische Ansatz zu einer Theorie der Erzählung bei Ricœur. Bochum: Universitätsverlag Dr. N. Brockmeyer, 1994.

Coover, Robert. “The Elevator.” Pricksongs & Descants: Fictions. 1969. New York: New American Library, 1970. 125-137.

Cortázar, Julio. Hopscotch. 1963. New York: Pantheon Books, 1987.

Lyotard, Jean François. “An Answer to the Question: What is Postmodernism?” 1982. Trans. Régis Durand. The Postmodern Condition: A Report on Knowledge. Minneapolis: University of Minnesota Press, 1984. 71-82.

Lyotard, Jean François. “Emma: Between Philosophy and Psychoanalysis.” 1989. Trans. Michael Sanders, Richard Brons and Norah Martin. Lyotard: Philosophy, Politics, and the Sublime. Ed. Hugh J. Silverman. New York: Routledge, 2002. 23-45.

Nietzsche, Friedrich. “On the Uses and Disadvantages of History for Life.” 1874. Trans. R. J. Hollingdale. Untimely Meditations. Ed. Daniel Breazeale. Cambridge: Cambridge University Press, 1999. 57-123.

Ondaatje, Michael. The Collected Works of Billy the Kid. 1970. New York: Vintage, 1996.

Plato. Theaetetus. Ed. and trans. John McDowell. Oxford: Clarendon Press, 1991.

Polti, Adolf. “Zur Rezeption und Kritik von Zeit und Erzählung.” Metageschichte: Hayden White und Paul Ricœur: Dargestellte Wirklichkeit in der europäischen Kultur im Kontext von Husserl, Weber, Auerbach und Gombrich. Ed. Jörn Stückrath and Jürg Zbinden. Baden-Baden: Nomos, 1997. 230-253.

Pynchon, Thomas. The Crying of Lot 49. 1965. London: Vintage, 2000.

Ricœur, Paul. “Erinnerung und Vergessen.” Erinnerung – Entscheidung – Gerechtigkeit. Ulm: Humboldt-Studienzentrum, 1999. 11-30.

Ricœur, Paul. “Life in Quest of Narrative.” On Paul Ricœur: Narrative and Interpretation. Ed. David Wood. London: Routledge, 1991. 20-33.

Ricœur, Paul. “Memory – Forgetfulness – History.” ZiF: Mitteilungen 2 (1995): 3-12

Ricœur, Paul. Memory, History, Forgetting. 2000. Trans. Kathleen Blamey and David Pellauer. Chicago: University of Chicago Press, 2004.

Ricœur, Paul. “Narrative Time.” Critical Inquiry 7.1 (1980): 169-190.

Ricœur, Paul. The Rule of the Metaphor: Multi-disciplinary Studies of the Creation of Meaning in Language. 1975. Trans. Robert Czerny. London: Routledge & Kegan Paul, 1978.

Ricœur, Paul. Time and Narrative. Vol. I. 1983. Trans. Kathleen McLaughlin and David Pellauer. Chicago: University of Chicago Press, 1984.

Ricœur, Paul. Time and Narrative. Vol. II. 1984. Trans. Kathleen McLaughlin and David Pellauer. Chicago: University of Chicago Press, 1985.

Silverman, Hugh J. “Introduction: Jean François Lyotard – Between Politics and Aesthetics.” Lyotard: Philosophy, Politics, and the Sublime. Ed. Hugh J. Silverman. New York: Routledge, 2002. 1-19.

Sukenick, Ronald. “The Death of the Novel.” The Death of the Novel and Other Stories. 1960. New York: Dial Press, 1969. 41-102.