Perception of duration presupposes duration of perception - or does it? Husserl and Dainton on time

1

Originally published in International Journal of Philosophical Studies 15/3, 2007, 453-471. Please quote from the published version. Dan Zahavi Danish National Research Foundation: Center for Subjectivity Research University of Copenhagen

Time is the substance I am made of. Time is a river which sweeps me along, but I am the river; it is a tiger which destroys me, but I am the tiger; it is a fire which consumes me, but I am the fire. Jorge Luis Borges

Perception of duration presupposes duration of perception – or does it?

Husserl and Dainton on time In the introduction to his Vorlesungen zur Phänomenologie des inneren Zeitbewusstseins, Husserl remarks that “we get entangled in the most peculiar difficulties, contradictions, and confusions” (Hua 10/4) the moment we seek to account for time-consciousness. I think most scholars of Husserl’s writings on these issues would agree. Attempting to unravel the inner workings of time-consciousness can indeed easily induce a kind of intellectual vertigo. In order to minimize the peril, I will in the following adopt a strategy of simplification. The literature abounds with very sophisticated and highly technical discussions of Husserl’s writings on time, but I think that it occasionally can pay off to return to some of the quite simple questions that motivated Husserl to commence his decades-long reflections on this topic. My somewhat indirect way of doing this will be by way of a critical discussion of Barry Dainton’s recent book The Stream of Consciousness. In his book, Dainton provides what must surely count as one of the most comprehensive discussions of time-consciousness in analytical philosophy, particularly within the last 10 years. In the course of doing so, he also challenges Husserl’s account in a number of ways. By looking at these challenges I think it is possible to gain a new perspective on why an analysis of time-consciousness is so important. Moreover, such a comparison might also provide us with an illuminating appraisal of the contemporary relevance of Husserl’s analysis. How does it measure up against one of the more recent analytical accounts? 1. Presenting the problem In his book Dainton sets out to provide an account of the unity and continuity inherent to our stream of consciousness. Experiences never occur in isolation, and a stream of consciousness is an ensemble of experiences that is unified both at and over time, both synchronically and diachronically. Dainton considers his approach phenomenological – in the loose sense of the word – since it is based on experiential findings; it is conducted from the first-person perspective, and it disregards the various subpersonal processes and mechanisms that might also be involved (Dainton 2000, xiii). Despite his appeal to phenomenology, Dainton defends a representative theory of perception and denies that veridical perception is a direct unmediated awareness of outer objects (Dainton 2000, 14, 18). According to the “projectivism” he favours, my object of perception is in fact a mental projection, i.e., an “internally generated and outwardly projected phenomenal image” (Dainton 2000, 16). As a consequence, “there is a sense in which we are all enclosed in spheres of virtual reality, phenomenal spheres somehow produced by activities within our brains: all we are directly aware of are the contents within these spheres” (Dainton 2000, 18). Needless to say, this is a position that classical phenomenology – with good reasons – would oppose. But although the merits of Dainton’s analysis of time-consciousness are to some extent independent of his commitment to this projectivism, we nevertheless need to keep this commitment in mind, if we are to understand Dainton’s somewhat idiosyncratic use of the terms phenomenal “content” and “object”. The terms are used more or less synonymously (and when accounting for Dainton’s view, I will do so as well) to refer to that which we are phenomenally aware of, be it melodies, spoken lines of poetry, the blue sky or approaching buses. All of these objects of perception seem to be out there in the world.

2

According to Dainton, however, they are in reality wholly experiential items; items that are parts and parcel of the stream of consciousness (Dainton 2000, 17, 24). Now, to get us started, let me quickly sketch out what might be considered a simple default account of time-consciousness. Pre-theoretically we all assume that we do have a direct experience of change and persistence. We can hear an enduring tone or a melody, just as we can see a stationary pyramid or the flight of a bird. However, if I at any given moment were only aware of what were perceptually present to me right then and there, how could I then ever perceive – in contradistinction to imagine, remember, or judge about - temporally extended objects? The solution is to recognize that our auditory and visual perceptions are themselves temporally extended processes. In fact, in the perception of a movement or a melody, there will be a succession of perceptual presentations that runs in parallel with the successive phases of the movement or melody. The perception of the melody starts when the melody starts, and comes to an end at exactly the same moment as the melody. Thus according to this account, we should subscribe to what has occasionally been called the Principle of Presentational Concurrence (Dainton 2000, 134). Unfortunately, however, things are not quite that simple. If a perception has duration and temporal extension, it will contain temporal phases of its own. But it is on closer consideration obvious that a mere succession of such conscious phases will not as such provide us with consciousness of succession. For that to happen, the succession of these phases must somehow be united experientially. The decisive challenge is then to account for this unification without giving rise to an explanatory regress. In order to avoid that, many have been tempted to adopt what Dainton calls the Principle of Simultaneous Awareness (Dainton 2000, 133). According to this principle, a sequence or succession of temporal phases is experienced only as a sequence or succession if it is apprehended simultaneously by a single momentary act of consciousness (Dainton 2003, 17). Why do we need to postulate a momentary act that embraces the full temporal sequence? We need a momentary act, because if it was extended, we would once again be confronted with the problem that an enduring consciousness is not as such a consciousness of duration. When we are aware of something temporally extended, something that includes the immediate past, the awareness itself must consequently be located in the present; it must be point-like and momentary (Dainton 2000, 133). The Principle of Simultaneous Awareness obviously doesn’t deny that there is a difference between hearing three succeeding tones and hearing the three tones simultaneously. The principle simply claims that the succession in order to be apprehended as a succession must be apprehended as a whole in a single momentary awareness; an awareness that is located in the pure now understood as an indivisible point or instant. If one opts for this model, one still has the choice between two different versions of it. One option is to hold that a momentary act of awareness apprehends a succession of content with real temporal extension. On this view, an act of awareness may be momentary, but its scope is not. Since this view holds that we are directly aware of temporally extended occurrences, one can call it a form of temporal realism. However, one can also take the view that the contents apprehended by momentary acts of awareness are themselves momentary, but that these contents simply appear as temporally extended. When one posits a momentary act, the content seized by this act must be given simultaneously. However, the different temporal phases of a temporally extended object are obviously not given simultaneously. Whereas the current phase of the object can be given perceptually, the former phase of the object is no longer present, and must therefore instead be re-presented when the current phase occurs. Thus, whereas we seem to be directly aware of temporally extended occurrences, we are in reality only aware of the representations of such occurrences (Dainton 2003, 8). One conclusion drawn by many advocates of this position – which might be labelled representational anti-realism – is that a perception of a temporal process is impossible. Our awareness of a temporal sequence is always representational. It is based on the simultaneous givenness of a manifold of contents that functions as representations of a temporally extended or distributed object. Our representational awareness of temporally distributed objects consequently lacks the directness and immediacy that characterize perceptual presentations. According to Dainton, the realist version of the Principle of Simultaneous Awareness is faced with a difficulty that he calls the problem of repeated contents. The scope of any act of awareness must be limited. Let us for the sake of the argument suppose that it is limited to the apprehension of two succeeding notes, and let us then take the awareness of the sequence of the three tones Do-Re-Mi as example. First, we will have an act that apprehends Do-Re and then another act that apprehends Re-Mi. If we in line with the Principle of Simultaneous Awareness suppose that these two momentary acts are distinct, we will have to accept that the same content ends up being experienced twice. But that is of course not true to experience. We don’t hear Re twice, we only hear it once (Dainton 2000, 141). When it comes to the anti-realist version of the principle, it can avoid the problem of repeated content by appealing to temporal modes of givenness. One and the same content is never given twice in the same

3

manner; rather every time it is given in different temporal modes, first as now, then as just-past, then as further-past etc., i.e., rather than repeatedly experiencing the same content in the same temporal mode of presentation, we will experience it as sinking smoothly into the past. However, despite this attempt at a solution, Dainton still argues that the anti-realist version must be rejected. It fails to provide us with a coherent and believable account of phenomenal temporality since it denies that we have a direct experience of change and succession (Dainton 2000, 115). Moreover, it is ultimately paradoxical. In order to be aware of successive objects, consciousness needs to compare the earlier and the later objects in an operation that makes the earlier and later simultaneous. But how can simultaneously presented objects be given as successive objects, how can they be both simultaneous and successive (cf. Gallagher 2003, 2)? 2. Dainton’s Husserl interpretation Dainton now continues his analysis by discussing Husserl’s theory of time-consciousness in some detail.1 To start with, Dainton readily acknowledges that the by far most sustained attempt to describe and understand temporal awareness can be found in Husserl’s various writings (Dainton 2000, 150). However, Husserl never managed to formulate a definitive account of his views on time-consciousness. His many writings on the topic are not easily summarized, some of them are markedly obscure, and as Dainton openly admits, he has had to abstain from expounding in any detail on those aspects of Husserl’s work that he has been unable to understand (Dainton 2000, 137). But Dainton then sets out to describe one of Husserl’s accounts, which he takes to exemplify a form of representational anti-realism (Dainton 2003, 53, 2000, 151). In his criticism of Brentano, Husserl had made it clear that we need to distinguish between directly experiencing change and duration and merely imagining or remembering it. There is a manifest phenomenological difference between seeing a shooting star, and remembering or imagining seeing a shooting star. At the same time, Husserl had also argued that Brentano’s theory failed to explain how a representation that is apprehended in the present can make us aware of something in the past. But as Dainton then points out, Husserl’s own theory is vulnerable to the very same criticism (Dainton 2000, 155). In his 1905 lectures, Husserl argued that the stream of consciousness consists of a succession of momentary experiences called primal impressions. According to Husserl, these momentary impressions were accompanied by simultaneously occurring sensory contents that were then supposed to be animated or imbued with sense by an act of apprehension in such a way as to appear just-past. Since Husserl took the animated content and the animating act to be simultaneous we would on his model not be directly aware of the past even though we seemed to be; rather each momentary experience would contain a representation of the preceding stretch of the stream. To put it differently, according to Dainton, Husserl took the scope of direct awareness to be confined to the momentary present (Dainton 2003, 54). In order to experience phenomenal duration and continuity we need the contribution of what Husserl calls the retention, but since the latter – again according to Dainton – is representational in character, Husserl is effectively denying that we are as directly aware of change as we are of, say, colour (Dainton 2000, 155). Dainton concludes his discussion of Husserl by conceding that the theory he has just described was in fact one that Husserl became unhappy with, and he admits that Husserl in his later writings abandoned the view that retentions are present-occurring representations of just-past contents and that he instead opted for the view that they provide us with direct, unmediated, access to the past (Dainton 2003, 55). But first of all, the retention of a past phase of an object is still supposed to exist in an experience which occurs after the phase in question, and how can something which is present present us with something in the past? Husserl’s answer, of course, would be that retentions are unique. Unlike primary impressions, retentions intend the past. Unlike memory, retentions present the past; they do not merely re-present it (Dainton 2000, 155). According to Dainton, this all sounds too good to be true and he argues that the explanation offered is verbal rather than real. In his view, Husserl simply stipulates that the retentions have the properties they need to have. Husserl tells us that something that occurs in the present can directly intend something in the immediate past, but he never explains in any detail or with any clarity how this is possible (Dainton 2000, 155-161). Moreover, according to Dainton, even in his later writings Husserl continues to ascribe a central role to durationless acts of awareness. This is testified in his diagrams of time, which always feature continuous series of momentary cross-sections of extended episodes of experiencing (Dainton 2003, 57). At each instant we are aware of some extended parts of the tone. Husserl does admit that these momentary cross-sections are nothing but ideal limits, but Dainton suggests that although Husserl might have held that momentary acts are dependent parts of extended phases of consciousness, for which reason they cannot exist in isolation, he still believed them to perform real functions within the extended acts. Thus, it is no

4

coincidence that Husserl continued to ascribe a central role to the momentary primal impressions. The primal impressions must be momentary, because if they had been temporally extended, there would have been no reason to introduce the retentions and protentions (Dainton 2003, 57). But if the primal impressions are momentary, they cannot provide us with an experience of change and duration. For that to happen, we need the full tripartite structure of retentions, protentions and primal impressions. But, according to Dainton, this prevents Husserl from doing full justice to the purely impressional continuity of consciousness. Primal impressions are originally present in a way that retentions and protentions are not. Consequently, on Husserl’s view, change and duration cannot be experienced as directly and immediately as colour or shape (Dainton 2003, 59). Finally, Dainton presents two further objections to Husserl’s account, which he calls the lingering content and the clogging of experience objection, respectively. In both cases, the basic problem concerns whether Husserl’s analysis of the intricate play between complexes of retentions, primal impressions, and protentions is the result of a proper phenomenological description or rather the upshot of a theoretical construction. Lingering content: If I snap my fingers, I hear the sound of the snap and then it is gone. The snap-sound does not linger on in my immediate experience. There may be a faint echo of the snap, but the echo is itself a sound that I am directly experiencing. According to Dainton, however, this is not what Husserl’s retentional model would lead us to expect. In Dainton’s reading, Husserl would argue that momentary experiences enjoy their moment of full consciousness, then they slowly slip away, they gradually become less and less present before they finally fade altogether. Only after they have left direct awareness altogether, can they appear in the guise of ordinary memory. But this does not seem to be true, since experienced content depart from immediate experience cleanly, leaving no residue, and it becomes immediately accessible to memory. Dainton admits that the lingering content objection is almost embarrassingly naïve, but he nevertheless insists that is should be taken seriously (Dainton 2000, 156-157). Clogging of experience: Husserl realized that we are continuously aware of the continuity of our experiences, and that this involves unity on two different levels. There is the continuity of the content of experience, and there is the continuity of the very awareness itself. Thus, at any given instant, we are aware not only of the present and past temporal phases of the object, but also of our present and past experience of the object (Dainton 2000, 154). In order to accommodate this fact, Husserl makes recourse to a very complex network of retentional continua, and Dainton claims that a consciousness containing this degree of internal complexity would be clogged with different contents to a nightmarish degree (Dainton 2000, 158). To put it differently, the account provided by Husserl of the most basic temporal structure of consciousness is in Dainton’s view a purely theoretical construction that goes far beyond the phenomenological data. 3. Dainton’s overlap model According to Dainton, most philosophers – and this includes Husserl – who have subscribed to the Principle of Simultaneous Awareness have also subscribed to the awareness-content model. In fact, the two views are natural partners. According to the awareness-content (or act-object) model of consciousness, consciousness is inherently bi-polar. Any experiencing has two components. On the one hand there is an empty and durationless pole of awareness, a pure locus of bare apprehension, with no phenomenal properties of its own. On the other hand, there is the phenomenal content that is presented to this awareness. These two components are situated on two different levels of consciousness, and the unity of consciousness is then said to consist in diverse contents falling under a single awareness (Dainton 2000, 41-42). The awareness-content model is not without problems, however. If we hold that phenomenal content cannot exist independently of awareness, this awareness must in some sense be directly responsible for bringing the diversity of different phenomenal characteristics into being. But it is difficult to understand how that is supposed to be possible, since the awareness is itself without any such characteristics. In fact, if awareness is wholly without phenomenal features, it is difficult to see what would be lost if it simply disappeared (Dainton 2000, 49-50). More generally, Dainton warns against the temptation to take experiences to be things we perceive or observe in essentially the same way as we observe and perceive ordinary physical things. Although we do have experiences, and although we are aware of them, we don’t become so by directing some special internal sensory-organ onto them (Dainton 2000, 45). This is where

5

Dainton presents his simple conception of experience. In his view, consciousness is inseparable from phenomenal content. Mental processes with discernible phenomenological characteristics are simply contents occurring in streams of consciousness. As he writes

when a given phenomenal item comes into being, it comes into being as a conscious experience; to be an experience it does not need to fall under any separate awareness [….]. In other words, contents are themselves intrinsically conscious, and hence – in a manner of speaking – they are self-revealing or self-intimating. […] I shall call this non-dualistic model of consciousness, the Simple Conception of experience (Dainton 2000, 57).

In order to understand the unity we find within experience we do consequently not have to look at anything above, beyond or external to experience itself (Dainton 2000, 236). A stream of water is a unified flowing whole, and so is the stream of consciousness. Since awareness and content are not separate, consciousness does not consist in an awareness of a passing stream; rather consciousness is the stream itself. As Dainton puts it, consciousness does not consist of a stream running beneath a spot of light, nor of a spot of light running along a stream; consciousness is the stream itself, and the light extends through its entire length (Dainton 2000, 237). We have a direct experience of temporally extended phenomena, and successive phases of the stream are welded together by nothing other than direct experience. This inter-experiential relation is sufficient; there is no need to introduce a separate act of awareness to bind the constituents of phenomenal presents into experienced unities (Dainton 2003, 26). According to the simple conception there is no distinction between consciousness and content (contents are intrinsically conscious), but since these inherently conscious contents are temporally extended, consciousness and content cannot fail to run concurrently.2 Thus, Dainton basically proposes that we should return to and adopt a modified version of the Principle of Presentational Concurrence, i.e., the principle that acts of awareness and their contents coincide in time (Dainton 2003, 33, 2000, 166). Only this principle can do justice to what Dainton considers a basic phenomenological axiom, namely that our experience of motion and change and endurance is just as direct and immediate as our experience of colour and sound (Dainton 2003, 5). To put it differently, Dainton takes it to be impossible to have a direct perceptual presentation of a temporal sequence if the content at the basis of this presentation as well as the perceptual presentation itself are not temporally extended. In short, Dainton takes us to be directly aware of duration and enduring content due to the enduring character of the acts themselves. When we hear the tonal sequence Do-Re, the content of this experience is the phenomeno-temporal pattern of Do-flowing-into-Re, and there is consequently no need to posit a point-like awareness which encompasses both tones (Dainton 2000, 180). This conception which argues that acts overlap to the same extent as their contents, supposedly allows Dainton’s realist model, which rejects the Principle of Simultaneous Awareness, to avoid the problem of repeated contents. Let us return to our awareness of a sequence of tones Do-Re-Mi. The first act apprehends Do-Re, the second act apprehends Re-Mi, but the sub-phase of the first act that apprehends Re is numerically identical with the sub-phase of the second act that apprehends Re. So Re is experienced only once (Dainton 2003, 18). Another way to make the same point is to call attention to the sequence of notes Do-Re-Mi; you hear Do-running-into-Re-running-into-Mi. Since we only hear Re once, we can conclude that the experiencing of Re in the earlier phenomenal present is numerically identical with the experiencing of Re in the later phenomenal present. And hence that we have just experienced two phenomenal presents that overlapped by virtue of possessing a common part (Dainton 2003, 39). It is consequently crucial to realize that the overlap model is not an overlap by superposition, but an overlap by sharing of common parts (Dainton 2003, 23).

Adapted from Dainton 2003

6

By arguing in this fashion, Dainton seeks to address the main weakness of the Principle of Presentational Concurrence: its failure to provide an explanation of how the unity of a temporally extended consciousness is established. Although the diachronic unity supposedly established by the overlap is a very short-term affair, it is, according to Dainton, capable of binding together the adjacent phases of the stream of consciousness – just like tiny links in a chain – thereby constituting the stream as a temporally extended whole (Dainton 2000, 113). 4. The temporality of time-consciousness In Dainton’s view, Husserl subscribed to the Principle of Simultaneous Awareness, and as result he had a hard time developing an adequate account of temporal experience (Dainton 2000, 136). But is this correct? If we look at Husserl’s early writings on time-consciousness, it is true that while criticizing Brentano’s account (Hua 10/18) and while advocating the possibility of an intuitive presentation of succession, Husserl also argued that the sensed contents are neutral with regard to time. The contents are devoid of temporal determinations, they are neither now, past, nor future, and they receive their temporal character from the acts of apprehension. Thus, the content might be considered as nontemporal material for the time-constituting apprehensions (Hua 10/417). Whereas the present phase of a temporal object is perceived by way of a present apprehension of a present content, the past and future phases of the object are co-perceived by way of present apprehensions of present, although modified and thereby no longer sensuously given, contents of apprehension. In short, whereas the perception of the now-phase of the object will be constituted through the animation of a certain sensory content by means of a “now-apprehension” (Hua 10/230), this perceptual consciousness will be accompanied by retentional and protentional apprehensions of modified content thereby providing a consciousness of those phases of the object that is no longer or not yet present. The weakness of this account is obvious. It remains too close to Brentano’s account, and is vulnerable to similar objections. The manifold of contents and apprehensions are all part of the actual phase of consciousness. But how, as Husserl himself asks, can a manifold of co-existing contents provide us with awareness of succession (Hua 10/323)? How can a present apprehension of a present content provide us with intuitive awareness of something just-past? Husserl eventually realized the deficiencies of his own model. While continuing to affirm the retentional-protentional structure of awareness, he ended up rejecting the content - apprehension schema. In order to appreciate Husserl’s later view, we first need to realize that the retention is not part of that which we are aware of, but a structural part of the very awareness itself. In short, the retention is not something we hear; rather we hear the just-past tone because it is retained by the retention. Moreover, the retention doesn’t accomplish this by retaining the tone in present consciousness. There is no simultaneity between the retention and that which is retained. The just-past tone doesn’t remain present in consciousness, like some reverberation; rather it is presented to consciousness as just-past, or as Brough has put it: “Retention does not transmute what is absent into something present; it presents the absent in its absence” (Brough 1989, 276). In short, the retention must be appreciated as a peculiar form of intentionality. It provides us with a direct intuitive grasp of the just-past, and is not a special apprehension of some present content. As Husserl writes, “retention is not a modification in which impressional data are really [reell] preserved, only in modified form: on the contrary, it is an intentionality—indeed, an intentionality with a specific character of its own” (Hua 10/118. Cf. 10/31). Thus, and this must be emphasized, being retentionally aware of the just-past phase of the object doesn’t entail having the just-past phase of the object sensuously co-present in some strange distorted way. To take a concrete example: If we look at a pedestrian who is crossing the street, our perception will not be restricted to capturing the durationless now-phase of his movement. Perceptually, it will not be as if he suddenly appeared as out of nowhere and we do not have to engage in an explicit act of remembering in order to establish the temporal context of his current position. Nor, however, will it be the case that all the previous phases of his movement are perceptually present in the same way as his current position. Had that been the case, the pedestrian would perceptually fill the entire space he had just traversed. But again, to exclude this is not to endorse the idea that the past phases of his movement remain visually present in some vague ghostly manner. Temporal “fading” into the past is not equivalent to the fading of an image that remains perceptually present. Rather, what we see is embedded in a temporal horizon. And its meaning is influenced by what went before and which is still intentionally retained. For the very same reason, and this is something that Gallagher has pointed out, Husserl is not vulnerable to the lingering content and the clogging of experience objections. Contrary to what Dainton is claiming, consciousness on the Husserlian model is not overloaded with sensory content. Rather, for the

7

just-past tone to be intentionally retained is for its meaning or significance to be retained as just-past (Gallagher 1998, 51, Gallagher 2003). In so far as retention is not representational, in so far as it provides us with an intuition of the just-past phase of the object, it can be considered part of perceptual consciousness (Hua 10/41). Husserl would agree with Meinong that the mere succession of conscious states doesn’t guarantee consciousness of succession, but this doesn’t entail the impossibility of a perception of duration and succession unless one also accepts the idea that perception is reduced to the grasping of a mere now-point, and that is an idea that Husserl categorically rejects. A perception cannot merely be a perception of what is now: rather any perception of the present phase of an object, includes a retention of the just-past phase and a protention of the phase of the object about to occur (Hua 11/315). Phrased differently (noematically), perceptual presence is not punctual, it is a field in which now, not-now and not-yet-now is given in a horizonal gestalt. This is what is required if perception of succession and duration is to be possible. Given that the retention constitutes the temporal horizon of the present it should be considered as part of perceptual consciousness rather than as a form of subsequent memory. As James once put it, memory proper entails a recollection of what once was present but which has subsequently been forgotten, and which is now “brought back, recalled, fished up, so to speak, from a reservoir in which, with countless other objects, it lay buried and lost from view” (James 1890, I.646). Given this narrow definition, retention cannot be a form of memory since it by contrast is involved in the very process of making something present for the first time. Since Husserl would deny the simultaneity between the retention and that which is retained, Husserl must be rejecting the Principle of Presentational Concurrence. But does this rejection entail that Husserl would defend a version of the Principle of Simultaneous Awareness? To put it differently, how would Husserl answer the following question? Is consciousness of a temporal process itself temporally extended? This is a deceivingly simple question. And it is a question that Husserl answered differently at different stages of his thinking. In 1904, Husserl’s answer was straightforward. As he writes in text nr. 21: “I see with evidence that the consciousness of a time itself <requires> time; the consciousness of a duration, duration; and the consciousness of a succession, succession” (Hua 10/192. Cf. 10/22). But of course, if the act of consciousness is not instantaneous and momentary, if it has duration of its own, if it contains temporal phases of its own, how then are these different successive phases synthesized in such a manner as to allow for an experience of succession? If the duration and unity of a tonal sequence is constituted by consciousness, and if our consciousness of the tonal sequence is itself given with duration and unity, are we then not forced to posit yet another consciousness to account for the givenness of this duration and unity, and so forth ad infinitum (Hua 10/80)? Husserl eventually became aware of these problems, and as he writes in text nr. 50:

Is it inherently absurd to regard the flow of time as an objective movement? Certainly! On the other hand, memory is surely something that itself has its now, and the same now as a tone, for example. No. There lurks the fundamental mistake. The flow of the modes of consciousness is not a process; the consciousness of the now is not itself now. The retention that exists “together” with the consciousness of the now is not “now,” is not simultaneous with the now, and it would make no sense to say that it is (Hua 10/333).

Even if we ascribe some kind of temporality to the stream of consciousness due to its dynamic and self-differentiating character, we should not conflate the temporality that is intrinsic to consciousness itself with the kind of temporality that pertains to the objects of consciousness. Husserl would reject the suggestion that there is a temporal match between the stream of consciousness and the temporal objects and events of which it is conscious. The relations between protention, primal impression and retention are not relations among items located within the temporal flow; rather these relations constitute the flow in question. Just as my experience of a red circle is neither circular nor red, there is a difference between the temporal givenness of the intentional object and the temporal givenness of the experience itself. They are not temporal in the same manner. It was against this background that Husserl eventually came to distinguish three different layers of temporality: the objective time of the appearing objects, the subjective, immanent or pre-empirical time of the acts, sensa, and appearances, and the absolute, pre-phenomenal flow of time-constituting consciousness (Hua 10/73, 10/76, 10/358). It has been a matter of controversy, how exactly one should understand Husserl’s account of the relationship between the subjective time and the absolute flow.3 But I think it is fair to say that most scholars agree on the fact that Husserl would take it to be misleading to describe the most fundamental level of time-consciousness as a temporally extended process, as if it was composed of a number of joined slices or building blocks:

8

Diese strömend lebendige Gegenwart ist nicht das, was wir sonst auch schon transzendental-phänomenologisch als Bewusstseinsstrom oder Erlebnisstrom bezeichneten. Es ist überhaupt kein „Strom“ gemäß dem Bild, also ein eigentliches zeitliches (oder gar zeiträumliches) Ganzes, das in der Einheit einer zeitlichen Extension ein kontinuierlich sukzessives individuelles Dasein hat (in seinen unterscheidbaren Strecken und Phasen durch diese Zeitformen individuiert). Die strömend lebendige Gegenwart ist „kontinuierliches“ strömendes Sein und doch nicht in einem Außereinandersein, nicht in raumzeitlicher (welträumlicher), nicht in „immanent zeitlicher“ Extension Sein (also in keinem Außereinander, das Nacheinander heißt, Nacheinander in dem Sinne eines Stellenauseinander in einer eigentlich so zu nennenden Zeit) (Hua 34/187).

Let us return to Dainton’s criticism: Does Husserl subscribe to the Principle of Simultaneous Awareness? Husserl does not defend the view that conscious awareness occurs in series of momentary glimpses. But he also denies that it makes sense to say of the time-constituting phenomena (the primal impressions, retentions, protentions) that they are present and that they have endured, that they succeed each other, or are co-present, etc. In short, in his view, they are neither “present”, “past”, nor “future” in the way empirical objects are (Hua 10/75, 10/333, 10/375-376). Thus, in a certain way inner time-consciousness is quasi-temporal (Hua 10/82) or atemporal (Hua 10/112), but only in the sense that it is not intra-temporal (innerzeitlich). Inner-time consciousness is not an object occurring in time, but neither is it merely a consciousness of time; rather it is itself a form of temporality,4 and ultimately the question is whether it makes sense to ascribe temporal predicates to time itself. Perhaps this worry can explain some of Husserl’s so apparently enigmatic and aporetic statements.

To recapitulate: According to Husserl, it is absolutely mandatory to distinguish sharply between the primal impression, retention, protention, i.e., the non-independent structures of time-consciousness, on the one hand, and the now-phase, the past-phase, and the future-phase, i.e., the phases of the temporal object, on the other hand. The primal impression, retention, protention are not related to each other as present, past, and future. Rather it is their conjunction which makes possible the senses of present, past, and future. In his writings, Husserl occasionally speaks of absolute time-constituting consciousness as an unchangeable form of presence (as a nunc stans) (Hua 34/384). It stands – to use James’ metaphor - permanent like the rainbow on the waterfall with its own quality unchanged by the events that stream through it (James 1890, I.630). But it is noteworthy that Husserl explicitly denies that this standing presence is to be understood as referring to merely one of the three temporal modalities (Hua 34/384). The presence in question is not the “now”, is not the “Gegenwart”, if one by Gegenwart means Gegen-wart, that is, a now that one stands over against (Hua 14/29). Inner time-consciousness is a field of experiencing, a dimension of manifestation, which encompasses all three temporal modes. And while it from a first-person perspective certainly makes sense to say that I had an experience of joy, or a perception of a flower, and that these experiences endured and have now ceased and become past – after all, otherwise it would hardly make sense to say that I can remember a former experience – the very dimension of time-consciousness with its threefold structure of protention-primal impression-retention, the very field of experiencing that allows for presence and absence, cannot itself become past and absent for me.

As Brough observes, the description of the absolute flow puts a fundamental strain upon language, since that which is to be described is unlike any object, unlike all other phenomena (Brough 1987, 23; cf. Hua 10/371). Presumably Dainton would disagree. After all, he argues that the temporality of, say, a perceived movement and the temporality inherent to the perception of that movement are one and the same. He would consequently want to deny the difference between what Husserl occasionally calls the noetic and the noematic-ontical temporalization (Ms. B III 9 23a). Moreover, Dainton even seems to think that one can measure the extension of the lived presence with a stop clock (Dainton 2000, 113, 171). Quite regardless of the extent to which such an attempt to quantify the most fundamental dimension of time-consciousness might miss its target, it is also quite questionable – as many phenomenologists have pointed out – whether the time of the clock can really do justice to lived time. To mention just one simple example: Think of the way in which the experience of time (for instance the interplay between the three different temporal dimensions) is differently articulated in such diverse states as hope, anxiety, insomnia and boredom. Think of the way in which the “same” 30 minutes can be experienced differently depending on whether you are anxious, bored or captivated. This is not to say that a stop clock cannot measure something, but the question is what precisely it is that is being measured. It is certainly possible to transform our experiences into mental objects and to posit or inject them into chronological time. But is this serial “time of the clock” a form of temporality

9

that is native to the experiences in question or is it rather derivative; the result of a subsequent objectification? 5. Conclusion To what extent is Dainton’s criticism to the point? As I have already admitted, I think his criticism of Husserl’s early theory might be justified. But when it comes to Husserl’s later view(s), Dainton’s criticism doesn’t strike me as being very perceptive. Contrary to what Dainton is saying, Husserl would not say that the scope of direct awareness is restricted to the momentary present, nor would he deny that we are as immediately aware of change or duration as we are of colour and sound.5 As for the idea that Husserl due to his concept of primal impression remained focused on durationless acts of awareness, I think this objection fails not only because Husserl is very clear about the fact that the primal impression cannot be thought independently of its horizon (Hua 11/315, 337-338) – it never appears in isolation and is an abstract component that, by itself, cannot provide us with awareness of a temporal object – but also because Husserl would ultimately argue that the very alternative between “temporally extended” and “durationless” is inappropriate when it comes to describing the fundamental dimension of time-consciousness. As for Dainton’s lingering content and clogging of experience objections, I think both objections miss their target, and that this should be clear the moment one recognizes the true intentional structure of the retention (cf. Gallagher 2003). What then about Dainton’s claim that Husserl simply stipulates that the retentions have the mysterious properties they need to have in order to make his theory work, and that he doesn’t really offer any explanation of how the retentions can provide us with an intuitive access to the past? As far as I can see, Husserl has in fact provided us with a painstakingly complex analysis of retentional consciousness. Moreover, his main aim is to provide an account of time-consciousness which is true to the phenomena. And if that calls for ascribing unique features to the retentions, then so be it. We should not forget that Dainton himself argues that the unity of consciousness is a primitive feature of experience, one which cannot be analyzed or reduced to anything else (Dainton 2000, 26). And if he can employ that kind of argument, I fail to see why Husserl can’t as well.

I have quite some sympathy for Dainton’s criticism of the awareness-content model and for his general thesis, namely that we should be wary of taking experiences to be things we perceive or observe in essentially the same way as we observe and perceive ordinary physical things. I would also endorse the claim that experiential processes are intrinsically conscious and hence self-revealing or self-intimating. In fact, I have elsewhere argued that this non-dualistic model of consciousness is one that Husserl himself occasionally favoured.6 But as should already be clear, I think there is a problem in the way Dainton cashes out his non-dualistic alternative. One can certainly reject the idea that phenomenal experiences become conscious by being taken as objects by some higher-order monitoring awareness and still argue that it is necessary to retain a distinction between the noetic and the noematic dimension. Thus, from a Husserlian point of view, Dainton’s attempt to dispense with conscious acts and to solve all problems by appealing to experiential content can be seen as an attempt to do away with noetic structures altogether in order to implement a purely noematic phenomenology. It also shares the familiar weaknesses of such an attempt. Moreover, Husserlian phenomenology would insist that we need to distinguish not only different levels in consciousness but also different forms of lived temporality. There is a difference between analysing consciousness in terms of different acts, such as acts of perception, judgment, imagination, etc., and analysing consciousness in terms of the structure protention-primal impression-retention. To simply collapse these different levels into one involves an oversimplification that is detrimental to a correct understanding of time-consciousness. To put it differently, although I have some sympathy for Dainton’s own simple conception of experience, I think he goes too far. To accept his argument would not merely be to opt for a simple conception of experience, it would be to opt for an oversimplified model, and I don’t think such a move would be phenomenologically warranted. What then about Dainton’s own alternative. How convincing is his overlap model? It is not a model that operates with anything resembling retentions and protentions, nor does it accept the idea of temporal modes of presentation. The overlap model would consequently deny that the first note in a tonal sequence will first be experienced under the temporal mode present, then under the temporal mode just-past, then further past, etc. Why does it deny that? Because if the note were given with different temporal characters as experience progresses, it would be incoherent to suppose that two subsequent experiences that both intended the same note could actually overlap in the sense of having common parts (Dainton 2000, 174). To put it differently, the overlap theory must reject the notion of temporal modes of presentation since such modes would prevent the overlap theory from accounting for diachronic unity by simply positing the numerical identity of the overlapping parts. However, this seems to confront Dainton’s theory with something of a problem. One might

10

ask, whether it can really do justice to the difference between synchronic and diachronic unity.7 Dainton himself argues that his aim is to establish that the diachronic unity of experience is not essentially different from synchronic unity. Given that Dainton considers diachronic unity to be a very short-term affair, spanning at most a second or so (Dainton 2000, 112), and given that it hardly makes sense to speak of an absolutely durationless synchronic unity, it is on his account indeed hard to distinguish the two. But is this really satisfactory? Moreover, how can it account for the dialectics between presence and absence that seem so indelible to mark our awareness of temporal objects and events? To solve the problem, Dainton occasionally appeals to what he calls phenomenal character. In his view, phenomenal contents all appear equally present as and when they occur, but the reason why they nevertheless have an experienced direction is because of their dynamic flow-character which must be considered a phenomenal attribute just like colour or timbre (Dainton 2003, 29).8 But is that convincing? What about our perception of an unchanging and enduring tone, which on Dainton’s account would presumably lack the phenomenal attribute of direction? This tone will certainly be perceived as temporally extended, as something that has already persisted for a while, and as something whose previous phases are no longer temporally present in the same manner as its current phase. But how is one supposed to do justice to this simple example without recurring to something like temporal modes of givenness? Ultimately, the overlap model’s inability to recognize temporal modes of givenness strikes me as a reductio ad absurdum of the theory. So I would obviously disagree with Dainton when he claims that the overlap theory is superior to all other accounts of temporal experience (Dainton 2000, 181). Dainton’s discussion of the structure of the stream of consciousness has many virtues, but in the end, I would bet my money on Husserl’s sophisticated account of the threefold structure of time-consciousness. It remains a source of profound inspiration. This is especially so, given that Husserl’s reflections on time-consciousness raise questions and address issues that are not even mentioned by Dainton.9 NOTES 1 Dainton discusses Broad’s theory in even greater detail, but I will ignore that part of his discussion. 2 Given Dainton’s projectivism, given that he considers ordinary perceptual objects to be part of the stream of consciousness, I find it hard to understand how he is able to avoid the threat of solipsism. I also think the extent of his temporal realism can be questioned, but these are all issues that I will be unable to pursue further in this paper. 3 For my own contributions, cf. Zahavi 1999, 2003, 2004. 4 Cf. Kern 1975, 40-41, Bernet 1994, 197, Merleau-Ponty 1945, 483, Heidegger 1991, 192. Although the field of experiencing has neither a temporal location nor a temporal extension, and although it does not last and never becomes past, it is not a static or momentary supra-temporal principle, but a living pulse (Lebenspuls) with a certain articulation and variable width, i.e., it might stretch. In fact, I would suggest that the metaphor of stretching – a metaphor used by both Husserl and Heidegger (Hua 10/376, Heidegger 1986, § 72) – might be quite appropriate as a characterization of the ecstatic self-differentiation of the constituting flow, since it avoids the potentially misleading talk of the flow as a sequence or succession of changing impressions, slices, or phases. To venture a more daring suggestion, perhaps a change of metaphor is really called for. Rather than likening time-consciousness to a river or stream, we should consider comparing it to a rubber band. As Claude Romano has pointed out to me, this suggestion recalls Augustine’s notion of distentio animi. 5 Regarding this specific objection, it might by the way be worthwhile recalling that everything that can be perceived has temporal duration and that this includes perceived color. 6 Cf. Zahavi 1999, 2003, 2004, 2005. 7 I am indebted to Joakim Quistorff-Refn for this critical point. 8 Did anybody whisper “metaphysics of presence”? 9 This study has been funded by the Danish National Research Foundation. REFERENCES Bernet, R. (1994), La vie du sujet. Paris: PUF. Brough, J.B. (1987), “Temporality and the presence of language: Reflections on Husserl’s phenomenology of

time-consciousness,” in Phenomenology of temporality: Time and Language, ed. A. Schuwer

11

(Pittsburgh: Duquesne University), 1-31. Brough, J.B. (1989), “Husserl's Phenomenology of Time-Consciousness,” in Husserl's Phenomenology: A

Textbook, ed. by J.N. Mohanty & W. R. McKenna. (Washington, DC: Center for Advanced Research in Phenomenology and University Press of America), 249-289.

Borges, J.L. (1970), Labyrinths. Harmondsworth: Penguin. Dainton, B. (2000), Stream of consciousness: unity and continuity in conscious experience. London:

Routledge. Dainton (2003). “Time in experience: Reply to Gallagher.” PSYCHE 9(10).

http://psyche.cs.monash.edu.au/symposia/dainton/gallagher-r.pdf Gallagher, S. (1998), The Inordinance of Time. Evanston: Northwestern University Press. Gallagher, S. (2003), “Sync-Ing in the Stream of Experience: Time-Consciousness in Broad, Husserl, and

Dainton.” PSYCHE 9(10), http://psyche.cs.monash.edu.au/v9/psyche-9-10-gallagher.html Heidegger, M. (1986), Sein und Zeit. Tübingen: Max Niemeyer. Heidegger, M. (1991), Kant und das Problem der Metaphysik. Frankfurt am Main: Vittorio Klostermann. Husserl, E. (1966), Zur Phänomenologie des inneren Zeitbewusstseins (1893-1917), Husserliana X. Den

Haag: Martinus Nijhoff; On the Phenomenology of the Consciousness of Internal Time (1893–1917). Trans. John Barnett Brough. Dordrecht: Kluwer Academic Publishers, 1991.

Husserl, E. (1966), Analysen zur passiven Synthesis, Husserliana XI. Den Haag: Martinus Nijhoff. Husserl, E. (1973), Zur Phänomenologie der Intersubjektivität II, Husserliana XIV. Den Haag: Martinus

Nijhoff. Husserl, E. (2002), Zur phänomenologischen Reduktion. Texte aus dem Nachlass (1926-1935), Husserliana

XXXIV. Dordrecht: Kluwer Academic Publishers. James, W. (1890), The Principles of Psychology I-II. London: Macmillan and Co. Kern, I. (1975), Idee und Methode der Philosophie. Berlin: de Gruyter. Merleau-Ponty, M. (1945), Phénoménologie de la perception. Paris: Éditions Gallimard. Zahavi, D. (1999), Self-awareness and Alterity: A phenomenological investigation. Evanston: Northwestern

University Press. Zahavi, D. (2003), "Inner time-consciousness and pre-reflective self-awareness." In The New Husserl. A

Critical Reader, ed. by D. Welton (Bloomington: Indiana University Press), 157-180. Zahavi, D. (2004), "Time and Consciousness in the Bernau Manuscripts." Husserl Studies 20/2, 99-118. Zahavi, D. (2005), Subjectivity and Selfhood: Investigating the first-person perspective. Cambridge, MA: The

MIT Press.