Does Dynamical Modelling Explain Time Consciousness
Total Page：16
File Type：pdf, Size：1020Kb
CHAPTER FIFTEEN
DOES DYNAMICAL MODELLING EXPLAIN TIME CONSCIOUSNESS?
PAAVO PYLKKÄNEN
Abstract
1. One of the fascinating features of conscious experience is its temporal structure. When I am listening to a song, I am conscious of the notes that I hear now for the first time. But I also seem to be conscious of the notes that I heard a moment ago, and through my anticipations I even seem to be conscious of the notes I have not yet heard.
2. Husserl’s famous model of time consciousness has three aspects. There is the “primal impression” directed to the notes heard 'now'. But there is also an awareness of the 'just past' ('retention') and even an awareness of the tones to come ('protention'). Husserl thought that retention and protention are perceptual processes. At the same time he realized that the idea that we would literally perceive the past is paradoxical. According to the usual view of time only the present and what is in it exists; the past no longer exists, the future does not yet exist. How could we possibly perceive that which does not exist?
3. van Gelder has proposed that time consciousness can be approached by considering a dynamical model of auditory pattern recognition. Here the state of the system at any given time models the awareness of the auditory pattern at that moment. Van Gelder suggests that this state builds the past and the future into the present, which is what Husserl required. However, van Gelder denies the idea that retention involves perception of the past. 219 Chapter Fifteen
4. I propose that even the dynamical approach fails to give an intelligible account of time consciousness. van Gelder just eliminates an essential aspect of time consciousness, namely the perception of previously experienced elements. A more adequate account can be provided within David Bohm’s 'implicate order' framework. This allows for a continued perception of previously heard notes, because these are understood to be 'enfoldments' that actually are in the present moment. Because Bohm’s approach builds upon a richer (quantum physically inspired) view of time and movement, it can better than Husserl’s make sense of the idea of retention as 'perception of the past'.
1. Introduction
Conscious experience has become a focus of intense study in recent years in philosophy, psychology, cognitive neuroscience, cognitive science, artificial intelligence, etc., so much so that it is fair to say that a new interdisciplinary field of 'consciousness studies' has been born (see e.g. Güzeldere 1997, van Gulick 2004). An important aim of consciousness studies is simply to explain the various puzzling aspects of consciousness. Following van Gulick’s (1995) classification, the puzzling aspects that need explanation include a) the difference between conscious mental states and nonconscious or unconscious mental states or processes; b) the distinction between conscious and nonconscious or unconscious creatures; c) qualia and the qualitative nature of conscious experience; d) subjectivity; and e) intrinsic intentionality / semantic transparency. There is yet another feature which van Gulick lifts up. While consciousness researchers sometimes use the term ‘phenomenal’ interchangeably with ‘qualitative’ (connected to ‘raw feels’), van Gulick prefers to reserve ‘phenomenal’ for a more comprehensive range of features: Current philosophical debate has focused heavily on raw feels, but they are just one aspect of our experienced inner life and thus only part of what we must deal with if we aim to describe the phenomenal structure of experience. In this sense the use of ‘phenomenal’ accords better with its historical use by Kant and later by the phenomenologists. The order and connectedness that we find within experience, its conceptual organization, its temporal structure, its emotive tones and moods, and the fact that our experience is that of a (more or less) unified self set over against an objective world are just a few of features other than raw feels that properly fall within the bounds of the phenomenal. All will need to be addressed if we take the phenomenal aspect as our explanandum. (1995: 64)
This article discusses the phenomenal structure of experience in light of Computation, Information, Cognition - The Nexus and the Liminal 220 some new theoretical developments. I will focus on the temporal structure of consciousness, in particular to a phenomenon known as 'time consciousness', characterized by Tim van Gelder as “...a special kind of awareness of temporal objects – an awareness of them as enduring ” (1999: 245). A good example of time consciousness is provided by considering what happens when one is listening to music. When I am listening to a song, I am conscious of the notes that I hear now for the first time. But I also seem to be conscious of the notes that I heard a moment ago, and through my anticipations I even seem, as least in some sense, to be conscious of the notes I have not yet explicitly heard. Husserl’s famous model of time consciousness has three aspects. There is the “primal impression” directed to the notes heard “now”. But there is also an awareness of the “just past” (“retention”) and even an awareness of the tones to come (“protention”). Van Gelder emphasizes that Husserl thought that retention and protention are perceptual processes. At the same time Husserl realized that the idea that we would literally perceive the past is paradoxical. According to the usual view of time only the present and what is in it exists; the past no longer exists, the future does not yet exist. How could we possibly perceive that which does not exist? Van Gelder (1999) has tried to resolve the above paradox by proposing that time consciousness can be adequately described by dynamical models (such as a dynamical model of auditory pattern recognition). Here the state of the system at any given time models the awareness of the auditory pattern at that moment. Van Gelder suggests further that this state builds the past and the future into the present, which is what Husserl required. However, van Gelder denies Husserl’s idea that retention involves perception of the past, in this way hoping to avoid the paradox. I propose in this article that even the dynamical approach fails to give an intelligible account of time consciousness. For it seems to me that van Gelder just eliminates rather than describes or explains an essential aspect of time consciousness, namely the perception of previously experienced elements. To understand the limits of the dynamical approach even better, I will consider the physicist-cum-philosopher David Bohm’s criticism of the differential calculus as a description of motion. I will finally consider Bohm’s own characterization of time consciousness within his ”implicate order” framework. I suggest that a Bohmian model of time consciousness allows for a continued perception of previously heard notes, because these are understood to be ”enfoldments” that actually are in the present moment. Because Bohm’s approach builds upon a richer (quantum physically inspired) view of time and movement, I suggest that it can better than that of Husserl make sense of the idea of retention as 221 Chapter Fifteen
“perception of the past”.
2. van Gelder’s dynamical model of time consciousness
When trying to tackle time consciousness in light of the dynamical approach van Gelder considers a dynamical model of auditory pattern recognition (the Lexin model, developed by Sven Anderson and Robert Port). The idea here is that the state of the system at any given time models awareness of the auditory pattern at that moment, and that state builds the past and the future into the present, just as Husserl saw was required. How is the past, according to van Gelder, built in to current awareness? He notes that in a dynamical system there is only one way in which the past can be retained in the present, namely by making a difference to the current state of the system, i.e. to the location of the current state in the space of possible states. The idea is that it is that location in its difference from other locations in the context of the intrinsic dynamics of the system, which ”stores” in the system the way in which the auditory pattern unfolded in past. It is how the system ”remembers” where it came from. In such an arrangement the past intrinsically and automatically flavours awareness of the current stage. In this kind of model, the momentary awareness (Husserl’s primal impression) is essentially shaped by retention of the past. How, then, is the future built in? Van Gelder notes that a dynamical system, by its nature, continues on a trajectory from any point, even when there is no external influence. The particular path it follows is determined by its current state in conjunction with its intrinsic dynamics. He suggests that there is a real sense in which the system automatically builds in a future for every state it happens to occupy. The system will automatically proceed on a path that reflects the particular auditory pattern that it has heard up to that point. For van Gelder this implies that protention, too, is a geometrically describable property of dynamical systems. What is relevant about the current state is the way in which, given the system’s intrinsic dynamics, the location shapes the future behaviour of system. The current location of the state of the system stores the system’s ”sense” of where it is going. As with retention, protention is an essential aspect of current awareness Van Gelder says, however, that if retention is the current location of system (considered insofar as that current location reflects the past inputs), it is hard to make sense of retention as perceptual and especially as perceptual with regard to something no longer exists. He is thus led to conclude that Husserl was mistaken in attempting to describe retention on a perceptual model. I think there are reasons to question van Gelder’s suggestion. Does a dynamical model really describe, say, our experience of listening to music Computation, Information, Cognition - The Nexus and the Liminal 222 adequately? When I am listening to a song, I hear some notes for the first time “now”, but the notes I have heard some time ago are typically still “reverberating” in my conscious experience. It seems obvious that I perceive them both, and in this sense it seems that Husserl was correct. Can a dynamical model, in the way characterized by van Gelder, thus really describe my experience. Van Gelder suggests, for example, that in a dynamical system there is only one way in which the past can be retained in the present, namely by making a difference to the current state of the system, i.e. to the location of the current state in the space of possible states. Presumably this means that the notes heard a little time ago no longer are present in experience but rather influence the only notes that exist, namely the notes that are heard “now”. But I think that in our actual experience the “past” notes make a difference to the current state of conscious experience by simply being present in experience, and thus they can be perceived. If this is correct, it seems that the dynamical model cannot really describe the past tones adequately. It is no wonder that van Gelder is led to give up the notion that retention of the past notes is a kind of perception. But it seems to me that when doing this he does justice to dynamical systems, not to actual conscious experience. I would like to suggest further that the above kind of troubles of trying to model time consciousness with dynamical models might connected with the troubles of using the differential calculus to model motion in general. Such troubles have been described in an illuminating way by David Bohm in the last chapter of his 1980 book Wholeness and the Implicate Order. In this context Bohm also proposed another way of characterizing motion in terms of a notion he called the “implicate order”. Further, he described time consciousness in light of his new theory. Assuming that van Gelder’s attempt to use dynamical modelling to describe time consciousness does not really work, it might be worth examining Bohm’s views. For on the one hand these might help us to get a deeper understanding of why dynamical modelling fails to capture the essence of time consciousness; on the other hand they might offer us an alternative, more fruitful way of describing time consciousness. In what follows I shall therefore first examine Bohm’s criticisms of using the differential calculus to describe physical motion in general (section 3). I will then briefly consider Bohm’s alternative approach to describe motion and consider what a Bohmian model of time consciousness might look like (section 4).
3. Bohm’s criticism of the differential calculus as a description of motion
Bohm starts his discussion of the limitations of the differential calculus in the description of physical motion by drawing attention to the way motion is usually 223 Chapter Fifteen thought of, i.e. in terms of a series of points along a line. If a particle moves, one typically assumes that at a given time t1 a particle is at a position x1, while at a later time t2, it is at another position x2. The velocity v of such a particle can be then be expressed as x2-x1/t2-t1.
Bohm then criticizes (in a way that resembles Bertrand Russell’s criticisms) this usual way of thinking: Of course, this way of thinking does not in any way reflect or convey the immediate sense of motion that we may have at a given moment, for example, with a sequence of musical notes reverberating in consciousness (or in the visual perception of a speeding car). Rather, it is only an abstract symbolization of movement, having a relation to the actuality of motion, similar to that between a musical score and the actual experience of the music itself. If, as is commonly done, we take the above abstract symbolization as a faithful representation of the actuality of movement we become entangled in a series of confused and basically insoluble problems. (1980: 201-2) Bohm is thus strongly underlining the difference between our immediate sense of motion and the commonly used abstract symbolization of motion. If one takes the abstract symbolization as a representation of motion, one is led to assume that the times t1 and t2 both exist. For presumably, if the abstract symbolisation is assumed to represent something real, the things it represents must exist. The symbolization refers to t1 and t2 at the same time, so presumably t1 and t2 must exist at the same time. Bohm emphasizes, however, that the assumption that both t1 and t2 exist at the same time is in contradiction with our actual experience, which indicates that “...when a given moment, say t2, is present and actual, an earlier moment, such as t1 is past. That is to say, it is gone, non- existent, never to return” (ibid: 202). The key trouble is that “...if we say that the velocity of a particular now (at t2) is (x2-x1)/(t2-t1) we are trying to relate what is (i.e., x2 and t2) to what is not (i.e., x1 and t1)” (ibid: 202). The usual view of time says that only the present and what is in it exists. The past is gone, the future is not yet. Yet the usual notion of velocity at a given now paradoxically involves both the present and the past. Bohm admits that we can use the above expression as long as we remember that we are using it abstractly and symbolically, as is, indeed, commonly done in science and mathematics. But he emphasizes that the abstract symbolism cannot comprehend that:
...the velocity now is active now (e.g., it determines how a particle will act from now on, in itself, and in relation to other particles). How are we to understand the Computation, Information, Cognition - The Nexus and the Liminal 224
present activity of a position (x1) that is now non-existent and gone for ever?
He notes that it is commonly thought that this problem is resolved by the differential calculus:
What is done here is to let the time interval, delta t = t2-t1 become vanishingly small, along with delta x = x2-x1. The velocity now is defined as the limit of the ratio delta x/delta t as delta t approaches zero. It is then implied that the problem described above no longer arises, because x2 and x1 are in effect taken at the same time. They may thus be present together and related in an activity that depends on both. (1980: 202)
However, Bohm is not satisfied with even this approach. He claims that “...this procedure is still as abstract and symbolic as was the original one in which the time interval was taken as finite. Thus one has no immediate experience of a time interval of zero length, nor can one see in terms of reflective thought what this could mean” (ibid: 202). The procedure is thus both empirically and conceptually/logically ambiguous! Bohm thus thinks that when taken as an abstract formalism, the differential calculus is not fully consistent in a logical sense. Of course, it is widely used in contemporary science, so it does work up to a point. What is particular important for our present discussion is that there is even a whole research programme in cognitive science, namely the above mentioned “dynamical approach” or “dynamical modelling” which makes essential use of the differential calculus when describing various cognitive processes (see e.g. Port and van Gelder eds 1995). However, Bohm underlines that the differential calculus has a limited range of applicability even for physical phenomena:
...it applies only within the area of continuous movements and then only as a technical algorithm that happens to be correct for this sort of movement. ...however, according to the quantum theory, movement is not fundamentally continuous. So even as an algorithm its current field of application is limited to theories expressed in terms of classical concepts ... in which it provides a good approximation for the purpose of calculating the movements of material objects. (1980: 202)
The above point might also encourage us to reflect upon the applicability of dynamical systems theory (insofar as it relies on the different calculus) to describe cognition and consciousness. To be sure, cognition and conscious experience involve the sort of movement which can conveniently be described in terms of the differential calculus. However, one sees sometimes a tendency to assume that this approach works for cognition and conscious experience in a 225 Chapter Fifteen very broad and comprehensive sense. But is such a tendency really justified? For example, we saw briefly above that van Gelder’s (1999) attempt to discuss time consciousness in terms of the dynamical approach seems very problematic. He ends up eliminating the idea of retention as perception, because there is no room for such “perception of the past” in the dynamical framework. But this may be more a reflection of the limits of the dynamical approach, than an indication that time consciousness does not involve perception of the past. I think Bohm’s above criticism of the differential calculus applies strongly also in the case of time consciousness, thus suggesting that there are important limits of applicability to the dynamical approach in the domain of cognition and consciousness. But how should we then think about time consciousness? Bohm proposed an alternative way of thinking about this, and we shall now move on to briefly consider this.
4. A Bohmian model of time consciousness
We saw above how for Husserl time consciousness involves ”perceiving the past”. He proposed that the “retentions” of the ”just past” tones ought to be understood as a kind of perception. At the same time he admitted that such “perception of the past” is paradoxical. Let us now see whether Bohm’s description of time consciousness can avoid this paradox, without eliminating important aspects of time consciousness, as van Gelder’s model seems to do. Bohm notes that when listening to music (e.g. a rapid sequence of C-D-E-F- G), in the moment we hear G for the first time, we also hear C-D-E-F reverberating in consciousness. Such reverberations are not memories nor representations but rather active transformations of the original sounds. They are all co-present but can differ in that they can have a different degree of enfoldment. We can measure this degree of enfoldment in terms of the time elapsed since the sound was first heard. Let us assume our unit of time to be 100 ms, and introduce an enfoldment parameter n. Cn then means that C was first heard n units ago, and is at the nth degree of enfoldment. We hear now G for the first time, and we have just heard the sequence of C-D-E-F-G (with 100 ms interval between the notes). We can now describe the conscious experience of the melody at that moment as a co-presence of Cn, Dn-1, En-2, Fn-3, Gn-4. That is, we have a co-presence of elements at different degrees of enfoldment When listening to music one is thus apprehending a set of co-present elements at different degrees of enfoldment. Such an experience has a certain order, which Bohm calls enfolded or “implicate” order. He suggests that when listening to music one is directly perceiving an implicate order (as opposed to Computation, Information, Cognition - The Nexus and the Liminal 226 thinking about such an order abstractly in terms of thought). Bohm further suggests that such a direct perception of an implicate order consists of an immediate sense of flow. In other words, our sense of flow is a certain mode of experiencing, namely a mode of directly experiencing an implicate order In my view, Bohm’s characterization helps to make sense of Husserl’s paradoxical view of time consciousness which involves ”perceiving the past”. For in Bohm’s model the ”past” elements are assumed to be in the present, as active transformations or enfolded structures. Thus, they can be perceived without such a perception involving any paradox. At the same time Bohm’s model does not eliminate important aspects of time consciousness, in the way van Gelder’s dynamical approach seems to do. Bohm arrived at the notion of implicate order when thinking about the problems of interpreting quantum theory. Although Bohm himself proposed in 1950 famous “hidden-variable” model, in which we can think of particles (such as electrons) as moving continuously, he admitted that it is not likely that they actually move continuously at very short time intervals (e.g. the Planck time of 10 to -33 s). The notion of a particle moving continuously thus has to be replaced by some model which gives rise to discrete movement. In one such discrete model we have “co-present elements (fields) at different degrees of enfoldment”, i.e. an implicate order in Bohm’s terms (see Bohm 1980: 179-86; Bohm and Hiley 1993: 367-8). The basic mathematical algorithm to describe an implicate order is a certain kind of algebra, not the differential calculus (Bohm 1980: 157-71). The above suggests, amazingly, that the implicate order model of an electron in Bohmian quantum theory is in some key ways analogous to a model that can describe the phenomenal structure of time consciousness. This opens up the possibility of a whole new research programme, with a wide range of applications. Of course, we have seen that even dynamical modelling has been applied in a wide range of fields, from the physical to the cognitive, all the way to time consciousness. What we have suggested in this article is that there may be important limits to the applicability of dynamical modelling, especially when it comes to phenomena which have a discrete structure, such as motion at the quantum level, or time consciousness. We have further suggested that it seems that the implicate order can do better. Much study is, of course, required in the future to establish in more detail the prospects of this new approach (for further discussion, see Bohm 1980; 1986, Pylkkänen forthcoming).
5. Concluding Remarks
Basic motivation for writing this paper was to show that the idea of logical 227 Chapter Fifteen pluralism could be applied to the notion of semantic information, and more specifically to that of informational content as it arises from the Inverse Relationship Principle. Thus far the following has been argued for: (i) logical pluralism arises in fairly simple communication contexts, and (ii) the formulation of a pluralistically inspired measure for informational content is necessary to get a firm grip on what is informative in such contexts. More generally it could be claimed that, on the one hand, logical pluralism provides an elucidation (if not a way out) of the problem of assigning informational content to logical truths, while on the other hand the informational perspective on logic makes the idea of logical pluralism more plausible (for it represents a useful application). Notwithstanding the fact that a comprehensive theory of informational pluralism is not a part of the present paper, its basic features have been outlined. Most importantly, it is shown to be not only a viable alternative to a monist understanding of information, but also a fundamentally superior one. The key argument in favour of informational pluralism could be summarised by appealing to the notion of a logic’s discriminatory power. The idea of the latter being that deductive strength varies with discriminatory strength: the more a logic proves, the fewer distinctions (or discriminations) it registers (Humberstone, 2005, 207), it appears that a weaker logic leads to a more discriminatory measure of informational content. Informational pluralism then, is precisely the idea that from a certain perspective (called global or external) these discriminations should partly be disregarded. As to the blindspots in this exposition, a few need to be highlighted. First and foremost, very little attention was paid to the requirement for semantic information to have an informee-independent meaning. Despite the non- pragmatic character of the last example’s informational pluralism, the meaning invariance of a message was not as such established. In other words, no argument was provided to show that a difference in informational content between internal and external perspective should not entail a change in meaning. This being a serious threat to a purported pluralist theory of semantic information, it should be treated in detail when designing such a theory. For the moment, such a defence could be restricted to pointing out that, when adopting a minimalist understanding of meaning, the meaning-invariance requirement can be fulfilled within the bounds of a pluralist understanding of information. Next, the problem of assigning informational content to contradictions — the so-called Bar-Hillel/Carnap-paradox from property 3.2 (see Floridi, 2004) — was not treated either. The semantics of relevant logic allowing for inconsistent situations, one might be tempted to think that a solution would arise in a very similar way to its dual problem (the content of logical truths). Computation, Information, Cognition - The Nexus and the Liminal 228
Inconsistent situations being very different from their incomplete siblings, it is doubtful whether they could count as genuine (that is, possibly factual) cases in the sense of definition 4 and the concept of information as being true in virtue of what is factual (Israel & Perry, 1990). Assigning less than maximal content to a contradiction, would require one to either adopt a (non-semantic) dialetheist position (the world is possibly inconsistent), or to reject the factuality requirement on semantic information. Both options need an extensive discussion which lies beyond present paper’s aim. Apart from mentioned omissions, further research on the topic of informational pluralism might include the formulation of an independent motivation of pluralism which relies on methodological considerations emerging from the philosophy of information. Reconsidering the notion of logical discrimination, one might easily start to think of a logic as an interface in the sense of Floridi & Sanders (2004), thus leading to a reformulation of informational pluralism in terms of levels of abstraction. This being only one possible suggestion for further research, the broad topic of informational pluralism as a part of the philosophy of information shows itself as a promising line of enquiry into the nature of semantic information.
References
Atmanspacher, H. and Ruhnau, E. eds, 1997. “Time, Temporality, Now. Experiencing Time and Concepts of Time” in an Interdisciplinary Perspective. Berlin: Springer. Bohm, D. 1980. Wholeness and the Implicate Order. London: Routledge. —. 1986. “Time, the implicate order and pre-space”, in D.Griffin ed. Physics and the Ultimate Significance of Time: Bohm, Prigogine and Process Philosophy. Albany: SUNY Press. Dainton, B. 2000. Stream of Consciousness. Unity and continuity in conscious experience. London and New York: Routledge. —. 2001. Time and Space. Chesham: Acumen. Güzeldere, G. 1997. “Introduction: The Many Faces of Consciousness: A Field Guide”, in Block et al. eds, 1997. The Nature of Consciousness: Philosophical Debates. Cambridge, Mass: MIT Press. Petitot, J. et al. eds, 1999. Naturalizing phenomenology, Stanford University Press. Port, R.F. and van Gelder, T. eds, 1995. Mind in Motion. Cambridge, Mass.: The MIT Press. Pylkkänen, P. (forthcoming) Time Consciousness. The Relevance of Bohmian Physics to Our Understanding of the Mind. Heidelberg: Springer, Frontiers Collection. 229 Chapter Fifteen
Ruhnau, E. 1995. “Time-Gestalt and the Observer”, in T. Metzinger ed, Conscious Experience. Schöningh: Imprint Academic. Van Gelder, T. 1999. “Wooden Iron? Husserlian phenomenology meets cognitive science” , in Petitot et al. eds, 1999. Van Gulick, R. 1995. “What would count as explaining consciousness?”, in T. Metzinger ed, Conscious Experience. Schöningh: Imprint Academic. — . 2004. “Consciousness”, The Stanford Encyclopedia of Philosophy (Fall 2004 Edition), Edward N. Zalta ed, URL