The Future of Phenomenology. Naturalization and Phenomenology of Perception

Carmelo Calì


One of the most hotly debated issues, from where the discussion about the features and the usefulness of phenomenology for contemporary scientific and philosophical research has evolved so far, has been the naturalization of phenomenology. The underlying assumption is that to settle the debate is to define to what extent phenomenology is suitable for the framework of the actual cognitive sciences and philosophy of mind paradigms. The debate, involving phenomenologists, post analytic philosophers, psychologists, cognitive scientists and neuroscientists, unavoidably represents a wide range of positions.
  Thus, I am going to outline some of these positions and then argue that Husserlian phenomenology does not have to wait for a radical theoretical reconstruction in order to be taken as able to integrate actual research, though it necessarily demands to let some of its claims drop or at least being reinterpreted in the light of current theoretical, experimental needs.

The Naturalization Problem Continuum

  To discuss some implications of the naturalization problem, I am going to see whether a few positions can be mapped onto a continuum ranging from the most favourable to the less consonant ones. It must be noticed that I am going to take into account neither the claims that phenomenology is something which might be directly ascribed to some brain mechanisms or to some properties of physical matter adequately arranged at the right level of low energy scales, nor the claims that state an absolute skepticism about the possibility of relating phenomenology to contemporary scientific research. I don't aim at covering every position in this controversy, but only to discuss some issues and objections to the naturalization position.
  At one end, we have the Petitot & Varela & Pachoud & Roy (1999) proposal. They claim that Husserlian phenomenology might be used to close the explanatory gap which affects the cognitive sciences, thanks to the descriptions and analyses Husserl carried out, often found to be supporting recent findings. The accounts provided by Husserl of the perspective dependence of the phenomenal world, of consciousness, of intentionality and time-space ordering of experience are supposed to be rich enough to give a satisfactory answer to the question about what kind of relationship holds between a computational and a phenomenological mind raised by Jackendoff (1987). But this could be the case only if Husserlian phenomenology is properly fitted to the explanatory framework of contemporary science, which stands on the claim that every property at stake must be continuous with properties admitted by natural sciences. This claim entails what is dubbed "naturalization of phenomenology" and requests that Husserl's anti-naturalism must be refused, the alleged impossibility of a mathematical formulation of phenomenological descritpions must be abandoned. In a nutshell, the transcendental dimension of phenomenology as it covers the descriptions and analyses with a philosophical interpretation unfit to integrate with those sciences which could turn them to profit must be dropped.
  On the other hand, we have Zahavi (2004) who points out, very clearly, some objections. He maintains that there are meaningful philosophical reasons for Husserl's (1987) antinaturalism to be upheld. Husserl stressed the difference between phenomenological psychology and transcendent phenomenology, because phenomenology doesn't simply contribute to positive knowledge but investigates its basis and possibility. The need for phenomenological reduction would be justified, for, it avoids any confusion with a natural and objectivist investigation, which would be to blame for treating consciousness as one object among the others in the world, whether it be taken as a psychical or a physical one. Hence, it is not possible to part the transcendental interpretation from phenomenology, because it not only prevents phenomenology from the "natural attitude", as is the case with the non-reductionist phenomenological psychology, but it also lets consciousness be the condition for any meaning, truth, validity, appearance.
  Thus Zahavi questions that a mutual exchange between phenomenology and cognitive sciences could result in a closure of the explanatory gap, and that a mathematical reconstruction would be of any profitable sense at all or that there might be a way to explain how experiences could be properties of the brain.
  Both Petitot & Varela & Pachoud & Roy (1999) and Zahavi (2004) recognize many kinds of relationships between phenomenology and cognitive sciences that could instantiate their own points of view. So, from the naturalization point of view, the instances amount to (a) the reductionism of the sort involved by the Identity Theory; (b) the "as if" strategy formulated by Dennett in his heterophenomenology; (c) the mutual constraining-variety. This ranges from (c1) the bridge locus argument, supporting the research of linking propositions between explanatory neural properties and phenomenal properties where the link is provided by a to be specified looking like-relation, to (c2) the isomorphist thesis, wherein the phenomenological descriptions are relevant in indentifying the right physiological mechanisms which in turn explains them, and finally to (c3) the operational generative thesis which allows for phenomenology and, say, neurobiology to share a common abstract and formal definition of properties that could belong to both at the same time, if considered at the right level of emergence. From the transcendentalist point of view, there is room for (d) the phenomenologists and scientists refusal of making their researches interact because of alleged independence for the former and the discredit of phenomenology by the latter; (e) one way relation from phenomenology, which would lay bare the foundation for other sciences, to empirical science, whose findings are not able to affect phenomenology; (f) the sharp distinction between a phenomenological psychology, which could contribute to empirical science, and the untouched transcendental phenomenology; (g) the mutual exchange between phenomenology and science only if transcendental phenomenology will change the very concept of nature and accordingly of naturalization.

A Mutual Constraining Isomorphism: The Case of Phenomenology as Formal Theory of Perception

  Exploiting the points above mentioned, I will argue that some difficuties stem from not considering phenomenology already as a descriptive science dealing with the structures of different kinds of appearances and providing a model of the various types of phenomenic manifolds from a phenomenological explanatory stance. Husserl (1973) provides a striking example of the way phenomenology can explain the laws ruling the visual world by appealing to concepts that are fit to a mathematical modelling. These analyses employ widely Riemann's concept of the n-dimensionality continuum and Weierstrass' concept of field, which prove to be profitable both in mathemathical analysis and in physical science.
  On the one hand, the visual object, with its aspects and phenomenal properties, is considered as a whole made up by parts being its variables varying along defined dimensions, corresponding to the visual and objective field multifariousness (Vielfältigkeit). Thus, the whole object is functionally equivalent to a manifold (Mannifältigkeit) constituted by groups of appearances which are ordered spatio-temporally by their positions and variations as to an inner manifold (the object field glanced at a current scrutiny), and a wider manifold (the neighbouring object fields). The relationships holding between these two manifolds are described then in terms of coeherent connections (Zusammenhänge) among the appearances and various kinaesthetic manifolds.
  On the other hand, the very concepts used by Husserl to designate the operations and the interconnections obtaining in the manifold system of vision have an intrinsic mathematical or geometrical meaning, such as congruency (Übereinstimmung), overlapping (Deckung), overlaying (Bedeckung), or admit a formal characterization, such as independence and non-independence. These observations make clear that it is neither necessary nor is it maybe desirable to narrow Husserlian phenomenology down to a mere philosophy of consciousness. Even though as an eidetic material science, as opposed to an eidetic formal one, phenomenology provides in a clear specifiable way the objective laws ruling a particular phenomenic dominion, giving the set of what pertains typically and generally to it. The possibility of a formal, even mathematical, formulation of the laws described by Husserl's analyses does not imply Computational Mentalism, according to which mental contents consist of mathematically definable operations on symbolic representations. It amounts only to saying that it is possible to build a mathematical model endowed with compelling phenomenological features, accordingly to what as been stated as the argument (c). In fact, this model might correspond to specific and non trivial organizational laws of the visual world and possess an explanatory or predictive power on its own. An example is the concept of double object elaborated by Husserl (1980) which has been given a great explanatory value in picture perception theory as Niederée & Heyer (2003) attest.
  This interpretation leads to the refutation of argument (d) and the assessment of argument (f) in a different way. To be sure, there remains in Husserl's view a difference between phenomenological psychology and transcendental phenomenology. However, one might assume a quite deflationary view about this distinction. If phenomenology is also a descriptive science of the phenomenal objective side of experience, then phenomenological psychology might be dealing with the phenomenal subjective side of it, thus contributing in psychophysics to relate phenomenological (how things look) and physical properties as Horst (2005) points out. The importance of phenomenological data for the psychophysical study of the Craik-O'Brien-Cornsweet effect plays against naturalization form (a) and (b). At the same time, the anti-objectivistic side of transcendental phenomenology could be rephrased in a more contemporary flavour as the compelling request for a pure theory of consciousness or perception, that is an abstract theory which employs only concepts derived from analyses of the intrinsic structure of perception. The reasons Mausfeld (2002) expounds for such a theory sound strikingly similar to those supporting Husserlian transcendentalism: phenomena must be studied in a non reductionist way. Therefore, a theory of perception must be formal in that it must not borrow its fundamental concepts primarily from physics or physiology, thus avoiding what Mausfeld calls the physicalistic trap. It is a pure theory of perception with the concepts mirroring the way the observer parses the world to specify the level and extent at which physical and physiological concepts might play an explanatory role.
  These considerations make the position (c) look like a plausible one, because they exclude an incommensurability between Husserlian phenomenology, mathematical modelling and the needs of a modern scientific perception theory. But how to fit phenomenology as a formal theory of perception with a mutual isomorphism constraint? The general principle could be shaped as the strong (c3), taking as example works such as Petitot (2003), Smith (1993), Petitot & Smith (1997), but for the time being it seems better to assume a balanced version of (c2), while admitting a variety of cross-talk cases between phenomenology and contemporary sciences. Overgaard (2004) challenges this possibility by requesting that the constraining must be fully reciprocal. I think this condition could be met. Phenomenology constrains cognitive sciences with its rich descriptions, fully specifiable at the desired formal level, thus letting models be built up and collecting richly defined data in order to find neural correlates which match the structure of appearances, as Todorovic (1987) suggests, whose claim makes room for a structural reinterpretation of (c1). Cognitive sciences constrain phenomenology in such a way that a phenomenologist is not forced to change her description only because a new brain area is found to be causally involved, but she does have to feel compelled to do that if a neurobiological study finds that some binding relations, structurally corresponding to those dependence relations held as fundamental ones, are a by-product of more fundamental ones.    This means that the isomorphism constraint must be kept at the relevant matching level, which causes a change in the phenomenological explanation of phenomenal relations. This last specification narrows a bit the (c2) argument and rests upon the conviction that phenomena are neither theoretical posits nor subjective qualia, but instead immediate, reproducible, undeniable facts of experience and hence a prime source of scientific investigation, as Ehrenstein & Spillmann & Sarris (2003) argue. Finally, this makes the (g) assumption unclear and unnecessary.


Literature

(1) Ehrenstein, W., & Spillmann, L., & Sarris, V. (2003). Gestalt issues in modern neuroscience. Axiomathes 13, 433 - 458.
(2) Horst, S. (2005). Phenomenology and psychophysics. Phenomenology and the Cognitive Sciences 4, 1 – 21.
(3) Husserl, E. (1973). Ding und Raum. Vorlesungen 1907. Den Haag: M. Nijhoff (Ed. U.Claesges in Husserliana XVI).
(4) Husserl, E. (1980). Phantasie, Bildbewusstsein, Erinnerung. Zur Phänomenologie der anschaulichen Vergegenwärtigungen. Dordrecht: Kluwer Academic Publishers (Ed. E. Mahrbach in Husserliana XXIII).
(5) Mausfeld, R. (2002). The physicalistic trap in perception theory. In Heyer D. & Mausfeld R. (Ed.), Perception and the physical world (pp. 75 - 114), Chichester: John Wiley & Sons Ltd.
(6) Niederée, R., & Heyer, D. (2003). The dual nature of picture perception: a challenge to current general accounts of visual perception. In H. Hecht & R. Schwartz & M. Atherton (Ed.), Looking into pictures. An interdisciplinary approach to pictorial space (pp. 77 - 98), Boston: MIT Press.
(7) Overgaard, M. (2004). On the naturalising of phenomenology. Phenomenology and the Cognitive Sciences 3, 365 - 379.
(8) Petitot, J. (2003), Neurogeometry of V1 and Kanizsa contours. Axiomathes 13, 347 - 363.
(9) Petitot, J., & Smith, B. (1997). Physics and thephenomenal world. In R. Poli & P.M. Simons (Ed.), Formal Ontology (pp. 233 - 254), Dordrecht/Boston/London: Kluwer Academic Publishers.
(10) Petitot, J., & Varela, F. J., & Pachoud, B., & Roy, J-M. (1999). Beyond the gap: an introduction to naturalizing phenomenology. In Petitot, J., & Varela, F.J., & Pachoud, B., & Roy, J-M. (Ed.), Naturalizing Phenomenology. Issues in contemporary phenomenology and cognitive science (pp. 1 - 80), Stanford: Stanford University Press.
(11) Smith, B. (1993). Common Sense. In B. Smith & D. W. Smith (Ed.), Husserl, Cambridge: Cambridge University Press.
(12) Todorovic, D. (1987). The Craik-O'Brien-Cornsweet effect: New varieties and their theoretical implications. Perception & Psychophysics 42, 545 - 560.
(13) Zahavi, D. (2004). Phenomenology and the project of naturalization. Phenomenology and the Cognitive Sciences 3, 331 - 347.