Meta-theory and Empirical Science

The second example of divergence regards the interpretation that Stumpf, Husserl and Kohler gave of the form that phenomenology as a philosophical theory should take in connection with empirical sciences. Stumpf and Kohler sided against Husserl on the question of the definition of phenomenology (cf. Fisette, 2009; Kaiser El-Safti, 2001; Rollinger, 1999: 83k). By analogy with the formal foundation of Riemannian geometry Husserl (1913) builds a top- down definition of phenomenology. As the ideal geometry of n-fold extended manifolds is distinguished from the geometry of physical space, so he con­tends that “pure” phenomenology is distinguished from psychology because the former deals with ideal phenomena in contrast with the “real” phenom­ena of empirical sciences posited in the space and time of natural sciences.[1] Pure phenomenology tries to discover the ideal invariant structures called “essence” in contrast with the “real” objects of empirical or experimental sci­ences (Hartimo, 2008, shows the group-theoretical inspiration of this notion). Therefore, it does not admit either the realistic thesis or its negation on the existence of anything (Gurwitsch, 1966: 110). This is also the aim of the bottom- up definition of phenomenology that guides the comparison between phe­nomenology and psychology (Husserl, 1911, 1916). However, the top-down definition does not start from the descriptions of common-sense experience and empirical sciences; hence, the tenet that phenomenology is not commit­ted to the ontology of common-sense and science is not presented as a “phe­nomenological distinction” that does not introduce new entities in the mind or in the world. Rather, it seems that phenomenology is possible merely by suspending or “parenthesizing” from the outset the belief in the validity of naive and scientific experience, and relying on the concept of consciousness intended as the ability to “have an insight” (Wesenschau) on the essential structure of phenomena ([1913] 1983: 34, 43k, 57ff.). According to this top-down definition, Husserl restricts the scope of Stumpf’s phenomenology and the im­portance of Gestalt psychology ([1913] 1983: 210, [1930] 1989: 424).

Stumpf and Kohler make meaningful points against this top-down defini­tion. Stumpf (1939: 192, 198) acknowledges Husserl’s attempt to build a theory of science as abstract as the theory of manifolds, but he criticizes the suspen­sion of belief in any scientific knowledge, which runs the risk of building a phenomenology without phenomena. He agrees with Husserl that the laws of structure are equivalent to “material axioms,” namely to propositions whose analytic truth is derived from experience. An example of such propositions is provided by the propositions on the dependence between surface and color or the belongingness of pitch and loudness as parts of sound. Stumpf (1873) had suggested that there are perceptual elements that by nature are dependent be­cause they cannot occur without others. For instance, it is impossible to hold brightness or loudness as they are while that instance of color or tone varies at will until it disappears. Stumpf calls such elements “partial contents,” that is to say parts that are inseparable from the elements to which they belong. There are also mutually dependent elements, such as color and extension, because each color is given immediately with and in any appearance of space and vice versa. On this basis, Husserl (1901) himself formulated a formal theory of parts, wholes and the relations of dependence and connection (“foundation”) among parts and among parts and wholes (infra § 7.4). Husserl calls necessary analytic propositions those sentences whose parts can be replaced by variables preserv­ing the truth for every value of the variables, while necessary synthetic propo­sitions those sentences whose truth is not preserved by replacing their parts at will by arbitrary abstract variables. The truth of the latter is indeed a function of the values corresponding to perceptual states of affairs. For example, the parts of sentences like “that note of the violin is a loud A4” can be replaced up to “every sound x has pitch and loudness” because any other substitution is false or inconsistent with respect to the nature of sound. This represents the common core of Stumpf’s and Husserl’s phenomenology, though Stumpf (1939: 201) rejects their interpretation as a new form of synthetic a priori laws.

Stumpf himself holds that it is desirable that the phenomenological laws on the structures and forms of order of appearances are expressed in algebraic and geometrical terms. Yet he contends that the propositions equivalent to the “material axioms” of perception do not require any suspension of scientific belief. The arbitrary variation of appearances cannot be free from any empiri­cal evidence. Phenomenology is a theory that should become scientifically complete through the experimental manipulation of psychic functions and appearances. Instead, the suspension of the validity ascribed to any science besides logic could turn pure phenomenology into the “armchair method” of philosophy (1939: 199, 319). Then there is nothing like a pure phenomenology intended as the theory of essence in principle separated from the empirical sciences of the various kinds of appearances (1939: 187, 191k).

Although he acknowledges that Husserl had clarified what counts as phe­nomenal data and how a theory could be constructed based on them, Kohler (1944: 203 n. 1) claims that putting scientific knowledge into “brackets” might turn a methodological tool into the “weapon[s] of an ontological prejudice,” because “certain phases of experience” can be neglected while “a first account of experience ought to be given and carefully studied without selections of any kind” (cf. also 1938: 47). Kohler (1938: 46, 52) concedes that the phenomenol­ogy of ideal structures fits the treatment of mathematical and logical objects. However, if applied to the forms of experience of the ordinary world, this theo­retical approach risks driving phenomenology away from empirical science in contrast with the tenet that phenomenology is the method that preserves the face value of data. In his lectures, Kohler advocates that phenomenology should specify the contexts of experience in which the things themselves allow for the deriving of laws from their phenomenal course in a reliable and consis­tent manner. Thus it provides the observational basis to test the theories that are not supported by direct phenomenal evidence. From this viewpoint Hus­serl’s distinction between theories of real and ideal objects is spurious (1938: 72; cf. Jaeger, 1994). There is only one world that poses an epistemological rath­er than an ontological question, which consists in the empirical justification of the models that are constructed to account for observable objects. The aim of phenomenology is to distinguish between the theories of what is directly observable and what is only posited to assess their models and laws. Finally, Kohler remarks on the counterintuitive consequence of Husserl’s arguments. Indeed, in order to show the autonomy of pure phenomenology Husserl ([1913] 1983: 104f.) conceives a thought experiment: were the physical world annihilat­ed, the consciousness and the correlated phenomena would remain the same. Husserl argues that this shows that nature is a theoretical construct of science, whose methods and procedure yet depend on the forms of cognition and the correlated objects. The structure of consciousness and appearances account for cognition and its objects and should be recognized as the source of the epistemological validity of any scientific endeavour. Kohler responds that were Husserl correct, nature would literally be reduced merely to the correlate of consciousness, contrary to the findings of biology, geology and astrophysics.

The criticism of Stumpf and Kohler emphasizes the problem posed by too an abstract account of the phenomenological research. Still, it is targeted at the shortcomings of Husserl’s top-down strategy. In fact, the ideal structures are not new ontological or phenomenal entities. “Matter of facts” and “essences” are inseparable, in the sense that the latter capture the repeatable contents of the former. Besides, the scientific research is not in want of or dependent on the conceptual definitions of pure phenomenology ([1913] 1983: 7, 17). Not­withstanding his disputable thought experiment, Husserl does not underes­timate the empirical content and the mathematical architecture of science. He acknowledges that pure and empirical sciences share the architecture of algebraic functions, calculus and analytic geometry, which permits a uniform, consistent and coherent treatment of all kinds of properties in nature (Hus­serl, 1976: 41; cf. Drummond, 1992). Pure and empirical sciences differ in their methods for building definitions and manipulating variables, but have in com­mon the mathematical form to represent the functional connection of mea­surable properties in specified coordinate systems. In this sense, Husserl finds it reasonable to claim that nature as a concept is correlated to consciousness. However, Husserl’s interest in the construction of a meta-theory resides in the identification of the epistemological problems that may arise in science if the nature of objects is reduced to its theoretical and mathematical structure. In the case of perception, if nature is construed as the collection of physical ob­jects or posits ruled by interconnected laws in mathematical form, perception is reduced to the subjective manifestation of the physical world intended as the directly unobservable cause that by hypothesis corresponds to the object of current physics. Perception is confined to subjective experience in contrast with what is believed on physicalistic grounds to be objective. The perceptual appearances become signs of physical objects. Accordingly, appearances are construed as inadequate or false representations of physical causes, while in fact being real means the possibility of appearing to anyone ([1952] 1989: 87).

Such arguments do not remove the differences between the accounts of Stumpf, Kohler and Husserl, but they restrict the scope of their disagreement to particular arguments that did lead Husserl to erroneous or cursory appraisal of other forms of phenomenology. For instance, Gurwitsch (1966: 113) shows that Husserl overlooks the refusal of Gestalt psychology to assume extraneous ontological commitments akin to the one that underlies his theory.

Source: Calì Carmelo (2017), Phenomenology of Perception: Theories and Experimental Evidence, Brill.

Leave a Reply

Your email address will not be published. Required fields are marked *