The Empirical Grammar of Perception in Brentano

Brentano’s phenomenological claims on perception can be found in the the­ory of perceptual reference as well as in the philosophical, that is, descriptive psychology (1874, 1982). Brentano contends that denoting something makes sense only if it implies an intentional reference by a subject to whom some­thing appears. If generalized, this claim means that something is impossible if by its very nature it cannot appear, and conversely that appearing with par­ticular features to a subject determines the epistemological and ontological characteristics of what is perceived, imagined, judged.

This contention is fundamental for Brentano’s theory of perceptual refer­ence which is built into his psychology and philosophy of perception. Unlike Kant, he holds that the term “phenomenon” is applied correctly only to a fact that appears, be it an ascertainable state, event or process. Instead, the term “object” has a general indeterminate meaning that needs to be specified by its phenomenal instances. Then, what the objects of perception are is specified by the appearances understood by the perceivers. The theoretical analysis of perception has to conform to the basic empirical rules of scientific research. A phenomenon is something that appears so that a full acquaintance of it is given, whence the laws of the phenomena of the same class are discovered and generalized by inductive, experimental and analytic procedure. Indeed, both philosophical and scientific theories have to adjust their methods to the nature of the object of investigation (Brentano, 1968: 78, uses the expression “sachentsprechender Methode”). All theories following this empirical method aim at the complete decomposition of phenomena, their classification accord­ing to inherent features, the discovery of their general laws, the specification of particular laws and their verification by controlled experience. Psychology differs from other sciences because the experience is not only its starting point but also its ultimate foundation. In particular for the psychology of perception, this claim extends to the experience of the perceptual reference.

Brentano (1874) claims that the experience of perceiving something implies being in the perceptual state of having such and such appearance, hence the perception of something is always accompanied by the incidental reference to that perceiver’s state. Therefore, Brentano decomposes perception into a primary reference to an object and a secondary reference to itself that is called “inner perception.” For instance, perceiving a color refers primarily to the occurrence of a shade of red, which is also located in the visual field, and concomitantly to the perception of that red as one of the perceiver’s states. Indeed, perceivers are aware of perceiving and able to distinguish perceptions of different objects simply because they know what it is to perceive something. However, this does not mean that the subjects perceive their states rather than objects. The secondary reference depends on the primary perceptual refer­ence to objects. No one could refer to his perception of sound, were he not actually perceiving a chord. Brentano (1874: 179, 180, 183) emphasizes that the distinction between primary and secondary reference is conceptual, because in ordinary experience they are merged into a psychological unity. Brentano suggests exploiting this twofold reference in order to observe the components of perception as phenomena, that is, as states whose meaning and appearanc­es are studied regardless of the presumed existence that the subjects ascribe to the referents of perception on the grounds of the commonsensical interpreta­tion of appearances in terms of their external causes.

Brentano (1874) holds that every perceptual occurrence is decomposable into two parts that require each other. For instance, the visual perception of x is decomposed into “seeing x” and the “x seen” as well as the auditory perception of y into “hearing y” and the “y heard.” The first term denotes a psychic phe­nomenon, the second a physical phenomenon. This decomposition does not mean that the object of perception is a non-ordinary entity, that is, a mental object or image of external things (cf. McAlister, 2004, for an epistemological rather than ontological interpretation of the intentionality thesis). Brentano (1874: 185) states that subjects don’t see seen colors and hear heard sounds, but always colors or sounds. The concept of color or sound is not relative, in the sense that it is consistent to conceive of colors or sounds regardless of their being actually perceived. Otherwise, perception would mean to be directed to the subject rather than to an object. Moreover, it is false to argue that percep­tion refers to seen colors and heard sounds, because it is impossible to ascribe properties to something other than one’s own appearances. Since the concepts of color and sound are not relative, ascribing existence to objects that fall into perception involves no contradiction. This distinction, then, is the result of the decomposition carried out when perception is analyzed rather than ordinar­ily experienced as the appearance of something. In fact, Brentano (1874: 129) claims that the objects of perception exist only phenomenally or intention­ally, but this claim can be construed to mean that the analysis of the physical phenomenon should not involve the naive realism associated with ordinary perception. The belief in the existence of objects in the external world is a characteristic of perception, but it has to be critically set aside because it im­plies the naive hypothesis that objects falling into perception have coincident physical and phenomenal properties. Yet physics suggests it can be dismissed as unlikely (1874: 129, 138).

The definition of the objects of perception as physical phenomena qualifies the analysis of perception that is carried out once appearances are concep­tually distinguished from common-sense and physical objects. In this sense Brentano (1874: 250) claims that the physical phenomena do not exist either inside or outside the mind. Physical phenomena are the objects of experience, which are provided to subjects by perception regardless of commonsensical, inductive, conceptual, theoretical knowledge. If psychic phenomena refer to something else, physical phenomena manifest themselves. To be sure, they may not exhaust their function in perception. Natural scientists take them as signs of unobservable physical or chemical states. However, if considered within the limits of the study of perception, both can be analyzed as pure phe­nomenal facts up to their constituents to find the laws that they obey.

Brentano’s Deskriptive Psychology (1982) refines the procedure to exploit the secondary reference in order to give psychology access to experience for studying the appearances in themselves. This refinement leads him to clari­fy the distinction between “descriptive psychology” and the natural sciences as well as “genetic” psychology. Brentano claims the empirical character of “descriptive psychology,” but if the natural sciences employ an indirect method of measurement and validate hypotheses by induction and the probability cal­culus, “descriptive psychology” derives the phenomena from what subjects are acquainted with in order to decompose them into constituents and describe their features and connections. Descriptive psychology starts from experienc­ing the reference to something through appearances as the empirical founda­tion of the analysis of phenomena. Brentano expounds a method whose stages depart progressively from experience through the ability of “noticing,” namely a perception that makes explicit what is already contained in appearances by making their relevant parts and connection accessible to observation (1982: 31-65; see also Baumgartner, 1986). Noticing a color perception means making the seeing and the color explicit as instances of a psychic reference and of an appearance. Seeing and color become the correlated parts of a whole that nec­essarily require each other. This whole does not exist as a new object along the color perception any more than a painted landscape exists along the existing depicted landscape. As painting brings about a whole that is a modification of the existing landscape through the pictorial representation, so noticing brings about a whole that is a modification of perception through making its constit­uents explicit. In both cases, no feature is either added or subtracted to land­scape and perception. The perception becomes an object of analysis of which seeing a color and the color appearance are only as abstractly separable parts as are the left and right sides of an atom.

Every perception can be modified in a phenomenal object of analysis. The modification does not change the function of reference and appearances in ordinary perception. It is necessary to carry out the analysis autonomously on strict phenomenological grounds, because as phenomenal correlates both perceptual reference and appearances are devoid of association with common- sense objects or any physical source exerting a causal power on the subjects.^[1] The phenomenological modification allows for treating the correlated parts of an instance of perception as repeatable data for analysis. For example, the appearances of colors are substituted for one another in order to show un­der which feature they are equivalent. The properties of phenomena are thus fixed and collected. Then, the method of descriptive psychology prescribes the inductive generalization of the features of phenomena and the deductive for­mulation of laws. This was a prescription common to empirical science, but Brentano characterizes it according to his epistemology. On the one hand, the inductive generalization is founded on intuitively grasping the fundamental features under which the phenomena under scrutiny are equivalent. From an empirical standpoint, the description aims at capturing the features of expe­rienced phenomena, hence at reading them off from particular perceptual instances. Thus the judgements for the inductive generalization are based on the recognition “at a single stroke” that a feature extracted from a particular instance specifies the nature of all phenomena of the same kind (1889: 22 n. 33, 93; see also Mezei and Smith, 1998: 33b). This is necessary to prevent the generalization being biased by the alleged priority of the subjective certitude or rationalistic argumentations that are shorn of reference to perceptual real­ity. On the other hand, the formulation of laws is characterized by Brentano’s interpretation of necessity. A feature is necessary if it is a part by means of which a phenomenon is a whole, such that its occurrence in experience with­out that part is self-contradictory. A sentence figuring in a law about necessary properties of phenomena might be paraphrased as a valid inference built upon correct negative existential judgements such as “there is the perception P and there is no P whose part x appears non-f” where “f” is a perceivable feature. This means that the experience P is dependent on x-appearances of f, but also that if f is altered or disappears, there will be no P. The method of descriptive psychology is designed for a theory of perception that is developed from with­in the perception itself. The epistemological priority that the natural sciences grant to experience is interpreted as the priority of the evidence about the in­trinsic properties of perception. Furthermore, the autonomy of such analysis of phenomena allows that the theory is a sort of formal theory of perception in the sense that the constituents, the features and connections of phenom­ena are treated like constants, variables, or syntactic rules in a logical language (1976; Smith, 1988b).

Indeed, Brentano (1982: 3) distinguishes between descriptive and experi­mental psychology as if between mathematics or mechanics and meteorology. In the first case, the propositions are expressed in rigorously defined and exact formulas, like those for the sum of the inner angles of triangles or the prin­ciple of inertia. In the second, the propositions are expressed in probabilistic formulas in which many variables are included, whose validity is bound to the determinable degree of confidence within the margin of error. Experimental psychology aims at finding the physical and the physiological conditions of perception through statistical correlations and generalizations of the physical causes.

Nonetheless, descriptive and experimental psychology are not opposed, but rather two sides of one and the same empirical science of perception. Brentano (1896/1897: 151-156) holds that explaining a fact is proving that it is a particular case derivable from a more general fact. Psychological and physi­ological explanations are different for the nature of the general fact taken as the antecedent. Descriptive psychology is not interested in the law of the sys­tematic association of phenomena with physiological antecedents. Yet, the abstract characterization of a descriptive theory does not rule out its empirical nature. It is subject to errors or different interpretations of evidence as in any other science ([1982] 1995: 32-34). Moreover it still admits of gaps, just as in the case of mathematics, while remaining internally consistent. Furthermore, if descriptive explanation is not required to specify all antecedent conditions of phenomena, then physiological findings may contribute to the psychologi­cal explanation. Brentano (1893, 1917) cites the law of specific sensory nerve energies, which is the physiological framework of his theory of “multiple quali­ties” meant to replace the construct of intermediate colors in von Helmholtz’s theory of color (for a complete account that involves a phenomenological reform of psychophysics, infra §§ 7.3, 7.3.1). However, it remains true that a physiological correlate may be a necessary but not sufficient condition, and the claim of its causal role will require the establishment of some restrict­ing assumptions (1982: 3). Therefore, descriptive psychology is still needed to discover independent findings that can even provide experimental psychol­ogy with well-defined objects of research (1896/1897: 153, 1982: 76, 129, 130). Indeed, the definition of the features of phenomena is obtained by crossing selected characteristics of phenomena with one another across paired differ­ent conditions, which is a method consistent with ordinary empirical science (1982: 49-54). Consequently, if a feature is proved to be a necessary part of a phenomenon, then one can design experimental settings to test it along with the role it plays in the connection of appearances (Baumgartner, 1986: 245).

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