Principles of Experimental Phenomenology: Complexes and Fusion

Qualitative perception of Gestalt has more recently gained an increased interest in the area of cognitive science, particularly by virtue of its non-reductive metaphysics against strong neurocentred and artificial intelligence models. This work on the structures of cognitive and perceptual experience is today being rediscovered and potentially provides immediate relevance to physiologically oriented cognitive science. The theory of Gestalt is, of course associated with the members of Berlin School such as Max Wetheimer, Wolfgang Köhler, and Kurt Koffka, and with the members of the Austrian school, Graz School, such as Benussi and Kanizsa, but its roots date back to the remarkable work of Brentano and his students, Christian von Ehrenfels, Edmund Husserl and Carl Stumpf, whose work we shall investigate here, particularly in what concerns the genesis of the concepts of complex and Gestalt.


1. Ehrenfels “On Gestalt Qualities”

The seminal paper by Ehrenfels “On Gestalt Qualities” (1890), is concerned with the reflections on the question “what complex perceived formations such as spatial figures or melodies might be”. The answer to this question requires the doctrine of intentionality by Brentano and the works on the ontology of the mind by Husserl. As it is well known, for Brentano, there are both simple and complex mental acts between the intuitive and the non-intuitive components of psychic phenomena of different sorts, between the various different sorts of phenomenally give boundaries and continuity.

The essay on “Gestalt qualities” consists in a conceptual proposal. Ehrenfels suggests that the German term ‘Gestalt’ which means ‘shape’, ‘figure’, ‘form’ should be generalized in a certain way. In his and Brentano’s view a spatial shape or Gestalt is perceived ⎯ is given a in visual presentation ⎯ on the basis of a complex of sensations of individual elements having ‘distinct spatial determinations’. In seeing the elements and their spatial determinations, one is able to apprehend the shape as an additional object (quality, attribute) as it were side by side with its associated elements. Our total experience is, therefore, something distinct from the experience of a mere sum or complex of sensory elements. What Ehrenfels proposes is that wherever we have a relation of this sort, between a complex of experienced elements on the one hand and some associated unitary experience of a single invariant structure on the other. This ‘structure’ should be conceived as the Gestalt. The unitary experience is structurally analogous to the experience of a spatial shape.

The Gestalt concept is then generalized further to embrace also complex objects of experience founded on inner perceptions ⎯ one’s presentations of one’s own elementary feelings, acts, or mental states. Sensory data from different sensory modalities may, according to Erenfels, combine together in such a way to provide the foundation for mixed Gestalt qualities of specific sorts. Having identified spatial shapes, melodies, chords, and complex taste as first order Gestalt qualities founded on given elementary sensations, Ehrenfels recognizes that these qualities, too, may combine together in such a way as to found new, second orer qualities which are themselves capable of founding third order qualities, and so on, in principle without limit (Smith, 1988).

As it is now explicit, Gestalt qualities, for Ehrenfels, are not wholes embracing their fundamenta ⎯ tones, colours, tastes, smells ⎯ as parts. Rather, they are additional unitary objects, existing alongside the unitary elements with which they are associated. The Gestalt quality is not a combination of elements but ‘something new in relation to these, which exists together with [their] combination, but is distinguishable from it’. It is a special sort of structure, a ‘positive content of presentation bound up in consciousness with the existence of complexes of mutually separable elementary presentations’. For Ehrenfels there are also unitary entities at successively higher levels, what one might call relative elements or ‘quasi-substances’, objects which, even though they do not belong to the ultimate wordly furniture, are yet given to consciousness in an unitary way and have to be recognized as such by any adequate theory. The quality in question here is associated with, but not reducible to, complexes of points, lines, symptoms and events, apprehended direct and immediately.

Ehrenfels  asks about the specific contents of the presentation, Is content a real entity? Something individual and spatio-temporal? Or is it rather an ideal or abstract universal, multiply exemplified in the acts of different subjets towards the same foundational elements? What is the structure of the “complex of presentations” that serves as the foundation or carrier of Gestalt quality?

Are we to acknowledge both Gestalt qualities and sui generis complexes, which they would be qualities of? And how is a complex of mutually separable elementary presentations related to those complex fusions of elements, which Ehrenfels also recognizes? 

For Ehrenfels, so also for Husserl, we grasp the configuration and its quality in one glance ⎯ not by collecting together in intuition a sum or a sequence of objects or relations, as occurs in those higher order articulated acts of counting and calculating which are the main subject-matter of Husserl’s early work. Husserl explains this through the notion of ‘fusion’, a notion he takes from Stumpf, signifying the absence of phenomenal discontinuities or boundary lines, as for example when one perceives an array of colour in which there is a gradual transition from red to blue or a glissando in which one musical tone passes continuously into another ⎯ the relations between these relevant parts become thereby fused together (in a figural moment, in Husserlian’s terminology). Husserl ontology plays an important role to reflections raised in Ehrenfels paper. The Gestalt problem is, in effect, a problem of unity, and Husserl suggest there are two ways in which the unity can come: either objects are such that they don’t need additional objects to glue them together into a whole; or they are such that they are in themselves not sufficient to make a unity and require the presence of some additional object to glue them together. Such additional objects may be of two sorts: independent objects (like a mass of glue), or dependant objects, capable of existing only in consort with the objects they serve to unify (moments of unity or figural moments).
​
The moments of unity in Husserl’s phenomenology are important because they serve to bridge the ‘phenomenological’ and the ‘objective’ spheres. The subject and the object do not simply co-exist but are in fact related together in a single unified whole.


Carl Stumpf: phenomena of ‘fusion’

Ehrenfels’s dichotomy between complex of experienced elements on the one hand and some associated unitary experience of a single invariant structure on the other, can be further understood through Carl Stumpf’s central ideia of  ‘fusion’ implies a anti-reductionist, descriptive attitude that represents an attempt to produce what we might call a natural philosophy of complex experiences, including not only the phenomena of fusion and purely aggregative phenomena, but also a range of different sorts of Gestalt phenomena considered to be lying between these two extremes. 

Stumpf draws the very important distinction between complex and Gestalt. Complex is a whole of sense contents; and the latter, the relational attribute, the network of relations between those contents. This network is somehow unitary because when we hear a chord or a melody we hear a relational attribute, not a complex or succession of dyadic relations. The network has a structural dimension because it can be transposed, transferred from one complex of relata to another. 

For this psychological mechanism to happen, Stump remarks, there must be something cognitive in our awareness. For to grasp a Gestalt is to grasp not merely an individual as such, but also the abstract net of transferable relations which is its essence (Stumpf, 1939/40, p. 229, 242). Thus, Gestalt can never be perceived of themselves but always only in and of some given formed material ⎯ a tone or phoneme or timbre are founding elements that may involve physical or physiological complexity, but it is phenomenologically non-articulated and therefore has no Gestalt. Because Gestalt phenomena are primarily given, we often see Gestalten without recognizing parts, in fact, it sometimes takes effort to delineate figures with their ground, to discriminate constituent parts (p. 247). A Gestalt is a whole of relations, but under certain circumstances only part of this whole may be perceived ⎯ and this part may be a Gestalt in its own right. 

One can apprehend the Gestalt of a melody only when one has heard the entire sequence of tones in such a way that a total impression has been gained through a discursive process (see also the post Gestalt and Qualitative Relations)

But the effect of the melody on our feelings does not begin only after it has been completed: we follow it through in its development from the very beginning, accompany this development with expectations, surprises, tensions, releases, for which foundation is provided by repetition, similarities, and so on.

References

Smith, B. (1988). Foundations of Gestalt theory. Munich and Vienna: Philosophia.

Stumpf, C. (1939/1940). Erkenntnislehre, 2 vol., Leipzig: J. A. Barth, 873 p.; 2nd ed., Lengerich: Pabst Science Publisher, 2011.

von Ehrenfels, C. (1988). On “Gestalt qualities.”. Foundations of Gestalt Theory. Munich: Philosphia Verlag.(Original work published in 1890).