Kant's Conception of the Development of Empirical Concepts

by Philip McPherson Rudisill

Written sometime before August 14, 2000
(with slight editing on 7/14/2013 and 2/27/2014)

A problem concerning the consistency of the Kantian system of recognition has been noted by Andrew Carpenter. The problem concerns the development and formation of empirical concepts. Briefly stated Kant has us understanding that envisagements/intuitions (Anschauung*) without concepts are blind.; and it is only by means of a concept that an envisagement/anschauung can be determined and a recognition (Erkenntniss) occur.** Empirical recognitions require an empirical concept for the empirical envisagement. The empirical concept arises through a procedure of comparing objects and focusing on their common elements and disregarding what is different, e.g., we might compare an oak with an elm and by ignoring what is diverse between them and retaining what is common we could arrive at an empirical concept of a tree as being a trunk with branches and foliage. But, as Professor Carpenter points out, it would seem that we would have to precedingly know and recognize the trunk and the other parts of the tree in order to be able to make this comparison and abstraction. But our recognition of these parts would be based on empirical concepts, and so they, too, would require us to have undertaken the same procedure; and so it seems that Kant has ensnared us in a vicious circle.

[* See Kant and the meaning of the Anschauung.]

[** I distinguish a determined envisagement from a recognition, and (if empirical) identify it rather with a perception. Accordingly we are to understand that a perception and a determined, empirical envisagement are one and the same thing. On its own the envisagement is a sheer sighting in light of which we are actually passive (and as I will discuss further below). It arises as it will. But then we also have the determined envisagement which we construct (in the case of pure envisagements, as when we put an imaginary circle into the space in front of the mime artist in order to make sense of the gyrations of his hand in drawing it***), and then we also have the determined, empirical envisagement where we are as certain of a shape or configuration of some sensations as we are of the pure circle of the mime artist, but which, unlike the pure circle, is given on its own (and not constructed by us) and so therefore requires a "second look" for this certitude. The "first look" is the appearance of the specter, but the "second look" is a careful, apprehending look for the purpose of being certain of the elements and arrangement and textures, etc., of the as yet unidentified object. For example: as an infant or todler I may happen to notice that the eyes of my mother and her nose (as mere shapes and without even names yet) seem to float around the room together, and I may then take a second look to be sure that that is actually happening; and in fact I may even pull at my mother's nose to see if I can remove it from such close proximity with the eyes.]

[*** See Circles In The Air.]

The relevant texts, I think, are essentially four, namely:

1. envisagements without concepts are blind (Critique of Pure Reason (CPR), A51/B75 [Page references to first (A) and second (B) editions of Kant's original work in German]),

2. empirical envisagements can be turned into perceptions by means of the categories (CPR, B Deduction, section 26, paragraph 4),

3. every recognition requires a concept, regardless of how imperfect or vague, and it is always something general with respect to its form, and which serves as a rule (CPR, A Deduction, Part II, Section 3, Paragraph 6), and finally

4. empirical concepts arise through comparison of objects of experience (Logik.3).

I propose a solution to this apparent inconsistency in the following manner. I submit that the envisagement/anschauung/intuition precedes the concept, and that the object of that envisagement, when empirical, is an empirical object, namely a specter/appearance (Erscheinung), and this is an unidentified or undetermined object (as Kant indicates in the very first section of the Aesthetic of the CPR). Now the envisagement is a grouping of a multiplicity which might very advantageously be called a “constellation,” namely a pattern or configuration which stands out to us as a singularity, e.g., the Big Dipper in the northern, night sky. In this regard and for our purposes here it might be worthwhile to consider the patterns formed by the tiles on bathroom or kitchen walls* and floors and the little blocks on checkered table cloths. These tiles present “invisible” mosaic patterns, e.g., X’s which suddenly appear and which just as suddenly vanish and are replaced with squares and which likewise vanish and are replaced in turn with rectangles and these in turn with parallel rows pointing this way and that, etc. This is exceedingly clear to anyone who has ever stared for a moment or two at such tiles. This configuring is accomplished as a reaction of our sensitivity to impressions of the objects about us, and are entirely independent of the understanding (although, of course, I will not recognize the X, for example, as a letter of the alphabet until I have come to learn the alphabet); and it is clear that the form of the specter/constellation (the various shapes and figures) resides in our way of looking at the tiles (which make up the material of these constellations), for obviously there is no X, no square, etc., in the tiles on their own at all. Furthermore and finally I see no reason to insist upon any predecessor for such sightings, e.g., I don’t have to ever have seen an X in order to see the tiles take such a shape (although I will not have a name to denote that particular pattern or shape).

[* In order for this to appear most dramatically the tiles should all be uniform in color.}

This suggests to me, in other words, that the sensitivity can give us objects (conglomerates, constellations) which have not yet been thought and so therefore for which no concept has been developed. This means in turn that while we see these patterns and configurations we are not able to tell what it is that we are looking at and so, in this sense, Kant is quite right to say that our envisagements, absent concepts, are blind.

These shapes can grow familiar to us, and while I will not say that we recognize them (for I wish to reserve recognition as an English rendering of “Erkenntniss”) nevertheless we are able to become comfortable with them, i.e., we reach a point where they are no longer novel. I would think that our early sightings of faces, e.g., our mother's face, would be of this sort. Likewise we can become familiar with the shape of a particular house without knowing what a house is, etc. But we have not yet introduced any notion of the recognition of an object, but rather merely the same sort of familiarity which we presume of the animals, namely a continuing strengthening of an impression through frequent and compounded exposure.

We are now ready to consider the perception of objects. It is when we come to the arena of perception that we can begin thinking of concepts and recognition. I will use Kant’s own example of a tree to present the development of an empirical concept. I assume that we have become familiar with a particular shape which we call Tree as a proper noun and without the least content, and where the term is nothing more than the name or identification of this particular envisagement. Now suddenly we hear the term Tree apparently being used for another shape and are imposed upon and confused by this, and we begin to wonder whether there is any connection between the term Tree and the shape as we had originally begun to think.* In an effort to work ourselves out of this unacceptable corner, i.e., where the sound is not associable with the sight, we undertake a perception of each of these objects (the original shape [the original tree] and the new, confusing shape [the other tree]) in that we gradually accumulate the elements making up each, i.e., the dark, thick, upright shape (an empirical sighting/envisagement of what we come to call the trunk), the dark, thinner shapes angling out from the first shape (later called the branches), and then the bushy green in and among the dark thinner shapes. Now in doing this we have both 1. perceived the elements of a tree and, at the same time, 2. we are comparing two shapes (trees) and it is then that the concept, and then also the recognition, of tree arises (where we are no longer dealing with merely with Tree as a proper noun of a specific configuration), for now we see indeed that a tree is a trunk (as we come eventually to name the dark, thick, upright shape) and some branches and foliage, and which can take various shapes and sizes.

[* This original identification of the sound with the picture/shape will also have been a work of the original, transcendental apperception in that we will have made a conscious unity of the two (the shape and the sound “Tree”) in order to obtain feedback from other people via a deliberate experiment; perhaps we will have pointed to the shape (tree) and said “Tree” in order to hear if others respond to us with approving voices.]

The concept of tree is now established within us and without need of any preceding concept regarding the elements, and so the problem is solved. As a recap and checklist we can follow the points which we listed earlier:

1. the original envisagement of the shapes of the tree was blind, for we could not tell what it was that we were looking at.

2. the tree developed from a mere envisagement (a mere shape) into a perception* (material for a concept and recognition) through the watchful- or careful-taking (wahr-nehmen), i.e., apprehension, of the multiplicity into consciousness (which is pretty much the thesis of the subjective portion of the A Deduction of the Categories in particular).

[* I identify a perception with a determined, empirical envisagement, and mean with that that a particular shape or pattern is established as fact, e.g., the configuration called Tree, but still where there is not yet a recognition of any object.]

3. by means of a comparison of the two objects (alleged to answer to the same name, e.g., Tree), a concept arises through a rule for the assembly of the parts (trunk, etc., but which themselves at first are also merely additional shapes) such that trees are recognized; and our envisagement is no longer blind, for now we can see trees, while before we could only see colorful shapes.*

[* Now once the terminology is developed, we can speak of the dark, vertical space as the trunk of the tree, and the long, dark spaces as the branches of the tree, etc.]

4. the concept of tree, which is empirical and is based on the empirical envisagement, was developed through the comparison and abstraction described in 3 above.

It seems then that Kant is quite consistent regarding blind envisagements and the formulation of empirical concepts which result then in empirical recognitions.

See also: Kant in a Nutshell.

To contact the author or translator, please e-mail: pmr@kantwesley.com
To The Table Of Contents on Kant Studies