Newton, A. (2015): Kant on the logical origin of concepts

PDF Link

  author = {Newton, Alexandra},
  title = {Kant on the Logical Origin of Concepts},
  shorttitle = {Kant on the Logical Origin of Concepts},
  journal = {European Journal of Philosophy},
  volume = {23},
  number = {3},
  pages = {456–484},
  year = {2015},
  abstract = {In his lectures on general logic Kant maintains that the generality of a representation (the form of a concept) arises from the logical acts of comparison, reflection and abstraction. These acts are commonly understood to be identical with the acts that generate reflected schemata. I argue that this is mistaken, and that the generality of concepts, as products of the understanding, should be distinguished from the classificatory generality of schemata, which are products of the imagination. A Kantian concept does not provide mere criteria for noting sameness and difference in things, but instead reflects the inner nature of things. Its form consists in the self-consciousness of a capacity to judge (i.e. the Concept is the ‘I think’).},
  file = {~/Library/Mobile Documents/iCloud~com~sonnysoftware~bot/Documents/be-library/newton2015_kant_on_the_logical_origin_of_concepts.pdf},
  doi = {10.1111/j.1468-0378.2012.00545.x},
  url = {},
  langid = {},
  location = {},
  keywords = {},

The dilemma:

Either the concept ‘tree’ is formed from the simpler concepts of ‘branches’ and ‘leaves’, which in turn are formed from more simple ones, all the way down to representations of common features that are simply given in sensibility (such as the ‘simple ideas’ of colours in empiricist accounts). Or the concept ‘tree’ is presupposed prior to its formation, and is somehow already involved in my perceptions of the spruce, willow and linden. [@newton2015, 457]

Though Newton takes Ginsborg and Longuenesse to opt for the circularity arm, she thinks both are objectionable and proposes an alternative:

I shall argue that Kant’s logical account of concept-formation does not confront the above dilemma, because he is not committed to its assumption that concepts reflect criteria for identifying and distinguishing things in one’s environment. Kantian concepts are not mere tools for classification, but reflect the inner nature (or logical essence) of things. The above interpretations do not sufficiently distinguish between the imagination and the understanding, or between the generality of schemata, as ‘products of the imagination’ (KrV A140/B179) and the generality of concepts, as products of the understanding (KrV A19/B33). (458)

However the rest of the paper reads as a series of equivocations and I can’t really make heads or tails of it. I’m not sure how the claim about internal natures resolves the problem with circularity. Also note that the appeal to a ‘logical essence’ is confused since Kant often uses this phrase to indicate a way of distinguishing a thing from other things, rather than getting at anything genuinely essential to it.

She says some things that are in the ballpark of things I think:

The concept ‘tree’, I argued, arises not from a comparison of different representations with regard to the ways they affect me (which requires no rational capacities at all), but from a reflection of the capacity to spontaneously use a representation in combination with different (possible) ones through a rational capacity for cognition (the understanding). Reflection on this capacity involves an appreciation of the rational relations between the representation and different ones in a ‘whole of cognitions’, since it involves an appreciation of the possibility of using the representation in combination with others in thought or cognition. (469)

The emphasis on concept formation as involving a link between actual and possible representation is right, as is the emphasis on this link being systematic. I think that this gets at an important feature of Kant’s account of generality. But the rest is pretty obscure.

One useful thing Newton does say is that there is a difference between using a representational mark to track a distinction—what she calls the ‘external’ use of a mark as criterion—and using a mark to track the thing distinguished—what she calls the ‘internal’ use of mark as criterion.

we must be careful to distinguish between two different senses of ‘because’. When we say that things have branches ‘because’ they are trees, we may mean that we can identify various things as having branches if we know they are trees. The mark ‘tree’ thus serves here in its external use as a criterion for picking out things with branches. But when we understand that this particular thing is a tree, our understanding of it as such is grounded in cognition of it as something that has branches; the mark ‘branches’ serves here in its internal use as a ground for cognizing the tree itself. This capacity to understand the ‘internal’ nature of the tree (as something that has branches, etc.) does not rest on any special faculty of intellectual intuition, but is characteristic of our discursive capacity to judge. (471)

This is right as far as it goes, and gets at an important difference between acquaintance and cognition.

Against positions like that of Longuenesse Newton writes,

I wish to suggest that marks do not have determinate positions in ‘logical space’ prior to logical reflection. Contrary to the above interpretation [i.e. Longuenesse], there is indeed something logically significant about self-consciousness, since marks acquire logical positions in relation to other marks only through reflection, and not prior to it. The schemata of the categories (as products of transcendental syntheses of the imagination) thus cannot be operative in logically ‘structuring’ intuitions prior to our acquisition of concepts (and in particular, of the categories themselves). (470)

A summing up:

I have argued that the form of concepts (their generality) does not rest on the formation of schemata of the imagination through which we represent objects in intuition. Empirical schemata can be formed prior to the formation of empirical concepts, but these schemata rest on a kind of reflection that plays no role in the logical origin of concepts. The transcendental schemata of the categories, by contrast, cannot be generated prior to the logical reflection that issues in the categories. Thus, neither of these products of the imagination contribute to the logical acts of concept-formation. … What I have shown here is merely that the form of concepts, as it is treated in general logic, rests on nothing other than the understanding as a capacity for (synthetic) judgement, since a concept’s form (generality) consists in the self-consciousness of this capacity. (475)

Icon by Nun from The Noun Project. Website built with Org-mode, Hugo, and Netlify.