Two Kinds of Content in Kant’s Theory of Concepts

The notion of conceptual 'content' (Inhalt) features conspicuously in the formulation of some of Kant's most signature doctrines. The distinction between analytic and synthetic judgments is frequently glossed by Kant as the distinction between those judgments that merely explicate the 'content' of given concepts and those that do not. It is in the context of articulating the requirements on cognition that Kant famously declares 'thoughts without content' to be empty. And Kant insists that the distinction between sensible and intellectual representation is a substantive, content-level distinction, rather than a merely formal, logical distinction between degrees of distinctness. Now, if it should transpire that Kant employs 'content' as a technical term in these passages, we can expect that a proper understanding of this technical usage will significantly improve our understanding of Kant's systematic project in the first Critique.
     The technical status of the term has recently been confirmed from two different directions. First, Clinton Tolley (2012) has alerted us to the fact that Kant's use of the term 'content' in the first Critique is systematically tied to contexts in which he discusses the 'relation' (Beziehung) of cognition to objects. On this basis, Tolley contends that Kant operates with a quasi-Fregean model of cognitive content in general and conceptual content in particular, on which the content of a concept is its relation to an object. Secondly, Lanier Anderson (2015) has made a compelling case that Kant's talk of conceptual 'content' betokens his membership in a logical tradition that stretches back at least to the Port-Royal Logique, on which the 'content' (or 'comprehension') of a concept is the set of partial concepts it 'contains', without which it would fail to be the concept that it is.
     We must take seriously both Tolley's and Anderson's findings. But a question that has gone unaddressed in the literature, and which I take up in the present talk, is whether their findings point towards a single, univocal notion of content. I argue that they do not. However, rather than jettisoning the findings of either author, we should respect both sets of findings by distinguishing two technical senses of 'content' within Kant's theory of concepts. I label the two notions 'Content-R' (content as relation to an object) and 'Content-C' (content as that which is contained in a concept), and advertise the textual grounds and exegetical benefits of this regimentation.
     As to textual grounds, a firm distinction between Content-C and Content-R is a precondition of making sense of the Table of Nothing in the Amphiboly. Kant there distinguishes two kinds of 'empty concepts' (A292/B349), corresponding to positions 1 and 4 of the Table. This distinction between two kinds of emptiness would be unintelligible if Kant did not employ a complementary distinction between two kinds of contentfulness. I argue that the distinction between Content-C and Content-R is precisely the relevant distinction: concepts at position 4 - whose logical constituents are mutually contradictory - are empty insofar as they lack both Content-C and Content-R, whereas concepts at position 1 - concepts with logically compatible constituents but whose objects cannot be counted as possible - possess logically coherent Content-C, but are empty insofar as they nevertheless lack Content-R.
    As to the exegetical benefits of drawing the distinction, I advertise two. First, it allows us to shine a light on Kant's doctrine of sensibility. Is Kant's doctrine that sensible and intellectual representation exhibit content-level differences to be understood in terms of Content-C or Content-R (or both)? Is the fundamental difference between the two kinds of representation one that surfaces at the level of their representational constituents or the kinds of relations in which they stand to objects? Second, it facilitates a fuller treatment of Kant's theory of cognition. Plausibly, those representations that fail to constitute cognitions are those whose conceptual constituents lack Content-R. If that is so, there will be as many kinds of cognition as there are kinds of concepts that can possess Content-R, amongst which we should include at least mathematical, natural scientific, observational, and psychological concepts. Exhibiting the different conditions under which these various types of concepts can possess content-R will accordingly clarify Kant's understanding of the nature, conditions, and scope of empirical cognition.

←zurück zum Programm.