Categories of the Understanding

Question

Kant's categories are supposed to tell us what kinds of judgments human minds are capable of making, but they are rather artificial. One commentator I've read says Kant was more concerned with filling out the structure of four groups of three than he was in coming up with something truly systematic. Has anyone taken on the task of really analyzing the forms of human judgment from a Kantian perspective, perhaps interacting with psychology and linguistics to find what really are the basic forms of judgment?

Answer 1

Kant's philosophical project was not primarily empirical, but rather, it was a critique of the limits and possibilities of human understanding. His categories aren't meant to be derived from empirical observations about how people actually think, but rather, they're intended to depict the a priori structures that, according to Kant, make human experience possible.

That being said, there have certainly been efforts to apply Kantian principles to fields like psychology and linguistics, though these are often complicated by the fact that Kant's philosophy is not primarily empirical. For example, cognitive linguistics, which examines how language reflects our underlying cognitive structures, has some parallels with Kant's project. However, it often diverges in significant ways, given its empirical basis and focus on specific linguistic phenomena, rather than the a priori structures of human understanding.

On the psychological side, the influential psychologist Jean Piaget drew heavily on Kant's work. Piaget's theory of cognitive development suggests that as children grow and learn, they develop new cognitive structures that shape their understanding of the world—structures that, in some ways, resemble Kant's categories of understanding.

Answer 2

Kant's categories are supposed to tell us what kinds of judgments human minds are capable of making, but they are rather artificial. One commentator I've read says Kant was more concerned with filling out the structure of four groups of three than he was in coming up with something truly systematic. Has anyone taken on the task of really analyzing the forms of human judgment from a Kantian perspective, perhaps interacting with psychology and linguistics to find what really are the basic forms of judgment?

Depends on what you mean by artificial.

Today, the prevailing view is not that there are perceptual-conceptual necessary rules that order experience. Today, broadly speaking, the view is exactly one that Kant inveighs against in his Critique of Pure Reason, namely that logical form, logical theorems, logic itself is an abstraction and formalization out of natural language/experience.

Natural languages have syntax, but that syntax is not thought of as constitutively structuring experience. Rather, syntax consists of regularities that human observers generalize from spoken and written natural language. Kant believed, roughly, the inverse.

However, to do justice to your question, there are schools of thought in the philosophy of mind that are in some affinity with Kant, at least as regards concepts (not sensibility). These would be computational theories of mind, and in particular, the Language of Thought Hypothesis, propounded most famously by Jerry Fodor. These theories essentially hypothesize that thought is akin to syntactico-logical manipulation. Namely, the mind computes much like a computer does: you have a semantics/data and that data is structured and manipulated by a set of rules supplied by the syntax.

These theories have fallen somewhat out of favour as explanations of the mind with the recent success of AI, which is not based on symbolic manipulation, but network structures and learning algorithms that do not require the specification of rules in advance. In neuroscience the prevailing view is that the mind works much closer to how artificial neural networks do; After all the latter are loosely modeled on the physical structure of the brain.

To return to the matter of the choice of Kant's Categories. The first two, quantity and quality, are not ad-hoc but correspond to pure intuitions, space and time respectively, and, at least superficially, could be said to apply to all possible propositions/judgments. Remember that Kant's definition of judgement does not correspond to the linguistic definition of sentence. Sentences are much broader, and include questions for example. Judgments are propositions that try to represent possible states of affairs in the world. Kant believed that Aristotle's logic was more or less complete, so it's only natural that he would include the two judgment forms that universally apply to all propositions in Aristotle's logic.

The other two, relation and modality, pertain to the dynamical composition of possible experience. These correspond to Kant's dynamical principles in his system of principles, i.e. for relation: substance, causality, and community; and for modality: possible states of affairs/experience, actual experience, necessary conditions of experience.

As relates to Kant's analysis of each of his categories into triads: he may have been motivated by symmetry to some extent, but Kant provides justifications for each of the triads in the text. In hindsight, the triads are logically redundant.

As relates to quantity, universal and particular jointly imply singular. As regards quality, affirmation and negation jointly imply what Kant means by infinite.

As regards relation, categorical and hypothetical are reducible to each other: you can express a hypothetical in categorical form and vice versa, but Kant didn't think so. Kant thinks that exclusive or (disjunction) is it's own logical form, but you can clearly express it with other logical operators, e.g. and and not, for example.

Finally, as regards modality, again you can logically reduce them to two: you can define necessity in terms of possibility and impossibility; impossibility in terms of necessity and possibility and so on and so forth.

So, at least three of the categories are unequivocally reducible to dyads (relation can be reduced to one).

But on the whole, Kant's motivations for the choice of categories and their three-fold structures are not logical (although to be precise Kant does think that the primitive logical forms correspond one-to-one to the a priori categories) but are intimately tied to his project of accounting for a priori synthesis. As you read the text, you'll see that the categories give us the system of a priori synthetic principles that structure all possible experience.

While Kant's project is not taken seriously by current science in its transcendental form, there are theories that are Kantian in flavour. e.g. Chomsky's universal grammar, and as the other poster noted, Piaget's genetic epistemology. These matters are still rigorously debated and there's no prevailing consensus about the "native ontology of the brain", so to speak.

Answer 3

Kant's table of categories constitutes a heuristic that guides Kant's three Critiques and the Metaphysical Foundations of Natural Science. It is not exhaustive by any means, and Kant never claims it to be, I believe, but it's not as bad as is sometimes thought. One distinction that Kant draws which is very significant is the distinction between mathematical/constitutive and dynamical/regulative categories. Mathematical categories make measurement and quantification of space and time possible and thus enable us to pursue empirical inquiry as such. Dynamical categories have only regulative use - their applicability to experience is never ultimately justified, as they posit explanatory connections which can be amended in the light of the future experience and provide only a building block of a complete system of natural science. Kant utilises some variant of this crucial distinction in all of his important works, even in his philosophical aesthetics, and it's quite independent from the exhaustiveness of the table.

One thing that is important to remember that, although Kant derives his table of categories from forms of judgement, the categories themselves actually are richer in content, often having to do with various modalities (the category of cause and effect, for example, doesn't describe a material implication, but a genuine causal connection). He thus doesn't depend on, now obviously outdated, logic of his times so significantly as one might think.

Also, while the categories provide a general hint according to which various parts of Kant's critical enterprise are structured, they're never applied thoughtlessly. Let's for example look at Kant's Metaphysical Foundations of Natural Science. The book is subdivided into four parts, each corresponding to three categories under one of four headings from the first Critique:

1. Metaphysical Foundations of Phoronomy - the categories of Quantity
2. Metaphysical Foundations of Dynamics - the categories of Quality
3. Metaphysical Foundations of Mechanics - the categories of Relation
4. Metaphysical Foundations of Phenomenology - the categories of Modality

But it is immediately clear that the second chapter, concerned with Dynamics, despite of this classification, involves not only mathematical construction of quantities in intuition, but also considerations about cause and effect (a relational category) and various qualitative categories. The relation of the table of categories to Phoronomy, Mechanics and Phenomenology chapters is (quite) strict but nevertheless the Dynamics chapter shows that Kant wasn't able to achieve 1:1 correspondence, even though many propositions of the first Critique make appearance in the Metaphysical Foundations of Natural Science as premises for reasoning which shows an intimate connection between the two works.

As a sidenote: almost all philosophers which in some sense saw their work as continuing Kant's critical project acknowledged the incompleteness of the table. Hegel, for example, objected to there being four classes of categories, as he thought a proper division must be triadic. He, however, acknowledged some truth in Kant's table, as he compared Kant's categories of Quality, Quantity and Relation to those of Quality, Quantity and Measure from the Doctrine of Being of the Science of Logic.