Old translations

A large number of Kant's publications are available for download at the Online Library of Liberty. They are available in HTML, PDF and E-Book format. All of them are rather old English translations, mostly from the late 19th and early 20th Century.

Freedom and two-aspects interpretations

At Kant blog, Andrew Roche is discussing how we can make sense of Kant's arguments regarding freedom within the framework of a two-aspects interpretation. His argument is based on distinguishing between the theoretical and practical perspectives that agents can take and arguing that freedom is only salient when we take up the latter perspective. Andrew and I are having an interesting discussion in the comments section.

More on noumena and phenomena

The noumena/phenomena debate is spreading through the blogosphere. Duck at DuckRabbit, has an interesting post discussing the relation between the Cartesian conception of objectivity and Kant's transcendental idealism. At Siris, there is a post suggesting that we have so little knowledge of noumena that we cannot decide between two-worlds and two-aspects interpretations. The idea is that the whole debate is misguided since the commentators do not really take seriously the limitations of our knowledge that transcendental idealism identifies. I have written a few short comments in reply to the latter post.

Kant song

An amusing song about Kant is available online.

The objective validity of 'judgements'

According to Kant, something is only a judgement if it is objectively valid. Otherwise, though it might have the grammatical form of a judgement, it does not classify as a 'judgement'. In the Prolegomena, he makes a distinction between 'judgements of experience' and 'judgements of perceptions'. Only the former properly classify as 'judgements' and it is this sense of the term that he mostly uses in the Critique of Pure Reason. This is the reason why Kant criticises logicians who define a judgement as a representation of a relation between two concepts. (cf. B140-141) Allison, following Prauss, interprets the need of judgements being 'objectively valid' as meaning that they must have a truth-value, not that they must be true. This interpretation seems unacceptable as Kant at several places equates objective validity with truth. For example, at A125 Kant explicitly equates objective validity with truth. Furthermore, in the Prolegomena (Prolegomena, 4:298) Kant states that "the objective validity of the judgement of experience signifies nothing else than its necessary universal validity" and states that this signifies that the "judgement agrees with an object", which is equivalent to the definition of truth given at A57/B82. In addition, Kant's classification does not commit him to the absurd conclusion that "every judgement is true, simply in virtue of being a judgement". (Allison: 2004, p. 88) Allison appears to invert the direction of reasoning - it is not that because it is a judgement that it is true, but that only because it is true that it classifies as a judgement. This implies that there are no false judgements in the sense that nothing that is false will classify as a 'judgement', but this does not bring with it the absurd conclusion that there can be no error.

A similar method of classification can be seen to be at work with respect to the notions of 'experience' and 'imperative'. The way Kant uses the term 'experience' differs from contemporary usage insofar as experience, understood in its modern usage, must be objectively valid in order to classify as 'experience' in Kant's sense. This can be seen when looking at the Transcendental Deduction since it is a strong notion of experience that constitutes the starting point of the Deduction. (cf. Ameriks: 1978) Similarly, every imperative can be expressed as an ought-statement but this does not imply that every ought-statement is an imperative. While all ought-statements are grammatical imperatives, they are not necessarily 'imperatives' in Kant's sense as these are universally valid principles that are necessitating for the will (this includes both hypothetical and categorical imperatives - the former are hypothetically necessitating, whereas the latter are categorically necessitating). Thus, there can be formulations that have the grammatical form of imperatives, such as imperatives of etiquette, without necessitating the will and, accordingly, such statements do not classify as 'imperatives'. We can thus see that Foot's criticisms are misplaced when she says that "[i]t follows that if a hypothetical use of 'should' gave a hypothetical imperative, and a non-hypothetical use of 'should' a categorical imperative, then 'should' statements based on rules of etiquette, or rules of a club would be categorical imperatives." (Foot: 1972, p. 309) No matter what form the should-statement has and no matter how it is used, it only expresses a categorical imperative if it is an objective principle that necessitates the will.

Search engine

A useful search engine of Volumes 1-23 of the Akademie Ausgabe is available online.

Kant poll

I recently stumbled across a poll over at Kant Blog. The question is: Do you believe in the 'two-worlds' or 'two-aspects' interpretation of transcendental idealism? I think that this is a nice idea and would like to encourage those who have an opinion on this matter to submit their views.

I personally favour an ontological reading of transcendental idealism. I do not think that an epistemic or methodological interpretation is sufficiently strong to do the work that Kant wanted it to do. This holds in particular with respect to freedom. Transcendental freedom plays a highly substantial role in the Third Antinomy, as well as in Kant's ethical theory. Both the Groundwork of the Metaphysics of Morals and the Critique of Practical Reason rely heavily on transcendental freedom and I do not think that a merely epistemic reading of transcendental idealism is able to fulfil this role.

I am slightly dissatisfied with the 'two-worlds' v. 'one-world' distinction. I think that there are quite a few ontological readings of transcendental idealism that are still in some sense 'one-world' theories. In particular, I have in mind van Cleve's theory that treats phenomena as virtual objects. Van Cleve notes that the virtual-object theory "does not give us a dualism of two sorts of existent. ... My interpretation is nonetheless dualistic in the following sense: the distinction between appearances and things in themselves is a distinction between two separate universes of discourse - not between two ways of discoursing about the same class of objects." (Problems from Kant, p. 150) Similarly, Langton's interpretation of transcendental idealism as involving a distinction between intrinsic and relational properties is an ontological reading that does not quite fit the 'two-worlds'-label. Moreover, 'two-worlds' talk has some rather unpalatable connotations. It seems to suggest that these worlds have the same status, but I believe that it is perfectly possible to combine an ontological reading with the claim that the noumenal realm is ontologically prior to the phenomenal realm. Accordingly, it is probably better to distinguish between 'ontological' and 'epistemic/methodological' theories, rather than between 'two-worlds' and 'one-world' interpretations.

Kantian ethics videos

Video recordings of the Kantian Ethics Conference that took place at the University of San Diego in 2003 are available online.

The keynote lectures were delivered by Henry Allison, Stephen Darwall, Thomas Hill Jr., Robert Pippin and Allen Wood. This a highly impressive set of speakers and I can assure you that the talks are well worth listening to.

The Two Assertive Frameworks

In Appendix K of Problems from Kant, van Cleve discusses the problem that anti-realists have in making sense of claims about the physical world that are supposed to be true in the absence of perceivers, given that they argue that the physical world is somehow mind-dependent. He asks: "How can you eat your cake empirically and still have it transcendentally?" (p. 249) The focus of the discussion is Foster's distinction between `two assertive frameworks'. According to Foster, one should distinguish between assertions made within the framework of physical theory and assertions made about this framework. Given this distinction, it supposedly follows that we can make sense of claims about physical reality in the absence of perceivers if these claims are seen to be made within physical theory. At the same time, we can accept the claim that physical reality is mind-dependent since it is a claim about the framework of physical theory. Van Cleve argues that there is no formal incompatibility between the two assertive frameworks, but that "inconsistency arises as soon as one adds that minds are necessary for the obtaining of the experiential constraints. A proposition that excludes the existence of minds cannot be made true by a state of affairs that essentially includes them." (p. 252) That is, the truth-makers of claims made within physical theory are experiential constraints. These experiential constraints only obtain as long as minds exist. Accordingly, we get inconsistency when the claim made within physical theory excludes the possibility of experiential constraints. The claim made within physical theory is such that it undermines the possibility of that very claim being true. In footnote 67, van Cleve refers to impure forms of phenomenalism that could possibly account for experiential constraints without the existence of minds. "Foster mentions the possibility of a phenomenalism that grounds physical reality in a 'causal field' defined by the constraints it would impose on minds if they existed (Case for Idealism, p. 240). Such a field would be similar perhaps, to Mill's 'permanent possibilities of sensation.' But Foster does not avail himself of this option, and it is not in any case a pure phenomenalism." (p. 315)

While I agree with van Cleve that pure phenomenalism faces this difficulty, I think that Kant's phenomenalism is not a form of pure phenomenalism, but a noumenally grounded phenomenalism. It is the noumenal sphere that grounds physical reality. According to transcendental idealism, the obtaining of experiential constraints is not dependent upon the existence of minds, but upon the existence of the noumenal ground. We can have claims made within physical theory that are made true by noumenal truth-makers which ensure that the experiential constraints obtain even though no minds exist.

What I am not so sure about yet is what implications this has for the two assertive frameworks. Does this mean that we can salvage the two assertive frameworks? In fact, is there any need need for these two frameworks? When appealing to noumenal truth-makers, do we still need the two frameworks? Or does the need for the frameworks result from a special problem that only faces pure forms of phenomenalism, which would imply that the two frameworks are irrelevant for impure forms of phenomenalism, such as transcendental idealism?

Oxford Scholarship Online

Oxford Scholarship Online is an online library that contains full-text versions of books published by OUP, including specially commissioned chapter abstracts. It contains the following books on Kant:

Interpreting Kant's Critiques by Karl Ameriks

Kant's Theory of Mind by Karl Ameriks

Kant's Theory of Knowledge by Georges Dicker

Kantian Humility by Rae Langton