It is not unusual, in discussions about Kant, to hear supporters of Kant emphasize that he is an empirical realist and a transcendental idealist. It is important to understand what Kant has in mind by empirical realism and why it is radically different than realist ontologies. At this late hour I will not do this issue the justice it deserves, but hopefully indicate some pointers that will help to clarify the issue. No one is forgetting that Kant claims to be an empirical realist in these discussions, above all those that advocate realist ontologies. Nor are realist criticisms of Kant based on the idea that somehow he is subjectivist or a subjective idealist. Empirical realism is something radically different than a genuine realist ontology. When Kant describes his position as an empirical realism, he is not asserting a realist ontology, but is making a claim about intersubjectivity. What Kant is saying is that the items that populate experience are “objective” in the sense that what we experience is intersubjectively communicable and universal by virtue of the transcendental structure of subjectivity or mind as outlined by Kant. In other words, for Kant we are entitled to say that when the sun warms the rock (here I’m drawing on his famous distinction between perception and experience in the Prolegomena), we’re entitled to claim that this causal relation is an objective truth, i.e., intersubjectively universal.

Nonetheless, while Kant is an empirical realist and this is a commendable thing (was it ever in dispute that he wanted to establish the objectivity of science and mathematics?), he remains a transcendental idealist. In short, Kant’s empirical realism only extends as far as the subject and humans. He nonetheless remains committed to the thesis that what objects might be independent of humans, and whether objects exist as our empirical claims portray them, is something that we can never know and which must be carefully excluded from philosophical discussion. For Kant, even in his empirical realism, there’s always an “asterisks” containing the qualification “for us and apart from us we can never know”.

read on!

One of the things I’ve learned in my discussions with defenders of Kant is just how extensive and far ranging the impact of Strawson’s reading of Kant in The Bounds of Sense is in Anglo-American circles. In many respects I think Strawson has been a silent interlocutor throughout the famous Kant wars, regardless of whether or not these interlocutors have actually read the text (the influence is still clearly there), and has caused a lot of confusion. Putting my cards on the table, I think Strawson is flat out wrong in his reading of Kant and has caused a tremendous amount of confusion in subsequent Anglo-American scholarship on Kant. Strawson’s reading of Kant is what might be called a “deflationary reading”. Strawson sets out to salvage what can be redeemed in Kant and abandon the rest. He downplays Kant’s idealism or thesis that objects conform to mind, not mind to objects, and emphasizes Kant’s transcendentalism, treating Kant’s thought as merely asking “under what conditions are such and such propositions possible?” Simplifying what Strawson is up to, this translates into the rather modest claims like objects must be situated in space and time, that they have causes, that they must endure (be substances), yadda, yadda, yadda. What falls out in this portrayal of Kant is that mind imposes space and time, the categories, etc., etc., such that we cannot claim that being independent of mind is spatial or temporal, that things-in-themselves are substances or have causes, and so on. When viewed through this lens– and again, I don’t think the interlocutors have to have actually read Strawson to more or less advocate this reading, they could have inherited in their studies and through the secondary lit they’ve mostly engaged with –I can see why these defenders of Kant would be perplexed by realist criticisms of correlationism, anti-realism, and Kantianism. As a matter of textual hermeneutics, however, I do not think this modest reading of Kant and its descendants can be supported. It’s simply not what Kant is claiming.

Now I’ve been criticized for using the language of “objects conforming to mind” on the grounds that “Kant doesn’t say this in these terms”. In my view, this sort of criticism is just plain stupid. On the one hand, it is certainly an entirely valid paraphrase of what Kant is arguing. On the other hand, these very same critics gush all over Lee Braver’s A Thing of This World who uses exactly the same language in discussing Kant’s epistemology. This is yet another example of the double standard and janus faced argumentative techniques of these interlocutors that treat certain forms of argument and thought as admissible in one context, only to turn around and decry them when the same lines of argument and phrasing are employed in another.

There is no resemblance between transcendental realism and empirical realism. Transcendental realism is an ontological thesis, whereas empirical realism is an epistemic thesis. Transcendental realism is a set of claims about what objects are regardless of whether anyone’s around to know them, whereas empirical realism is a thesis about intersubjective universality or what can be communicatively shared among human subjects. Empirical realism is a thesis about our access to beings, whereas transcendental realism is a set of claims about the being of objects regardless of whether we know them or have access to them.

Now for the anti-realist such distinctions cannot but seem perplexing as the anti-realist does not believe we can draw a distinction between theory and the world. Of course, if this is the case, then one wonders why the anti-realist bothers to engage in discussion at all as all philosophical differences become a matter of taste or personal preferences as there’s no independent world, existence, or being that could disconfirm any of these positions. Apparently for the anti-realist the only theory that is prohibited is realist theory, despite the fact that we have no independent criteria for excluding any particular theory. This aside, transcendental realist ontology is based on a series of epistemological considerations. Thus, for example, if the world were not differentiated and structured our empirical inquiry would be impossible because there would be no mind-independent differences to disconfirm theories and hypotheses. Since inquiry is possible, it follows that the world is structured and differentiated independent of minds. This simple point already undermines any holist tripe where being in-itself is portrayed as a holistic continuum that subjects somehow carve apart. There must already be differences and structures at work in the world for inquiry to take place. It cannot all come from subjects. Likewise, it must be possible to form closed systems for inquiry to be possible insofar as it would not be possible to isolate variables were it not possible to form more or less closed systems.

Finally, third, it must be the case that causal mechanisms and objects must be capable of being active and acting without producing any sort of observable effect. Here is one of the major differences between Kant’s empirical realism and transcendental realism. Kant’s empirical realism is a thesis about regularities of sensation. Let us return to his famous distinction between judgments of perception and judgments of experience in the Prolegomena. A judgment of perception says something like “the sun shines, the rock is warm”. It posits no causal or objective relation between the sun shining and the warmth of the rock. A judgment of experience says “because the sun shines the rock grows warm.” Here the two sense-events (the shining sun, the warmth of the rock) are linked by a causal relation imposed by mind. For Kant causal claims are regularities of sensation or impressions that constantly occur together and that can be experienced by a subject with the same transcendental constitution.

The crucial point however, is his emphasis on regularities of sensation or constant conjunctions of sensation. Not so for the transcendental realist. First, the transcendental realism notes that there are all sorts of situations where causal mechanisms are present without the consequent event being produced. Given that the causal relation is still operative, this is enough to undermine the empiricist thesis that causality consists in regularities of sensation or a constant conjunction of sensations. Second, the transcendental realist notes that there are all sorts of constant conjunctions of sensations that don’t involve a constant conjunction of sense-events. I make my morning coffee every morning and the sun rises. My making of the morning coffee does not cause the sun to rise. Kant however is committed to the thesis that there must be a causal relation here because of the constancy of the conjunction of sensations at the experiential level. He is unable to distinguish between these constant conjunctions and genuine causal relations. Finally, he is unable to explain why it should be necessary to carefully construct experimental closed systems to isolate variables and discover causal mechanisms as all conjunctions of sensations are, for Kant, equal. Kant suffers from the problem of epistemic actualism, basing his epistemology on the actuality of experienced sensations and their conjunctions, and treating this as the sole domain of access. Yet without making a genuinely ontological move, dispensing with questions of access for the moment, all sorts of epistemological issues become entirely unintelligible.

Above all it should be remembered that OOO and transcendental realism are fallibilisms. I get the sense that the anti-realists among us think that us realists are somehow saying that we can sit in our armchairs and just know reality without having to investigate it. This seems to be how, based on a rather superficial reading, they understand the signifier “speculative” in “speculative realism”. Like so many things in this world, the situation is precisely the reverse. The object-oriented ontologists are the ones who are modest in what we can know and can’t know, emphasizing that knowing takes work, that we must actually grapple with the world, and that we only get knowledge in bits and pieces, and that hypotheses about how things are are often mistaken. It is difficult to see how this sort of fallibility is even conceivable or possible within an anti-realist framework. Given the impossible of a disadequation between world and theory is it really a surprise that philosophy has degenerated into discussion of texts and a war among rival aesthetic tastes and normative commitments? Where real being has disappeared as even a possibility all we can do is talk about talk or talk about how people have talked about the world and talked about talk, and engage in struggles over taste and normative commitments. In claiming that the world is differentiated and structured the object-oriented ontologist doesn’t claim to know a priori what structures and differences compose the world. This is something that can only be discovered through long and laborious collective inquiry (the idea that knowledge is a relation between a subject and an object rather than a collective product is one of the biggest bits of nonsense in all of philosophy), it cannot be discovered from the armchair or through “speculation”. All the object-oriented ontologist is committed to is that the world is differentiated and structured and that this differences and structures aren’t merely a product of mind, language, society, etc.