I’m feeling pretty demoralized this evening, so the only thing to do is try and distract myself so I don’t have to think about things.
Lately I’ve been thinking a lot about transcendental arguments and their status. I have written this post a few times already in the past, but like a person working through a trauma who must repeat it endlessly in the form of nightmares or neurotic symptoms, I believe I must go over this ground once again. And if I must repeat, then this is because I am myself a reformed transcendental idealist who must therefore marshal arguments convincing to myself. In many respects, the transcendental argument is Kant’s central contribution to thought, his ultimate secret ninja judo move. Outside of philosophy I get the sense that there’s a lot of confusion as to what a transcendental argument is. I often hear it confused with an appeal to the transcendent. However, in many respects, the transcendent and the transcendental are opposites. When we appeal to, for example, Platonic forms to account for justice or to God to account for moral laws we are making an appeal to the transcendent or that which is beyond and independent of both the world and the subject or mind. Take the standard Platonic argument for the existence of the forms (and here I’m presenting a vulgar, cereal box version of Plato).
The argument runs something like this: When faced with all the the things to which justice pertains, we note that they are very different and share no resemblance to one another. For example, in what way do serving one’s function within the polis and getting a coke out of a coke machine when you put a dollar in the machine resemble one another. Both of these events are instances of justice, yet when we examine the properties or qualities of these events, we find no quality shared by the two. Similarly, when we enter into debates and discussions about justice, we seem to all approach justice in different ways. However, apart from the crassest Protagorean relativist, we all nonetheless agree that while we might not know what it is, there is a truth of the matter or fact of the matter concerning justice. In short, justice is not simply a subjective sentiment or opinion, but something real that exists in its own right. But what is this real thing that exists in its own right? Plato’s proposal is that the just is a form, a universal, that exists in its own right, independent of all instances of the just and all opinions about the just. The form of the just is ideal, but its ideality is not a subjective ideality. Indeed, as Derrida likes to point out, the ideal is the most objective of all. It is neither an object in the world (a material object), nor an idea in the mind (a subjective entity), but an ideal entity that is entirely real, eternal, universal, and so on. For Plato, even if all humans ceased to exist, even if there were no individual objects in the world, the form of the just would continue to exist for all eternity. The forms are thus transcendent to subjects and objects. They are the most real things of all.
read on because having arguments for abstruse and abstract issues is concretely important!
In this way Plato is able to secure a referent for claims about the just that would allow for a selection between the true and the false. Interestingly, for Plato, it is not simply beliefs or propositional attitudes that can be true or false, but objects themselves can be true or false. The measure of the truth or falsity of an object will be a function of its ontological distance from these ideal forms.
The transcendental, by contrast, is something entirely different. Where things like Platonic forms (and God) are transcendent or independent of both subjects and objects, the transcendental is, strictly speaking, immanent. There have been a number of proposals as to just what the transcendental is immanent to. Kant’s famous answer was that the transcendental is immanent to mind. Where Platonic forms are mind independent, Kant’s transcendentals are a priori structures of mind. On the strong reading, they do not exist independent of mind, but are structures of mind, not unlike rules governing a computer program, that structure human experience. On a weak reading, we cannot determine whether or not the world itself, independent of humans, is structured according to these a priori structures, but we do know that mind is structured in this way. If Husserlian phenomenology is a transcendental phenomenology, then this is in the sense that it is structured by the intentional structure of consciousness. However, in Husserl the issue is vexed insofar as the relationship between Husserl’s eidetic phenomenology, constitutional phenomenology, and genetic phenomenology is rather Byzantine and difficult to untangle. In Habermas, the transcendental is immanent to communication or dialogue.
Why, then, appeal to the transcendental? What work is it doing? The Internet Encyclopaedia of Philosophy gives a better gloss on the nature of transcendental arguments than I can must right now in my demoralized state. As the author, Adrian Bardon, so nicely puts it:
Transcendental arguments are partly non-empirical, often anti-skeptical arguments focusing on necessary enabling conditions either of coherent experience or the possession or employment of some kind of knowledge or cognitive ability, where the opponent is not in a position to question the fact of this experience, knowledge, or cognitive ability, and where the revealed preconditions include what the opponent questions. Such arguments take as a premise some obvious fact about our mental life—such as some aspect of our knowledge, our experience, our beliefs, or our cognitive abilities—and add a claim that some other state of affairs is a necessary condition of the first one. Transcendental arguments most commonly have been deployed against a position denying the knowability of some extra-mental proposition, such as the existence of other minds or a material world. Thus these arguments characteristically center on a claim that, for some extra-mental proposition P, the indisputable truth of some general proposition Q about our mental life requires that P. (my emphasis)
In short, a transcendental argument is an appeal to a condition that explains how a certain form of knowledge, judgment, experience, or activity is possible. The example of causal judgments is perhaps clearest. Here we are not talking about any particular or specific causal judgment. For example, we are not talking about judgments of the form “the match lit the paper”. What is at issue here is not the question of how we have a knowledge of the myriad causal phenomena we make judgments about. Rather, what is at issue is the question of how it is possible for us to make causal judgments as such. Put differently, what are the conditions under which causal judgments are possible?
So why is this an issue? Well, when we make a causal judgment we’re not simply making a judgment about temporal succession, but about a necessary relation between two events. To understand this point, consider the difference between the following two judgments: “I made my coffee and the sun rose.” “I lit the paper and it caught on fire.” Both judgments involve succession in time, but in the case of the first judgment there is no dependency relation between coffee being made and the sun rising. The sun would rise regardless of whether I made my coffee. The case is not the same when discussing the relationship between lighting the paper and it catching on fire. Here the paper would not catch on fire unless it was lit by the match. In short, there is a relationship of necessity in the latter event.
So why the problem? Well the issue here is that of how we are able to cognize this necessity. Note that from the standpoint of sensation our two judgments are indistinguishable. Both experiences consist in one event following another. What I do not have is a sensation of the necessity of the relation itself. So the question is, “how can I make judgments about empirical matters (matters pertaining to sensations) that involve necessity?” If we begin from a Humean conception of sensations (and the thesis that experience is structured as Hume or the classical empiricists described it is tendentious), then any judgment involving necessity should be impossible. How would the idea of necessity (truths who’s denial would imply a contradiction) possibly arise from disconnected (for Hume) sensations? This seems impossible.
It is precisely here that we get the super-ninja transcendental judo move. Kant’s move– and here I’m simplifying tremendously –is to point out 1) that we do make such judgments, 2) that we can’t account for how such judgments are possible on empirical grounds or arising from experience, therefore 3) since 1 and 2, the condition under which such judgments are possible must be an a priori category in the mind, specifically the category of cause-and-effect, that the mind imposes to experience. In short, since we can’t get the idea of necessity from our sensations (there’s no intrinsic connection between the redness of an apple and its delicious causal powers as I discovered much to my dismay at my grandfathers in an unfortunate encounter with a very realistic wax apple), but we do make judgments of necessity, therefore the capacity to make these judgments must arise from a structure of the mind, not something we learn from the world. These categories, of course, are strictly immanent to the mind. Voila! We have grounded necessity.
Now note, once again there is a strong and a weak reading of this thesis:
Weak Reading: The transcendental argument is simply explaining where our idea or concept of necessity arises from. Since nothing in our experience of particular entities and events itself suggests or grounds the idea of necessity, and since the idea of necessity must come from somewhere, and since it cannot come from sensation, it must come from mind.
I am not sure whether anyone advocates this sort of weak reading, though when I reflect back on my discussions with Alexei-Tuffini (who really knows his Kant) before they went sour, I can only make sense of his objections to my characterization of Kant based on a weak reading such as this. While I might disagree with this solution to the problem of how we get the concept of necessity given that I hold that developmental or what are called “genetic” accounts in philosophy (not to be confused with what takes place with DNA) are more plausible, I would not a priori exclude such a thesis as there’s nothing particularly anti-realist about this thesis. In short, here one is simply making claims about how we get the concept of necessity, not making the claim that mind constructs reality. Perhaps Graham or someone else can explain to me why this weak reading is objectionable.
However, I do not think this weak reading really gets at the spirit, letter, or manner in which Kant’s texts have been appropriated. This for two reasons: First, in responding to Hume’s skepticism I take it that Kant wasn’t simply attempting to account for our ability to make judgments of necessity, but rather was seeking to account for what renders our causal judgments reliable. Second, I do not believe that the letter or subsequent appropriation of Kant’s text supports this reading. Ergo the
Strong Reading: Mind structures the world through its a priori structure, such that we are forever prohibited from speculating as to whether the world itself, independent of humans, has this structure, but are instead restricted to how the world is for-us.
Note that the weak reading does not really address Hume’s skepticism. Hume had argued that based on sensation alone, we are unable to arrive at anything more than a probabilistic grounding of causal claims. Moreover, we are unable to establish a rational ground for our belief in relations of necessity in causal judgments. From an Enlightenment perspective this must have been devastating as so much of the critique of superstition was premised on causal claims. If the weak reading does not address Hume’s skepticism or the thesis that we have no grounds for establishing that the future must be like the past, then this is because while it establishes how we might make judgments of necessity, it does not establish why relations between events are structured in a necessary fashion. In other words, while the Newtonian convert endorses the thesis that Newton has stripped nature bare, revealing her alluring secrets, we are still left in the position of being able to rationally demonstrate that the future must be like the past. The weak reading remains in this quandary.
It is only a strong reading that holds that the mind actually structures reality that is able to avoid Humean skepticism. Hume’s thesis was never a thesis aimed at undermining our ability to make judgments of necessity. No. Hume’s skepticism was directed– in a very healthy fashion –at our ability to know necessary relations. And if we adopt the weak reading of Kant, we have no more responded to this quandary than we have solved the problem of cold fusion. It is only if we adopt the thesis that the mind structures reality, that reality is like a turtle’s shell that we carry about on our back, that we can respond to Hume’s challenge. This would be the anti-realist version of the transcendental argument, for now the claim is that we have no access to reality as it is in-itself, but only reality as it is for-us. And why? Because if we are the one’s structuring reality then we can never speak of what that reality might be apart from us. Therefore all of our claims about reality necessarily possess an “asterisk” in the sense that they are necessarily for-us. We can see why Meillassoux will be lead to claim that anti-realist arguments necessarily founder on physics and evolutionary theory, for insofar as these domains require us to talk about a time of being prior to the human we are necessarily prohibited from such claims as we can only speak of what being is for us. Ergo creationism.
Having unfolded the transcendental argument in crude outline, here I would like to shift gears a bit and raise the question of why, precisely, transcendental arguments are not circular or question begging. In Deleuze’s Logic of Sense we encounter precisely this criticism. Deleuze writes:
The presupposed primacy of signification over denotation, however, still raises a delicate problem. When we say “therefore,” when we consider a proposition as concluded, we make it the object of an assertion. We set aside the premises and affirm it for itself, independently. We relate it to the state of affairs which it denotes, independently of the implications which constitute its signification. To do so, however, two conditions have to be filled. It is first necessary that the premises be posited as effectively true, which already forces us to depart from the pure order of implication in order to relate the premises to a denoted state of affairs which we presuppose. But then, even if we suppose that the premise A and B are true, we can only conclude from this the proposition in question (let us call it Z)– we can only detach it from its premises and affirm it for itself independently of the implication –by admitting that Z is, in turn, true if A and B are true. This amounts to a proposition, C, which remains within the order of implication, and is unable to escape it, since it refers to a proposition, D, which states that “Z is true if A, B, and C are true…,” and so on to infinity. (LS, Continuum Edition, 19 – 20)
This argument sounds obtuse, but is basically Deleuze’s argument against transcendental arguments of a particular variety. In Difference and Repetition Deleuze will refer to this as “tracing the transcendental from the empirical”. Now why is this problematic. Remember that the whole point of a transcendental argument, whether in the domain of physics, mathematics, moral theory, communicative theory, or political theory was to ground the necessity and universality of its claims. It does this by recourse to a shared condition that must be presupposed to render a particular experience or activity or capacity intelligible.
However, if transcendental arguments trace their conditions from the empirical, if they simply double what we encounter in recognition by pushing a formal and abstract version of this empirical actuality back into mind, consciousness, or communicative interaction, then transcendental arguments beg the question. Why? They assume the truth and universality of the experience from whence they begin and then simply double it. But this necessity, truth, and universality was precisely what was at stake. Consequently, if we arrive at the transcendental through an exercise of tracing it from the empirical, we risk simply enshrining prejudice, habit, recognition, rather than getting at real necessity. This would be especially the case in normative transcendental theories in politics and ethics where we sense, all too clearly, that it’s simply a question of grounding specific prejudices about what one believes ought to be right and just, rather than genuinely grounding the right and just. This would be the reason that Deleuze was led to claim that thinkers like Kant and Habermas are defenders of the State on the grounds that they simply present a mythological ground for the “right” and “justice” of the State rather than a genuine philosophical grounding of the right and the just. In other words, what we get is an argument to the effect “we make these arguments, we can’t ground them by recourse to experience, therefore there must be a necessary a priori structure at work in x” (where x is mind, consciousness, communication, and so on). Yet if this is the case, the question that inevitably arises as to why this isn’t the worst form of speculative dogmatism or merely a version of the standard informal fallacy of wishful thinking and rationalization (which, no doubt, advocates of transcendental approaches will vigorously reject on the grounds that “they teach critical thinking courses too”– not well, we’d respond).