Perhaps because I spent fifty years in the classroom, I think of this blog as an open-ended continuing class, with no registration limits, no prerequisites, no constraints on subject matter, and of course, no exams or grades. It calls to mind the great line from one of the songs in Guys and Dolls, sung by Nathan Detroit: "It's the oldest established permanent floating crap game in New York."
Thinking of the blog in this way has many advantages, but there is one downside, I find. It makes me feel that I ought to be able to answer any question posed by one of the "students," especially when it is asked by someone who actually was my student in real life. Andrew Blais asks several questions provoked by my lengthy post on the relationship between Kant's First Critique teaching and the theses of his moral philosophy. Now, Andrew really was my student, back in the days when I was still a Professor of Philosophy. I directed his fine doctoral dissertation, which then became a very good book. So when he asks me about Plato and possible worlds and one thing and another, I feel that is incumbent upon me to have answers. Unfortunately, I am really, really out of touch not only with the very latest journal article of 2014, but also with the very latest journal article of 2004, 1994, 1984, and even 1974. So my reply to his questions, which I shall give here, are offered with more than the usually cautionary caveats. [I am reminded of a practice of the Vietnamese restaurants in Paris. They put mouth-watering photographs of some of the dishes on their big, glossy menus, and then add in small letters the warning that the actual dish may not actually look like that!]
After an opening clearing of his throat, Andrew asks: "An often mentioned metaphor or allegory for the appearance/reality cut is Plato's cave. So, why is it that the contradiction in Kant's understanding doesn't have a counterpart in the cave allegory? If there is no counterpart, what has to be added to get the analogous contradiction?" The distinction between appearance and reality, on which Kant's entire philosophy is founded, does indeed have its origin in Plato's Allegory of the Cave, as Andrew says. The difference is this: Plato [in his guise as Socrates] argues that the philosopher, by a process of rational inquiry and critique, can ascend from the experience of appearances to a knowledge of reality, while still, of course, retaining an awareness of appearances and an understanding of their relation to reality. Having achieved that higher [or deeper] knowledge, the philosopher can then act in the world in accordance with it, even though he will to be sure be scorned, as Socrates was, by those still trapped in the realm of appearance. That is the allegorical meaning of the journey out of the cave into the sunlight. Something like this notion is present in the teaching of the Inaugural Dissertation of 1770, although in that work Kant does not talk about principles of right action. But as a consequence of his encounter with Humean scepticism, Kant gave up the position of the Dissertation, and instead embraced the view that our knowledge is forever confined to the realm of appearance, for all that we can formulate an empty idea of independent reality. Hence for Kant, but not for Plato, there is a problem in making sense of the claim that the philosopher acts in the realm of appearance as one who has a knowledge of independent reality. There is also the historically more complicated difficulty that the Greeks, Plato included, had no clearly formulated notion of laws of nature, and hence no philosophically usable idea of causal determinism.
Andrew goes on to make a second point that raises much more complicated questions. Here is what he says: "The relation between the noumenal ethical self and the phenomenal physical self is similar to the counterpart relation between objects in different possible worlds. Socrates in one world is F, but in another, he is not F. The conundrum is how to understand how it could be that it is the same Socrates that is and is not F. The Kantian conundrum is how to understand how it could be that Socrates is a causally caught up object and yet also a free agent. If there is an way to understand the modal case, isn't it clear that there should be a similar way to understand the Kantian case?"
I am now going imitate to Wiley Coyote and rush pell mell off the edge of a cliff, my legs pounding fiercely until I pause, look down, and discover that I am no longer on solid ground, at which point I will plunge to my destruction. I am counting on the many serious philosophers who read this blog to weigh in and correct me.
For the first two millennia and a bit more of Western Philosophy, metaphysics took primacy of place over epistemology. Questions about being, about what is, were thought of by virtually all the great philosophers as primary, questions about what we can know being considered as secondary. Aristotle distinguishes in the Physics between things that are first in the order of knowing [such as the nature of substance] and things that are first in the order of knowing [such as sensory accidents.] It is for this reason that he called the group of essays dealing with Being and associated matters "First Philosophy." [It was only a medieval accident that these essays came to be called Metaphysics. In the manuscript of Aristotle's works with which the Scholastics were working, those essays came after the Physics -- ta meta ta physika.] His speculations about the processes of our knowing were consigned to a relatively secondary work, De Anima.
At the beginning of the seventeenth century, René Descartes dramatically reversed this age old prioritization with his proclamation that the one absolutely certain truth on which he was able to found his reasoning was cogito, I think. The power of Descartes' arguments swept away two thousand years of belief in the priority of the order of being over the order of knowing. For the next two centuries, every great philosopher devoted his most intense speculations to the question, "What can I know?," and all of them followed Descartes in taking as their primary task an analysis of the cognitive powers and limits of the human mind. It is for this reason that we find works with such titles as Essay Concerning the Human Understanding, Principles of Knowledge, A Treatise of Human Nature, and, of course, Critique of Pure Reason.
Even in the middle of the twentieth century, when I was an undergraduate, epistemology took pride of place over metaphysics, which in the analytic circles I frequented, was viewed with suspicion. All of this changed shortly thereafter. As a consequence of the work of Saul Kripke and a number of other philosophers, questions of being once more strode forward to the head of the class and took the place of questions about the cognitive capacities of the human mind. [This may help to explain why Saul, then an undergraduate, stopped coming to my course on the Critique after a few weeks, apparently having decided that nothing was happening there that he needed to pay attention to.] All of this was tricked out with some very fancy modal logic, and eventually gave us this talk about possible world semantics and such.
Which brings me to Andrew's question. Why can we not make use of the modern discussions of possible worlds and so-called "counterpart theory" to resolve Kant's problem of making his talk about a causally determined phenomenal self and a morally free noumenal self compatible? Because those modern discussions assume that one can talk about different possible worlds and their relation to one another without first determining what the subject of these discussions, the self doing the reasoning, can know about itself. Take any text on possible world semantics and ask the simple question, "Who is doing the writing?" Who is the self making those assertions? How can that self arrive at the knowledge claimed in those assertions? Kant grounds his entire philosophy on the proposition, "The 'I think' can be attached to every one of my representations," which is his version of Descartes' cogito.
To put the point in a suitably Greek manner, there is no pou sto[a place 'where I may stand'] from which the modern possible worlds semanticist can contemplate both this world and other possible worlds equally so as to formulate propositions about them -- or at least, so Kant would argue, if he were to encounter this modern philosophical school.
Andrew concludes: "This similar way to understand the Kantian case is adumbrated in your point about how the author of one narrative can't, or perhaps can, appear in another author's narrative. A narrative picks out a set of possible worlds. Suppose that the characters of our respective narratives pick out objects that are identical in the way that the Socrates who is F is identical with the Socrates is not F. To borrow a phrase, they are transworld characters. What is needed to resolve the contraction, in sum, is an understanding of how the same thing can have contradictory descriptions such as causally networked and free, but expressing the problem in terms of narrative already points the way to a resolution?"
Once again, note Andrew's precise wording. "A narrative picks out a set of possible worlds." But that is, strictly speaking, not true. It is the narrator who picks out a possible world, and the narrator does this by creating the possible world by his or her words. On this matter, I must refer anyone who has made it this far into this post to my essay, "Narrative Time" [which Andrew has clearly read], archived on box.net and accessible through the link at the top of this page.
Well, I am looking down, and I have not started to fall yet, but that may just be because the Road Runner is not up on possible world semantics. On to Timothy.