My Recreation of a Caspar David Friedrich

The painting of which this photo is an unconscious imitation is perhaps the most common cover illustration for books of philosophy. Coming in second is probably that Vermeer with the woman holding the scales.

Bentham's Mummified Head

While visiting Bentham's Auto-Icon at University College, London, I learned that it is a MYTH that King's College students kidnapped the head and played a game of soccer with it.

It turns out that they only kidnapped the head and demanded that a ransom of 10 pounds be donated to charity.

The head is now kept in the college safe.

What else do they have in there?

The Problem of Cognitive Dynamics

On an episode of the Simpsons, Homer says the following to Marge:

"First you didn't want me to get the pony. Now you want me to take it back. Make up your mind!"

Homer does not grasp the way the expression of Marge's want has to change to accommodate changing circumstances. That's why it's funny.

One-Upmanship and Folk Psychology

From Stephen Potter, One-Upmanship, p. 22:

"After this opening treatment, Doctor may, under certain circumstances, ask Patient his symptoms. But he will let it be seen that he is not listening to what Patient is saying, and may place his hand on Patient's wrist, or better, stomach, as if to suggest that he as Doctor can tell more through the sensitive tip of one finger than from listening to the layman's self-deceiving, ill-observed, and hysterically redundant impressions of what is wrong with him".

Millikan, Images of Identity: In Search of Modes of Presentation

Strawson's "dot" model of co-referential thinking is anti-Fregean because:

(1) Thinking of an object as the same involves eliminating multiple modes of presentation
(2) A single dot can equivocate in its reference (so sense doesn't determine reference)

Millikan endorses the dot model and argues that Evans's dynamic thoughts are equivalent to the dot model.

Worries about the argument: there are functionally significant reasons for hanging on to multiple modes of presentation for an object, rather than eliminating them when you learn they co-refer. You might, e.g., want to hang on to a variety of modes of presentation for an object because other people refer to the object by way of those modes and don't realize they co-refer.

Steven Gross, Essays on Linguistic Context-Sensitivity

Summary of Stephen Gross, Essays on Linguistic Context Sensitivity

While this is about Stephen Gross’s Essays on Linguistic Context-Sensitivity, I want to make a few comments about a philosopher who plays an important role in Gross’s book. Charles Travis has developed a radical context-sensitive view in philosophy of mind and philosophy of language to which Gross is “very much indebted” (p. 3). Instead of sticking with the most commonly discussed context-sensitive singular expressions (demonstratives, indexicals, definite descriptions) Travis focuses on simple predicate expressions like “is blue” and “is a philosopher”. These kinds of expression display a much richer and more multifarious form of context sensitivity, which appears (prima facie) more difficult to handle using the devices of formal semantics. Gross gives a perspicuous catalogue of the forms of context sensitivity these predicate expressions display in chapter one of his book. One of the forms of context-sensitivity he calls context-sensitivity of “aspect”, which is that which Travis concentrates on. “Aspect”, Gross says, is a “catch-all [covering] the various parts, capacities, roles, states, activities, etc. that a thing can possess, play, be in, engage in, etc” (p. 4). A statement like “The apple is green” Gross says is aspect-dependent, which means that whether it is true or not depends on which aspect of the apple we care about being green on some occasion. Even if you hold the state of the apple constant, uttering “The apple is green” can be true on some occasions and false on others. For example, if someone asks me what kind of apple I’m looking at, I may truly reply “The apple is green”, thereby referring to its skin (which is what someone who wanted to identify the type of apple would presumably be interested in). But if someone asks me whether the apple I’m looking at is moldy inside, and I say, “The apple is green”, then I have thereby (according to Gross) said something false. Gross and Travis do not want to claim that the speaker has said the same thing in both cases, but implicated something false in the second. Gross worries, plausibly, about responses based on implicature that they require making seemingly arbitrary decisions about what cases will count as strictly speaking true or false and which cases will be merely implicated. The trade-off of ruling out rampant context-sensitivity is ruling in rampant falsehood.

Gross also gives, in the first chapter, a few positive arguments in favor of rampant context sensitivity: (i) it enables us to explain how “light dawns gradually over the whole” by making room for the role of exemplars in language learning, before a learner acquires full mastery of a concept (mastery that meets Evans’s steep requirement of satisfying the Generality Constraint); (ii) context sensitivity increases the efficiency of communication (p. 16), in that we don’t have to spend lots of time spelling out in full what aspect of the thing under discussion is important to us; and (iii) the most interesting suggestion Gross makes concerns the way in which context-sensitivity allows some of the parameters of content to remain open so that we can sharpen things up as we go along. This last argument in favor of rampant context-sensitivity plays an important role in the discussion of vagueness in chapter four. Chapter functions as a nice, clear contextualist manifesto.

Chapter two is a survey of three different conceptions of context, attributed to Lewis, Stalnaker and (roughly) Davidson. “Lewis’s conception is an ordered triple: the ‘location—time, place and possible world—where a sentence is said”; “Stalnaker favors representing such a context as the set of possible worlds compatible with what is presupposed”; “On the Davidsonian conception, a context comprises the features adverted to by a truth-theory that assigns truth conditions for the sentences of a given language” (p. 30).

Chapter three is an examination of what a context-sensitive Davidsonian truth-theory for a language would look like. The issues are complex, but Gross argues that it may not be possible to eliminate context-sensitivity on the right hand side of the truth theorems. Gross draws some tentative conclusions from this result, like the impossibility of a theorist completely specifying the linguistic knowledge of a speaker of the object language, since the theorist will be relying on his own context in formulating the truth theorems. I would have liked a little bit more hand-holding at this stage of the argument (that's more a statement about me than the book), but I’m sympathetic to Gross’s profound-sounding conclusion, that there is “an in principle limit—if you will—on our capacity for theoretical self-knowledge” due to the ineliminability of context even for the theorist. (It would be an interesting project to connect up this work on the ineliminability of context with the literature on the possibility of an “absolute” point of view). In a footnote, Gross works hard to qualify this claim, but one might wonder: what’s the point of doing all the technical work if you can’t say something big, exciting and Kantian (as Gross does)?

Chapter four proposes an interesting solution to the sorites paradox: there is a general condition on the expression of propositions that is not met in the case of the formulation of the sorites paradox, so when a skeptic tries to generate the paradox, he is not actually expressing any propositions. He is speaking nonsense. The discussion is provocative, but I’m queasy about accepting a general constraint on sense of whatever form—I’d prefer to hang on to the idea that “(If everything in the symbolism works as though a sign had meaning, then it has meaning.)” (TLP, 3.328). That Wittgensteinian criterion of sense and nonsense strikes me as much more consistent with the spirit of the context-sensitive project than the general principle Gross proposes.

The final chapter of the book is a free-wheeling, fun account of contemporary philosophical accusations of nonsense in discussions of ontology (Gross makes a point of citing Kim, van Fraassen and Searle rather than Carnap, Austin or Wittgenstein [p. 111]). Anyone who has wrestled with Travis’s pioneering work on context-sensitivity will enjoy Gross’s book for its clear defense of certain central contextualist commitments as well as its willingness to engage head on with philosophers outside the Wittgensteinian-Austinian camp.

Logic and Luck

Some reflections on Sorenson, "Logical Luck":

If we reject subjectivism about logical responsibility (Boghossian) as well as "strict liability" (Millikan), we can adopt a moderate view of responsibility, according to which sometimes we hold a subject responsible for irrational inferences he wasn't in a position to recognize as such. But most of the time we do not hold him responsible for irrational inferences he wasn't in a position to recognize.

This reasonable conception of rationality should go hand in hand with a revision of our understanding of cognitive content, if we understand cognitive content to be governed by norms of rationality. Cognitive content should not be understood as constrained either by the mere possibility of doubt (as the subjectivist would claim) or as constrained merely by the world (as the radical externalist would claim).

Cognitive content (sense) then becomes sensitive to contextual assessments of rational responsibility. As long as conditions are normal, our inferences can be fast and loose without getting us into trouble. When conditions aren't very good, we should be more cautious in our reasoning. If we aren't, then we can be subject to criticism even if our inferences continue to yield truths.

This kind of revision of the constraints on cognitive content should make room for dynamic thoughts, if conditions are good enough. It should also make coordination of indexical contents easier (since at least sometimes, they won't be so finely individuated).

Queen of Hearts Psychology

A tag for extreme externalism & rejection of transparency of sense:

Queen of Hearts Psychology.

From Alice in Wonderland:

"The Queen had only one way of settling all difficulties, great or small. 'Off with his head!' she said, without even looking round."

Someone who thinks that rationality "ain't in the head" might be described as following a similar policy.

Millikan provides a suggestion in "WQP" for developing a moderate view of meaning rationalism, but dismisses it on the grounds that the meaning rationalist is committed to meaning and rationality being completely in the head. Her suggestion is roughly that we can count as knowing sameness and difference of thought content a priori so long as we inhabit a favorable environment. But since we cannot know a priori that we are in a favorable environment, it seems that we can't know sameness and difference of content a priori.

That argument assumes that if we know a priori that two contents are the same or different, then we know a priori that the environment is favorable. Why not reject that assumption?

Rejecting it would go along with denying that to know a priori whether two contents are the same or different requires excluding all possible doubt that they are different or the same.

In favorable circumstances, we could then credit a subject with a priori knowledge of the contents of his thoughts. In unfavorable circumstances (twin-Earth switching cases, the presence of lots of barn facades, zookeepers painting donkeys to look like zebras), one's ability to tell sameness and difference of content a priori breaks down.

Adopting this approach would even make dynamic thoughts possible: the standard worries about dynamic thoughts (e.g., that it is never in principle impossible that one is mistaken when one reidentifies an object at different times) could be handled. A subject could maintain a single singular sense for an object through time and across different sensory modalities.

This is a way of working out the view Campbell argues for in "Is Sense Transparent" and that Evans takes for granted in parts of VoR (in the discussion of "here" thoughts, e.g. in Ch. 7).

The point of all of this is to make externalism compatible with a circumscribed version of what Millikan calls "meaning rationalism".

White Queen Psychology

Two extreme, opposed views of rationality and externalism: Boghossian (pro-transparency, anti-externalism); Millikan (pro-externalism, anti-transparency).

Moderate views: Campbell, Evans.

Is the moderate position tenable? I think you could occupy such a position if you thought:

(1) Transparency is not a universal condition; it is only the normal condition for thought

and

(2) Though we cannot rule out all possible doubts about failure of judgements about sameness and difference, we can usually rule out all relevant doubts.

Both extreme positions (radical externalism and radical anti-externalism) rely, I think, on a skeptical assumption: that we must be able to exclude all possible doubt about sameness and difference of contents before we are entitled to claim transparency for thoughts.

This puts a new twist on these positions: you can have transparency, if you are lucky (if you are in the right environment, etc.). We can preserve the old-fashioned, Cartesian understanding of rationality, provided we aren't in a bad epistemic situation (which isn't completely up to us).

Cognitive Access and Semantic Puzzles

Kripke's Puzzle About Belief:

Pierre assents to both:

(1) London is pretty

and

(2) London is not pretty.

And, by stipulation, Pierre is rational--he doesn't let contradictory beliefs pass.

Options to resolve the puzzle:

(A) Reject disquotation: Though he assents to them, Pierre doesn't actually believe (1) or (2) (Marcus).

(B) Reject assumption of Cartesian access: Pierre can believe both without impugning his rationality if he doesn't have immediate access to sameness and difference of content (Owens, Brown).

(C) Adopt subjective senses: "London" has different senses associated with it in (1) and (2), so the associated contents of belief are different. Pierre is not irrational.

Or:

(D) Reject assumption of Cartesian rationality: Pierre is not perfectly rational. He holds two contradictory beliefs. Rationality is not completely the subject's responsibility: he can be irrational without recognizing it (Millikan). Thinking (with Boghossian, Owens, Kripke, others) that rationality is completely within the subject's control is another version of Quine's dogmas of empiricism: it represents a fundamental cleavage between the domain of rationality and the domain of empirical inquiry. Maintaining such a cleavage requires an unacceptably formal understanding of rationality. (Marcus argues plausibly against the rationality of someone who holds contradictory beliefs.)

There is luck involved in rational culpability just as there is luck involved in claims to know, and in the ethical assessment of actions.