Reading Notes on John Hawthorne’s “Direct Reference and Dancing Qualia” In (T. Alter and S. Walter) Phenomenal Concepts and Phenomenal Knowledge (OUP, 2006).
Hawthorneâ€™s main thesis: Direct reference theory for phenomenal concepts is incompatible with the dualist possibility of dancing qualia.
What is the gist of Hawthorneâ€™s argument?
Hawthorne wants to argue that if qualia can dance, then there are certain situations in which one might not be sure whether one concept and another concept refered to the same qualia. Further, Hwthorne wants to argue, in such situations, it looks like what is going on is more like wondering whether â€œHesperus = Phospherusâ€ than being unsure about the a priori knowable â€œHesperus = Hesperusâ€. Thus, if qualia can dance, Fregeanism, not Russellianism is a better theory of phenomenal concepts.
What is direct reference?
A direct reference theory is a theory about the content of a term. On a direct reference theory, the content of a term just is the object the term refers to. Since Russell like this sort of theory, it is also called â€œRussellianâ€. According to the Russellian, thereâ€™s no difference between the content of â€œHesperusâ€ and â€œPhosphorusâ€ for they just refer to the same object, namely Venus. Fregeans, following Frege, disagree. â€œHesperus = Hesperusâ€ is a priori knowable but â€œHesperus = Phosphorusâ€ is not. Fregeans conclude that there is more than just reference to content. Content also includes sense. So, â€œHesperus = Hesperusâ€ is a priori knowable because the terms flanking the equals sign share a sense and â€œHesperus = Phosphorusâ€ is not a priori knowable because of a difference in sense between â€œHesperusâ€ and â€œPhosphorusâ€. One way of thinking of Russellians is that they think sense and reference are exactly the same. Thus, they would have to hold that â€œH=Hâ€ and â€œH=Pâ€ are both a priori knowable.
What are dancing qualia?
According to contemporary dualists, a being physically just like you can differ from you in whether they have inverted qualia, no qualia, or dancing qualia. When you look at an apple, you have red qualia. When your physical twin with dancing qualia looks at a red apple, the qualia â€œdanceâ€ from red qualia to blue qualia and back again. One crucial thing to keep in mind regarding the way dancing qualia are discussed in the Hawthorne paper, is that it is assumed that beliefs are fully determined by physical properties. So you and a person physically just like you would have the same beliefs, and this would be true even if that person differed from you in having dancing qualia. This means that, according to the dualists in question, a personâ€™s qualia can dance without the person noticing, because noticing would entail a change in belief which would entail a physical change.
Whatâ€™s the deal with Hawthorneâ€™s central thought experiment?
The crucial thing to get out of the thought experiment about Fred and Twin Fred is that they are both thinking â€œthus = thusâ€ and attempting to think about the same qualia each time. But they have been told that at some time their qualia will dance. So they are unsure whether â€œthus = thusâ€ is true. Suppose the first thus, call it â€œthus1â€ refers to a red quale and â€œthus2â€ refers to a blue quale. Since qualia can dance without you noticing, this is like pointing in the dark. You wouldnâ€™t be sure whether you are pointing to the same thing twice. Suppose you were looking at a red screen and God told you that in a few moments the qualia on one side will start dancing. So you are now wondering whether the qualia on the left are the same as the qualia on the right. Would you be able to know a priori whether â€œthus is thusâ€ is true? It seems not. But if Russelianism about phenomenal concepts is right, you would be able to figure out a priori whther â€œthus is thusâ€ is true, and your temporary lack of certainty is just a minor breakdown in your ability to figure out stuff a priori. According to the Russellian, it is NOT due to â€œthus is thusâ€ being not a priori.
What the Russellian needs to try to do
The Russellian needs to explain how it is possible for something that looks like a simple identity statement (like â€œ2 = 2â€ as opposed to â€œ214 x 72 = 15,408â€) can simultaneously be a priori knowable and yet not a priori known. If the Russellian canâ€™t do this, then the Fregean wins and a simple identity statement that is not a priori known must be not a priori knowable and thus the whole â€œthus is thusâ€ qualia thing is much more like â€œHesperus is Phosphorusâ€ than â€œHesperus is Hesperusâ€
Why Hawthorne thinks the Russellian will fail.
The Russellian needs a model of a priori justification. There are three possible models, but none of them seems to adequately describe what is going on with Fred and Twin Fred.
What the three models are and why they wonâ€™t work for Fred/TwinFred
Model 1: Platonic acquaintance.
WHAT IT IS: On this model, one is cognitively acquainted with a property and then just knows whatâ€™s what. Oneâ€™s a priori justifications come from just staring up into Platonic heaven.
WHY IT WONâ€™T WORK: Just seems unhelpful.
Model 2: De jure linkages between terms
WHAT IT IS: As a matter of semantic rule, two terms are linked, like perhaps â€œbachelorâ€ and â€œunmarried maleâ€. Oneâ€™s judgments are sensitive to the linkages.
WHY IT WONâ€™T WORK: The only plausible de jure linkages for thus 1 and thus2 would link via a semantic rule in connection with some kind of scale (involving ordered relations like â€œlighter thanâ€, â€œredder thanâ€ etc.). But when reference is secured via linkages to a scale, direct reference is ruled out. Analogy: if I thought of the height of a chair as 6 feet, it looks like the association with the scale in my mind, not the actual height of the chair, determines the referent of the thought.
Model 3: Dispositional linkages between terms
WHAT IT IS: While there are no de jure linkages between terms, as a matter of fact, one has a behavioral disposition to apply two different terms to one and the same thing. One gains a priori justification here by imaginatively going through cases and checking what one would be disposed to say.
WHY IT WONâ€™T WORK: When dancing qualia are a live threat, it may be unclear that you are able to discriminate, even in imagination, the denotata and supply them with neurtral descriptions and ask â€œwould â€˜thus1â€™ apply?â€ â€œwould â€˜thus2â€™?â€
Rationalism and the un-twin-earth-ability of phenomenal concepts
A concept like â€œwaterâ€ is twinearthable because we can easily imagine an epistemic counterpart that is epistemically just like us but locks onto a different property by â€œwaterâ€ than we do (XYZ instead of H2O). A concept is not twinearthable when beings epistemically just like us would lock onto the same properties that we do.
Non-twinearthable properties are so fully present to the mind that epistemic possibility is a guide to metaphysical possibility. Call this latter view â€œrationalismâ€. Rationalism, when combined with the view that phenomenal concepts are not twinearthable, would seem to rule out antiphysicalism since it would seem to rule out dancing qualia. However, both rationalism and the nontwinearthability of phenomenal concepts have been crucial premises in antiphysicalists arguments.
Problems not just for antiphysicalists
The Hesperus-phosphorus type of problem raised for direct phenomenal concepts does not arise only for antiphysicalists. Three observations:
First: The dancing qualia case can be raised for certain physicalists who identify qualia with physical properties and judgments with functional properties.
Second: If distinct types of phenomenal character may be indiscriminable then a subject may not know a priori that A is not identical to B.
Third: a less theoretically laden challenge: Identity claims about the empirical external world, like â€œwater = H2Oâ€, can be necessarily true without being a priori resolvable. Failure to know the identity does not imply breakdown in the grasp of the associated concepts. Why think the situation is any different with phenomenal concepts? Is it jus part of our competence with phenomenal concepts that we have them only when it is a priori resolvable what is identical to what? It would look like, then, that no one has phenomenal concepts. Is having the phenomenal experience itself evidence for the associated identity claims? It would be weird if it were true. The fact that Hesperus is Phospherous couldnâ€™t be evidence for the claim that Hesperus is Phosherous for a person who couldnâ€™t discriminate that fact. It is bizarre to think otherwise for phenomenal concepts.