Comments on Pryor’s “Externalism about Content and McKinsey-Style Reasoning”

William S. Larkin

Conference on Self-Knowledge

University of Utah: 4/5/02

____________________________________________________

 

 

I.  Pryor on McKinsey:

 

A.  Pryor’s Version of McKinsey-style Reasoning

1.  Given authoritative self-knowledge, I can usually tell the contents of my own thoughts just by introspection.  So I can know the following claim on the basis of reflection alone:

                               

                McK-1:  I am thinking a thought with the content water puts out fires.

 

2.  Given content externalism, which I can know to be true on the basis of armchair philosophical reflection on Twin Earth thought experiments, I can know the following claim on the basis of reflection alone:

 

McK-2:  If I am thinking a thought with the content water puts out fires, then E is true.

(Where E expresses some fairly specific claim about my environment being a certain way, e.g., that samples of water must have at some time existed and someone in my community must have causally interacted with them.)

 

3.  So I can put these two pieces of reflective knowledge together to deduce, and thus know on the basis of reflection alone, the following:

                                               

McK-3: E is true.

 

B.  Possible Responses:

                1.  Transcendental Response to Skepticism

                2.  Incompatibilism

                                a.  reject authoritative self-knowledge

                                b. reject content externalism

                3.  Something is wrong with the McKinsey-style reasoning.

 

C.       Prior’s Diagnosis in a nutshell:

Either McK-1 or McK-2 is not knowable on the basis of reflection alone because I cannot know on the basis of reflection alone whether those who fixed the reference of my word “water” succeeded in their intention to pick out a genuine natural kind (i.e., a substance with a stable underlying essential nature).

 

                D.  Fuller Diagnosis:

1.  There are 2 possible interpretations of Dry Earth scenarios—So a putatively external term/concept c is such that either

a.  Fallback (F)-interpretation: c either expresses an external concept or some qualitative fallback concept.

b.  External or Bust (EB)-intepretation: c either expresses an external concept or nothing at all.

                               

2.  Given F-interpretation: McK-2 is not knowable on the basis of reflection alone.

a. Knowing McK-2 requires antecedently knowing that water is an external concept.

b.  But knowing that would require knowing that I am not in a Dry Earth scenario.

c.  And I cannot know by reflection alone that I am not in a Dry Earth scenario.

 

                                3.  Given EB-interpretation: McK-1 is not knowable on the basis of reflection alone.  

a.  Knowing McK-1 requires antecedently knowing that I am thinking a thought with some content or other.

b.  But knowing that would require knowing that I am not in a Dry Earth scenario.

c.  And I cannot know by reflection alone that I am not in a Dry Earth scenario.

 

 

II.  Me on Pryor on McKinsey

                A.  McKinsey-style reasoning should replace MCK-1 with the following:

                               

McK-1*:  I am thinking that water puts out fires.

 

                B.  Significance:

1.        McK-1 mentions the content water puts out fires.

[Compare McK-1 to: I am thinking a thought with the content that I would express using “water puts out fires”.]

 

2.  McK-1* uses the content water puts out fires.

                               

3.  McK-1 is a conceptually more sophisticated judgment than McK-1*. 

 

                C.  Implications for Pryor’s Diagnosis

1.        It may be plausible to say that in order to know McK-1, I need to antecedently know that I am thinking a thought with some content or other.

 

For it may be that in order to know McK-1, I need to antecedently know that I actually succeed in referring to some content with water puts out fires (or with “the content I express using ‘water puts out fires’”).

               

2.        But I think it is not plausible to say that in order to know McK-1*, I need to antecedently know that I am thinking a thought with some content or other.

a.  Since I am not trying to refer to some content, I do not have to antecedently know whether I succeed in doing so in order to know McK-1*.

b.  A small child might be able to knowledgeably judge (on the basis of introspection) that she is thinking that water puts out fires without even having the conceptual sophistication to judge that she is thinking a thought with the content water is wet.

c.  Normal adults may always know that they are thinking a thought with some content when they know they are thinking that water puts out fires.  Still, they do not need to antecedently know that they are thinking a contentful thought.  They know they are thinking a contentful thought by knowing that they are thinking that water puts out fires.

 

 

III. Me on the McKinsey-Style Reasoning

                A.     The following is not knowable on the basis of reflection alone:

 

                                McK-2*:  If I am thinking that water puts out fires, then E.

                               

1.        Knowing McK-2* requires knowing that my thought that water puts out fires is externally individuated—or, in other words, that my concept water is externally individuated.

 

2.        I cannot know on the basis of reflection alone whether any of my actual concepts are externally individuated.  

a.        If content externalism is the thesis that some of my actual concepts are such that possessing them requires being appropriately related to a certain environment, then we cannot know on the basis of reflection alone that content externalism is true.

b.       Twin Earth thought experiments and the like either do not establish content externalism (understood as a claim about some of our actual concepts rather than about possible concepts) or they assume certain empirical knowledge of the actual world.

 

                B.  Argument

P1.  To know that one of my actual concepts C is externally individuated requires knowing that there is something about C in particular that makes it such that I must have been appropriately related to a certain kind of environment in order to acquire it.

 

P2.  So to know that my actual concept C is external requires knowing that there is some possible world w where I could not have possessed C because w differs from the actual world with respect to certain features that are relevant to the possession of C in particular (i.e., because w is merely relevantly distinct from the actual world).

 

P3.  I cannot know on the basis of reflection alone whether any world w is merely relevantly distinct from the actual world.

3a.  I may be able to know, on the basis of semantic intuitions or something of the sort, that I could not have possessed C is some possible world w, if that world is described to me in sufficient detail.

 

3b.  But I cannot know on the basis of reflection alone that w is merely relevantly distinct from the actual world.

i.  I cannot know that w is distinct from the actual world in some relevant respect without knowing how the actual world is in that respect.

 

ii.  Since C is in fact an external concept, that relevant respect will be a specific feature of my external environment.

 

iii.  I cannot know about such specific features of my external environment on the basis of reflection alone

 

C:  So I cannot know on the basis of reflection alone that one of my actual concepts C is externally individuated.

 

 

                C.  Summary of my position: 

1.        Regardless of what the correct account of external content is, there is a way of purely reflectively knowing what we are thinking that does not antecedently require knowing that we are thinking a contentful thought. 

 

2.        To block the McKinsey-style reasoning, then, we must reject that we can know on the basis of reflection alone that being able to think some of our thoughts entails something specific about our external environment.

 

 

D.      Boghossian-Style Objection:

1.        On the EB-interpretation, one can know McK-2* on the basis of reflection alone.

2.        For one can know purely reflectively whether a term like ‘water’ expresses an external content.

a.        One can know purely reflectively that ‘water’ is intended to express an external content.

b.       One can know purely reflectively that terms that are intended to express an external content either do so or express no content at all. 

Boghossian: Argues that the externalist cannot accept the fallback interpretation of Dry Earth.

c.        One can know purely reflectively that ‘water’ expresses some content.

d.       So one can know purely reflectively that ‘water’ expresses an external content.

 

E.       In most cases, I doubt that I can even purely reflectively know that a term is intended to express an external content.  For this would require knowing the intentions of the reference fixers, and in most cases this is not me.

 

F.       But I definitely have to say: We cannot know purely reflectively that any word intended to express an external content either does so or expresses no content at all.  So I have to respond to Boghossian.

1.        Boghossian Quote:  “...the Twin Earth externalist is committed...to holding that ‘water’ expresses an atomic concept under conditions where it has a non-empty extension...That is one of the presuppositions of the Twin Earth thought experiment.  But, then, how can the very same word, with the very same functional role, express an atomic concept under one set of external conditions and a compound, deompositional concept under another set of external conditions?  A concept’s compositionality is exclusively a function of its internal ‘syntax’, and can’t be contingent upon external circumstances in the way that the present proposal would require.

 

2.        Reply:  The present proposal does not require a single concept to be atomic under one set of external circumstance but compound under another.  It only requires that a single word express an atomic concept under one set of external conditions and a different compound concept under another set of external conditions.  The single word in question will only have to express the same concept on views where having the same of functional role is sufficient for expressing the same concept—but the externalist will of course deny any such view.  

 

William S. Larkin

Southern Illinois University Edwardsville

wlarkin@siue.edu