Re: BHA: truth

[Louis Irwin <lirwin1@xxxxxxxxxxxxx>]

Ruth,

Nice to have you back with us.  I think the deflationary theory refers to
the view that the concept of truth as a property of propositions is
dispensable in principle.  The basic idea is that calling, say, the
proposition that snow is white true means no more than saying that snow is
white.  Calling a specific proposition true seems to be no different from
asserting the proposition itself, so why do we need a concept of truth at all?

Higher order attributions of truth, where a specific propositions are not
available, cause a difficulty with this view.  For example, how are we to
restate "Any proposition that Jack believes is not true" without reference
to truth?  Well, we could say that anything Jack believes is not the case,
or not a fact,  but that move deflates truth by inflating facts.  Saying
that a proposition is the case or is a fact is not an advance over saying
that a propositon is true.

Other theories deepen deflationary theories by attempting to eliminate
propositional truth by defintion.  This is famously associated with
statements like: "Snow is white" is true if and only if snow is white.
These statments are basically the same as the deflationalry ones above,
however they do not constitute the DEFINITION of propositional truth; they
are merely the CRITERIA that a definition must satisfy.  The idea is that
if you define truth as a predicate, then the criteria of success is whether
the deflationary statements logically follow from the definition.  Any
definition that succeeds in doiing this yields an
adequate theory of propositional truth.  However, the deflationary
statements do not comprise the definition, because there are infinitely many
of them, and a defintion must be finite.  So what you do is to use the
deflationary statements for atomic sentences and then define truth for all
logically complex sentences.  For example, p&q is true if and only if p is
true and q is true, where p and q are sentences whose truth has already been
defined.  Voila, and welcome to Tarski-land!  You have to have a structured
language with a finite number of atomic elements and logical devices from
which all other sentences are built.  It works great for purely logical
languages, and a lot of work has been done trying to extend to natural
languages.  I've passed over the issue of the structure of atomic sentences,
which break down into predicates, names and variables.

I note in passing that definition of a concept does not really eliminate
it.  In fact, quite the reverse: a definition shows that the concept is
always there to be used.  In other words, reduction does not mean
elimination.  (Definitional reduction should be resisted on grounds other
than that the concept defined is eliminated.)

The relation of all this to critical realism is of great interest.  How does
the requirement of atomic element plus logical structure match up with the
critique of ontological extensionalism?  (Some non-extensional contexts like
necessity can be treated to a certain extent.)  Is the notion of a truth
predicate entirely atualistic? (I'm thinking of the way "is true" and "is a
fact" match up.)

Louis Irwin

>I'm wondering if any of you who know anything at all about analytic
>epistemology could help me out, off-list is fine, with some questions I'm
>having reading through a few books in the area.
>
>The first, really basic question is this:  what EXACTLY is the deflationary
>theory - or approach, if there are a lot of versions - of truth, and how
>EXACTLY is it different from correspondence?
>
>I have a few questions beyond this, but this would be a good start (and I
>can even link it all back to realism and Bhaskar...). I have asked everyone
>I can think of to explain this to me, but my department is a bastion of
>continental marxists, none of whom know a *THING* about this stuff.




     --- from list bhaskar@xxxxxxxxxxxxxxxxxxxxxxxxxx ---