THE MEANING OF TRUTH

1
THE MEANING OF TRUTH

• SUSAN HAACK
• ltshaack_at_law.miami.edugt

2
True is a word we all understand, but if we
try to explain it, we can easily get involved in
a maze of confusion. Frank Ramsey
3

• CORRESPONDENCE THEORIES
• THE SEMANTIC THEORY
• THE LACONICIST THEORY
• CONSEQUENCES OF LACONICISM

4
1. Correspondence Theories

• some correspondence theories are toothless,
  saying no more than it is true that p if and
  only if really, in fact, p
• more serious correspondence theories turn those
  emphatic adverbs into serious metaphysics/philosop
  hy of language giving them real bite

5
for example
the Logical Atomist correspondence theories of
Wittgenstein and Russell
6
these require

• heavy metaphysical apparatus of facts (atomic
  molecular, positive negative)
• propositions with a specific logical form
• a relation of structural isomorphism (which
  proved very difficult to spell out)

7
then there was
J. L. Austins version, with correspondence as a
coincidence of demonstrative and descriptive
conventions but this applies only to indexical
statements
8
many still find correspondence attractive

• but it has (I believe) never been spelled out in
  a way
• that gets beyond the toothless correspondence
  idea, and
• doesnt lead to excessive metaphysical
  commitments, or restricted applicability

9
2. The Semantic Theory
has been enormously influential, thanks in part
to the support of Popper, Quine, and Davidson
but not always well-understood
Alfred Tarski
10
Tarski proposes

• Formal Adequacy Conditions
• the definition must not be circular
• should not use semantic primitives
• can be given only for a language that is formally
  specifiable
• semantically open

11
the last formal requirement

• is imposed to avoid the Liar Paradox
• This sentence is false
• by requiring truth-in-O (the object language) to
  be defined in a meta-language

12

• because of the Formal Adequacy conditions, Tarski
  can define truth only for syntactically
  characterizable formal languages
• can define only true-in-L, not true

13
more famous is

• Tarskis Material Adequacy Condition
• any acceptable definition of truth must have as
  consequence all instances of
• (T) S is true iff p
• (where S is the name of the sentence on the
  right)

14
the Material Adequacy Condition

• is not a definition of truth (but a condition on
  acceptable definitions)
• (according to Tarski) cannot simply be
  generalized
• so his definition takes an indirect route

15
Tarski defines

• satisfaction of atomic open formulae (a relation
  to infinite sequences of objects) --
  enumeratively
• then satisfaction of molecular open sentences --
  recursively
• then truth of closed sentences satisfied by
  all infinite sequences

16
misunderstandings

• this is not, as Popper suggests, a version of the
  correspondence theory Tarski wants to
  articulate Aristotles Insight WITHOUT relying on
  correspondence or facts
• nor is it a disquotationalist theory, as Quine
  suggests if it were, Tarski could simply drop
  his last 100 pages!

17
for, according to Tarski

• the quotation-mark name of an expression is a new
  word, of which the contained expression is not
  semantically a part
• you cant quantify into quotation marks
• (p) (p is true iff p) makes no sense the
  T-schema cant be generalized in this way

18
ironically enough

• this is the logical block view of quotation --
  which Quine himself once accepted!
• when he wrote that a word in quotation marks is
  not semantically a part of the whole expression
  any more than rat is of Socrates

19
moreover

• Tarskis is not, as Soames suggests, a theory of
  the truth of propositions
• propositions dont have syntactic structure (the
  same proposition can be expressed by sentences
  with different structures)
• but Ts definition relies on structure of wffs

20
Tarskis approach has real limitations

• it stratifies the truth-concept (treated not as
  one concept, but many)
• its application is limited to formally
  specifiable semantically open languages
• which is why Tarski says truth cannot be defined
  for natural languages

21
Tarski himself is ambivalent

• he says he doesnt claim to have captured the
  real meaning of true he would be willing to
  use the word frue instead!
• and yet goes on to say BOTH that he doesnt say
  the semantic theory is right, AND that he cant
  imagine what it would mean to say it is wrong
  (!!!)

22
furthermore
the failure of the Davidson Program confirms
that Tarski was right in thinking his methods
apply to formal, but not natural, languages
Donald Davidson
23
not to mention the failure of Poppers theory
of verisimilitude, or the casualness of his
assumption that Tarskis theory applies to the
consistent parts of natural language
Sir Karl Popper
24
I am inclined to conclude

• that Tarskis work, though a very impressive
  technical achievement
• is not, in the end, fully satisfying
  philosophically

25
3. The Laconicist Theory

• laconicist is a better name for what is usually
  called the Redundancy Theory
• A belief is true if it is a belief that p, and
  p. . A belief that Smith is either a liar or a
  fool is true if Smith is either a liar or a fool,
  and not otherwise. Frank Ramsey

26
Ramsey is well aware

• that while it is true that can be eliminated
  from (is redundant in), e.g., It is true that
  Hannibal crossed the Alps
• it cannot be eliminated form, e.g., Plato said
  some true things and some false things

27
the new name

• coined by Dr. Kiriake Xerohemona
• derives from the English word laconic, which
  means short, terse
• itself derives from the Greek word Laconia,
  the name for the ancient city-state, Sparta

28
to call something spartan (in English) means
that it is austere, simple
29

• laconicism seems to capture the core meaning of
  true
• to conform precisely to the Aristotelian Insight

30
still, as Ramsey is aware

• the theory is incomplete, requiring
• an account of propositional quantifiers (which
  may not itself use the concept of truth)
• to explain, e.g. Plato said some true things
  -- (Ep) (Plato said that p, and p)

31

• an account of representation
• an understanding of reality
• which Ramsey takes to be involved in saying that
  this is the belief that p representation, and p
  reality

32
on propositional quantifiers

• neither an objectual nor a substitutional account
  will do
• because both involve the concept of truth,
  explicitly or implicitly
• the inference-ticket approach (suggested by
  Arthur Prior, C. J. F. Williams, María-José
  Frápolli) might work

33
on reality

• laconicism is entirely compatible with my
  Innocent Realism (itself fairly laconic!)
• according to which there is one real world
• largely but not wholly independent of us

34

• a real world that includes
• natural things, events, etc.
• human artifacts
• social institutions, roles, and rules
• mental states, processes, events
• imaginative creations such as novels, plays,
  cartoons, etc.

35
4. Consequences of Laconicism

• that when we say that it is true that p, what we
  say about p is the same, whatever kind of
  proposition p is (scientific, historical,
  literary, legal, etc.)
• this is one way to put what I mean by speaking of
  the unity of truth

36

• that truth is objective
• i.e. (normally), whether or not it is true that p
  does not depend on whether you, or I, or anyone
  believes it is true that p
• or on whether we agree that p, or know that p,
  etc.

37

• that truth is not relative
• for (just as there is no reference in the account
  of truth to what anyone believes, etc.), there is
  no relativization to culture, community, theory,
  or even (as in Tarski) language

38
moreover

• laconicism suggests a plausible understanding of
  observations like
• a genuine inquirer seeks the truth
• a scientific theory is successful (in prediction,
  technology) because it is true

39
a genuine inquirer seeks the truth

• means, not that there is a kind of Holy Grail,The
  Truth, that every genuine inquirer seeks
• but that someone who is genuinely inquiring into
  whether p wants to end up believing that p if p,
  that not-p if not-p

40
e.g.

• James Watson really wanted to discover the truth
  about the structure of DNA
• means Watson wanted to end his investigation
  believing that DNA is a double-helical,
  backbone-out macromolecule with like-with-unlike
  base pairs iff DNA is a double-helical, ...,
  macromolecule (etc.)

41
as, of course, it is, and he did!
42
Watson (L) Crick (R) with their model of DNA,
1952 a model Watson described as too pretty
not to be true.
43
again

• this scientific theory works because it is true
• means this theory works because it says that p
  and (in fact!) p

44
for example

• why does the plane not burst at the seams?
• because it is built using the assumption that
  metal m can withstand pressure p, and this
  assumption is true
• i.e. because m can withstand pressure p

45
furthermore

• laconicism can readily explain why true plays
  the pragmatic roles it does
• e.g., the use of thats true to express
  agreement is just a short way to say yourself
  what the other person just said

46
Rorty suggests he is following Ramsey

• when he identifies truth with here-and-now
  agreement
• but this is a big muddle

47

• thats true often has the force of I agree
• but (of course) we can agree that p when p is
  not true
• (of course) we may not agree that p when p is
  true
• so it is true that p doesnt mean we agree
  that p

48

• laconicism can also explain the concessive use of
  true
• A says The price of gold will probably rise B
  replies True, but later it will fall
• yes, the price of gold will probably rise, but
  after that it will fall

49
next time, I will explore

• the many different kinds of true proposition (the
  plurality of truths)
• specifically, how truths in science differ from
  truths in history, in law, in literature

50
but for now, Id like to hear what you think