(1) possible worlds framework
[ ]Pa

(2)  Argument that proper names do rigidly designate
    must say how issue of rigid designation arises, stuff on counterpart theory vs. identity theory
(3)  Argument that descriptions at best fix reference and don't fix meaning
    3b--why this argument requires proper names to rigidly designate (above argument is essentially that if we take descriptions to be the meanings of names, then we must give up the view that names rigidly designate)
(4)  Whither Frege?

3 reasons for Frege-Russell view

Searle argument

synonymy vs/ reference fixing

ftnote 10 another person could have done all Moses did.

begin a priori/necessity/analyticity discussion

(note that his characterization is different from Russell's)

35 bottom

reasons people thought necessary = a priori

Quine's arguments about the number of planets
and rigid designation to the rescue

important insight using historical essentialist thing

identity stuff

bare particulars vs. bundles (Locke vs. Berkeley)

big issue is same object in different possible worlds or counterparts kripke points out that counterpart view is really unintuitive
blames tendency to think in terms of counterparts on thinking of possible worlds the wrong way (Lewis' way--like you could have a telescope)

So, we do not begin with worlds (which are supposed somehow to be real, and whose qualities, but not whose objects, are perceptible to us),  and then ask about criteria of transworld identifiation; on the contrary, we begin with the objects, which we have, and can identify, int he actual world.  We can then ask whether certain things might have been true of the objects. (Kripke, (1980, p. 53))

again description theory as theory about meaning or merely as theory of reference
x is one meter long =def x is as long as the standard meter stick in paris
does not ``give the meaning'' but only ``fixes the reference.
for one could heat the standard meter bar so that the something is one meter long and not as long as the standard meter stick in paris.  So the definiens and definiendum are not synonymous (the two phrases don't have the same meaning).  Even if the phrase works to fix the reference.

Again, the theory of rigid designation allows us to express this well.
`one meter' designates rigidly while `the lenght of S at t0'' does not.

epistemologically ``the standard meter stick is one meter long'' is a priori, at least for anyone who uses the meter stick to fix the reference.
metaphysically ``the standard meter stick is one meter long'' is coningent

So definitions which fix a reference are (if true) contingent and a priori knowable while definitions which give the meaning are necessary and a priori knowable.

return to theory of proper names
(a) argues that if we take descriptions to give the meanings of names, then names do not rigidly designate.

seems question begging.

Argues that when we fix the reference via description, it is usually possible that the description ends up being false of the object in question.  Say the reference of ``hesperus'' was fixed by the the description ``the heavenly body appearing at yonder position in the sky.''  We still want to be able to say ''it might have been the case that Hesperus appeared in a different position.''  [David Chalmers views Kripke as having essentiall bifurcated sense into the reference fixing aspect (which is how we locate obects in the actual world) and the intension fixing aspect, (which is what makes an object the same object in all possible worlds).  The second aspect may be totally unknown to speakers of the languge, it is metaphysical, not epistemelogical. (the remarks on page 59 support Chalmers' view.] and the best way to accomodate this insight is if we take names like ``Hesperus'' to be rigid designators and take the reference fixing description not to be a rigid designator.

Note- Donellan's might be viewed as arguing that the reference fixing description is sometimes not only false in some possible worlds, but actually false in the this world.

again marks difference between him and Russell/Frege
If 'Moses' means the same as a description D, then saying 'Moses didn't exist' is the same as saying 'the existence and uniqueness conditions for a certain description are not fulfilled,' and to say that Moses did exist would be to say that they were so fulfilled.  As with the argument Searle gave, Kripke seems to be strong here.

arguments agains the cluster theory
Kripke restates Searle's argument against Searle's own theory.  Couldn't we discover that Aristotle didn't do anything we think he did?

important point that Kripke only needs one case where we could say that Johah still exists even  though all the things we believe about him are false.

circularity stuff (pretty confusing)