It seems trivially obvious that a property such as “going to the cinema” is not a
necessary property of a person, since she/he could very well have done
It is also important to carry this ordinary intuition back to situations in the past.
When one does htis extrapolation, it becomes clear that no ordinary property one
can ascribe to, say, Aristotle, is a necessary property. It is neither necessary for
Aristotle to have written Politics, nor was it necessary for him to have taught
Alexander the Great, etc.
If one thinks that the very meaning of the name is necessarily associated with a
definite description, then one is really asking for a necessary property that can
always be ascribed to the name. Kripke says that this is not possible.
Names are rigid designators.
Descriptions are generally NOT rigid designators.
Ex. The name “Aristotle” is a rigid designator, and the description “the teacher of
Alexander the Great” is not.
This is the formalization of our regular intuition that, generally, rigid designators
pick someone out uniquely in the ambient world.
This is why the first Frege thesis – you can give the meaning of a name with a
definite description – fails.
It is possible to have a necessary but a posteriori truth.
Ex. through empirical observation, we can determine that Hesperus is
Furthermore, it is also possible to have a priori but contingent truths.
Ex. Say we designate one metre to be the length of rod S at t0. In fact, one metre
will refer to lengthof rod S at t0. We can say that “one metre is the length of S at
t0” is an a priori truth. However, this is also a contingent truth, because we can
imagine possible worlds of t1, in which case the sentence would no longer be true.