---------------------------------

Ravi asked:

What is the nature of necessity involved in physical laws? In what way it is distinct from logical
necessity? What role do counterfactuals play in distinguishing law statements from accidental
generalizations?

============

Laws of nature are governed by natural necessity and we come to know such laws by observation
and confirmation in particular instances. When natural law is understood as causal, one event is
supposed to be necessary and sufficient for another, although the predictability which this implies is
called into question by quantum physics. However, at an ordinary macro-physical level, if water is
heated to 100C it will be boiling. This is necessary because it would not be boiling without having
reached this temperature and sufficient because the water's being at this temperature guarantees that
it is boiling. We understand this sort of necessity as exceptionless and lawlike, given the proper
conditions.

Logical necessity, as it relates to counterfactual situations, is that which we cannot imagine to be
otherwise, or that which is true in all possible worlds. It is not logically necessary that when water is at
100C that it is boiling. This is a law of nature, determined by the actual world. An example of logical
necessity understood as what we cannot imagine to be otherwise, is that an object cannot be red and
green all over. This is because we cannot perceive two different colours at the same time nor imagine
what it would be to have a sensation of something's being red and green at the same time. This is not
just true for secondary qualities, but also primary qualities. An object cannot be both round and
square: We cannot imagine what such a counterfactual state of affairs would be. This is an a priori
approach to identifying logical necessity.

However, Kripke has argued that the identity of names determined by reference, when considered in
terms of counterfactual situations, is a natural necessity. It is arguable whether the identity of
Hesperus and Phosphorus — the 'evening star' and the 'morning star' — is logically necessary. If
logical necessity is what we cannot imagine to be otherwise, it is not logically necessary because it is
conceivable that each name identifies a different star. However, for Kripke, a counterfactual state of
affairs is something we posit within our language and when we refer to Venus we mean our Venus.

The laws of nature can be held to be contingent, i.e. it is possible that water doesn't boil at 100C.
However, on Kripke's view of natural kinds such a state of affairs wouldn't involve what we call water,
but rather something that appears to be 'water', but isn't. I think the role of a counterfactual is to test
our conceptual commitments. We won't give up our commitment to water's boiling point when we
consider a possible world. If we consider the state of affairs in which water boils at 35C we suppose
that it is not water. If, in fact, we found water was boiling at 35C we would still take the stuff to be
water if it was H2O and wonder what would explain this change in natural law. Counterfactuals
highlight our commitment to essentialism for natural kinds insofar as we take water to be necessarily
H20 and the contingency of the laws of nature which seem to be exceptionless and lawlike, and so
we take them to be necessary in this sense, but they could change. Our notion of causation is our
way of understanding the world but we have no grasp of the significance of causation or whether the
world could change.

Kripke's view of counterfactuals and necessity is not subscribed to by everyone. There is a lot of
debate about counterfactuals and the nature of possible worlds. Since you have been interested in
counterfactuals for some time, I wonder if you have read An Introduction to Philosophical Logicby A
C Grayling and Causation and Conditionalsedited by E Sosa.

Rachel Browne