beyond wishful thinking 9boundaries of the confi guration; there is nowhere outside the universe
to stand. He cannot observe or prepare copies of the states of aff airs
that he investigates; there is only one universe, or at least one observ-
able universe, at a time.
Th e second limitation of the dominant practice of natural science as
a model of cosmological inquiry is that it assumes a historically provin-
cial view of how nature works. It pictures the relatively settled and
cooled- down universe. In this universe, the constituents of nature, as
described by the standard model of particle physics, are unchanging
and, for all practical purposes, eternal. States of aff airs can be clearly
distinguished from the laws governing them. We can think of the laws
of nature as the indispensable warrants of all our causal explanations,
and of causal connections as par tic u lar instances of the workings of
these unchanging laws. Th e range of the adjacent possible is tightly
drawn: the ways in which, and the extent to which, some things can
turn into others.
What science has already discovered, however, suggests that nature
did not, and does not, always appear in this form. It has another, fi ery
and unsettled variant, in which it presented itself in the very early his-
tory of the universe and may present itself again. In this variant, what
we now think of as the elementary and eternal constituents of nature
did not yet exist, or were not or ga nized distinctly, as they now are, as a
diff erentiated structure. Th e laws of nature may not have been distin-
guishable from the states of aff airs that they governed. Indeed, causal
connections or successions may not have assumed a law- like form at
all. Th e susceptibility of the phenomena to transformation may have
been much greater than it subsequently became in the relatively settled
and cooled- down universe that the science founded by Galileo and
Newton takes for granted.
When we cast aside feel- good metaphysics, with its disposition to
claim more than we can pretend to know, recognize the incomplete-
ness of that scientifi c tradition as a basis for thinking about the uni-
verse, and nevertheless attend to the revolutionary empirical discov-
eries of twentieth- century science, we reach a view reaffi rming our
speculative groundlessness rather than overcoming it. According to
this view, everything changes sooner or later: the types of things that
exist as well as the regularities connecting them. Change changes.