21. To come then to Berkeley's development of his laws of nature.
The laws of nature are the way in which sensations succeed one another
and are distinguished from the arbitrary ways in which images succeed
one another. Now in the first place, Berkeley says that knowledge of
these laws is necessary for the conduct of worldly affairs. It gives
us a sort of foresight which enables us to regulate our actions for
the benefit of life. (XXXI). We discover, he says, “That food nourishes,

  ― 40. ―
view facsimile

sleep refreshes, and fire warms us”, and we make various discoveries
of this kind not by finding any necessary connection between our
ideas but only by observing the settled laws of nature, that is, the
order in which our ideas come. Now one of the greatest difficulties
for Berkeley arises precisely in this suggestion of his that knowledge
of the laws of nature is necessary for the conduct of worldly
affairs. The laws of nature being the order in which sensations are
presented, and sensations being those ideas which come to us whether
we will or not, it follows that no amount of knowledge of these laws
can enable us to make any sort of provision for the future; all that
we can effect by our wills are those reproduced ideas or images
whose order is different from that of nature. Accordingly the only
way in which we could conduct worldly affairs would be by producing
one idea rather than another, and this reproduction would not in any
way be assisted by a knowledge of the laws of nature while, on the
other hand, nature would proceed quite independently of those reproductions
on our part. But obviously this is not in the least what
is meant in the conduct of worldly affairs. In conducting our worldly
affairs we do determine, as far as we can, what objects we shall perceive;
we make a difference in the perceptual situations confronting us.
Now if that is admitted, then the division between sensations and
images, which Berkeley has erected, breaks down; that is to say, some of
the things which we know by means of sensation are actually produced
by us. In fact that is what we really mean by an act of will. But if
in that way we can act upon the objects of our perception, it follows
that these objects must have an independent existence in order that
we amy act upon them and it is perfectly possible that they should
act upon us in turn causing us to perceive the changes that have
taken place as a result of our actions. At any rate we do have, in
this instance, in the admission that there are practical affairs of
life, an admission that the objects of knowledge have an existence of
their own.

Now, again, in speaking of a knowledge of the laws of nature
as enabling us to foresee certain events, Berkeley shows that his division
of ideas into sensations and images is not in accordance with
the actual facts of knowledge. The argument is: we have learned from
past experience that that visual object, which we call a fire, is accompanied
by that other idea which we call heat. We express that by saying
that it is a law of nature that fire should warm us or, in general,
should be accompanied by heat. Now if we are to exercise foresight
as a result of this knowledge, that means that on a subsequent occasion
we might, on seeing a fire, predict that it would give heat; but
in order that this prediction should be verified, we should require
to go on and obtain the sensation or perception of heat. Before
obtaining that sensation, we had an expectation of heat, that is, we
had a certain image in our minds; but if there is to be true prediction
then that which we find, as the experience goes on, must be
exactly the same as that which we had anticipated to begin with; unless
the object of our perception is the very object of our expectation,
then our expectation was not correct. But that means that unless
something that Berkeley calls an image is identical with something
which he calls a sensation, our prediction could never be correct;
the most that we could have would be a certain image, followed
by a sensation which was very like it but which was an entirely
distinct idea, so that nothing that could really be called prediction
had taken place.

  ― 41. ―
view facsimile

Unless, therefore, he gives up the division that he has made,
Berkeley can give no account of successful prediction and therefore
of the conduct of worldly affairs. And this particular difficulty
is important as bringing out the point that even in imagining,
whatever may be the peculiarities of that attitude to things, the
things that we are dealing with are still independent existences of
the same sort as those we perceive and act upon.

Now another point arising in connection with the notion of
prediction and the conduct of affairs is that our knowledge, in order
to be of any value, must enable us to secure certain consequences
and to avoid certain others; for example, in the case of the heat of
a fire, our knowledge of this law would lead us, upon seeing the fire,
not to approach too near it, if we did not desire its warmth at that
particular time. But in terms of Berkeley's theory of the orders of
sensations, this avoidance would mean that, in this particular case,
the vision of a fire was not followed by the perception of heat and
therefore the supposed law was not actually true. Now the fact that
we do not argue in this way, the fact that we do use our knowledge
of the laws of nature in order to avoid what is unpleasant to
us, shows that we regard the laws of nature not as holding among
sensations but as holding among things, that is, we assert that the fire
is hot even although we have not perceived that heat; we assert its
? existence in the thing even while and of wishing to avoid perceiving it. So that
here again the conduct of worldly affairs and the conception of laws
of nature indicatesthat we have relations with actual things which
we are capable of perceiving but which are not dependent upon our perceiving
them and which are not to be described as ideas, sensations,
impressions, or in any of those terms. This same difficulty in regard
to laws of nature comes out in connection with the theories of Hume.