24. (We come now to one of the most important parts of Berkeley's
theory, giving the key to the elucidation of his philosophy, namely
his theory of universals and his criticism of abstract ideas.)

The essay of Locke was expressly put forward as indicating
or endeavouring to indicate the limits within which human understanding
had to operate; to show, in other words, what we could possibly know
and what we could not possibly know, granted that our knowledge was
based upon certain original data and had to be developed from them.
According to Locke, these original data are ideas and ideas are the content
of any individual act of thought or any individual cognitive act.

Now this at once raises the difficulty of showing how more
articulated and systematised knowledge can be derived from simple
elements and further, granted that ideas are whatever we think about,
then it would seem that any knowledge we gain, however systematised it
might be, would also be an idea or a number of ideas; and therefore
there would be no clear distinction between what was supposed to be
the basis of our knowledge and what was supposed to be the further
development of our knowledge. And again, in holding that our knowledge
was limited by the original ideas that we had received, that these
constituted the sphere within which our knowledge was possible and
were, so to speak, the boundaries of our thought, Locke not merely goes
beyond the original data themselves, assuming that there are such, in
calling them limits or boundaries, but he also makes the unjustified
assumption that there must be limits of our knowledge.

Now if we recognise development in knowledge, then we are
admitting that it is possible to know more about the supposed data
than we originally knew; and consequently, these are by no means such
rigid limits as Locke had originally supposed them to be. And again
in making the assumption that there must be certain original knowledge
to be contrasted with certain derived knowledge, Locke is simply
following Descartes' rationalistic procedure even while he criticises
the Cartesian doctrine of innate ideas. The assumption that there are
original objects from which further knowledge can be unfolded really
depends upon a doctrine of the essence type, that is, that there is a
pure or ultimate knowledge on which all particular or special knowledge
is based. And if this theory of a general type of knowledge is
adopted, it becomes difficult, if not impossible, to explain the process
of derivation; it will be found that the knowledge which is said to be
derived is really just as original, if we have it at all, as the knowledge
which is said to be original. In the case of Descartes, we find
him professing to derive a knowledge of the various mental faculties
from the original essence or identity of thinking. In the case of Locke
we find a similar attempt to derive a knowledge of external things
from the contemplation of idea. But actually, in his argument, Locke

  ― 46. ―
view facsimile

has to assume that we already know external things; and similarly as
regards Descartes' argument, if it were not assumed that we are
acquainted with the various mental faculties just as directly as we
can possibly be acquainted with thinking as such, it would be impossible
for us to understand what those various forms of mental operation
were and, in particular, how they differ from one another.

Now just in the same way, Locke's theory of propositions, which
he attempts to derive from a knowledge of ideas, really implies that
we have a knowledge of propositions from the beginning, and if this
were not so, it would be impossible to derive this knowledge from a
knowledge of single and separate ideas. Accordingly, while we can distinguish
the position of Locke from that of Descartes in that Locke
professes to begin with a plurality of objects of knowledge and is
therefore more empirical then Descartes who endeavours to begin with
the one thing certain and to derive all other knowledge from that, they
are both alike in assuming that there must be some original entity or
entities from which scientific knowledge somehow emerges, and also in
assuming that our actual knowledge is advanced by a criticism of mental
faculties, showing, as Descartes thinks, that there is one superior faculty,
or, as Locke thinks, the extent to which our knowledge can be developed
and the limits beyond which it cannot go.

Now these are both rationalistic assumptions and the objection
to them is that there cannot be this contrast between original or
essential, and derived knowledge, since the process of derivation would
have to belong to both spheres and would really break down the division;
and, in the second place, we cannot develop knowledge by a criticism
of faculties because that criticism can be undertaken only by
our faculties themselves, and therefore if there were any doubt about
their findings in regard to things in general, there would be equal
doubt about their findings in regard to themselves or one another; and
if, on the other hand, there were any general reason for casting doubt
upon the findings of our faculties, then the fact, insisted on by Descartes,
that the findings of reason are gauranteed by reason itself
would really be of no account because we should naturally expect reason
to support the findings of reason but this would never be a proof of
the soundness of reason if once doubt had been cast upon the value
of its findings.

Accordingly, we find that Berkeley takes up a more empirical
view in objecting, as he does at the beginning of the Introduction, to
this method of criticism of the faculties, the point being that we can
develop any theory only by asserting what we find to be the case; and
the fact that we require to have certain faculties in order to make
these assertions could never be recognised as a reason for casting
doubt on any assertion of that kind. Only if we found the opposite to
be the case, that is, only if we had direct evidence, in regard to the
subject under consideration, which conflicted with some assertion that
we or others had made, would we think it desirable to call that assertion
in question. So far, then, Berkeley is in the right in suggesting
that we can rely only on what we find to be the case and that
criticism of faculties is not a method by which scientific knowledge
can be developed. And Berkeley sees that, according to the assumptions
of Descartes and Locke, some sort of representational theory is necessarily
arrived at because, when we contrast original and derived
knowledge, or essence and accident, we are really contrasting the inner
with the outer, and looking with the inner (knowledge), which, because
it is inner, is supposed to be peculiarly certain, we can at most take
it as representative or symbolic of outward affairs since we cannot

  ― 47. ―
view facsimile

have the same certain knowledge of the latter. But even Berkeley
points out that we cannot say with certainty that there are such
outward things and therefore still less can we say whether the inner
knowledge represents the outer rightly or wrongly. We can put this
point again in the form of a dilemma and say that when we are aware
of representation, then either that object of knowledge is inner, and
therefore can give no clue to anything external, or else it is outer
and therefore is just as uncertain as the thing which is supposed to
be represented and again it is impossible for us to say under those
circumstances that any object of inward knowledge has the relation of
representation to something outside.

Now what Berkeley has specially to insist upon in the Introduction
is that Locke tries to overcome this division but only succeeds
in concealing it by his use of propositions and by the theory of
abstraction in terms of which he gives an account of propositions.
According to Locke we can be aware not only of ideas but also of
relations of agreement and disagreement among ideas, and these relations
make it possible for us to formulate propositions which may be
true or false. But since, according to his theory, whatever is in the
mind is thereby known, it follows that if two ideas in the mind agree
we cannot possibly think that they disagree and therefore we cannot
possibly be wrong. That is apart from the difficulty that, strictly
in terms of the theory of ideas, we should never have in the mind a
knowledge of the agreement between two ideas, A and B; we should only
have the three ideas, A, agreement, and B, and we should still have to
discover how these three were related to one another, and if we did
make any such suggestion the same difficulty would again arise.

Now that being the case Locke is forced to develop a theory
of correspondence between what is in our minds and what is in reality.
The question of truth, he says in Book IV. Of the “Essay”, is the question
of agreement or disagreement of our ideas with reality. Now that
raises the difficulties of representationalism in an acute form, and it
is against that presentation of the case that Berkeley argues, the point
being that if we are to know that a certain idea agrees with a certain
reality, or even that there is a reality for it to agree or disagree
with, we must have before our minds, at one and the same time,
and in connection with one another, the idea and the reality. Now that
means that the contrast between the external and the internal cannot
be maintained for that knowledge---the knowledge of the contrast
between idea and reality---could neither be internal nor external.