at a distance' and so forth--and attention to the sphere
of visible phenomena. The excessive pretensions of the framers of
metaphysical systems had led to hopeless puzzles and merely verbal
solutions. Locke, therefore, insisted upon the necessity of ascertaining
the necessary limits of human knowledge. All our knowledge of material
facts is obviously dependent in some way upon our sensations--however
fleeting or unreal they may be. Therefore, the material sciences must
depend upon sense-given data or upon observation and experiment. Hume
gives the ultimate purpose, already implied in Locke's essay, when he
describes his first treatise (on the title page) as an 'attempt to
introduce the experimental mode of reasoning into moral subjects.' Now,
as Reid thinks, the effect of this was to construct our whole knowledge
out of the representative ideas. The empirical factor is so emphasised
that we lose all grasp of the real world. Locke, indeed, though he
insists upon the derivation of our whole knowledge from 'ideas,' leaves
reality to the 'primary qualities' without clearly expounding their
relation to the secondary. But Berkeley, alarmed by the tendency of the
Cartesian doctrines to materialism and mechanical necessity, reduces the
'primary' to the level of the 'secondary,' and proceeds to abolish the
whole world of matter. We are thus left with nothing but 'ideas,' and
the ideas are naturally 'subjective' and therefore in some sense
unreal. Finally Hume gets rid of the soul as well as the outside world;
and then, by his theory of 'causation,' shows that the ideas themselves
are independent atoms, cohering but not rationally connected, and
capable of being arbitrarily joined or separated in any way whatever.
Thus the ideas have ousted the facts. We cannot get beyond ideas, and
yet ideas are still purely subjective. The 'real' is separated from the
phenomenal, and truth divorced from fact. The sense-given world is the
whole world, and yet is a world of mere accidental conjunctions and
separation. That is Hume's scepticism, and yet according to Reid is the
legitimate development of Descartes' 'ideal system.' Reid, I take it,
was right in seeing that there was a great dilemma. What was required to
escape from it? According to Kant, nothing less than a revision of
Descartes' mode of demarcation between object and subject. The 'primary
qualities' do not correspond in this way to an objective world radically
opposed to the subject