volition,
sensation, or association, are seldom perfect copies of their correspondent
perceptions, except in our dreams, where other external objects do not
detract our attention.

4. Those ideas, which are more complex than the natural objects that first
excited them, have been called compounded ideas, as when we think of a
sphinx, or griffin.

5. And those that are less complex than the correspondent natural objects,
have been termed abstracted ideas: thus sweetness, and whiteness, and
solidity, are received at the same time from a lump of sugar, yet I can
recollect any of these qualities without thinking of the others, that were
excited along with them.

When ideas are so far abstracted as in the above example, they have been
termed simple by the writers of metaphysics, and seem indeed to be more
complete repetitions of the ideas or sensual motions, originally excited by
external objects.

Other classes of these ideas, where the abstraction has not been so great,
have been termed, by Mr. Locke, modes, substances, and relations, but they
seem only to differ in their degree of abstraction from the complex ideas
that were at first excited; for as these complex or natural ideas are
themselves imperfect copies of their correspondent perceptions, so these
abstract or general ideas are only still more imperfect copies of the same
perceptions. Thus when I have seen an object but once, as a rhinoceros, my
abstract idea of this animal is the same as my complex one. I may think
more or less distinctly of a rhinoceros, but it is the very rhinoceros that
I saw, or some part or property of him, which recurs to my mind.

But when any class of complex objects becomes the subject of conversation,
of which I have seen many individuals, as a castle or an army, some
property or circumstance belonging to it is peculiarly alluded to; and then
I feel in my own mind, that my abstract idea of this complex object is only
an idea of that part, property, or attitude of it, that employs the present
conversation, and varies with every sentence that is spoken concerning it.
So if any one should say, "one may sit upon a horse safer than on a camel,"
my abstract idea of the two animals includes only an outline of the level
back of the one, and the gibbosity on the back of the other. What noise is
that in the street?--Some horses trotting over the pavement. Here my idea
of the horses includes principally the shape and motion of their legs. So
al