energetically we pursue the intellectual work of abstraction the
further we get from the fact itself from which we began. The point of
grouping facts into classes, whether by analysing them into qualities
or classifying them into "real things," is that we can then apply to
the particular fact all that we know to be true in general of whatever
belongs to these various classes: in a word, once we have fitted a
fact into a class we can apply to it all the general laws which are
known to apply to that class.

Common sense, as we saw, tells us that when we distinguish qualities
in any given fact we obtain fuller knowledge than was given in the
mere unanalysed fact, and this knowledge is supposed to become fuller
still when we go on to classify these qualities into "real things."
Bergson, on the contrary, says that common qualities are arrived at by
leaving out much of the fact originally known, while each successive
stage in the process of abstraction by which we explain facts, though
it enables us to apply more and more general laws, yet leaves out more
and more of the actual fact itself. Analysis begins this whittling
away of the actual fact by confining our attention to qualities which
do not exhaust the whole content of the actual fact. At this
preliminary stage, however, though we concentrate our attention on the
quality, we still remain aware of the whole fact in which the quality
has its setting. Classification carries the work of limitation a stage
further. "Things" are a stage further removed from actual fact than
qualities are since, while qualities are classes of facts, "things"
are only classes of qualities. For classification into "things"
therefore only the qualities in a fact will be of any use, and so,
when we have reached the stage of classification, we need no longer
burden our attention with the actual facts themselves in their
entirety, we need pay attention only to the qualities which