_]
by the arrangement.... That for working purposes we treat, and do well
to treat, some relations as external merely, I do not deny, and that
of course is not the question at issue here. That question is ...
whether in the end and in principle a mere external relation [_i.e.,
a relation which can change without forcing its terms to change their
nature simultaneously_] is possible and forced on us by the facts.'[1]

Mr. Bradley next reverts to the antinomies of space, which, according
to him, prove it to be unreal, although it appears as so prolific a
medium of external relations;

[Footnote 1: _Appearance and Reality_, 2d edition, pp. 575-576.]

and he then concludes that 'Irrationality and externality cannot be
the last truth about things. Somewhere there must be a reason why this
and that appear together. And this reason and reality must reside in
the whole from which terms and relations are abstractions, a whole in
which their internal connexion must lie, and out of which from the
background appear those fresh results which never could have come
from the premises' (p. 577). And he adds that 'Where the whole is
different, the terms that qualify and contribute to it must so far be
different.... They are altered so far only [_how far? farther than
externally, yet not through and through?_], but still they are
altered.... I must insist that in each case the terms are qualified by
their whole [_qualified how?--do their external relations, situations,
dates, etc., changed as these are in the new whole, fail to qualify
them 'far' enough?_], and that in the second case there is a whole
which differs both logically and psychologically from the first whole;
and I urge that in contributing to the change the terms so far are
altered' (p. 579).

Not merely the relations, then, but the terms are altered: _und
zwar_ 'so far.' But just _how_ far is the whole problem; and
'through-and-through' would seem (in spite of Mr. Bradley's somewhat
undecided utterances[1])

[Footnote 1: I say 'undecided,' because, apart from the 'so far,'
which sounds terribly half-hearted, there are passages in these very
pages in which Mr. Bradley admits the pluralistic thesis. Read, for
example, what he says, on p. 578, of a billiard ball keeping its
'character' unchanged, though, in its change of place, its 'existence'
gets altered; or what he says, on p. 579, of the possibility that
an abstract quality A, B, or C, in a thing, 'may throughout remai