Andreas asked:

I have some questions about British empiricism:

(1) How could Russell claim that 'Every proposition we can understand must be composed wholly of
constituents with which we are acquainted' (The Problems of Philosophyp. 91) and in the same
book claim that we are not acquainted with physical objects, for instance a table? How could Russell
then claim that we could understand propositions of the external world?

(2) Why was Russell's principle of acquaintance in The Problems of Philosophyso influential when
his only argument for the principle was, 'We must be able to attach some meaning to the words we
use' (p. 91)?

============

Russell believed that we are 'acquainted' with particulars, such as a particular sense datum of red, as
well as universals, such as red. The crucial premiss required for his argument for the principle of
acquaintance is that knowledge of meanings must be invulnerable to sceptical attack.In other words,
even if all my beliefs about the external world are false, I still understand what I mean when I state
those beliefs. The meanings of my words are immediately present to me.

Because my knowledge of meanings must be invulnerable to sceptical attack, everyday names for
objects in the external world must ultimately derive their meaning via 'logically proper names' for the
immediate data of consciousness. I can be wrong about the external world in lots of ways, but I
cannot be wrong about the way things seemto me here and now. So although we talk of being
'acquainted' with a place, or a physical object, or a person, that is not a case of genuine
acquaintance. Our relation to these things is indirect. Russell does not consider it an objection to his
theory that most of us have never attempted to 'name' our sense data.

So according to Russell, when I speak of 'my wife', or 'the Meadowhall Shopping Centre in Sheffield',
what I really 'mean' by my words is certain patterns of sense data, which I believe will be repeated in
the future. That belief could turn out to be false. It could turn out that the whole of my life has been a
dream, and neither 'my wife' nor 'The Meadowhall Centre' actually exist.

I would agree with Russell that our knowledge of the meanings of our words is not merely factual
knowledge. The reason his principle of acquaintance has been so influential is that it is difficult to see
a third alternative. If knowledge of meanings is not factual knowledge, then it looks as if meanings
must be immediately present to consciousness, in a way that is completely independent of how the
facts might turn out to be.

I think that Russell was wrong about this. As Wittgenstein showed in the Philosophical Investigations,
knowledge of meanings is a practical abilitythat is made possible through a shared social life with
other language users. There can be no 'private language'.

Geoffrey Klempner