CHAPTER X
ON OUR KNOWLEDGE OF UNIVERSALS
Note 1:
Let us consider first the knowledge of
universals by acquaintance. It is obvious, to begin with, that we are
acquainted with such universals as white, red, black, sweet, sour,
loud, hard, etc., i.e. with qualities which are exemplified in
sense-data.
This may seem obvious, but it hard to believe this on Russell's
account. For he said that our experiences are all of particulars, and
therefore what is in our experience are, then, only this white
and that white, this loud noise and that loud
noise, and so on. If these are all particulars, we cannot be
acquainted with universals, in Russell's own sense of acquaintance. Of
course, Russell has his answer to this, which we will remark on when
coming to it below.
On my account, on which all our
experiences are universal or of kinds of things
("things" used in the broadest possible sense, of anything
we may think of, whatever it may be, so that the hypothesis may be
sloganised as: for the human mind, every thing is of some kind),
there is no problem. What is more problematic on my account are
particulars, which for me come by assumption: those kinds of things we
assume to be unique i.e. to occur at any time on one place only. Back.
Note 2:
(..) we easily learn to abstract the whiteness
which they all have in common (..)
This is Russell's account of how we arrive at universals: by
abstracting something that is common to particulars we experience. But
this involves several problems.
First, what we abstract is not a
particular, and since all experiences are particular, on Russell's
account, what we abstract cannot be an experience. In cases involving
white or whiteness, and many other similar cases, this seems false.
Second, Russell does not explain what
he means by abstracting. It may be fairly inferred he meant something
like: selecting something by disregarding something else that comes
together with it, but if so, the problem is that again, in the cases
under consideration, what we select seems of the same kind (particular
or universal) as what we select from.
Third, it seems that in case of such
universals as Russell mentions, which are all of the kind of what he
calls sense-data, one needs not perform any mental operation, such as
abstraction of patches of white, to perceive a universal whiteness.
One simply sees several patches of white, which are several instances
of a universal, since the universal white recurs identically in them. Back.
Note 3:
Universals of this sort may be called 'sensible
qualities'. They can be apprehended with less effort of abstraction
than any others, and they seem less removed from particulars than
other universals are.
On
my account, sensible qualities are simply given and need not be
abstracted at all. Insofar as mental operations are required for their
cognition, experience and recognition seem to be all that need be
involved. Back.
Note 4:
The process of
abstraction in this case seems to proceed somewhat as follows: I see
successively a number of sense-data in which one part is to the left
of another; I perceive, as in the case of different white patches,
that all these sense-data have something in common, and by abstraction
I find that what they have in common is a certain relation between
their parts, namely the relation which I call 'being to the left of'.
In this way I become acquainted with the universal relation.
It
seems to me that there are sensible relations in the same way as there
are sensible qualities: we simply experience or recognise them. Back.
Note 5:
From either of these sources I can abstract the
universal relation of before and after, just as I abstracted the
universal relation 'being to the left of'. Thus time-relations, like
space-relations, are among those with which we are acquainted.
Again it seems to me that before and after in time are simply given in
experience and do not involve some highly dubious process of
abstraction. Back.
Note 6:
In
this way I become acquainted with the universal resemblance
or similarity.
I
don't think so since the relation of resemblance differs from the
universals we have considered so far.
One way of bringing out the difference
is thus: so far we've considered claims to the effect that "there
are several things which have something in common, and it is
such-and-such", where such-and-such is a sensible quality like
whiteness or sensible relation like left of. At present, we deal with
the somewhat more general or slightly different: "there are
several things, namely this and that, which have something in
common".
Thus, the recognition of some
resemblance or similarity is involved in the recognition of any
sensible quality or relation, but one can truly claim that several
things resemble each other without saying what the resemblance is, or
without claiming they resemble each other in respect of some sensible
attribute.
So it is here, in judgments to the
effect that things resemble each other, without thereby judging what
the resemblance is, or without thereby judging the resemblance is a
sensible attribute, that it seems a genuine process of - what may be
called - abstraction is used. The reason to qualify the term
"abstraction" is that it seems less dependent on some form
of disregarding as it does depend on language: it is easy to
see how we arrive in this way at "this triangle and that circle
have something in common" since they are both geometrical shapes,
and very difficult to see how one can abstract the notion of shape
from seeing a triangle and a circle.
In other words, the process of
abstraction Russell really used seems to me to be different from what
he thought it was, and to involve my earlier "to the human mind,
every thing is of some kind" and language, that generally makes
it easy - for one familiar with a natural language - to supply various
words for qualities and relations that are (supposed to be) true of
some things one is thinking of. Back.
Note 7:
Here we are
dealing with a relation, namely 'greater than', between two relations.
Our knowledge of such relations, though it requires more power of
abstraction than is required for perceiving the qualities of
sense-data, appears to be equally immediate, and (at least in some
cases) equally indubitable. Thus there is immediate knowledge
concerning universals well as concerning sense-data.
It
seems to me my previous note explains the reason for the knowledge we
have of relations between relations better. It seems in most cases to
depend on language. Indeed, what is involved if we for the moment
indulge in a few elementary formalities, seems to be this: Given two
pairs of things, a and b and c and d, and two relations R and S
between these, so that a and b stand in relation R and c and d in
relation S, we may also know that there is some respect T (whether
quality or relation) that is true both of the relation a and b have
and the relation c and d have. This respect T, if it is itself a
relation, may be considered a relation between relations.
In most cases where human beings
believe themselves to know that several relations are related this
belief is due to their linguistic skills, if only because it is far
easier to perceive a relation between the terms used for things (and
having isolated it attribute it to the things the terms stand for)
then to perceive a relation between things.
Indeed, this is part of the reason
for the cognitive successes of the human mind: in general it is far
easier to think about the terms for things than about the
things themselves, since the terms and their relations are far simpler
than what they stand for. Back.
Note 8:
This suggests a
proposition which we shall now endeavour to establish: namely, All
a priori knowledge deals exclusively with the relations of
universals. This proposition is of great importance, and goes a
long way towards solving our previous difficulties concerning a
priori knowledge.
The
proposition Russell wants to establish is false. To show this all we
need to establish is that there is - in Russell's or our opinion - a
priori knowledge that does not deal exclusively with the relations of
universals, and one such proposition is "There is an
independently existing reality"; another is "There is
experience" and one Russell ought to insist on, given his earlier
text, is "I exist". Back.
Note 9:
If we can show that such statements as this
really deal only with universals, our proposition may be regarded as
proved.
This is shoddy reasoning. First, Russell does not consider the denial
of what he seeks to prove, but a special case of it, the proof of
which he declares here ought to establish the more general
proposition. This way of arguing is capable of proving anything. Back.
Note 10:
(..) to ask ourselves what words we must
understand -- in other words, what objects we must be acquainted with
The understanding of words does not necessarily involve an
acquaintance with the objects they denote, as ought to be obvious when
we consider words that denote things we do not have experience of
either because we did not have the chance, so far, or because this is
impossible as the objects do not or cannot exist, or can be grasped
only linguistically (such as "all natural numbers" or
"all human beings, whether past, present or future").
Most of the words we understand depend
on the understanding we have of other words, and thus on our capacity
to think of things or imagine them (where both processes are equally a
matter of fantasy, but only the latter involves sensible possible signs
of things: we can think of far more things in terms of symbols than we
can imagine in terms of signs, and that is another reason for the
cognitive successes of the human mind.). Back.
Note 11:
As soon as we see what the proposition means,
even if we do not yet know whether it is true or false, it is evident
that we must have acquaintance with whatever is really dealt with by
the proposition.
Often that is impossible, or at least very problematic: what if the
things meant by a proposition are impossible or non-existent? It is
easy to think of people who believe (what we think are) manifest
absurdities and contradictions, but if this is true we are thinking of
something that cannot exist. This shows the power of symbols, in that
symbols may represent what cannot possibly exist (if only to deny
existence to it), which is something that should be accounted for. Back.
Note 12:
It is quite unnecessary to know all the couples
in the world: if it were necessary, obviously we could never
understand the proposition, since the couples are infinitely numerous
and therefore cannot all be known to us.
In
one way, this is true, but in another way gravely misleading: I insist
that we do know something about all the couples in the world,
namely that their number is two, and that, therefore, the number of
any two couples is four. And the reason we do know this, on my account
of universals, is that they all have the same attribute, i.e. what is
(a) universal is something that identically recurs. Back.
Note 13:
The statement made is about 'couple', the
universal, and not about this or that couple.
This is misleadingly expressed. First, the statement is not about the
term 'couple' but about anything the term may be true of, and
secondly, since it is about any thing that is a couple, it is also
about this and about that couple (which are two particulars - this and
that - that have a universal - being a couple - in common). Back.
Note 14:
It must be taken as a fact, discovered by reflecting upon our
knowledge, that we have the power of sometimes perceiving such
relations between universals, and therefore of sometimes knowing
general a priori propositions such as those of arithmetic and
logic.
Sure, but this way of putting it, and especially the term
"perceiving" doesn't make this clearer. And rather than
ordinary perception linguistic skills seem involved, that easily move
from "this is green" and "that is green" to
"there is something this and that have in common", and from
the additional premise "green is a colour" and "if this
is a colour, this is a quality" to "this and that have some
quality in common". Back.
Note 15:
The thing that seemed mysterious, when we
formerly considered such knowledge, was that it seemed to anticipate
and control experience.
On
my account of universals, that is based on the assumption that the
human mind reasons in terms of kinds, this is not mysterious. Back.
Note 16:
No fact concerning anything capable of
being experienced can be known independently of experience.
Really? What about the very proposition Russell gives? And my point is
not that one can somehow know truths of experience without experience,
but is rather that we do seem to know or assume quite a few general
things about things capable of being experienced, and about our
capacities to experience, without it being necessary to experience
those things. (Thus I know that anything I will see but have so far
not seen will have some sort of shape and colour; that if I keep being
conscious I may judge and classify anything I experience; that if I
cannot assign any kind to what I experience it must remain a riddle,
and many more propositions of the same kind.) Back.
Note 17:
(..) we
do not know a priori that if Brown and Jones are
two, and Robinson and Smith are two, then Brown and Jones and Robinson
and Smith are four.
This seems wholly false to me. We may not know there really is anyone
named Brown etc., but that is not relevant, for we are considering a
conditional proposition for which such knowledge is not required. Back.
Note 18:
(..) the
difference between an a priori general proposition and an
empirical generalization does not come in the meaning of the
proposition; it comes in the nature of the evidence for it.
No, and in view of Russell's notion that universals are a special kind
of thing in their own world it is odd he says this. In any case, the
difference between an a priori general proposition and an empirical
generalisation is one of meaning, and therefore the two kinds of
proposition have a different relation to evidence. The difference is
that the a priori general proposition generally assumes a relation of
inclusion between kinds ('human beings are animals'), that, if true,
imply empirical generalisations ('everything that is a human being is
an animal'), whereas empirical generalisations do not assume a
relation of inclusion between kinds ('everything that is a human being
has been born on the planet Earth") and thus are compatible with
the notion that the kinds involved in the empirical generalisation may
cease to be empirically included. Back.
Note 19:
We believe
that all men are mortal because we know that there are innumerable
instances of men dying, and no instances of their living beyond a
certain age. We do not believe it because we see a connexion between
the universal man and the universal mortal.
I
don't believe this. What I believe is that men are so constituted that
they must die; that they cannot grow taller than 100 feet; that they
cannot fly by flapping their arms; that they are able to learn natural
languages; and many more such propositions, and that, men being thus
constituted i.e. being or not being of such and such kinds, that therefore
we are able to see innumerable instances of the truth of these
propositions.
The mere occurrences of instances that
may illustrate certain general propositions is no reason at all to
formulate or believe such general propositions, especially not if we
do not believe that these instances belong to the kind of pattern we
can explain by assigning them to some relation of inclusion between
kinds. Thus - for example - being Dutch, my whole life is replete of
instances of women who happen to be white, incapable of speaking
Chinese, and not fond of Italian music, yet this has never moved me to
any general proposition to the effect that all women are thus
disposed.
So unlike Russell I believe we do
formulate such principles because we believe there is some sort
of connection between two kinds of things. Back.
Note 20:
But although
this gives a greater degree of certainty, it does not give a
different kind: the ultimate ground remains inductive, i.e.
derived from instances, and not an a priori connexion of
universals such as we have in logic and arithmetic.
The foregoing note entails I also disagree with this. It seems to me
that our inductive generalisations about the real world, if they do
not explicitly concern mere correlations or coincidences, are all
based on some presumed constant relation between kinds of
things, for which presumption we try to find empirical evidence.
One way of phrasing a pertinent
difference is that I hold that we - often - do not move from the perception of
instances to a general law, but rather from a guessed general law to a
gathering of instances of it. It is true that such a general law may
be guessed from a few instances of it, but this need not be so, and in
any case any finite number of instances may be explained by any number
of different generalisations about them, each of which is supported by
those instances. Back.
Note 21:
Here
is a general proposition of which the truth is undeniable, and yet,
from the very nature of the case, we can never give an instance;
because any two numbers we may think of are excluded by the terms of
the proposition.
This
possibility, of knowledge of general propositions of which no instance
can be given, is often denied, because it is not perceived that the
knowledge of such propositions only requires a knowledge of the
relations of universals, and does not require any knowledge of
instances of the universals in question. Yet the knowledge of such
general propositions is quite vital to a great deal of what is
generally admitted to be known.
Indeed. And let us note carefully that, on our theory of universals,
the fundamental reason we may have such knowledge is that kinds
of things may be thus related, and if they are all the things
which are of those kinds will be thus related, and that we do not need
to know any specific such things to know this, but that all we need to
know is some true knowledge about their kinds. This type of knowledge
is of the same kind as what makes us so confident that all widows of
the year 2300 AD will have been married: we assume the kinds to be
thus related so as to make this true of whatever things are of the
kind, wherever, whenever. Back.
Note 22:
Hence we can never know any proposition of the
form 'this is a physical object', where 'this' is something
immediately known.
I
don't believe this, and doubt it is really polite. In any case, I feel
pretty certain that I am myself, like my readers, among other things,
a physical object.
Russell's reason for believing that he
could never know anything is a physical object is that it is a
judgment about something not given in experience.
I have commented on this before, but
repeat that if there are physical objects we may have experiences of,
we may validly infer their existence from the experiences of them,
just as we infer the existence of objects photographed from their
photographs.Back.
Note 23:
(..) all our knowledge concerning physical objects is
such that no actual instance can be given. We can give instances of
the associated sense-data, but we cannot give instances of the actual
physical objects.
Here we are firmly within the region of philosophical dreams, it seems
to me. I agree there is a real underlying problem, but the way Russell
tries to solve it leads into illusions and odd philosophical seeming
profundities.
The underlying fact of the problem (or
so we shall assume) is that the experience of X is not necessarily the
same as X, whereas the only way of knowing an empirical X, apart from
correctly guessing it or deducing it from other knowledge, is by
experiences of X. This I grant, but it proves as little that we cannot
possibly know X as it follows from the fact that the term for X is not
necessarily the same as X proves that we cannot know anything about X
from the term for X.
It so happens that humans know by
means of signs and symbols, and inferences based on them and framed in
terms of them, and that the only thing that occurs in a human being's
experience are signs and symbols. However, if human beings do know
anything at all about the real things the signs and symbols in their
experience are signs and symbols of, which Russell calls 'physical
objects', they know something about physical objects.
Finally, there is also this
consideration. Again I grant that all humans know of any real thing X
are the experiences they have concerning X, and that the two do not
necessarily coincide. Even so, just like I may know a lot about real
hands from inferences about the behaviour of gloved hands, it seems I
may know a lot about real things from inferences about my experiences
of them. Back.
Note 24:
And the same applies to our knowledge of other
people's minds, or of any other class of things of which no instance
is known to us by acquaintance.
As
pointed out before, this denies ESP, and also denies imaginative
(partial) identification with other people, sympathy, role-playing and
other similar means to know or guess at another person's feelings and
beliefs.
Besides, it seems a fair guess that
human beings are born with a special facility to understand the
private experiences of others, since so much depends on their having a
fair understanding of it.
In any case, to settle possible
misunderstandings: I believe ESP is logically possible but does not
exist i.e. all apparent evidence for it is misleading or involved
trickery, and I believe human beings have some facility for
understanding others, and lots of reasons to develop it, and in many
cases it is quite easy and natural to make a fair guess about what
another human being in certain circumstances does or would experience. Back.
Note 25:
We may now take a survey of the sources of our
knowledge, as they have appeared in the course of our analysis.
I
shall not extensively comment on Russell's summary, since I have
commented on all relevant points in it before. The only thing I do
want to point out is that the main differences between my analysis and
his concern that of universals, and my related thesis that all our
experiences are experiences of kinds of things, and thus of
universals, rather than of particulars (even though I agree with
Russell that each of our experiences is a particular experience, since
it happens to a specific person at a specific place and time). Back.
Note 26:
Our
immediate knowledge of truths may be called intuitive
knowledge, and the truths so known may be called self-evident
truths.
Since we are going to be involved with analysing what Russell calls
'intuitive knowledge', it is well to repeat the gist of a remark I
made before: There is very good evidence to the effect that many of
the things people believe to be intuitively true in fact are false,
and that there are some necessary and, apart from that, many common
illusions and delusions.
The reason to stress this is that
these falsities are often overlooked, and that it certainly seems at
least a fair possibility that human beings not only may know some
truths but may need to be misled (flummoxed, deluded, flim-flammed) by
the nature of some of things they try to know.
It may be that such falsehoods are
especially those that concern the qualities of our leaders, friends,
enemies, and cognitive and moral capacities (our leaders invariably
are Great Men, our friends Excellent People, our enemies contemptible
idiots, our own cognitive capacities even if not of the very best are
still quite able to see and appreciate the fundamental verities just
mentioned, while our moral capacities and intentions are, of course,
always among the best and the noblest) but it may also be that they
concern non-human aspects of reality, such as the structure of space
and time, or the relation of evidence to guesses (which tends to be
much more dependent on our likes and dislikes than we like to think:
very often people believe things to be true because they desire them
to be true, and almost as often people bend, bias or selectively prune
whatever evidence they have to conform to their desires).
These possibilities will concern us
later. Back.
Note 27:
If the above account is correct, all our
knowledge of truths depends upon our intuitive knowledge. It therefore
becomes important to consider the nature and scope of intuitive
knowledge
It
was already clear to Aristotle that all our reasoning must be based on
assumptions, and that therefore we have to start with some
assumption(s) that cannot themselves (at that point) be explained or
entailed by yet other assumptions.
To call these assumptions that are
necessary to base any conclusion upon 'intuitive knowledge'
seems to involve a rather self-congratulatory account of human
capacities. That is: Why style it 'knowledge' if it is clear that
often it is at best some guess based on some particular evidence?
And in any case, a philosopher should be aware that most
assumptions men have framed to account for their commonsense experiences have been
refuted in the course of time, for which reasons it seems better to
avoid the term 'knowledge' in the present context, and to stick to
'belief' or 'guess'. Back.
Note 28:
This problem
does not arise with regard to knowledge by acquaintance, for, whatever
may be the object of acquaintance, even in dreams and hallucinations,
there is no error involved so long as we do not go beyond the
immediate object: error can only arise when we regard the immediate
object, i.e. the sense-datum, as the mark of some physical object.
This makes sense, but it also involves a problem, which may be brought
out thus: People who lost a limb often complain about feelings of pain
in the limbs they lost, and also often are given some sort of mental
representation of the limb they lost that endures for a long time
after they lost the limb.
Strictly speaking, this is not in
contradiction with Russell's words, but it does show that it is very
natural to pass from the immediate object of our experience to the
supposed object they are an experience of, and it also shows that at
least the experiences we have of our own bodily organs is somewhat
peculiar and special. Back.
