The space and time connections of things are special examples of the connections of parts in systems, and the questions which come up with them, for the detail of winch we have to go to mathematics, are outside the range of these pages. But the senses of name and of the other words which may take its place are part of Logic. With the discussion of them we come to the third division of this book, the Theory of Instruments.
The reader will have taken note of a number of words which have been used at important points without any special account of their sense being given. Among them have been agreement, same, different, property, quality, relation, general, special, degree, sort and is. What comes now is an attempt to put our uses of these words in a clear light. I take them together in a separate division under the name of Theory of Instruments because -- more, it may be, than with any other words -- we have to take care not to put the wrong questions about them. All words are instruments with winch we keep control of their senses : but these words are instruments with whose senses in addition to this we keep control of the senses of other words. It will be best to make a start with Property.
Property 12-1 . In the simple everyday sense, a property is that of which some man is the owner, that which some man has. 12-2. In a sense taken from 12-1,
anything which anything (see thing, 2-2)
may be said to be or to have (see is 13-1). For
example : mountains are high -- have the property of being high; 6 is greater than
5 -- has the property of being greater than 5 ; men are not plants -- have the property
of not being plants. No limit is necessary to the things we may, truly or falsely, say
are properties of things. Everything is what it is -- has the property of being the
same as itself. The moon has the property of having the property . . . of
not being made of green cheese. This last is only another way of saying that the
moon is not made of green cheese.
But there is a danger of our taking
properties as something more than other
ways of saying what may be said without them. Experts in mathematics have said,
for example, that "even if there were no things at all, there would still be the
property of being seven in number -- though nothing would have this property".
This danger may be overcome by keeping in mind that all properties are like this
property of the moon -- that it is not made of green cheese.
In other words a property is an instrument for making comparisons between
statements, between thoughts, between the senses of words. But we commonly
make use of properties as if they were something more, part of the structure of
what we are having thoughts about or making statements about. ' X is green '
and X has the property of being green do not say different things -- the second is
not a fuller account than the first. But they say the same thing in ways which
are of use for different purposes, and these
purposes for which the word property is of use have now to be taken into view.
Here comparison may be made with
our use of the words being and existence.
To say. ' There are cats and dogs ' and to say ' There are beings which are cats and
dogs ' or to say ' Cats and dogs have existence ' or to say ' Cats and dogs are
things (2-1) ', or to say ' Cats and dogs have the property of having existence ' are
only different ways of saying the same thing. Things, beings, existences, are
words of use when we have a need to put into words thoughts which are not only
about cats or only about dogs, or only about any limited sort of thing, but are
about anything. In the same way property
is a word of use when we have to do with thoughts not about green things only, or
about coloured things only or about limited sorts of things, but about the ways
in which things are in sorts. The danger
is that we make use of property when there
is no need for it, and so get the idea that
' X is green ' and ' X has the property of being green ' say different things.
This will be the best place in which to
take the word is. It has three chief senses.
Is 13-1 . Existence . To say ' A is ' or ' There
is an A ' is the same as saying ' A has
existence ' or ' A has being ' or ' Something is A '.
13-2 . Part and Sort . The is of connection.
To say ' A is B ' puts A into the greater sort B. This connection between A and B
may be taken in two ways
13-21 . A is a part of the sort B. For example, ' Gold is a metal ' --
gold is part of the sort, metals : or
13-22 . The properties common to B are part of the properties of A. Gold
has the properties common to all metals and some more -- those which make it
different from all other metals. These two ways of taking ' A is B ' are of use for
13-3 . Completely the same (in Logic ' identity ').
To say ' A is B ' is the same as saying ' A has all the properties which
B has and no others '. When this is so completely, and A and B are not different
in any way, then they are not two things, but one thing. ' A ' and ' B ' are two
names for one thing, and the statement ' A is B ' then becomes a bad way of saying
this, because it is hard to keep in mind that A and B are not names for different
things. We will see (pp. 101,119) some of the dangers which come from this.
It is clear from this that same has two
important and different senses ; one in which to say that two things (or groups of
things) are the same is to say that in certain ways they are not different,
another in which we are only saying that one thing, or group of things, has two
names, or is being taken in two sorts.
The three words property, sort and same, it will be
noted, do almost the same work. So does the word general, to which
we may now go on-coming back later to the discussion of the forms of properties.
General This word has one very important sense -- and some others which in different ways
have come from it but are now almost opposites to one another. This makes it a
very interesting example of the ways in which senses become changed.
First take a look at these marks on the paper.
14-1. They are all marks. What they
all are is what is general about them -- a
general property of them. In addition to
being marks they are all marks-used-in-printing. This is another general property
of them, equally general in this sense. It
comes from the sense of all -- which is all
of a sort. But in another sense it is less
general because marks-used-in-printing axe
only a part of all marks. In the same
way marks which are not marks used in
printing may be on paper. To be a mark-
on-paper is a more general property than
to be a mark-used-in-printing and to be
a mark-on-something is more general than
to be a mark-on-paper. So, to be coloured
is more general than to be green. This is
a more general (14-11) sense of general, but
one which comes from 141, for marks on
paper are a part only of all marks.
14-11. A thought (or a statement)
which is about a greater group of things
(having a group G as part of it) is more
general than a thought about the things in
group G or the things in a part of G.
14-12. A thought of anything as being
of a sort is a general thought : it is more
or less general as the sort is greater or less.
14-2 . Most of these marks here are full
stops. We frequently say that something
is generally so, when the sense of general is
in most, but not all examples. This sense is
an opposite of 14-l, where all examples
have to be so for the statement to be< i>generally true.
14-21. A statement about a great
number of a group of things is a more
general statement than one about a smaller
number of them.
14-22. Even more loosely generally may
have the sense of frequently only.
Some of these senses of general are taken
by the word common. What is common
to a group is what all the things in the
group have (14-1). A property common
in a group is one frequent in it (14-22). But
common has two other important senses.
A number of persons have something in
common (a property 12-1) when no one of
them is the owner, but they all taken
together are the owner. And, a thing is
said to be common when it has no special
value, when it is so frequent that to have
it does not make a man different from
The word used as an opposite to general
is special. As some is to all, so special is to
general in this first root sense. Special
goes with general in 14-11. To be marks on
paper is a more special property than to be
only marks. So with 14-2 and 14-21. A
special example is one which is different
in some way from* most examples ; in this
sense special is frequently used to give a
suggestion of value. A special statement
it may be noted is frequently one made
for a special purpose.
We saw in the discussion of sense 2-2 of
thing, and in connection with the word
' abstract ' that a thing without its
properties and a property without its things
are equally incomplete, and that there are
two ways of taking parts of the sense of a
word from the other parts of its sense. An
'abstract' thought is (i) of a quality or
property taken from the thing which has
it ; or (ii) it is of a thing without the
properties which it has. For example it
may be a thought (A) of a leaf of a tree,
as being green, soft, full of holes or folds,
from a certain branch, and now in my hand
and so on ; or we may have a thought (B)
of it as being only a leaf, as having only the
properties which are necessary if it is
to be a leaf. To have a thought of it of
the (in comparison) complete sort (A)
is to have a ' concrete ' thought. A
thought like (B) is of the opposite sort :
it is an ' abstract ' thought.
A ' concrete ' (or taken together) thought
of a thing puts it with a great number of
sorts. It puts this leaf here into the sort
of green things, the sort of soft things,
the sort of here-things, now-things,
-- and so on. Every one of these
sorts may be more or less general, 14-11.
In place of taking it as a green thing, I
might have taken it as a coloured thing, or
as an able-to-be-seen-thing, for example,
or as an apple-green thing, to go the other
way and put it in a small special sort. And
so with the other sorts I put it in. Now if
the sorts I put it in are like the picture, so
that it is the only thing which may be in
(or of) these two lines or sorts, then we get
an account of it
[ diagram of an 'X'.]
which is as special as possible.
The simplest way of making a thought
special is by using the sorts here and now,
or then and there -- by giving the thing
which the thought is of a position in time
and space, because our knowledge of the
order of the positions of things in time
and space is so complete that we are
certain that one thing and only one thing
has a given position. In Logic a thing of
which we have a thought which is completely
special -- that is, which is not a
thought of any other thing in addition --
is sometimes named a 'particular', but
there are other senses given to this word.
Our only way to have a 'particular' is
to put it into a number of sorts which
have no other example in common.
This and That, the most frequent names
of particulars get their power by putting
the things they are used for into groups
of sorts which have only one example in
From a completely ' concrete ' thought
of a thing we may go in two ways, (i) to
a thought of one or other of the sorts if
is in, the properties it has, -- for example,
to a thought of green or soft or thin, (ii) to
a thought of it as something which has
properties (see 2-2).
Whichever way we go we come in the
end to having a thought of it only as a
thing (2-1). For example, from a thought
of a bit of chalk as a white,
powder-covered, dry, used-for-writing,
in-my-hand, here, now, thing we may
go on to thoughts of it as :--
only a bit of chalk
a bit of calcium carbonate (CaCO3)
a bit of a chemical substance
a substance (that is, a thing 2-1)
Or we may go off to thoughts of it as :--
able to be seen
able to be sensed (8-2)
-- that is, a thing, 2-1
So ways of ' abstracting ' which seem at
the start to be going in opposite directions
come in the end to the same completely
Sometimes it is not hard to see that one
thought is more 'abstract' than another.
But with 'This is green' and 'This is
coloured', for example, no answer is at
present possible. Our knowledge of how
the mind does its work is not complete
enough. We are able, on the other hand,
to say that 'This is green' is less general
than 'This is coloured' because green
things are a part of coloured things. And
this question 'How general is a thought ?'
is much more important than the question
'How abstract is it ?'
It is not hard to make completely
general true statements : 'At all times
something is going to take place' is an
example. Equally it is not hard to make
completely special true statements : 'This
is quite this green'. What is hard is to
make true general statements about special
examples, statements true of all of a
certain limited sort of things. In this
attempt one danger is so common that an
account of it may be of value. It is the
danger of sorting things in the same way
with two different forms of words, and
then taking ourselves to have said something
which is important about them --
not only about the senses of our words.
For example, a number of writings may
have some effect upon us which is the
same. By this they are sorted. They
are the writings which have the effect X
on us. We then give them a name.
They become, say, Poetry for us. We then
put the question, ' What have these
examples of Poetry in common? ' We
tightly say, 'If they have no property in
common then we have no reason to give
them a common name'. But in saying
this we frequently do not keep in mind that
they have in common the property of
causing-the-effect-X-in-us, and that this
property has been used in sorting them.
We go on to make attempts at the discovery
of another common property.
After much time and trouble and work
we may be able to come upon some other
common property which they all have, and
which no other things have. But with
most groups of writings to which we give
the name Poetry this is not very probable.
The only common property which they
will probably have is the one we have used
in sorting them at the start -- that of
causing the effect X. Great numbers of
serious persons have given this property
two different names and said 'All poetry
is beautiful ', or 'All poetry has form ',
with an idea that they were saying some-
thing more than ' What gives the effect
X, gives the effect X '.
But, someone may say, ' They would
not be the cause of the effect X without
having some other common property by
which they give it. For the same effect
to take place there has to be the same
cause !' This is the point at which the
danger comes in. There is a common
belief that for the same effect there has to
be the same cause, but this is false, if we
take cause and effect in the only sense in
which it is possible to take them here.
That is, in sense 9-11, -- events which will
be widely separate in time and space.
Much takes place between the writing of a
verse and its effect in the mind. A very
great number of chains of events come in
from other directions. For a parallel
example, take as an effect, Death. What
property is there which all things which
are causes of death have in common?
Poison, blows, knives, not enough air,
falls, . . . ? Or what is the common
property of all the causes of pleasure?
Or of angry or delicate feelings ? We will
be in agreement that the causes (9-11) of
these will frequently be as different from
one another as a smell from a noise.
Only very simple persons make this
error with Death or Pleasure, but numbers
of persons who are not at all simple (or
only simple in their ideas of logic) have
made it with Poetry, Art, what is Beautiful,
what is True and such things. The parallel
with Death they do not see ; but they do
see a parallel with the examples of Red,
Sweet, Sharp and Long. Red things have
a common effect upon us, and they have a
common property (in almost all examples)
by which they have this effect. The time
and space distance between cause and
effect is here very small. So it is with
and Sharp and, in less degree, with
Long. So here the argument from a
common effect to another common property
as the cause is a good one. But it is a
very bad one with Beautiful Things or
True Things' or Poetry or Art, because
the chains of causes in between the things
we say are Beautiful, are Poetry or Art,
the other causes, and the effects in us
are so long, complex and delicate. Generally
(14-2) the properties which we take as
the causes in works of art of their effects
of this sort in us, are more probably these
same effects themselves put by us back
into our picture of the things. They are
effects in us which we see as properties
With this we come back to property,
quality and relation. Property is the
general word. Under it may be put a
division between qualities and relation-properties,
though quality is frequently
used in the same sense as property.
Quality Ab Relation-Property A-B
| | | |
Necessary May be (possible) Necessary May be (possible}
Quality 15-1 . A quality
is a property of which
a complete account may be given
naming any other thing (2-2) than the one
which has the quality. A picture of a
quality is Ab. Examples of Qualities :
A is Red, A is Beautiful, A is Wise....
Quality has other senses:
15-11 . It is used to give a suggestion of
value -- as short for 'good quality'. In
addition it is used in judging paintings and
music when we have nothing more special
to say -- as a sort of space in our statements
into which we let other persons, if
they are able, put something.
Relation 16-1 . A relation is a property which
puts two or more things (2-2) into connection
with one another. Pictures of relations are A -- B
A< A --< B -- A -- C
Examples of Relations : A is greater
than B ; A is the father of B. A is between
B and C. A is later than B. A is better
A has much knowledge,
A does the right thing at all times.
By comparison of the last examples, and . . .
others like them, with ' A is wise ', we see
that the question, 'Is this a quality or a
relation-property ?' is frequently one
which may be answered in the two ways.
The history of the language frequently
makes us take a sense as a quality when a
little more attention would make it clear
that it is better to take it as a relation-
property. Sometimes this has very serious
effects upon thought. Most of the troubles
which made the theory of relativity1 so
slow in coming, and so hard for the general
public to take in, came from the way in
which position and time were taken to be
qualities of things, not relations which
they had to other things. A great
number of dangers to thought of this sort
are still strong and damaging in the
present day theories of the mind. Young
boys and girls before they are 7 years old
generally take right and left (hand) as
qualities. They have trouble in seeing
how of three things :--
A ----- B ----- C
B is right of A and left of C. If B is
right how may it not be right ? In the
same way some experts in the theory of the
Beautiful have great trouble in seeing how
anything may .be beautiful and not
beautiful. It seems probable that qualities
are a special form of fiction whose purpose
is to give us stronger feelings for things
and keep us from wasting time unnecessarily.
1 The theory that things are not in fixed times
and places in a fixed time and space but that all
things may only be measured in their relations to
An interesting and natural question
comes up here. May the same fact be
looked on as a quality and as a relation-
property ? May Ab and A--B be the
same as facts ? May they be two ways
of taking one fact, or are they such that
one has to be false if the other is true ?
The answer seems to be that, if we are not
attempting to say something about the
structure of the fact, but are only sorting
facts then they may be the same. If we
are going into the structure of the fact, our
question becomes one about fictions -- that
is, about the use of our language machine.
But a statement using qualities frequently
seems very different from one using
relation-properties. For example, 'X is
beautiful' may sometimes seem very
different from ' X has certain effects in
Relations between relations are an
important group of relations. For example,
take A B C.
(1) B is between A and C..
(2) B comes before C and after A.
(3) B is the son of A and father of C.
Between these three -- a space, a time,
and a cause relation -- there is a relation
of being the same in one way. Let us
make a comparison between these relations
and A == B. The comparison gives us
a different relation between relations.
Relations, like other properties, may be
general or special , being in a box, being in
a certain part of a box ; they may be simple
or complex: being the father of A, being
the hated father of A , they may be ' abstract'
or ' concrete' (see in comparison,
pp. 34, 97) : being in a space relation to A,
being in A. The detailed theory of
relations has important developments in
Logic and Mathematics.
Necessary Properties (qualities and relations) are
of two sorts : necessary and chance.
17-1 . A necessary property is one which
a thing has under all conditions. 17-2 . A chance property is one which a
thing has only under some conditions. Examples :
Necessary. Straight lines from the
middle point to the limiting line of a circle
are all equal. A is not not-A. What
I have done I have done. The past may
not be changed.
Chance. Balfour had on a tall hat at the
Versailles Conference, Clemenceau a round
one. Every year millions of good apples
are put in the water to keep the prices
up. Death comes to all.
The two words necessary and chance
have a number of other senses which have
to be kept out here, as the examples will
17-11. We say something is necessary
without which under present conditions
things would not be as they are. Even
when we see that it is not probable that
conditions will ever be changed (as with the
last example 'Death comes to all') the
something is not necessary (in the 17-1
sense) if a. possible change in conditions
might make things different. It is only
where no change of conditions, however
great, makes the property different that it
is necessary in sense 17-1.
17-12. A special and very common
sense of necessary is that in which, if we
have a desire to do something and there
is only one way of doing it (or one way
is better than any other) that way is said
to be necessary.
17-3. We say something is a chance
when we have not enough knowledge of the
conditions under which it takes place, or
when (17-31) such knowledge as we have
makes us ready for another thing to take
place. But in 17-2 the degree of our
knowledge of the causes of the thing does
not come in. If with a change in conditions
a change might come in the property,
then it is a chance property.
Another way of saying that a property
of a thing is necessary in sense 17-1 is to
say that it is not possible for this property
of the thing to be changed. If it was
changed the thing would then not be that
thing. It would become another thing.
So the question 'Is this a necessary (17-1)
property ?' is a question about the senses
of the words with which the thing and
its property are being named -- that is
a question about our thoughts of them.
Possible 18-1. Something is possible when its
being so does not make any change in our
language machine necessary (17-11).
This has to be taken with care. Any
one language machine (any one group of
senses) we make puts limits to the things
which may said with it. What comes
outside these limits is ( in this sense) not
possible. By making a change in the
machine (changing the senses of some of
the words) we are frequently able to take in
some of these not possible things and make
them possible. But there are certain
limits to all language machines, certain
ways of working which are necessary (17-12)
for all of them, however different the details
of their working may be. For example,
in all language machines, A is not not-A.
If a sense is A it is not possible for it to be
not-A. If we do not keep this rule the
working of the machine goes wrong everywhere.
In the chief sense of possible, the
limit between what is possible and what is
not possible is put by these rules, which
are necessary for all language machines
But we have not at present enough knowledge
about the senses used in our language
machines (if, for example, some words have
one sense or a number of others which we
have not at present given attention to)
to be certain where these limits come,
so a good working sense of possible is
18-2. That is possible which the rules of our
language machine seem to let as say. Other senses are common :--
18-3. That is possible which is in
agreement with our knowledge of the way things
are, of fact (4-2).
In this sense it is possible for a man to be
running at 20 miles an hour, but not at
30 miles an hour. We are certain of
this because we take the conditions under
which he is running as fixed. But with a
change in these conditions (running on the
moon, for example) we would not be certain.
It might be possible for him to take
a run at 30 miles an hour then. So we
18-31. That is possible which, with a
change of conditions, would not be out of
agreement with our knowledge. We are still certain that with no change
of conditions would he be able to take a
run at 10,000 miles an hour. To do so he
would have to become some different
thing ; he would no longer be a man, or no
longer be running.
With this we come to probable and some
of the most troubling questions in the
theory of science. We have seen (p. 100)
that it is not hard to make true general
statements if we do not put anything
special into them. 'Something will take
place now' was the example And
equally it is not hard to make completely
special true statements : 'This is quite so'.
We may take these two statements as
necessary -- made true by the working of
the language-machine and such that it is
not possible for them to be false. Taking
them so and putting them as the limits of
the completely general and the completely
special, all statements between these
limits (and all knowledge) are only
probable more or less. To give an account
of probable we have to go back to the
conditions of thought and knowledge.
In knowledge (5-1) effects take place
in our minds and their causes are what we
have knowledge of. This is cause in
sense 9-5. The general form of the
conditions of knowledge (in 5-1) is the
same as the general form of any other
conditions. And the way in which knowledge
takes place is the same as the way in
which any other effect takes place. See
the picture after 9-1. The questions
'What is a thought about ?' and 'How is
it true ?' (in the senses which go with 5-1)
are the same questions in different words.
The questions 'How probable is it that the
thought is true?' or 'that it is knowledge?'
or 'that what it is of is a fact?'
are again the same. The use of the word
probable, as if a quite new question was
coming in is the cause of most of our
trouble here. Generally (14.2) the way
we have of taking up the same question
in new words without seeing that the
question is the same, and the way we have
of not seeing when the question is changed
without the words being changed are
the two chief causes of our errors. It is
much harder to keep these changes in
control than to give a right answer, in
agreement with our knowledge, to any
Any picture of cause and effect is (for
this sense of knowledge) a picture of
knowledge. But our use of other senses
of knowledge and other ways of putting
questions about it makes this hard for us
to see. What we have to do here, is to
put away other theories for a time, and go
to the roots of our everyday ways of saying
things about our knowledge.
Take a simple example first.
The noise of a bell comes to your ears.
Something then takes place in your mind --
a thought of the bell. Here the bell is the
cause of your thought of it in a clear way.
(You may have in place of this, a thought
of the noise of the bell not of the bell, or
the noise may come to you through the
radio.) Take note here that your thought
Now take a more complex example.
Someone says the word bell to you or you
see the word printed on this paper. Again
a thought of a bell takes place in your
mind. The effect of the word in your
mind is the same in some ways as the
effect of the bell itself would have been.
The word is a sign or representative of the
bell. How this comes about is a question
to which we have at present only parts of
the answer. A great amount of work is
being done upon it, and it may be hoped
that we will before very long have much
more knowledge about it.
To say that the word is a sign of the bell
is to say that when the effect of the word
comes into the mind, it is joined by chains
of events which come from bells. The
special property of the mind is that in it
these joining of chains of events some
times take place. The word bell has in the
past come to our ears together with the
noises of bells, the seeing of bells, the
touching of bells and so on. This coming
together of a number of events in the mind
makes it possible for the mind to go
through effects (have thoughts) the cause
of which is not in fact present. The other
cause, here the word bell, joined with it in
the past, takes its place and so is its
representative. Other things than words
are representatives in this way, though
words are our most ready examples. Here
take note again that your thought is true.
There are or have been bells, that is a
condition of your being able to have
thoughts about them. So, because your
thought says no more than this, it is true.
This joining together of causes, so that
in the future one of them becomes the
representative of the other when the other
is not present, only takes place under
special conditions of attention and interest
-- about the detail of which our knowledge
(5-12) is small. It only takes place sometimes.
If this were not so, we would at all
times be having all thoughts -- which
would be as bad as not having any at all.
In addition there are certain regular forms
under which causes are joined together.
The seeing of a bell is joined with the
seeing of the word bell ; but, with the seeing
of a bell, has in the past been joined the
touching of it, the touching of other things,
the hearing of it and of other things all
through the history of our minds and
back, probably, far into the early history
of man, and the animals from which he
came. So there is a fixed order in the
ways in which causes of thoughts are
joined together and this order is what we
get in our picture of the way things are.
A quite small change in our interests may
be able to make this picture very different,
and there is some reason for the view that
the picture is being changed quickly at
Now take an even more complex example.
You see the words ' All bells are made of
metal '. This group of words has an
effect in your mind which is a representative
of the effects of :--
(2) Metal things
(3) Anything made of something
(4) Any groups with which you have
had to take note of all of the things
All these are joined together in your
mind in a way which you see, without
trouble, is very different from, for
example, the way of 'Some things are
metal bells ' or ' All metal things are bells '
or ' Some bells are made of metal '. Your
thought has a form, the effects the words
have (effects which are representatives of
other causes than the words-of bells,
made-of-metal things, and groups of alls)
come together in a certain way. Now, if in
fact (4-4) all bells are made of metal, if,
that is to say, these causes are together in
the way in which the form of your thought
takes them as being, then your thought is
true. Here, in fact, it is false : in China
and Japan some bells are made of wood !
You have a way out. You may say
that you were not taking into account any
wood things with which persons may make
noises. That for you a bell was a bell and
a metal thing, that these wood things are
only given the name bell because the right
name was not at hand. In your sense of
bell, ' All bells are made of metal ' is true
because anything not made of metal would
not be a bell.
This is a very frequent position in
arguments. The important question then
is : 'Is your statement about the senses
you are giving to your words or is it
about the things you are having thoughts
of ?' If it is taken to be about the senses
of words, then it might give us knowledge
and may be true or false. If if is taken
to be about things, then it gives us no
knowledge (the senses being fixed) and so
is not true and not false. It only says
again what has been said before with the
word Bell. (See in Comparison pp.101-2,
'Poetry is the cause in us of the effect
caused by poetry because it is Poetry ')
All thought is sorting. We have thoughts
of things only as being things of a sort.
This is another way of Saying that
thoughts are of their causes (9-3). When
things have been sorted in one way with
bells (as made of metal), if we go on sorting
them in the same way only, by saying
' All bells are made of metal ' we are saying
nothing. To say something we have to
put them into a new sort. So, if we take
bel1sto be instruments used for making
certain noises for certain purposes in
certain ways, and then say ' All bells are
made of metal ', we are saying something
because we are sorting them in a new way --
we are putting these things into the sort
of things made of metal. In so doing we
come into danger. What we say may not
Every statement not dependent only
upon the working of the language machine
(as with ' All bells (metal) are made of
metal ') is in this danger, it may not be
true. We may put this by saying that
every statement is only probable in a high
or low degree. This is again only a way
of saying that the laws of cause and effect
-- thoughts coming under their control --
are not certain. Probable is the word
used when we are attempting to give a
measure of the degree in which thoughts
may safely be taken as true. Even the
best tested laws of science, when they
have been made separate from the rules of
the language machine of science, are only
probable in a high degree. And those of our
thoughts which have not been taken into
science and tested, are only probable in a
degree which by comparison will seem very
low. Most of them are general in senses
14-11 and 14-2, not in l4-1, which gives
them a better chance of being true --
because any one of a number of different
facts will make them true. But even so
the space and time between the thoughts
and their causes (what they are of) and
the complex way in which the effects of
different things are joined in them makes
error probable. Error is the natural
condition of man.
This is the reason why the man of
science, whenever he is able to do so, puts
man-made instruments, whose tendencies
to error he is able to keep in control, in the
place of his mind. The camera for the
eye, the balance for his muscles, and so on.
And why, in addition, he makes use of
mathematics -- much better language
machine than everyday words because the
discovery and control of its rules is not so
hard. With the use of mathematics, the
language machine for measuring, comes the
tendency to put questions about degrees
in place of questions about sorts.
This division between degrees and sorts
is not as important as it has seemed to
some writers. It is parallel to one between
amounts and qualities. Science, it is
sometimes said, is only able to give
accounts of those things which may be
measured and says nothing about qualities.
This is only true of those sciences which
have been able to make statements about
amounts do the work which in every-day
language is done by statements about
qualities. When this is not, at present,
possible science makes use of qualities.
Science does not put amounts in place of
qualities because it has a desire (from a
hate of living things and so on) to take
qualities out of thought, but because
statements about amounts may be made to do
the work of thought better `1n some ways,
though possibly not in others, Amounts
are relations which are not so hard to put
in order and keep in control as qualities.
In the same way, degrees are a network of
sorts which may be measured. A degree
is a fixed amount by which examples in a
general sort may be different from one
another. In a general sort we put special
sorts which are different by equal amounts.
The question ' Are A and B different in
degree or in sort ?' is common at certain
turning-points of discussion. Put in this
way it frequently seems a very deep
question to which no certain answer may
be given. For example, ' Is man different
in sort from the animals or only in degree ?'
Are the animals different in sort or degree
from the plants, and the plants from the
earth, red from blue, 'Communism ' from
'Capitalism ', seeing from hearing . . . and
so on ? Some persons have a feeling that
an answer one way will be very different
In its effect from an answer the other way.
These questions seem hard and deep and
important because we have not first taken
a decision as to what we are talking about.
And we have not kept in view that a
degree is one division of a sort -- a division
by which comparisons between examples
of the sort may be made. To make such
division in a sort we have first to come to a
decision as to which the sort is. A sort
is a group of things with some property
in common. And only when we have
this property clearly in view are we able
to go on to attempts to make divisions of
amounts (degrees) of the property. These
attempts are dependent upon our power of
measuring and there are a great number
of things which, at present, we have no
power of measuring. Which is not to say
that we will not ever have it.
To give an answer to any question of the
form ' Are A and B different in sort or in
degree ?' it is necessary to have a clear
idea of the properties which A and B have
in common -- because it is not possible to
put two things in comparison in any way
if they have no property in common.
(See in comparison change p. 83). Then
we have to see if we have any way of
measuring some property which they have
in common. If we have no way, we have
to say that, as far as our knowledge goes,
they are different in sort and not only in
degree. To say this is to say that we have
not got the power, at present. We have
such need of more powers in this troubled
existence that this would, if we were wiser,
seem a sad thing to have to say. But
numbers of persons take great pleasure in
saying, whenever possible, that things are
different in sort and not only in degree.
It is as if they had the thought that to say
this makes things more free, that when a
thing may be measured it is of less value.
They seem to have the idea that more
power over things would make our condition
worse, not better ! The less foolish
are probably in fear of the use to which
other persons might put greater powers.
The more foolish are in fear of having their
thoughts about things put in better order.
The end of this discussion is near, but
the reader may be waiting for an answer to
a question which has been in his mind
from the first page. What is this Agreement
on which almost everything in this
apparatus of divisions seems to be
dependent ? It came at the start in the
account given of our purpose , it came
again in the senses of thought, fiction, fad,
knowledge, true, sense, and in the senses of
of and about. It came, but not so openly,
in the chief senses of cause and law. It is
at the back of all our discussion of change,
same, property, general, necessary, possible
and probable. No other word seems so
important, but no special discussion of it
has been attempted till now in these pages.
We have seen -- with true, with sense ;,
with property, general and sort ; with
possible and probable -- how the same
questions may be put again and again in
different words ; how a way, a form, a
sort, a group, a property are all ways
(sorts, forms, groups, properties) of things ;
and how a thing may be a law, and a law
again a way. The words with which
discussion goes on are more in number --
though every word has its group of senses
-- than the senses they are used to put in
order. And at more than one place the
trouble and danger to thought which come
from our way of taking an answered question
as a new one have been pointed out.
Is this question ' What is Agreement ?'
only the other questions ' What is a sort ?'
'What is a way of being the same ?'
'What is a general property ?' 'How are
thoughts true ?' 'What is knowledge ?'
and 'What is a cause ?' put in another
form ? As questions -- as forms of words
to which, when senses for them have been
fixed, answers may be given -- these are
or may be made, clearly different. But the
fact (4-2) about which we put them seems
to be one fact -- a very complex fact, of
which a number of views, of parts of it,
may be taken. The part which may not
be clear -- for which a separate account of
the senses of Agreement might be a help --
is covered by senses of the words general,
property and cause. It is possible and
not hard, to give a list of them by using
these words. But then, someone might
say 'Ah!I you are saying what Agreement
is by using cause and general, and you said
what cause and general are by using
Agreement ! You are moving in a circle
and your account of these things is only a
trick !' If, on the other hand, I took
some new words, say X and Y, with
which to give an account of the Agreement
which has been used in talking about
knowledge, then someone would say,
'Ah I he has given no account of X and Y,
the senses upon which everything in his
system is dependent ; so it is not complete
and has no base !' These two protests
would equally be signs that the purpose
of these pages has not been rightly taken.
As was said at the start, this apparatus of
senses is to be tested by the help it gives
us in putting our thoughts in order, In
letting us say what we have a need to say
and keeping us from saying other things
which will get in the way of our purposes.
If it is a help, that help is its base. What
the purposes are for which the machine
may be a help is only made clear by the
range of its uses. We are able to give an
account of a purpose only by saying in
detail what it is a purpose to do. A purpose,
in this sense, is not something different
from the way in which it may be worked
There are other purposes than those
whose limits are put by the language
machine of which a rough account has
been given in these pages. At certain
points suggestions of these other purposes
have been put in the form of possible
senses for some of the words (see fact 4-3,
knowledge 5-3, true 7-21) senses of which
I have made little use in the later pages.
With a farther working out of these
purposes -- and of the purposes of this
language machine -- it might be possible
to see more clearly into their connections.
It is the hope of everyone in whom thought
about thought and its ways is a strong and
frequent interest that by making one
language-machine more complete, more
clear, more delicate, room will be made for
the senses and divisions of the other
language-machines which are the servants
of other purposes. Every system of
thought makes some attempt to take into
account other and different systems.
Its power to do this is limited by
the errors in itself -- in addition to the
errors in them. This system here, this
apparatus of senses, has in it a great
number of errors. Some of them I see
and would put right if, in my attempts at
present to do so, new errors did not come
in their place. It is printed because
anything which may be of help, however
little, may in these years be a step to
A last word. If you do not clearly get the sense of what I have said you will, in making the attempt to get it, have gone through a great number of the possible senses of these words in your mind. It is not very important that you get my sense and no others, and not important if what I say is true or not. And it is not important for anyone to have any belief that what has been said here is so. What is important is to see that the sense of words may be taken in groups, and that if the form of one group of senses becomes clear to us, the form of other groups of senses, which we may not ever have put in connection with them, may become clear at the same time. This gives us new chances for the control of our thought and for taking over the knowledge we have of one field into other fields. As Colegidge said, ' that only is learning which comes again as power '. And to see how any sense is in relation to any other is to get a sort of learning which comes again as power.