You are on page 1of 133

Commentary on Wittgenstein's

Philosophical Investigations
by Lois Shawver

Shawver Commentary
One of the most difficult or misleading aspects of Wittgenstein's
Philosophical Investigations is the way in which he uses multiple
voices to converse with himself. To have a sense of understanding
Wittgenstein you need to be able to hear these different voices.The
Philosophical Investigations is written in aphorisms short numbered
passages that are loosely tied together in terms of theme. !e often
begins an aphorism with a "uoted passage. #or e$ample he begins the
first aphorism with a "uotation from %ugustine.
&ost "uoted passages are not actual "uotes however but rather
Wittgenstein's construction of a 'ind of interlocutor. This interlocutor
might be thought of in terms of %ugustine Plato characters in Plato's
dialogues (ertrand )ussell or even early Wittgenstein or perhaps *ust
a vague composite of these various figures. %t any rate this voice +and
it is not always in "uotes, represents the problem that Wittgenstein tries
to thin' through. I will call this voice whatever seems most
appropriate to the passage such as the voice of %ugustine early
Wittgenstein but the label I use is somewhat arbitrary in most
instances. What is important is that you notice that this is the voice that
provides the conte$t for Wittgenstein's response.
In addition to the interlocutor it is useful to thin' of there being two
additional voices. One is the voice that discovers perple$ities or
aporia. This voice is often but not always introduced with a dash and
it often but again not always begins with the word -(ut-. I will often
call this the voice of aporia.
Then there is a third voice in which Wittgenstein ma'es an incisive
point in the face of the tradition and aporia. .ou might thin' of this as
the -voice of clarity.-
The basic format then is/

voice of the
0verything has an essence.-
voice of aporia (ut is this true1
voice of clarity
It seems that this notion has been a
Of course this greatly simplifies the content of what Wittgenstein is
saying and not every passage has "uite this form. (ut if you loo' for
these different voices it should assist you ma'ing sense of what you
find in these pages.
I suggest that you never presume that these voices are all there in any
given passage. !e sometimes introduces for e$ample a thought
e$periment that he calls language games and in those cases it does not
ma'e much sense to spea' of these three voices. (ut you might
e$amine a passage to see if thin'ing of it in terms of these voices helps
that passage ma'e sense to you. If it does then you're probably right in
presuming that the passage in "uestion adopts this standard format.
%nd for my part when I see this format being used I will often call
your attention to it referring to it at times as -LW's standard format.-
%phorism 2324 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance
Shawver commentary:
2. -When they +my elders,
named some ob*ect and
accordingly moved towards
something I saw this and I
grasped that that the thing was
called by the sound they uttered
when they meant to point it out.
Their intention was shown by
their bodily movements as it
were the natural language of all
peoples5 the e$pression of the
face the play of the eyes the
movement of other parts of the
body and the tone of the voice
which e$presses our state of
mind in see'ing having
re*ecting or avoiding
something. Thus as I heard
words repeatedly used in their
This is a "uotation that
Wittgensteinn has ta'en from
%ugustine +Confessions I.6.,.
7isuali8e %ugustine's picture of
how language is learned and
notice how natural and complete
it sounds as a total e$planation
for how language is learned.

proper places in various
sentences I gradually learnt to
understand what ob*ects they
signified5 and after I had trained
my mouth to form these signs I
used them to e$press my own
These words it seems to me
give us a particular picture of the
essence of human language. It is
this/ the individual words in
language name objects--
sentences are combinations of
such names33In this picture of
language we find the roots of the
following idea/ !very word has
a meaning "he meaning is
correlated with the word It is
the object for which the word
9ow Wittgenstein is beginning
his commentary. The emphasis
is mine. It is the deconstruction
of %ugustine's picture of
language that is the focus of this
entire boo'. +%lthough I should
say that many others beside
%ugustine have shared this
picture of language. %s we will
see it is a cultural illusion,
Once deconstucted new and
stri'ingly different ideas about
language begin to emerge.
%ugustine does not spea' of
there being any difference
between 'inds of word. If you
describe the learning of language
in this way you are I believe
thin'ing primarily of nouns li'e
'table' 'chair' 'bread' and of
people's names and only
secondarily of the names of
certain actions and properties5
and of the remaining 'inds of
word as something that will ta'e
care of itself.
!ere the deconstruction begins.
Loo'ing at the %ugustinian
picture of language we see that
%ugustine has e$plained only
one type of word.

9ow thin' of the following use
of language/ I send someone
shopping. I give him a slip
mar'ed 'five red apples'. !e
ta'es the slip to the shop'eeper
who opens the drawer mar'ed
'apples' then he loo's up the
word 'red' in a table and finds a
colour sample opposite it5 then
he says the series of cardinal
This scenario is a thought
e$periment. To what e$tent do
you thin' the language in this
scenario is e$plained by
%ugustine's picture of language1
Thin' of the shop'eeper
counting out the apples one
through five. :id he learn to do
this by someone pointing to five
apples1 !ardly. The teaching of
numbers33I assume that he
'nows them by heart33up to the
word 'five' and for each number
he ta'es an apple of the same
colour as the sample out of the
drawer.33It is in this and simlar
ways that one operates with
words33-(ut how does he 'now
where and how he is to loo' up
the word 'red' and what he is to
do with the word 'five'1- 333Well
I assume that he 'acts' as I have
described. 0$planations come to
an end somewhere.33(ut what is
the meaning of the word 'five'1
339o such thing was in "uestion
here only how the word 'five' is
language by pointing cannot
e$plain learning to count. What
about using written languge to
communicate what is wanted1
Someone had to teach him how
to read before he could ma'e
sense of the note and translate it
into a order. %nd to follow the
order he had to 'now much
more than was specifically
contained in the note 3 which *ust
said 'five red apples.' The
shop'eeper had to be able to find
the apples even to 'now to loo'
for them and also to 'now to
put them in a sac' and accept
money in e$change for them. !e
had to be able to recogni8e
various coins our bills and add
them together. It would be hard
to e$plain all of this within the
%ugustinian picture of language.

;. That philosophical concept of
meaning has its place in a
primitive idea of the way
language functions. (ut one can
also say that it is the idea of a
language more primitive than

(y -that philosophical concept
of meaning- Wittgenstein means
the %ugustinian picture that he
gave us above. Loo' at
%ugustine's picture again/

"he individual words in
language name objects--
sentences are combinations
of such names !very
word has a meaning "he
meaning is correlated with
the word It is the object
for which the word stands
This concept of meaning
Wittgensein says has its place in
helping us understand primitive
language language more
primitive than 0nglish <erman
#rench etc. It is also the case
Wittgenstein e$plains that there
are regions of our developed
language in which language
wor's *ust as %ugustine portrays
Let us imagine a language ...The
language is meant to serve for
communication between a
builder % and an assistant (. % is
building with building3stones5
there are bloc's pillars slabs
and beams. ( has to pass the
stones and that in the order in
which % needs them. #or this
purpose they use a language
consisting of the words 'bloc''
'pillar' 'slab' 'beam'. % calls
them out5 33( brings the stone
which he has learnt to bring at
such3and3such a call. 33 =onceive
this as a complete primitive

This is an important thought
e$periment. %lthough he does
not call it a language3game in
this passage it will become clear
shortly that this passage
describes the prototypic
primitive language3game. !e
will refer to it often sometimes
in its present form or in one of a
multitude of variations he will
give us shortly.
We will often refer to this as
language game +;, using the
number of the aphorism to inde$
the number of the language
game. I picture a wor'
supervisor at the front of a site
with a wor'er responding to the
supervisor's commands. There
are piles of pillars slabs bloc's
and beams. The supervisor calls
out -Slab>- and the wor'er
brings a slab and sets it at the
supervisor's feet. Pretty simple.
Wittgnstein puts forth language3
game +;, in order to try to
envision a language in which
%ugustine's picture of language
:oes %ugustine's picture of
language wor' here1 !ow did
the wor'er learn this language
by teachers pointing and naming
the slabs and beams as
%ugustine suggested1 %n
e$ercise li'e %ugustine suggests
might e$plain how the wor'er
'new which ob*ect to fetch but
how did the wor'er learn to
fetch1 %s opposed say to
ta'ing ob*ects behind the fence1
=rushing them1 Or tapping them
with a stone1

?. %ugustine we might say does
describe a system of communication5
only not everything that we call
language is this system. %nd one has to
say this in many cases where the
"uestion arises 'Is this an appropriate
description or not1' The answer is/
'.es it is appropriate but only for this
narrowly circumscribed region not for
the whole of what you were claiming to
It is as if someone were to say/ -%
game consists in moving ob*ects about
on a surface according to certain
rules...- 33and we replied/ .ou seem to
be thin'ing of board games but there
are others. .ou can ma'e your
definition correct by e$pressly
restricting it to those games.
Somehow %ugustine's
picture of language
although appropriate for a
subsection of langauge is
not as all inclusive an
e$planation of language as
we are at first glance
inclined to believe.

%s Wittgenstein says in
+2, we tend to sweep
under the rug all the uses
of language that do not fit
the %ugustinian picture
that seems to capture our
%lthough language3game
+;, restricts the
vocabulary to words that
seem to refer to ob*ects
the %ugustinian picture
cannot e$plain everything
that happens.

@. Imagine a script in which the
letters were used to stand for
sounds and also as signs of
emphasis and punctuation. +%
script can be conceived as a
language for describing sound3
patterns., 9ow imagine someone
interpreting that script as if there
were simple a correspondence of
!ow might this be1 Suppose
we taught a parrot to say -Polly
wants a crac'er- and whenever
it says it we gave the parrot a
crac'er. On the surface this
loo's li'e language. The parrot
is as'ing for and receiving a
crac'er. !owever on closer
e$amination it is not. We could
letters to sounds and as if
the letters had not also
completely different functions.
%ugustine' conception of
language is li'e such an over3
simple conception of the script.

have taught the parrot to say
-<et lost>- and give it a crac'er
each time it does. Then it
would not have loo'ed as
though the parrot were spea'ing
To thin' that simply saying the
words -Polly wants a crac'er-
constitutes -language- is to have
this sort of over3simple
conception of the language.
Something profound is missing
from this conception although it
is not yet clear e$actly what this
is. Still it is a beginning to say
that when the parrot says -Polly
wants an crac'er- he doesn't
"uite 'now what this sentence
means in 0nglish. It amuses us
because nevertheless it seems
as though he does.
The same would be true if we
taught a two year old to answer
the "uestion -What is @A4
divided by ?BB1- by saying
-One point two three.- It would
be a correct answer in 0nglish
but the child would not 'now
what she was saying because
she would not 'now how to
count 'now wha this number
means or 'now what division
means. There is more to
language than stringing together
correct words.
A. If we loo' at the e$ample in +2, we
may perhaps get an in'ling how much
this general notion of the meaning of a
word surrounds the wor'ing of language
with a ha8e which ma'es clear vision
impossible. It disperses the fog to study
the phenomena of language in primitive
'inds of application in which one can
(ut although the
parroted sentences are
not language in the
richest sense of the term
they help us to
understand how
language begins the
roots of language.
command a clear view of the aim and
functioning of the words.
% child uses such primitive forms of
language when it learns to tal'. !ere the
teaching of
language is not e$planation but training.

B. We could imagine that the
language of +;, was the whole
language of % and (5 even the
whole language of a tribe. The
children are brought up to
perform these actions to use
these words as they do so and to
react in this way to the words of
others. %n important part of
the training will consist in the
teacher's pointing to the ob*ects
directing the child's attention to
them and at the same time
uttering a word5 for instance the
word -slab- as he points to that

%lthough the word -slab>- is not
tied to any particular activity in
0nglish in the language we are
imagining in +;, it is always a
command to fetch a slab. What
tends to confuse us is that we
can imagine something li'e this
ta'ing place in 0nglish. It is
*ust that the word -slab>- would
not be confined to only this use.
!owever in the community we
are imagining this is the only
use for the term -slab>- %nd
how might children be taught
the use of the term1 We can
well imagine that the
%ugustinian picture of language
training might be involved. The
child's attention will be directed
to the different shapes and the
child will learn to e$pect each
shape to be associated with a
particular sound.

+ I do not want to call this
-ostensive definition- because
the child cannot as yet as' what
the name is. I will call it
-ostensive teaching of
words-.33333I say that it will form
an important part of the training
because it is so with human
beings5 not because it could not
What is the difference between
ostensive teaching of words and
ostensive definitions1 In
ostensive definitions someone
points and gives a name of
something and this serves to
ma'e clear how the term is to be
used. When someone points to
a crac'er and says -crac'er-
be imagine otherwise.,

those who 'now what a crac'er
is +but not the name for it, can
receive this as an ostensive
definition. (ut if a child has not
yet learned language it is li'e
the parrot. It does not 'now
what is being pointed to on what
the word crac'er means.
+&aybe the word -crac'er-
means -s"uare- or -salty-. Or
maybe it means -food-.,
!owever the child understands
the term the child can be taught
to say it in assocition with the
ob*ect. %s %ugustine imagined
things in +2, . %s %ugustine
imagined things the child
without any language was able
to -grasp-
This ostensive teaching of words
can be said to establish an
association between the word and
the thing. (ut what does this
mean1 Well it can mean various
things/ but one very li#ely
thin#s first of all that a picture
of the object comes before the
child's mind when it hears the
word $ut now% if this does
happen---is it the purpose of
the word&
The emphasis here is mine. I
want to show what I will call
Wittgenstein's aporetic voice.
!e is reminding us of the
cultural ways we thin' so tht he
can deconstruct them. !ere
Wittgenstein is tal'ing about the
cultural illusion that is related to
%ugustine's picture of language
and what we are li'ely to say
that supports this illusion. it can be the purpose.333I
can imagine such a use of words
+of series of sounds,. +Cttering a
word is li'e stri'ing a note on the
'eyboard of the imagination., (ut
in the language of +;, it is not the
purpose of the words to evo'e
images. +It may of course be
discovered that that helps to
attain the actual purpose.,
(ut although language may
create images for us remember
the language in +;, was not
re"uired to create images for the
wor'ers. The wor'er in +;,
would understand what was
being said to him if he simply
fetched what was called for
whether or not he had images of
what called for when it was
called or not.
(ut if the ostensive teaching In +;, one understands the call
has this effect 333am I to say that
it effects an understanding of the
word1 :on't you understand the
call -Slab>- if you act upon it in
such3and3such a way1 33
:oubtless the ostensive teaching
helped to bring this about5 but
only together with a particular
training. With different training
the same ostensive teaching of
these words would have effected
a "uite different understanding.
-Slab>- if one brings it when it
is called. Pointing to slabli'e
ob*ects and saying -slab- might
have faciliated this teaching but
one could also imagine learning
to ta'e the slab behind the fence
when it is called. % different
training would have resulted in
the wor'er doing different
things with the slab hitting it
hiding it burying it and so
-I set the bra'e up by
connecting up rod and lever.-333
.es given the whole of the rest
of the
mechanism. Only in con*unction
with that is it a bra'e3lever and
separated from its support it is
not even a lever5 it may be
anything or nothing.
Cnless one 'nows how to weave
the word into some form of
human activity the saying of the
word is not yet language. It is
li'e a brea' that is not yet
connected with the entire
mechanism. The parts seem to
be there but it does not yet have
the connections to function as it

D. In the practice of the use of
language +;, one party calls out
the words the other acts on
them. In instruction in the
language the following process
will occur/ the learner names the
ob*ects5 that is he utters the
word when the teacher points to
the stone.333%nd there will be
this still simpler e$ercise/ the
pupil repeats the words after the
teacher33333both of these being
processes resembling language.
%ll of this sounds li'e
%ugustine's picture of learning

We can also thin' of the whole
process of using words in +;, as
one of those games by means of
which children learn their native
language. I will call these games
-language3games- and will
!ere Wittgensein introduces the
concept of a language game but
he will amplify this concept later
so that it does not merely apply to
language learning e$ercises. To
anticipate this amplification of
sometimes spea' of a primitive
language as a language3game.

the meaning of this term we
might sometimes distinguish this
meaning of the term by calling
these language games -primitive
language games.-
%nd the processes of naming
the stones and of repeating
words after someone might also
called language3games. Thin' of
much of the use words in games
li'e ring3a3ring3a3roses.
In ring3a3ring3a3roses the child
learns the phrases without
'nowing what they mean as a
parrot might learn to say -Polly
wants a crac'er.-
I shall also call the whole%
consisting of language and the
actions into which it is woven%
the 'language-game'
So -the language game- is not
merely speech. In +;, he whole
activity of fetching the ob*ects
was part of the -language game-
of +;,.
6. Let us now loo' at an
e$pansion of language +;,.
(esides the four words -bloc'-
-pillar- etc. let it contain a
series of words used as the
shop'eeper in +2, used the
numerals +it can be the series of
letters of the alphabet,5 further
let there be two words which
may as well be -there- and
-this- +because this roughly
indicates their purpose,that are
used in conne$ion with a
pointing gesture5 and finally a
number of colour samples. %
gives an order li'e/ -d333slab333
there-. %t the same time he
shews the assistant a colour
sample and when he says
-there- he points to a place on
the building site. #rom the
stoc' of slabs ( ta'es one for
each letter of the alphabet up to
-d- of the same colour as the
sample and brings them to the
place indicated by %.333On other
In +6, LW creates a new language
game that is a variation of +;,.
9ow we will be able to spea' of
bringing E number of slabs and
we will be abe to indicate where
we want the slab to be put. We
understand these concepts LW
e$plains because they e$ist in
0nglish. 9otice however that
LW does not say that the slabs
will be counted with numbers
but with the letters of the
alphabet. This helps us get into
the feel of what it would be li'e
if we had a more primitive
system of counting one in which
there was no arithemetic
possisilibities for e$ample.

occasions % gives the order
-this333there-. %t -this- he points
to a building stone. %nd so on.
F. When a child learns
this language it has to
learn the series of
'numerals' a b c ... by
heart. %nd it has to learn
their use.333Will this
training include ostensive
teaching of the words1333
Well people will for
e$ample point to slabs
and count/ -a b c
slabs-.333Something more
li'e the ostensive
teaching of the words
-bloc'- -pillar- etc.
would be the ostensive
teaching of numerals that
serve not to count but to
refer to groups of ob*ects
that can be ta'en in at a
glance. =hildren do learn
the use of the first or si$
cardinal numerals in this

!ow can we imagine the people of +6,
learning language1 =an they learn it
ostensively as %ugustine imagined1
Ta'e the learning of numbers. We could
imagine them learning to distinguish
numbers ostensively as we might learn
to distinguish two from three by
distinguishing these configurations of
two and three/
o o o o
(ut this would be of limited use. We
cannot learn to distinguish apparently
much larger numbers in this fashion.
Thus we count.
%re -there- and -this
also taught
how one might perhaps
teach their use. One will
point to places and
things333but in this case
the pointing occurs in the
use of the words too and
not merely in learning the
!ow will -there- and -this- be taught1
This is tric'y and LW does not answer
the "uestion for us. :o you point to
-this- and say -this-1 :oes that clarify
the use of the word -this-1 !ardly.
24. 9ow what do the words of
this language signify1333What is
supposed to shew what they
signify% if not the #ind of use
What does -two signify-1 :oes it
signify any two ob*ects1 Say two
bloc's1 Well we 'now what the
word -bloc' signifies.- It signifies
they have& %nd we have already
described that. So we are as'ing
for the e$pression -This word
signifies this- to be made a part of
the description. In other words the
description ought to ta'e the form/
-The word . . . .signifies . . . .-

each of the two bloc's. :oes
-two- signifiy something other
than what -bloc' signifies-1 There
are conceptual pu88les here.
%nd what does -this- signify. It
signifies what I point to. (ut that
can be anything. !ow can a child
learn to associate the naming of
anything by one term1
(ut do we need to say what these
words -signify-1 Isn't everything
clear already1 Since we 'now
their use1 Why would we re"uire
that all words -signify-1
Of course one can reduce the
description of the use of the word
-slab- to the statement that this
word signifies this ob*ect. This
will be done when for e$ample it
is merely a matter of removing the
mista'en idea that the word -slab-
refers to the shape of building3
stone that we in fact call a
-bloc'-333but the 'ind of 'refering'
this is that is to say the use of
these words for the rest is already
In language3game +;, pointing and
saying -slab- may be helpful to
show which slab is to be fetched
but pointing and naming would not
show that the slab is to be fetched.

0"ually one can say that the
signs -a- -b- etc. signify
numbers5 when for e$ample this
the mista'en idea that -a- -b-
-c- play the part actually played
in language by -bloc'- -slab-
-pillar-. %nd one can also say that
-c- means this number and not that
one5 when for e$ample this
serves to e$plain that the letters
are to be used in the order a b c
d etc. and not in the order a b d
In other words we might want to
e$plain that -c- is not *ust another
ob*ect li'e -slab- or -bloc'- and so
we might need e$plain -a- -b-
and -c- signify numbers. (ut
where does this leave us1 :oes it
teach the child in +6, to learn to use
numbers +by counting things, and
until the child learns to count does
the child really 'now what
-numbers- means1
(ut assimilating the
descriptions of the uses of the
words in this way cannot ma'e the
So although we can find a way to
say that -a- -b- -c- signify
something assimilating these
themselves any more li'e one
another. #or as we see they are
absolutely unli'e.

different 'inds of words to the
same e$pression +they are
instances if -signifying- hides the
enormity of the difference and
creates a over simplified picture
language and how language is
%phorism 223;4 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance
Shawver commentary:
22. Thin' of the tools in a tool3
bo$/ there is a hammer pliers a
saw a screw3driver a ruler a
glue3pot glue nails and
screw.333The functions of words
are as diverse as the functions
of these
ob*ects. +%nd in both cases
there are similarities.,
%ugustine was struc' by the
similarities of different words and
failed to note their differences.
Such an understanding would be
as superficial as learning that all
the ob*ects in the toolbo$ were
-tools- but not 'nowing any of
their different functions.
Of course what confuses us
is the uniform appearance of
words when we hear them
spo'en or meet them in script
and print. #or their application
is not presented to us so clearly.
0specially when we are doing
Loo' at the words on this page.
:on't they loo' ali'e1 They loo'
so much more li'e each other
than they loo' li'e your 'eyboard
or your hand. This is what
confuses us.

2;. It is li'e loo'ing into the cabin of a
locomotive. We see handles all loo'ing
more or less ali'e. +9aturally since they
are all supposed to be handled., (ut one is
the handle of a cran' which can be moved
continuously +it regulates the opening of a
valve,5 another is the handle of a switch
which has only a bra'e3lever the harder
one pulls on it the harder it bra'es5 a
fourth the handle of a pump/ it has an
effect only so long as it is moved to and
We are mesmeri8ed by
the similarity in the
appearance of words.
This 'eeps us from
noticing the vast
differences in their

2?. When we say/ -0very word in language signifies something-
we have so far said nothing
whatever5 unless we have e$plained e$actly what distinction we wish
to ma'e. +It might be of course that we wanted to distinguish the
words of language +6, from words 'without meaning' such as occur in
Lewis =arroll's poems or words li'e -Lilliburlero- in songs.,

2@. Imagine someone's saying/
-%ll tools serve to modify
something. Thus the hammer
modifies the position of the nail
the saw the shape of the board
and so on.-333%nd what is
modified by the rule the glue3
pot the nails1333-Our 'nowledge
of thing's length the temperature
of the glue and the solidity
of the bo$.-33333Would anything
be gained by this assimilation of
It seems we loo' for ways to
disguise the differences in
different 'inds of terms. We try
to assimilate them all to a
particular way of describing
them. (ut the fact that we can
find an e$pression that treats
them all the same +e.g. all
words are made of characters,
does not mean that they are as
similar as we thin'. We fail to
notice their differences and this
undermines our philosophy
about language.

2A. The word -to signify- is
perhaps used in the most
straight3forward way when the
signified is mar'ed with the
sign. Suppose that the tools %
uses in building bear certain
mar's. When % shews his
assistant such a mar' he
brings the tool that has that
mar' on it.
It is in this and more or less
similar ways that a name
means and is given to a
thing.333It will often prove
useful in philosophy to say to
ourselves/ naming something
is li'e attaching a label to a
Well does the word -signify- mean
anything at all1 There is a
e$emplary case of our using this
term. It is used best when we mar'
ob*ects with a sign. Sometimes it
is useful to use such a model in
understanding language.


2B. What about the colour
samples that % shews to (/ are
they part of language1 Well it
is as you please. They do not
belong among the words5 yet
when I say to someone/
-Pronounce the word 'the' -
you will count the second -the-
as part of the language3game
+6,5 that is it is a sample of
what the other is meant to say.
It is most natural and causes
least confusion to rec'on the
samples among the instruments
the language.
++)emar' on the refle$ive
pronoun -this sentence-. 3
There is a certain analogy
between saying -This is the color
pillar I want you to bring- and
-This is the way I want you to
pronounce the word 'the.'- We
sometimes give samples of how to
say things or what to call things
with words and sometimes we
use supplementary techni"ues
such as color samples.
Wittgenstein is urging us to count
all of these techni"ues regardless
of whether they consist of words

2D. It will be possible to say/ In
language +6, we have different
'inds of word. #or the functions
of the word -slab- and the word
-bloc'- are more ali'e than those
of -slab- and -d-. (ut how we
group words into 'inds will
depend on the aim of the
classification333and on our own
Thin' of the different points of
view from which one can classify
tools or chess3men.
Treat this as an e$ercise. What
'ind of words are there in +6,.
The way to classify words in 6
will vary but one way that
suggests itself is we can count
some words as names some as
numbers and some as
pronouns. (ut couldn't we also
classify these words according
to whether they are one syllable
or two1 %ren't there other ways
to classify them1

26. :o not be troubled by the
fact that languages +;, and +6,
consist only of orders. If you
want to say that this shews them
to be incomplete as' yourself
whether our language is
complete5333whether it was so
before the symbolism of
chemistry and the notation of the
infinitesimal calculus were
incorporated in it5 for these are
so to spea' suburbs of our
language. +%nd how many
houses or streets does it ta'e
before a town begins to be a
town1, Our language can be seen
as an ancient city/ a
ma8e of little streets and s"uares
of old and new houses and of
houses with additions from
various periods5 and this
surrounded by a multitude of
new boroughs with straight
regular streets and uniform
%t what point does a language
become complete1 Was our
language complete before we
introduced the speciali8ed
language of psychoanalysis1
(efore we introduced the 8ero
into our counting system1 %nd
for that matter is our language
complete now1
We have no way to evaluate the
completeness of language. 0ach
language is more or less rich but
the ways that it is rich are
different from that in other

2F. It is easy to imagine a
language consisting only of orders
and reports in battle.333Or a
language consisting only of
"uestions and e$pressions for
answering yes and no. %nd
innumerable others.33333(nd to
imagine a language means to
imagine a form of life
Wittgenstein has already told
us that language games are not
not *ust to be -words- and our
ways of responding with
words. The language game in
+;, for e$ample was woven
into a culture that fetched slabs
and bloc's. Their words were
woven into their activity their
forms of life.
(ut what about this/ is the call
-Slab>- in e$ample +;, a sentence
or a word1333 If a word surely it
has not the same meaning as the
li'e3sounding word of our
ordinary language for in +;, it is a
call. (ut if a sentence it is surely
!ow can it be an elliptical
sentence1 There are no words
possible in language3game +;,
e$cept -slab- -bloc'- -pillar-
and -beam.-

not the elliptical sentence/ -Slab>-
of our language.
33333%s far as the first "uestion
goes you can call -Slab>- a word
and also a sentence5 perhaps it
could be appropriately called a
'degenerate sentence' +as one
spea's of a degenerate hyperbola,5
in fact it is our 'elliptical'
sentence.333(ut that is surely only
a shortened form of sentence
-(ring me a slab- and there is no
such sentence in e$ample +;,.333
$ut why should I not on
contrary have called the
sentence '$ring me a slab' a
lengthening of the sentence
0ven in 0nglish it is biased to
say that -Slab>- is an elliptical
form of -(ring me a slab.- If
we began by learning the
command -slab>- +and maybe
we did, then wouldn't -(ring
be slab>- be a lengthened form
of -Slab>-1

(ecause if you shout -Slab>- you
really mean/ -(ring me a slab-.333
!ere is LW's aporetic +or
%ugustinian voice,. Let's
unpac' what we mean by
-really mean.-
(ut how do you do this/ how do
you mean that while you say
-Slab>-1 :o you say the
unshortened sentence to yourself1
%nd why should I translate the call
-Slab>- into a different e$pression
in order to say what someone
means by it1 %nd if they mean the
same thing333why should I not say/
-When he says 'Slab>'-1 %gain if
you can mean -(ring me the slab-
why should you not be able to
mean -Slab>-1 33333(ut when I call
-Slab>- then what I want is that he
should bring me a slab>33333
=ertainly but does 'wanting this'
consist in thin'ing in some from
or other a different sentence from
the one you utter1333
%nd here are some
observations that are meant to
shed clarifying light/
!ow do you have this other
meaning -(ring me a slab>-
going on1 In what way is this
what we really mean1 We
don't say -(ring me a slab>- to
ourselves while we say
-Slab>- Why not say that
-(ring me a slab>- really
means -Slab>-
This notion -really mean- is
confusing here. We do not
-really mean- a particular
sentence in this case. Or we
might *ust as well say that we
really mean -slab>- as to say
that we really mean -(ring me
a slab>-

;4. (ut now it loo's as if
when someone says -(ring me a
slab- he could mean this
e$pression as one long word
corresponding to the single word
-Slab>- 3333Then can one mean it
sometimes as one word and
sometimes as four1 %nd can one
mean it sometimes as one word
and sometimes as four1 %nd
how does one usually mean
%nd when a person says -(ring
me a slab>- it is not the same as
if a peson said -bring3me3a3
slab>- as if it were *ust one
word. What is wrong with our
analysis here1
When is -(ring me a slab>- four
words and when is it one1
I thin' we shall be inclined to
say/ we mean the sentence as
four words when we use it in
contrast with other sentences
such as -!and me a slab-
-(ring him a slab-. -(ring two
slabs- etc.5 that is in contrast
with sentences containing the
separate words of our
command in other
When we have a variety of
sentences that use most of the
same words but are variations on
a theme then we will say that the
sentence has four words.

(ut what does using one
sentence in contrast with others
consist in1 :o the others
perhaps hover before one's
mind1 %ll of them1 %nd while
one is saying the one sentence
or before or afterwards1333
9o. 0ven if such an e$planation
rather tempts us we need only
thin' for a moment of what
actually happens in order to see
that we are going astray here.
We say that
we use the command in
contrast with other sentences
because our language contains
The clarifying voice/
Our temptation to use an
e$planation that re"uires us to
thin' of the other sentences
-hovering- is instructive. It
teaches us to stop and loo' and
not base our conclusions on
-what must be.- When we stop
to loo' we see that the other
the possibility of those other
sentences. Someone who did
not understand our language a
foreigner who had fairly often
heard someone giving the order/
-(ring me a slab>- might
believe that this whole series of
sounds was one word
corresponding perhaps to the
word for -building3stone- in his
language. If he himself had then
given this order perhaps he
would have pronounced it
differently and we should say/
he pronounces it so oddly
because he ta'es it for a single
sentences are no in anyway
hovering in our minds. What
ma'e one way of saying -(ring
me a slab>- a sentence and the
other way -(ring3me3a3slab>- a
word has something more to do
with the fact that we can ma'e
sentences that are variations on
the theme -(ring me a slab>-

(ut then is there not also
something different going on in
him when he pronounces it333
something corresponding to the
fact that he conceives the
sentence as a single word133333
(ut what is going on with him1
&ust he be picturing the -slab-
when he hears it1 Or must he
say this sentence to himself
-(ring me a slab>-
0ither the same thing may go on
in him or something different.
#or what goes on in you when
you give such an order1 %re you
conscious of its consisting of
four words while you are
uttering it1 Of course you have a
mastery of this language333
which contains those other
sentences as well333but is this
having a mastery something that
happens while you
are uttering the sentence1333%nd
I have admitted that the
foreigner will probably
pronounce a sentence differently
if he conceives it differently5 but
what we call his wrong
conception need not lie in
anything that accompanies the
utterance of the command.
We we issue a command -slab>-
what goes on in us1
Introspectively need there be
anything private1 There might
be something present when we
utter the command but there
need not be.

The sentence is 'elliptical' not
because it leaves out something
that we thin' when we utter it
but because it is shortened333in
comparison with a particular
paradigm of our grammar.333
In our culture we create the
paradigm of the full sentence as
the -real.- Therefore we say
-Slab>- is a shortened form and
not -(ring me a slab>- is a
lengthend form. (ut this
paradigm that calls the longer
form the real form is arbitrary.
Of course one might ob*ect here/
-.ou grant that the shortened
and the unshortened sentence
have the same sense.333What is
this sense then1 Isn't there a
verbal e$pression for this
%nd if they have the same sense
then isn't one form of the
sentence the -right- or -real-

(ut doesn't the fact that
sentences have the same sense
consist in their having the same
use1333+In )ussian one says
-stone red- instead of - the stone
is red-5 do they feel the copula
to be missing in the sense or
attach it in thought1,
&aybe not. &aybe we say that
the sentences have the same
-sense- only because they have
the same use in the language3
game. They cause one person to
fetch the ob*ect and both the
same regardless of which form
we use.
%phorism ;23?4 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold in Wittgenstein's
te$t has been inserted by Shawver to
enhance commentary.,
Shawver commentary:
;2. Imagine a language3game in
which % as's and ( reports the
number of slabs or bloc's in a pile
or the colours and shapes of the
building3stones that are stac'ed in
such3and3such a place.333 Such a
report might run/ -#ive slabs-. 9ow
what is the difference between the
report or statement -#ive slabs- and
the order -#ive slabs>-1333
That is what is the difference
between -#ive slabs- +in
language3game ;2, and
+-#ive slabs>- in language
game 61 ,

Well% it is the part which uttering
these words plays in the language-
0mphasis mine. (ut isn't the
important thing that -#ive
slabs>- in +6, causes the
wor'er to bring A slabs1
While -five slabs- in +;2,
only causes the supervisor to
have information1
9o doubt the tone of voice and the
loo' with which they are uttered
and much else besides will also be
different. (ut we could also imagine
the tone's being the same333for an
order and a report can be spo'en in
a variety of tones of voice and with
various e$pressions of face333the
difference being only in the
application. +Of course we might
use the words -statement- and
-command- to stand for
grammatical forms of sentence and
intonations5 we do in fact call
'Isn't the weather glorious to-
day&' a *uestion% although it is
On the surface the difference
might be a matter of how it is
voiced. (ut we could
imagine them being voiced
with the same intonation.
The intonation is after all
only a clue as to what the
differences are not the
difference itself.

(esides we could imagine a
used as a statement., We could
imagine a language in which all
statements had the form and tone of
rhetorical "uestions5 or every
command the form of the "uestion
-Would you li'e to. . .1-. Perhaps it
will then be said/ -What he says has
the form of a "uestion but is really a
command-333that is has the
function of a command in the
techni"ue of using the language.
+Similarly one says -.ou will do
this- not as a prophecy but as a
command. What ma'es it the one or
the other1,
language in which everything
stated or commanded was put
in the form of a "uestion.

;;. #rege's idea that every
assertion contains an
assumption which is the thing
that is asserted really rests on
the possibility found in our
language of writing every
statement in the form/ -It is
assert that such3and3such is the
case.-333 (ut -that such3and3
such is the case- is not a
sentence in our language333so far
it is not a move in the language3
game. %nd if I write not -It is
asserted that . . . .- but -It is
asserted/ such3and3such is the
case- the words -It is asserted-
simply become superfluous.
Still there is the dream +such as
#rege had, of including some
sort of notation in the body of
the sentence saying how it was
used. #or e$ample one might
include a statement such as -It is
asserted that- and complete the
sentence any such way. Or
alternatively one might do the
same thing by saying -It is
asserted/- and complete the
sentence any way.
(ut isn't it clear at least in the
last case that the notation -It is
asserted/- is superfluous1

We might very well also write
every statement in the form of a
"uestion followed by a
instance/ -Is it raining1 .es>-
Would this shew that every
statement contained a "uestion1

(esides there is nothing to
guarantee that a notation -It is
asserted/- will in fact be attached
to an assertion. %fter all don't
we use "uestioning grammatical
forms to ma'e statements1
:on't we say -It is a wonderful
day isn't it1- 0ven when we use
formulations that seem to tell us
how a sentence is being used

they need not accurately do so.
Of course we have the right to
use an assertion sign in contrast
with a "uestion3mar' for
e$ample or if we want to
distinguish an assertion from a
fiction or a supposition. It is
only a mista'e if one thin's that
the assertion consists of two
actions entertaining and
asserting +assigning the truth3
value or something of the 'ind,
and that in performing these
actions we follow the
prepositional sign roughly as we
sing from the musical score.
)eading the written sentence
loud or soft is indeed
comparable with singing from a
musical score but 'meaning'
+thin'ing, the sentence that is
read is not.

#rege's assertion sign mar's the
beginning of the sentence. Thus
its function is li'e that of
full3stop. It distinguishes the
whole period from a clause
within the period. If I hear
someone say -it's raining- but do
not 'now whether I have heard
the beginning and the end of the
period so far this sentence does
not serve to tell me anything.
(ut we can try to construct
language so carry such a notation
accurately. The mista'e is in
thin'ing that it is the notation
that ma'es it so. What is in
"uestion is whether the sentence
is a "uestion and the notation
does not ma'e it so.
The notation is only a label and a
label can be correct or
This means when we determine
that a statement is an assertion or
a "uestion it is not enough to
loo' to see what the notation +or
punctuation, tells us. This
information is not contained in
the words but in the way these
words are being used in the
#rege's notation that a sentence
is an asssertion is li'e the full
stop of a period at the end of
string of words. Gust as a period
does not assure you that the
sentence functions as a
statement however so #rege's
notation does guarantee that the
sentence functions as the
notation says.

See inserted comment of LW's.

;?. (ut how many 'inds of
sentence are there1 Say
assertion "uestion and
command1333 There are
The rules of language games are
not unchangeable laws. There is
a continuous evolution not only
in how many language games
countless 'inds/ countless
different 'inds of use of what we
call -symbols- -words-
-sentences-. (nd this
multiplicity is not something
fi+ed% given once for all, but
new types of language% new
games% as we may say% come
into e+istence% and others
become obsolete and get
forgotten +We can get a rough
picture of this from the changes
in mathematics.,
there are but evolution too as
to the 'ind of language games
thee are.
!ere the term -language3game-
is meant to bring into
prominence the fact that the
spea'ing of
language is part of an activity or
of a form of life.

We have seen this concept of
langauge being woven in a form
of life before. In +2F, he said
that to -imagine a language
meant to imagine a form of
life.- %nd in +;, he pointed out
that the slab language of that
language3game involved not
only words but activities
specifically the activity of
fetching ob*ects on command.
)eview the multiplicity of
language3game in the following
e$amples and in others/

9ow that LW has taught us
something about -language3
games- he is going to give us
samples to count. This serves as
a 'ind of ostensive definition of
language games although note
these e$amples differ from the
primitive language games he
tal'ed about in D +which was
illustrated by the slab language
of ;
H <iving orders and obeying
H :escribing the appearance of
an ob*ect or
giving its measurements333
H =onstructing an ob*ect from a
description +a
It is a useful e$ercise to imagine
a sentence of any sort
functioning in several of the
different language games. When
it does this it ta'es on a different
meaning. #or e$ample -There
was a storm today.- Imagine
H )eporting an event333
H Speculating about an event333
H #orming and testing a
H Presenting the results of an
e$periment in
tables and diagrams333
H &a'ing up a story5 and reading
H Play3acting333
H Singing catches333
H <uessing riddles333
H &a'ing a *o'e5 telling it333
H Solving a problem in practical
H Translating from one language
into another333
H %s'ing than'ing cursing
greeting praying.
333It is interesting to compare
the multiplicity of the tools in
language and of the ways they
are used the multiplicity of
'inds of word and sentence with
what logicians have said about
the structure of language.
+ Including the author of the
Tractatus Logico3

how a sentence li'e this might
function in -reporting an event-
-speculating about an event-
-presenting results from an
e$periment- -play acting-
-singing catches- and so forth.
Some sentences of course do
not ma'e sense in all language
games but whenever they do
they mean something different in
different language games.

Of course Wittgenstein is
himself the author of the
%nd in that boo' as well as in
wor's by other authors of that
era +e.g. )ussell, language was
seen as much more stable and

;@. If you do not 'eep the
multiplicity of language3games
in view you will perhaps be
inclined to as' "uestions li'e/
-What is a "uestion1-333Is it the
statement that I do not 'now
such3and3such or the statement
that I wish the other person
would tell me. . . .1 Or is it the
description of my mental state
Iuestions such as these LW tells
us come about from the
%ugustinian +Platonic and
confused, understanding of
language that is our heritage.
Why is this confused1

In my reading LW it is because a
-"uestion- is *ust a grammatical
form. It does not get at the
uncertainty1333%nd is the cry
-!elp>- such a description1

activity of -as'ing-. We can as'
with cries such as

Oh I wish I had someone
to go to the movies with>
%nd a sentence in the form of of
a "uestion might not be an

Would you mind going to
get me a slab1
We want to get beneath such
grammatical form +which LW
calls -surface grammar-, and
move down to the depth that is
something more important than
language than the form we use to
e$press it. %s'ing -What is a
"uestion1- betrays a concern
with the way things loo' on the
page or sound in the voic and
not a concern with the deep
structure that is the way the
language is wor'ing and having
an impact on what is happening.
Thin' how many different 'inds
of thing are called -description-/
description of a body's
position by means of its co3
ordinates5 description of a facial
e$pression5 description of a
sensation of touch5 of a mood.
If as'ing what a "uestion is
reveals a hidden confusion what
about as'ing what a description
Of course it is possible to
substitute the form of statement
or description for the usual form
"uestion/ - I want to 'now
whether . . . .- or -I am in doubt
whether . . . .-333but this does not
bring the different language3
!ere too with descriptions we
find there is a surface form that
does not tell us much about how
the sentence is being used. Gust
as practically anything can be
put in a "uestioning formt so
practically anything can be put in
a descriptive format.
games any closer together.
The significance of such
possibilities of transformation
for e$ample of turning all
statements into sentences
beginning -I thin'- or -I
believe- +and thus as it were
into descriptions of my inner
life, will become clearer in
another place. + Solipsism.,
LW gives an account of pain
language later that I thin' this
refers to but it is too early to get
into this now. The important
thing now to feel at home in his
distinction between the surface
of language +such as -What is a
"uestion-, and the "uestions
about the depth of language
+how is the sentence functioning
in the language game1,

;A. It is sometimes said that
animals do not tal' because they
lac' the mental capacity. %nd this
means/ -they do not thin' and that
is why they do not tal'.- (ut333
they simply do not tal'. Or to put it
better/ they do not use language333
if we e+cept the most primitive
forms of language333
=ommanding "uestioning
recounting chatting are as much a
part of our natural history as
wal'ing eating drin'ing playing.

!ere LW is loo'ing bac' at this
cultural imagery that he has been
deconstructing. %ccording to this
imagery to be able to -tal'- one
must be able to thin' 33 because
-tal'ing- is the e$pression of our
internal ideas.
:on't try to deconstruct this
imagery at this moment. Gust
notice that it is a natural thing to
thin' here. :ogs do not tal'
because they do not thin' internal
(ut note the parenthetical that I
have emphasi8ed. % dog can be
taught to fetch on command *ust
as the wor'er in +;, could fetch
slabs on command. Why are we
leaving this 'ind of language
outside the scope of -language-1
(ecause this is an aporetic voice
the voice of the fly3bottle.
Still we are indlined to say that
-dogs do not tal'- and by this we
mean that they also -do not thin'.-

;B. One thin's that learning What is naming a preparation
language consists in giving names
to ob*ects. 7i8 to human beings to
shapes to colours. to pains. to
moods to numbers etc. To repeat3
naming is something li'e attaching
a label to a thing. One can say that
this is preparatory to the use of a
word. (ut what is it a preparation
for1 Imagine a culture that could
only name. It had no other use
for language. People simply sat
around and named things or else
they did things without
language. %ll that this culture
would lac' in its language is
what naming is a preparation for.

;D. -We name things and then
we can tal' about them/ can refer
to them in tal'.- '%s if what we
did ne$t were given with the
mere act of naming. %s if there
were only one thing called
-tal'ing about a thing-. Whereas
in fact we do the most various
things with our sentences.

Isn't this e$actly what the
%ugtinian picture of language
in +;, implies1 We name things
and then we can tal' about
them. It is as though this is all
that is re"uired.
(ut naming things we have
come to see does not show us
what to do with them. The
wor'ers might be able to name
the beams pillars bloc's and
-slabs- and still not 'now to
fetch them. Language is not
*ust the uttering of words. It is
the use of words in the activity
of language.
%lso the illusion that all we
need to do to be able to tal' is
name things neglects how few
of the words we use are actually
Thin' of e$clamations alone
with their completely different
H Water>
H %way>
H Ow>
H !elp>
H #ine>
H 9o>
%re you inclined still to call
these words -names of ob*ects-1
Loo' at e$clamations. %re
these *ust names of ob*ects1 :o
you want to say that there is
something internal that these
words name1 Of course
someone uttering an
e$clamation li'e this might
have a image but are they

In languages +;, and +6, there
was no such thing as as#ing
something's name "his% with its
correlate% ostensive definition%
is% we might say% a language-
game on its own That is really
to say/ we are brought up
trained to as'/ -What is that
called1-3upon which the name is
given. %nd there is also a
language3game of inventing a
name for something and hence of
saying -This is ....- and then
using the new name. +Thus for
e$ample children give names to
their dolls and then tal' about
them and to them. Thin' in this
conne$ion how singular is the use
of a person's name to call him>,

In +;, and +6, the wor'er simply
brought the ob*ects re"uired.
There was no language for
as'ing what something was
called. Pointing and naming is
a language game of its own.
One must learn how to do this.
%nd in addition to learning to
give the e$isting name of an
ob*ect one can learn how to
invent names.

;6. 9ow one can ostensively
define a proper name the name
of a colour the name of a
material a numeral the name of
a point of the compass and so
on. The definition of the
number two -That is called
'two' -33pointing to two nuts3is
perfectly e$act. 33(ut how can
two be defined li'e that1 The
person one gives the definition
to doesn't 'now what one wants
to call -two-5 he will suppose
that -two- is the name given to
this group of nuts> !e may
suppose this5 but perhaps he
does not. !e might ma'e the
opposite mista'e5 when I want
to assign a name to this group
of nuts he might understand it
as a numeral. %nd he might
e"ually well ta'e the name of a
person of which I give an
ostensive definition as that of a
Where for e$ample is this hand
pointing1 Is it pointing to both of
the diamonds1 Or one1 Or is it
pointing to the color red1 Or is it
pointing to the side of one of the

Wittgenstein says that in every

case the ob*ect being pointed to is
ambiguous. =an you thin' of an
e$ception1 If not does this not
undermine %ugustine's picture of
how we learn language1
colour of a race or even of a
point of the compass. That is to
say/ an ostensive definition can
be variously interpreted in
every case.

;F. Perhaps you say/ two can only
be ostensively defined in this way/
-This number is called 'two' -. #or
the word -number- here shews what
place in language in grammar we
assign to the word. (ut this means
that the word -number- must be
e$plained before the ostensive
definition can be understood.

2 ; ?
@ A

This number is
called -two-.
:oes that solve the problem
of how we might ostensively
define ;1 There are several
problems with it. #irst the
child must learn what
-number- means in order to
understand what is being
pointed to.
33The word -number- in the
definition does indeed shew this
place5 does shew the post at which
we station the word. %nd we can
prevent misunderstandings by
saying/ -This colour is called so3
and3so- -This length is called so3
and3so- and so on. That is to say/
misunderstandings are sometimes
averted in this way. (ut is there only
one way of ta'ing the word -colour-
or -length-13Well they *ust need
defining.3:efining then by means
of other words> %nd what about the
last definition in this chain1 +:o not
say/ -There isn't a 'last' definition-.
That is *ust as if you
chose to say/ -There isn't a last
house in this road5 one can always
Still you might say the ; is
in the right place. One can
see where ; sits in the series
of numbers. %nd
misunderstandings can
sometimes be averted by
pointing li'e this. (ut how
can we define number1
=an we do it by e$ample1
Should we use a figure li'e

build an additional one''.,

This number is called
Or will the student be
confused by this ambiguity
too1 %nd if we tried to get
around this problem of
ambiguity by defining the
words how shall we define
them without their being
ambiguous too1
Whether the word -number- is
necessary in the ostensive definition
depends on whether
without it the other person ta'es the
definition otherwise than I wish.
%nd that will depend on the
circumstances under which it is
given and on the person I give it to.

(ut perhaps someone learns
what two means in a
particular conte$t even
without a completely
ade"uate e$planation for all
conte$ts. I as' for a ball and
the child learns to fetch a

Then I as' for two balls and

the child learns to fetch two
balls. This always pleases

%nd how he 'ta'es' the definition
is seen in the use that he ma'es of
the word defined.
(ut if he -ta'es- it in the
right way it will become a
powerful and reinforcing
?4. So one might say/
the ostensive definition
e$plains the use33the
meaning33of the word
when the overall role of
the word in language is
In ?4 Wittgenstein continues to
investigate the %ugustinian model and
its problems as the total e$planation for
our developing language. This model
you'll recall is based on the picture of
words being defined ostensively that is
clear. Thus if I 'now that
someone means to e$plain
colour3word to me the
ostensive definition -That
is called 'sepia' - will help
me to understand the
by naming and pointing.
33%nd you can say this so
long as you do not forget
that all sorts of problems
attach to the words -to
'now- or -to be clear-.

Someone from another country wants
to teach you a word in her native
language. She points to a pillow and
ma'e a strange sound -upapal- and
your "uestion is -What is she pointing
to1 Is it the pillow or the shape of the
pillow or what1- (ut if you 'new
somehow that she was pointing to the
color of the pillow then that would
ma'e all the difference in the world.
(ut that is because you 'now what
-color- means. Imagine then how
difficult it must be to learn a color
word from an ostensive definition if
you don't even have a concept of color.
%nd of course all of us were in that
place initially. isn't it remar'able that
we learned anything at all from the
One has already to 'now
+or be able to do,
something in order to be
capable of as'ing a
thing's name. (ut what
does one have to 'now1

If I already am "uite clear about what a
color word is then I can begin to as'
what the color of something is. If I
'now the term for color and my teacher
'nows the term for -color- too then I
am indeed a smart student. Gust
pointing and saying -that is the color
sepia- should surely do it. (ut without
those tools things are going to be a lot

=ould one define the word -red- by pointing to something that was
not red1 That would be
as if one were supposed to e$plain the word -modest- to someone
whose 0nglish was wea' and one pointed to an arrogant man and
said -That man is not modest-. That it is ambiguous is no argument
against such a method of definition. %ny definition can be
(ut it might well be as'ed/ are we still to call this -definition-133
#or of course even if it has
the same practical conse"uences the same effect on the learner it
plays a different part in the calculus from what we ordinarily call
-ostensive definition- of the word -red-.
%phorism ?23?6 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis is bold is inserted by
Shawver to enhance
Shawver commentary:
?2. When one shews someone
the 'ing in chess and says/ -This
is the 'ing- this does not tell
him the use of this piece3unless
he already 'nows the rules of the
game up to this last point/ the
shape of the 'ing. .ou could
imagine his having learnt the
rules of the game without ever
having been strewn an actual
piece. The shape of the chessman
corresponds here to the sound or
shape of a word.
Suppose someone showed you
an Eray and said to you -see
that tumor1- It might be evident
to all who have learned to read
Erays but *ust pointing to it is
not enough to enable this 'ind of
seeing. So it is with handing a
child a chess piece and saying
-This is a 'ing.- The
bac'ground for ma'ing sense of
this pointing and naming has not
be laid down.
One can also imagine
someone's having learnt the
game without ever learning or
formulating rules. !e might have
learnt "uite simple board3games
first by watching and have
progressed to more and more
complicated ones. !e too might
be given the e$planation -This is
the 'ing-33 if for instance he
were being strewn chessmen of a
shape he was not used to. This
e$planation again only tells him
the use of the piece because as
we might say the place for it
was already prepared Or
even/ we shall only say that it
tells him the use if the place is
already prepared. %nd in this
case it is so not because the
The emphasis in this passage is
mine. It represents a 'ey
concept the concept of an
ostensive definition being made
possible by the place for the
definition being prepared.
(ut the primary point I believe
is that if we 'new the rules of
the chess game 'new that losing
your 'ing meant that you lost the
game for e$ample or how the
'ing can move within the rules
of the game then having
someone say -This is the 'ing in
a chess set- would mean a lot
more would clarify more than
if you had never heard of chess
or board games. Sometimes
one does not 'now enough about
person to whom we give the
e$planation already 'nows rules
but because
in another sense he is already
master of a game.

a sub*ect to even as' useful

=onsider this further case/ I am
e$plaining chess to someone5
and I begin by pointing to a
chessman and saying/ -This is
the 'ing5 it can move li'e
this .... and so on.- 33 In this case
we shall say/ the words -This is
the 'ing- +or -This is called the
''ing' -, are a definition only if
the learner already ''nows what a
piece in a game is'. That is if he
has already played other games
or has watched other people
praying 'and understood'3and
similar things. #urther only
under these conditions will he be
able to as' relevantly in the
course of learning the game/
-What do you call this1-33that is
this piece in a game.
We may say/ only someone who
already 'nows how to do
something with it can
as' a name.

There are a family of ways one
might go about preparing a
person to understand -This is a
'ing- when showing them a
chess piece. It would help
perhaps if a person 'new how to
play chec'ers and 'new in
addition that in chess losing the
'ing meant losing the game.
Still this would not prepare the
listener to understand his
statement as much as if he
learned to play chess with pieces
that had a different 'ind of 'ing.

%nd we can imagine the person
who is as'ed replying/ -Settle
the name yourself-3and now the
one who as'ed would have to
manage everything for himself.

If you did not have the concept
of what is being named that is
if the place for this name is not
prepared then perhaps it would
be as well for you to name it for
yourself. Learning the -name-
of something +instead of naming
it, is important precisely in those
cases that learning the name will
connect with what we already

'now and allow us to learn what
we are seeing more completely.
Say you go to the doctor with a
s'in rash and as' -What is this
called1- %nd suppose the doctor
gives you an unintelligible
technical name. 9ot helpful.
(ut suppose the doctor says
-This is a measles rash.- Then
because you have an idea as to
what measles is you have
learned "uite a bit. (ut if you
didn't have the concept of
measles things would be
different. .ou could call it
whatever you wanted. It would
be *ust as meaningful to you.
!owever it might prepare you
less well for tal'ing with others.

?;. Someone coming into a
strange country
will sometimes learn the
language of the
inhabitants from ostensive
definitions that they give him5
and he will often have to 'guess'
the meaning of these definitions5
and will guess sometimes right
sometimes wrong.
I remember !arry describing
learning a foreign language li'e
this. !e was in a foreign country
and people would teach him the
names of things by pointing and
naming. This seems li'e a very
easy way to learn the names of
things in a foreign tongue.
%nd now I thin' we can say/
%ugustine describes the learning
of human language as if the
child came into a strange
country and did not understand
the language of the country5 that
is as if it already had a
language only not this one. Or
again/ as if the child could
already thin' only not yet
Isn't it so1 %ugustine describe
this 'ind of pointing and naming
as the way that the child learns
language1 (ut we have been
wor'ing on why this e$plains so
little in the learning of language
and noticing the limits to this
'ind of learning for e$ample
that pointing and naming -blue-
doesn't mean that the hearer
spea'. %nd -thin'- would here
mean something li'e -tal' to

recogni8es what we are naming
33 even if the hearer then can
point at the blue ob*ect and say
%lso such an ostensive
definition can hardly e$pain how
we learn the word -the- or -for-
or in fact most words. Loo'
bac' at this paragraph and see
how many words could be taught
to the child by ostensive
The problem is that the young
child in the beginning +picture
baby %ugustine, does not have a
place prepared for learning by
What 'ind of bac'ground is
necessary to prepare such a
place1 !ow would you train a
child so that it understood that
you are naming a chess piece for
e$ample1 Or the color -blue-1

??. Suppose however
someone were to ob*ect/ -It
is not true that you must
already be master of a
language in order to
understand an ostensive
definition/ all you need 33of
course>33 is to 'now or guess
what the person giving the
e$planation is pointing to.
That is whether for e$ample
to the shape of the ob*ect or
to its colour or to its number
and so on.- 33 %nd what does
'pointing to the shape'
'pointing to the colour'
consist in1 Point to a piece of
!ere LW is luring us bac' into the
muddle and it is good to let
ourselves go there for a moment
'nowing it is a muddle but letting
ourselves feel the pull. In this
muddle he continues to as' how can
an ostensive definition teach the
meaning of a term1 !ow does the
student 'now what we are pointing
to. There is ambiguity in the
pointing in every case we can
paper. 33%nd now point to its
shape 33 now to its
colour 33 now to its number
+that sounds "ueer,. 33!ow
did you do it1 33.ou will say
that you 'meant' a different
thing each time you pointed.
%nd if I as' how that is done
you will say you
concentrated your attention
on the colour the shape etc.
(ut I as' again/ how is that
Suppose someone points to a
vase and says -Loo' at that
marvellous blue3the shape
isn't the point.- 33Or/ -Loo'
at the marvellous shape3the
colour doesn't matter.-
Without doubt you will do
something different when
you act upon these two
invitations. (ut do you
always do the same thing
when you direct your
attention to the colour1
Imagine various different
cases. To indicate a few/
What we do when we -attend to the
color' of something seems when
you thin' about it rather nebulous.

H -Is this blue the same as
the blue over there1
:o you see any
H .ou are mi$ing paint and
you say -It's hard
to get the blue of this
H -It's turning fine you can
already see blue
s'y again.-
H -Loo' what different
effects these two blues
H -:o you see the blue boo'
over there1 (ring
=onsider all these conte$ts in which
you -attend to the color- of blue.
Isn't there something different about

it here. -
H -This blue signal3light
means ....-
H -What's this blue called.'3
Is it 'indigo'1-
.ou sometimes attend to
the colour by putting your
hand up to 'eep the outline
from view5 or by not loo'ing
at the outline of the thing5
sometimes by staring at the
ob*ect and trying to
remember where you saw
that colour before.
.ou attend to the shape
sometimes by tracing it
sometimes by screwing up
your eyes so as not to see the
colour clearly and in many
other ways. I want to say/
This is the sort of thing that
happens while one 'directs
one's attention to this or that'.
(ut it isn't these things by
themselves that ma'e us say
someone is attending to the
shape the colour and so on.
Gust as a move in chess
doesn't consist simply in
moving a piece in such3and3
such a way on the board3nor
yet in one's thoughts and
feelings as one ma'es the
move/ but in the
circumstances that we call
-playing a game of chess-
-solving a chess problem-
and so on.
%lthough there are surely typical
things you actually do when you
attend to the color it is not the
things you actually do that are in
fact what we mean by the -attending
to the color.- There are a variety of
things people might actually do in
the process of -attending to the

?@. (ut suppose someone said/
-I always do the same thing
when I attend to the shape/ my
In ?@ the "uestion is/ -!ow does
the student 'now what the
teacher is pointing to1 What if
eye follows the outline and I
feel....-. %nd suppose this
person JwereK to give someone
else the ostensive definition
-That is called a 'circle' -
pointing to a circular ob*ect and
having all these e$periencesJ
J cannot his hearer still interpret
the definition differently even
though he sees the other's eyes
following the outline and even
though3he feels what the other
the teacher points to the shape
and says 'This is the shape1'
!ow will we 'now that the
teacher is not pointing to the
color1 Would it help to notice
that the teacher ma'es some
moves of her hand to suggest she
is pointing to the shape1

That is to say/ this
'interpretation' may also consist
in how he now ma'es use of the
word5 in what he points to for
e$ample when told/ -Point to a

0ven when you point at the blue
circular image to me and say
-circle- very carefully and
follow the edge of the circle with
your eyes maybe even run your
finger around the edge of the
circle and even when you are
possessed of a 'circle3feeling' I
can still misinterpret what you
are pointing to. Is that not true1
#or neither the e$pression -to
intend the definition in such3
and3such a way- nor the
e$pression -to interpret the
definition in such3and3such a
way- stands for a process which
accompanies the giving and
hearing of the definition.

If you intend your pointing to the
shape to be a definition of the
circle that is all well and good
but there is no mental
accompaniment of this act that
we call -intention- that is
re"uired for it to be an ostensive
definition. Ostensive definition is
*ust the pointing and naming of
something. It is pointing to the
blue circle and saying -circle-
regardless of inner intention.
+Thin' of someone who does this
so routinely that it can be done
'without thin'ing about it' in the
moment., %nd the same is true
for the student's interpretation of
the ostensive definition. Imagine
the student paying meager
attention to the teacher and
neverthless pic'ing up on the
definition correctly or as another
e$ample incorrectly. If the
student failed to understand
correctly would that ma'e the
definition any less of a

?A. There are of course what
can be called -characteristic
e$periences- of pointing to +e.g.,
the shape. #or e$ample
following the outline with one's
finger or with one's eyes as one
points. 33(ut this does not
happen in all cases in which I
'mean the shape' and no more
does any other one characteristic
process occur in all these cases.
33(esides even if something of
the sort did recur in all cases it
would still depend on the
circumstances 33that is on what
happened before and after the
pointing 33whether we should
say -!e pointed to the shape and
not to the colour-.

#or the words -to point to the
shape- -to mean the shape- and
so on are not used in the same
way as these/ -to point to this
boo' +not to that one, -to point
to the chair not to the table- and
so on. 33Only thin' how
differently we learn the use of
the words -to point to this
thing- -to point to that thing-
and on the other hand -to point
to the colour not the shape- -to
mean the colour- and so on.
Wittgenstein is distinguishing
two related language-games of
pointing -ne in which you
point to the thing and give its
name% and another related one
in which you point to the shape
or the color and give its name
(oth cases re"uire only that you
point in the same physical way.
There may be differences in the
way people point in these two
language games but these
differences only help us
distinguish between them. These
different ways of pointing are
not inevitable and they are not
To repeat/ in certain cases
especially when one points 'to
the shape' or 'to the number'
there are characteristic
e$periences and ways of
pointing3'characteristic' because
they recur often +not always,
when shape or number are
'meant'. (ut do you also 'now of
an e$perience characteristic of
pointing to a piece in a game as
a piece in a game1
%ll the same one can say/ -I
mean that this piece is called the
''ing' not this particular bit of
wood I am pointing to-.
+)ecogni8ing wishing
remembering etc. ,

!ere LW is saying that the
sentence -/ -I mean that this
piece is called the ''ing' not this
particular bit of wood I am
pointing to- is itself ambiguous.
-&ean- can mean -reccogni8ing
wishing remembering etc.- #or
e$ample the above sentence
might be paraphrased -I
recogni8e that this piece is called
the ''ing'...- or -I wish this piece
were called the ''ing'... and so
forth. %ll these different
paraphrases have different
Thus this concept of
introspective pointing to the
shape or color to teach shape and
color remains a pu88le.

?B. %nd we do here what we
do in a host of similar cases/
because we cannot specify
any one bodily action which
we call pointing to the shape
+as opposed for e$ample to
the colour, we say that a
spiritual +mental intellectual,
activity corresponds to these
?B. When we point to the ball there
is a physical ob*ect we are pointing
to. When we point to the color
what we are pointing to is much
more nebulous. In these cases LW
says we tend to do something "uite
peculiar. We imagine that there
must be something that we are
pointing to even though it is hard to
see or even imagine and this
Where our language suggests
a body and there is none/
there we should li'e to say is
a spirit.
-something- we imagine ourselves
pointing to is -spirit.-
I don't thin' this concept of -spirit-
necessarily implies anything
religious although it sometimes
might. What he means by -spirit- is
more subtle and available only by
introspection. One points to the
blue circle and mean -blue-. !ow
does one do this. LW is saying that
it feels li'e we are doing it
-spiritually-. )emember LW is not
saying that we are doing it
spiritually. !e is saying that we all
have a tendency to thin' of it this
way. It is as though there is
something -spiritual- involved in
forming a -meaning- in our minds
and that this -meaning- that we
form in our spirit somehow
corresponds to the words that we
are thin'ing.
When do we do this1 !e says we
tend to do it when our language
says there is a body we should be
referring to and where in fact
there is none. The language
suggests that -blue- is a body but
in fact it is not so it seems we are
pointing spiritually.
Let's imagine another e$ample. I
-It is raining.-
Our language suggests there should
be a body to correspond with the 'it'
in this sentence. 9otice however
that it is hard to find a body
although our language suggests that
there is one. !ere is a case then
that we might be tempted to say that
the -it- that is raining is spirit.
!ere are some more e$amples/
H I have a hard time 'eeping all
these numbers
in my mind.-
H What about the word -numbers-1
H It's time to go.
Is there a body to correspond to
these nouns1
What about the word -mind-1 Is
there a body to correspond with
that1 What about -numbers-1 Or
the word -It's-1 :o you want to say
that -it- is -time- in this sentence1
Then as' yourself what you oint to
when you point to time.
In cases li'e this LW is saying we
are inclined to thin' that what is
being referenced is spirit or
something spiritual or mental.

In L?B Wittgenstein noted that we cannot identify
a distinctive action that we call pointing to the
shape +or pointing to the color, and because of
that we tend to see this 'ind of pointing as
?D. What is the
between name
thing named1
Well what is
it1 Loo' at
+;, or at
MWhen we consider the matter more
imaginatively as %ugustine did in L2 when he
imagined that he had been taught language by
being taught to name things we might well thin'
of the name bringing up a mental image of that
originary lesson. Supposedly according to this
imaginative picture we 'now what the other
person is tal'ing about +e.g. a slab, because
having learned the name of slab ostensively we
another one/
you can see the
sort of thing
this relation
consists in.
This relation
may also
among many
other things in
the fact that
hearing the
name calls
before our
mind the
picture of what
is named5 and it
among other
things in the
being written
on the thing
named or
when that thing
is pointed at.
now have mental images of a -slab- every time
we hear the word. This is particularly compelling
because we have all e$perienced mental images
when things are named. Still a little
introspection shows that we do not have a mental
image for every word we hear.
%lternative to the theory of mental images
assisting understanding we sometimes imagine
ob*ects having labels attached. Still we do not
often write the word -chair-on our chairs. So in
the end these two theories of language do not
wor' very well when we thin' about them.
(ut that does not mean we give them up. What
we do sometimes is imagine that the images +or
the labels, are there but in a fu88y and spiritual
way. In this fu88y and spiritual way we point to
things and name things in our mind.
(ut then LW as's us to loo' at L;. .ou remember
in L; we had the simple game of the wor'er and
his supervisor. The supervisor called out -beam>-
and the wor'er brought it. What is the
relationship between the name and the thing in
that particular instance1 It simply causes the
wor'er to fetch what the supervisor wants. 9eed
there be mental images here1 )emember our
tal'ing about the way I might teach a gorilla to
hand me a banana when I said -banana-1 %nd
that this would be a 'ind of tric'. It wouldn't need
to be the case that the gorilla actually understood
what the banana was apart from this particular
conte$t of handing one to me. !ere we might
say that the 'name' of the ob*ect does not function
merely as a name. It functions more as a
command although the word we thin' of as a
name has a role in ma'ing the command clearer.
So can you see that in spite of our models of
language +pointing spiritually or attaching a label
spiritually, these models do nt seem entirely
satisfactory. %side from the problematic
metaphysics of a spiritual pointing and naming
we have the fact that in the language game the
term -slab- is not *ust a name of an ob*ect. It is a
command to fetch a slab. That activity around
which the word gets pronounced is not accounted
for by naming and pointing.
%re the mental images re"uired for this activity of
fetching1 9o. 9ot logically. The wor'er is *ust
trained to do something at the sound of the name.
The supervisor does not re"uire him to create a
mental image of the ob*ect first. Of course he
might do so anyway but this is not re"uired.
This shows how problematic our notion of naming
is and how much we try to patch it up with
notions of fu88y spirits doing the wor'.

?D. We have been tal'ing about
the relationship between a name
and the thing named and we have
studied two cultural models. In
one the name is metaphorically
-attached- to the thing +li'e a
label might be inscribed on the
thing it names, and in the other
model the word we use -points-
spiritually to the thing it names.
These are the vague models we
use for how words -attach- to
things. (ut Wittgenstein is
leading us through a critical
reflection on these models
because these models lead us to
thin' we have the problem
solved when in fact they are in
many ways unsatisfactory
models that lead us astray.
Wittgenstein continues to
deconstruct these old models of
language. !ere in ?6 he is going
to remind us again that the
models are only satisfactory
when we thin' of certain 'inds of
words. Then he points to terms
for which it is hard to use one of
the two models above.
?6. (ut what for e$ample is the
word -this- the name of in
language3game +6, or the word
-that- in the ostensive definition
-that is called ....-1

-This- and -that- are very
difficult words to understand if
we stay within the models
above. of teaching something by
attaching labels or pointing.
!ow could you attach the word
-this- to everything you call
-this-1 %nd if you point
spiritually to a particular -this-
with your hidden soul then what
on earth does this -pointing-
have to do with the word -this- in
a more general sense. One might
illustrate an apple or a dog by
pointing to one but can one
illustrate a -this- *ust by
33If you do not want to produce
confusion you will do best not
to call these words names at
all.33 .et strange to say the
word -this- has been called the
only genuine name5 so that
anything else we call a name
was one only in an ine$act
appro$imate way.
This "ueer conception springs
from a tendency to sublime the
logic of our language3as one
might put it.
If we call -this- a name then it is
a name that can be applied
everywhere. It offers no
specificity at all. .et at a certain
point in doing philosophy it
seems li'e the only legitimate
name. To call something a
-chair- classifies it with other
often dissimilar ob*ects. (ut
what can be purer than *ust
calling it a -this.-
This is a way of trying to ma'e
our logic more lofty our
statements more pure. %nd when
we do this it leads to "ueer

The proper answer to it is/ we
call very different things
-names-5 the word -name- is
used to characteri8e many
different 'inds of use of a word
!ere LW is introducing us to an
important pu88le that he will
clarify later. !e wants us to
notice that diverse 'inds of
things are called -names' and that
related to one another in many
different ways53but the 'ind of
use that -this- has is not among

we have no golden thread to tie
them all into a neat conceptual
%nd at the same time he is
showing that it will be
problematic for us if we try to
include -this- and -that- within
this diverse bundle of words that
we call names.
It is "uite true that in giving
an ostensive definition for
instance we often point to the
ob*ect named and say the name.
%nd similarly in giving an
ostensive definition for instance
we say the word -this- while
pointing to a thing. %nd also the
word -this- and a name often
occupy the same position in a
sentence. (ut it is precisely
characteristic of a name that it is
defined by means of the
demonstrative e$pression -That
is 9- +or -That is called '9' -,.
(ut do we also give the
-That is called 'this' - or -This
is called 'this'-1
This seems to devastate the
notion that you can ostensively
define -this- and -that-. !ow
can one point to anyplace and say
-that- is -that-. Or if one does
how does this e$plain to the
hearer what -that is.-

This is connected with the
conception of naming as so to
spea' an occult process.

When LW tal's of the notion of
naming as a 'ind of occult
process he is critici8ing the
picture of naming that he feels
our culture teaches us. It is the
picture of naming being a 'ind of
spiritual pointing.
9aming appears as a "ueer
conne$ion of a word with an
ob*ect. 33%nd you really get
such a "ueer conne$ion when
the philosopher tries to bring out
the relation between name and
thing by staring at an ob*ect in
This sentence -#or philosophical
problems arise when language
'goes on holiday'- is a famous
sentence in Wittgenstein. It
means that language is ta'en out
of conte$t and philosophi8ed
about it becomes -confusing-. It
front of him and repeating a
name or even the word -this-
innumerable times. .or
philosophical problems arise
when language goes on
holiday. %nd here we may
indeed fancy naming to be
remar'able act of mind as it
were a baptism of an ob*ect.
%nd we can also say the word
-this- to the ob*ect as it were
address the ob*ect as -this-3a
"ueer use of this word which
doubtless only occurs in doing
reminds me of a time when I was
a child that I said -butterfly- over
and over. Isn't it strange I
thought that we say -(utter3fly-
as though butter were to fly
away or -but 3er 3fly- and by the
time that I had said this 2A times
or so the word no longer seemed
to mean -butterfly- in the simple
way it had. Often when one
philosophi8es about a concept the
concept has -gone on holiday-.
We have lost our grounding in
concrete e$amples. We 'now
very well how to use the word
-virtue- in a sentence for
e$ample but when we scratch
our heads and wonder what
-virtue- really means then the
word -virtue- is on holiday. We
are *ust thin'ing about the word
not using it in the natural way
that our language allows us to
use it.
:o you have any e$perience with
language going on holiday1 0ver
said a word a few times a
familiar word and then sort of
lose the meaning of it as you
reflect on what this word means1
%nd what do you thin' about
-that- and -this-1 :o they seem
li'e names to you1

What is it to mean the words
-That is blue- at one time as a
statement about the ob*ect one
is pointing to 33at another as an
e$planation of the word -blue-1
Paraphrase li'e this can help us
be clearer about what language
game is being played.
Well in the second case one
really means -That is called
'blue' -. 33Then can one at one
time mean the word -is- as -is
called- and the word -blue- as -
'blue' - and another time mean
-is- really as -is-1
It is also possible for someone
to get an e$planation of the
words out of what was intended
as a piece of information.
J&arginal note/ !ere lur's a
crucial superstition.K
Of course. I might say -!ow do
you li'e my new sepia couch.-
This might give you an
unintended e$planation of the
word 'sepia.-
Can I say 'bububu' and
mean 'If it doesn't rain I shall
go for a wal#'& --It is only in
a language that I can mean
something by something "his
shews clearly that the
grammar of 'to mean' is not
li#e that of the e+pression 'to
imagine' and the li#e
This is a critical point that should
be pu88led about at this moment
rather than clarified. =an one say
-hello- to mean goodbye1
Without somehow creating a
special code for others to
interpret1 Or does the meaning
that we spin with our words have
to cooperate somehow with their
more standard cultural use1
%phorism ?F3A4 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis is bold is inserted by
Shawver to enhance
Shawver commentary:
?F. (ut why does it occur to
one to want to ma'e precisely
this word into a name when it
evidently is not a name13That is
*ust the reason. #or one is
tempted to ma'e an ob*ection
against what is ordinarily called
a name. It can be put li'e this/ a
name ought really to signify a
simple. %nd for this one might
perhaps give the following
reasons/ The word -0$calibur-
say is a proper name in the
ordinary sense. The sword
0$calibur consists of parts
combined in a particular way. If
they are combined differently
0$calibur does not e$ist. (ut it is
clear that the sentence
-0$calibur has a sharp blade-
ma'es sense whether 0$calibur
is still whole or is bro'en up. (ut
if -0$calibur- is the name of an
ob*ect this ob*ect no longer
e$ists when 0$calibur is bro'en
in pieces5 and as no ob*ect would
then correspond to the name it
would have no meaning. (ut
then the sentence -0$calibur has
a sharp blade- would contain a
word that had no meaning and
hence the sentence would be
In +?F, LW introduces the
"uestion of whether comple$
ob*ects have simple components.
We discuss whether 0$calibur
+the sword of Ning %rthur,
disappeared when it is bro'en into
a blade and a handle. %nd if it
does then how can we spea' of
0$calibur having a sharp blade1
If the blade is re"uired to be
attached to the handle in order for
0$calibur to e$ist then the blade
is part of 0$calibur and that
means that 0$calibur is the
handleOblade combination so to
say that 0$calibur has a sharp
blade is to say that this
handleOblade combination has a
sharp blade 33 which ma'es no
sense. +!ence our aporia.,

nonsense. (ut it does ma'e
sense5 so there must always be
something corresponding to the
words of which it consists. So
the word -0$calibur- must
disappear when the sense is
analysed and its place be ta'en
by words which name simples. It
will be reasonable to call these
words the real names.
@4. Let us first discuss this point
of the argument/ that a word has
no meaning if nothing
corresponds to it.3It is important
to note that the word -meaning-
is being used illicitly if it is used
to signify the thing that
'corresponds' to the word. That is
to confound the meaning of a
name with the bearer of the
name. When &r. 9. 9. dies one
says that the bearer of the name
dies not that the meaning dies.
%nd it would be nonsensical to
say that for if the name ceased
to have meaning it would ma'e
no sense to say -&r. 9. 9. is
!ere is a digression as to whether
a word has a meaning if nothing
corresponds to it.

@2. In L2A we introduced
proper names into language +6,.
9ow suppose that the tool with
the name -9- is bro'en. 9ot
'nowing this % gives ( the sign
-9-. !as this sign meaning now
not.13What is ( to do when he is
given it13We have not settled
anything about this. One might
as'/ what mill he do1 Well
perhaps he will stand there at a
loss or shew % the pieces. !ere
one might say/ -9- has become
meaningless5 and this e$pression
would mean that the sign -9- no
longer had a use in our language3
This continues with the digression
of whether names ma'e sense
once the ob*ects disappear. In 2A
we are tal'ing about one of the
building site language games.
The wor'er is fetching pillars and
bloc's. If the pillars and bloc's
have proper names does it ma'e
sense to refer to them if they have
no ob*ect to reference1

game +unless we gave it a new
one,. -9- might also become
meaningless because for
whatever reason the tool was
given another name and the sign
-9- no longer used in the
language3game. 33 (ut we could
also imagine a convention
whereby ( has to sha'e his head
in reply if % gives him the sign
belonging to a tool that is
bro'en.3In this way the
command -9- might be said to
be given a place in the language3
game even when the tool no
longer e$ists and the sign -9- to
have meaning even when its
bearer ceases to e$ist.
@;. (ut has for instance a name
which has never been used for a
tool also got a meaning in that
game1 Let us assume that -E- is
such a sign and that % gives this
sign to ( 33 well even such signs
could be given a place in the
language3game and ( might
have say to answer them too
with a sha'e of the head. +One
could imagine this as a sort of
*o'e between them.,
Say that the E is -tree-. The
supervisor as's the wor'er to
bring a (loc'2 Pillar? and then
-tree- and all the wor'ers laugh.
Or instead of -tree- the
supervisor might say -angel- and
this too might provo'e a laugh
even though no angel
corresponded to it. Or the wor'
supervisor might say -pillar B-
even though both supervisor and
wor'er 'now that -pillar B- was
crushed recently and so cannot be
brought because it -no longer
@?. #or a large class of cases3
though not for all3in which we
employ the word -meaning- it
can be defined thus/ the meaning
of a word is its use in the
%nd the meaning of a name is
sometimes e$plained by pointing
to its bearer.
The e$amples in L@; that I gave
illustrate ways in which words
can have a use in the language3
game even when they do not have
a referent that we can point to and
name. This settles the "uestion
introduced in ?F. .es a word can
have a meaning even if it does not
have a -bearer- +something to
point to,. Its meaning is
e$plained by its use in the
+=lic' here for an e$planded
commentary on this aphorism.,
@@. We said that the sentence
-0$calibur has a sharp blade-
made sense even when 0$calibur
was bro'en in pieces. 9ow this
is so because in this language3
game a name is also used in the
absence of its bearer. (ut we can
imagine a language3game with
names +that is with signs which
we should certainly include
among names, in which they are
used only in the presence of the
bearer5 and so could always be
replaced by a demonstrative
pronoun and the gesture of
In @@ LW uses the point
established in L@? that a name can
ma'e sense even in the absence of
its bearer. (ut now he wants to
reflect on the possibility of having
a language in which words only
made sense when they have a
bearer that is when the names
could be replaced with the
pronoun -this- as in -bring this>-
+Imagine the wor' supervisor
wal'ing over and pointing to the
pillar that he wanted ta'en over to
the pile. We can hardly imagine
this wor'ing if the pillar wasn't
@A. The demonstrative -this-
can never be without a bearer. It
might be said/ -so long as there
is a this the word 'this' has a
meaning too whether this is
simple or comple$.- (ut that
does not ma'e the word into a
name. On the contrary/ for a
name is not used with but only
e$plained by means of the
gesture of pointing.
Imagine someone pointing to
person and saying -This is
Goseph.- The -This- is not a
name. It is a way of e$plaining
who Goseph is.

@B. What lies behind the idea
that names really signify
simples1 33Socrates says in the
Theaetetus/ -If I ma'e no
mista'e I have heard some
people say this/ there is no
definition of the primary
elements 33 so to spea' 33 out of
which we and everything else
are composed5 for everything
that e$ists in its own right can
only be named no other
determination is possible neither
that it is nor that it is not..... (ut
what e$ists in its own right has
!ere LW shows us how deep the
roots of the ideas of simples is.
The idea is that everything is
either a simple thing or a comple$
thing where a comple$ thing is a
composite of simples things.

to be ....... named without any
other determination. In
conse"uence it is impossible to
give an account of any primary
element5 for it nothing is
possible but the bare name5 its
name is all it has. (ut *ust as
what consists of these primary
elements is itself comple$ so the
names of the elements become
descriptive language by being
compounded together. #or the
essence of speech is the
composition of names.-
(oth )ussell's 'individuals' and
my 'ob*ects' +Tractatus Logico3
Philosophicus, were such
primary elements.
@D. (ut what are the simple
constituent parts of which reality
is composed1 33 What are the
simple constituent parts of a
chair1 33 The bits of wood of
which it is made1 Or the
molecules or the atoms1 33
-Simple- means/ not composite.
(nd here the point is: in what
sense 'composite'& It ma#es no
sense at all to spea# absolutely
of the 'simple parts of a chair'
+The emphasis is mine., When he
says it ma'es no sense to spea'
-absolutely- of the simple parts of
something he means that it ma'es
no sense to spea' of -parts-
without some 'ind of conte$t that
defines what a -part- is.

%gain/ :oes my visual image
of this tree of this chair consist
of parts1 %nd what are its simple
component parts1 &ulti3
colouredness is one 'ind of
comple$ity5 another is for
e$ample that of a bro'en outline
composed of straight bits. %nd a
curve can be said to be
composed of an ascending and a
descending segment.
This is the gestalt notion that the
perception consists of more than
the sum of its parts. If you loo' at
a particular person you do not see
*ust a collection of parts. %nd if
you loo' at a curved line
you do not *ust see the elements
of that curve. .ou see it as a
If I tell someone without any The "uestion re"uires a conte$t.
further e$planation/ -What I see
before me now is composite- he
will have the right to as'/ -What
do you mean by 'composite'&
#or there are all sorts of things
that that can mean> 33
Otherwise we don't 'now what to
count as -parts.-
The "uestion -Is what you see
composite1- ma'es good sense if
it is already established what
'ind of comple$ity 33 that is
which particular use of the word
33 is in "uestion. If it had been
laid down that the visual image
of a tree was to be called
-composite- if one saw not *ust a
single trun' but also branches
then the "uestion -Is the visual
image of this tree simple or
composite1- and the "uestion
-What are its simple component
parts1- would have a clear
sense3a clear use. %nd of course
the answer to the second
"uestion is not -The branches-
+that would be an answer to the
grammatical "uestion/ -What are
here called 'simple component
parts'1-, but rather a description
of the individual branches.
That is we can create a language
game in which we count
-branches- as parts and say that a
tree is a composite +imagine a
s'etched tree, if it has branches.
(ut without such a conte$t the
"uestion -Is this tree composite1-
doesn't ma'e much sense. If there
is no such conte$t then the
answer to the "uestion -What are
its parts- is an answer as to what
to count as parts in this conte$t
not an answer about what the
parts are aside from the conte$t.
In other words to say that -the
branches- are the parts is an
answer to the grammatical
"uestion as to what to count as
parts not an answer about the
component parts in this tree aside
from conte$t. If we wanted to
tal' about this particular tree +and
not *ust negotiate what are to
count as parts, we will want to do
something closer to describing
what we see as its parts +which is
arbitrary outside of a negotiated
language game,.
(ut isn't a chessboard for
instance obviously and
absolutely composite1
9otice the word -absolutely-
here. It has the special meaning
of -absolutely and irrespective of
33 .ou are probably thin'ing of
the composition out of thirty3two
white and thirty3two blac'
s"uares. (ut could we not also
say for instance that it was
composed of the colours blac'
This is the "uestion again as to
whether there are ever absolute
parts of anything. The chessboard
is the e$ample he chooses that
seems most compelling. :oesn't
it seem in some natural sense
and white and the schema of
s"uares1 %nd if there are "uite
different ways of loo'ing at it
do you still want to say that the
chessboard is absolutely
'composite'1 33
that there are absolute parts of a
chessboard1 %nd these parts are
the s"uares on the chessboard1
What conte$t could change the
answer to that1
%s'ing -Is this ob*ect
composite1- outside a particular
language3game is li'e what a
boy once did who had to say
whether the verbs in certain
sentences were in the active or
passive voice and who rac'ed
his brains over the "uestion
whether the verb -to sleep-
meant something active or
This is Wittgenstein's emerging
philosophy. It says that
everything we say ma'es sense
only within a language3game that
establishes the rules and sets the
meaning of the terms. The
distinction between -active- and
-passive- is different when we
thin' of sleeping than when we
thin' of grammar. In grammar if
that's our language game at the
moment the passive voice has
nothing to do with being sleepy
or passive in that sense of the
%nd Wittgenstein is suggesting it
is the same with -parts.- What
counts as -parts- depends on the
We use the word -composite-
+and therefore the word
-simple-, in an enormous
number of different and
differently related ways. +Is the
colour of a s"uare on a
chessboard simple or does it
consist of pure white and pure
yellow1 %nd is white simple or
does it consist of the colours of
the rainbow1 33 Is this length of
; cm. simple or does it consist
of two parts each P cm. long1
(ut why not of one bit ? cm
long and one bit I cm. long
measured in the opposite

To show that things do not have
-absolute- parts but only parts
relative to the language game we
are playing he is now showing us
some of the different ways we
define the parts in different
language games.
I consider the last e$ample the
one of lengths most compelling.
What are the parts of a length that
is two inches1 %re there two
parts each one inch long1
Wouldn't this be different if we
measured the ob*ect in
To the philosophical "uestion/
-Is the visual image of this tree
composite and what are its
component parts1- the correct
answer is/ -That depends on
what you understand by
'composite'.- +%nd that is of
course not an answer but a
re*ection of the "uestion.,
%gain this is not Wittgenstein's
aporetic voice but his clarifying
voice. This is his own philosophy
which says that we can only
answer the "uestion -What are its
parts1- once we have negotiated
the meaning of -part- in a
particular language game.
@6. Let us apply the method of
+;, to the account in the
Theaetetus. /et us consider a
language-game for which this
account is really valid. The
language serves to describe
combinations of coloured
s"uares on a surface. The
s"uares form a comple$ li'e a
chessboard. There are red
green white and blac' s"uares.
The words of the language are
+correspondingly, -)- -<-
-W- -(- and a sentence is a
series of these words. They
describe an arrangement of
s"uares in the order/
9otice the statement -Let us
consider a language3game for
which this account is really
valid.- This is most e$plicit. This
is what he is trying to do trying to
find an illustration in which the
theory is really valid. What
theory is that1 The theory of
simples the theory that
Wittgenstein had in the Tractatus
and is also )ussell.

2 ; ?
@ A B
D 6 F

%nd so for instance the sentence
-))(<<<)WW- describes an
arrangement of this sort/

!ere the sentence is a
comple$ of names to which
corresponds a comple$ of
elements. The primary elements
are the coloured s"uares. -(ut
The sentence is
-))(<<<)WW.- It describes
the way in which the s"uares are
colored. :oesn't it seem natural
to call these different s"uares the
are these simple1-3I do not 'now
what else you would have me
call -the simples- what would
be more natural in this language3
game. (ut under
parts1 This is LW's aporetic voice
ta'ing us bac' into the fly3bottle.
other circumstances I should call
a monochrome s"uare
-composite- consisting perhaps
of two rectangles or of the
elements colour and shape. (ut
the concept of comple$ity might
also be so e$tended that a
smaller area was said to be
'composed' of a greater area and
another one subtracted from it.
=ompare the 'composition of
forces' the 'division' of a line by
a point outside it5
%nd here he ta'es us bac' out of
the fly bottle. !e is pointing to a
way to see the components of the
above figure differently. We may
see F if we insist that each part is
a s"uare but we could see the
continugous colors as constituting
a part.

So there would be two red parts
as the following figure helps to

these e$pressions shew that we
are sometimes even inclined to
conceive the smaller as the result
of a composition of greater parts
and the greater as the result of a
division of the smaller.
When I read this I see a mista'e
that I overloo'ed before. The
smaller is a division of the greater
+the smaller s"uare is half of the
larger s"uare, and the larger is a
composite of two small s"uares.
This is what I ta'e him to mean.
In other words we sometimes
divide up a part to ma'e smaller
parts or combine parts to ma'e
larger parts.
(ut I do not 'now whether to
say that the figure described by
our sentence consists of four or
of nine elements> Well does the
If the parts are determined by the
colors then there are @ parts. (ut
if the parts are determined by the
shape +s"uare, then there are F.
sentence consist of four letters or
of nine1 %nd which are its
elements the types of letter or
the letters1 :oes it matter which
we say so long as we avoid
misunderstandings in any
particular case1
Which way you count them
depends on how you define
-part.- %nd the same thing is true
for the sentence/
.ou will say there are F words if
you count each appearance of a
character as -a word.- (ut if you
count the second appearance of
each character merely a copy of
the same word then you will
count a different number of
@F. (ut what does it mean to
say that we cannot define +that
is describe, these elements but
only name them1 This might
mean for instance that when in
a limiting case a comple$
consists of only one s"uare its
description is simply the name of
the coloured s"uare.
!ere he ta'es us bac' to @B. +Cse
your ordinary way of returning
from a lin' to get bac' to this
comment after you clic' on the
above @B to pea' at @B., The
point is that if we are thin'ing of
the s"uares as the -parts- then
when we loo' at a single s"uare
we can no longer name the parts.
We can only describe the s"uare.
Isn't this the dilemma that Plato
was noticing in the Theaetetus1
!ere we might say 33 though
this easily leads to all #inds of
philosophical superstition 33
that a sign -)- or -(- etc. may
be sometimes a word and
sometimes a proposition. (ut
whether it 'is a word or a
proposition' depends on the
situation in which it is uttered or
written. #or instance if % has to
describe comple$es of coloured
s"uares to ( and he uses the
word -)- alone we shall be able
to say that the word is a
description 33 a proposition. (ut
if he is memori8ing the words
and their meanings or if he is
teaching someone else the use of
the words and uttering them in
the course of ostensive teaching
I have emphasi8ed the
parenthetical -though this easily
leads to all 'inds of philosophical
superstition- because I want to
show you how LW shows us
which voice he is using the voice
that leads us into aporia or out of
it. !e does not really e$pand on
this aporia but you can note it.
The "uestion is when is
something a sentence or a word1
We 'now but it is hard to say.
We could say that it is a sentence
when it ma'es complete sense
but a sentence does not always
ma'e complete sense and a word
sometimes does. :oesn't it1

we shall not say that they are
propositions. In this situation the
word -)- for instance is not a
description5 it names an element
but it would be "ueer to ma'e
that a reason for saying that an
element can only be named> #or
naming and describing do not
stand on the same level/ naming
is a preparation for description.
9aming is so far not a move in
the language3game 33 any more
than putting a piece in its place
on the board is a move in chess.
We may say/ nothing has so far
been done when a thing has
been named. It has not even got
a name e$cept in the language3
game. This was what #rege
meant too when he said that a
word had meaning only as part
of a sentence.
Wittgenstein steps out of this
aporia by saying that naming and
describing do not stand on the
same level that naming is
preparation for describing it is
not a move in the langauge game.
It is li'e setting up the pieces in a
game of chess.
Still this is confusing because we
don't 'now how to tell at times
what constitutes the langauge
game. It is easier when we thin'
of chess.
A4. What does it mean to say
that we can attribute neither
being nor non3being to
elements1 33One might say/ if
everything that we call -being-
and -non3being- consists in the
e$istence and non3e$istence of
conne$ions between elements it
ma'es no sense to spea' of an
element's being +non3being,5 *ust
as when everything that we call
-destruction- lies in the
separation of elements it ma'es
no sense to spea' of the
destruction of an element.
This fu88y word -being- is really
necessary here. It is the concept
that we are reaching for when we
are in an %ugustinian frame of
mind and trying to ma'e sense of
things. The idea is that if you
destroy something by brea'ing it
into its parts then the e$istence of
that thing is destroyed because its
e$istence consisted in the
relationship between its parts.
#or 0$calibur to be 0$calibur the
blade of the sword has to have a
certain relationship to the handle.
(ut what about the little piece of
the handle @ cm above the blade
does it have to have a relationship
to the rest of the handle1 There is
a way in which we cannot spea'
of the destruction of the handle.
One would however li'e to
say/ e$istence cannot be
attributed to an element for if it
(ut if the handle has to be in a
relationship to the blade in order
for 0$calibur to e$ist then
did not e$ist one could not even
name it and so one could say
nothing at all of it.
0$calibur is a handleOblade in a
certain relationship. %nd what
sense would that ma'e1 When
the blade bro'e off we would
have to say that the handleOblade
+that is 0$calibur, no longer has a
33(ut let us consider an
analogous case. There is one
thing of which one can say
neither that it is one metre long
nor that it is not one metre long
and that is the standard metre in
Paris.3(ut this is of course not
to ascribe any e$traordinary
property to it but only to mar'
its peculiar role in the language3
game of measuring with a metre3
rule.3Let us imagine samples of
colour being preserved in Paris
li'e the standard metre. We
define/ -sepia- means the colour
of the standard sepia which is
there 'ept hermetically sealed.
Then it will ma'e no sense to
say of this sample either that it is
of this colour or that it is not.
!ere he gives us two e$amples of
an ob*ect becoming the paradigm
we use to ma'e *udgments. If we
say that the standard meter in
Paris is one meter long it isn't the
same sense of -one meter- as
when we say this cloth is -one
meter long.- The standard meter
sets the standard. What would it
mean to say that it is inaccurately
measured1 It is what sets the
standard of perfection. On the
other hand we can say that the
cloth was inaccurately measured.
%nd the same is true when we
define -sepia- by giving a sample
that we will 'eep as being -sepia.-
We can put it li#e this: "his
sample is an instrument of the
language used in ascriptions of
colour In this language3game it
is not something that is
represented but is a means of
representation.33 %nd *ust this
goes for an element in language3
game +@6, when we name it by
uttering the word -)-/ this gives
this ob*ect a role in our
language3game5 it is now a
means of representation. %nd to
say -If it did not e$ist it could
have no name- is to say as much
and as little as/ if this thing did
not e$ist we could not use it in
our language3game.33
!ere he is tal'ing about the way
in which we negotiate the
meaning of the terms of our
language game. One way we do
it is by using an e$ample to define
the meaning of the term. When
we utter the word -)- in +@6, this
is actually a way of negotiating
the meaning of the term. We are
giving the ob*ect a name and a
role in our language game. It is as
though someone were to place a
stic' in Paris and say -This is a
meter- or -this is a length we shall
call 'finger'.- It sets up a meaning
for this term.
What loo#s as if it had to e+ist%
is part of the language It is a
paradigm in our language-
game, something with which
comparison is made. %nd this
may be an important
observation5 but it is none the
less an observation concerning
our language3game3our method
of representation.

It had loo'ed as though we could
not brea' the ob*ect up into
smaller components. (ut on
reflection it is *ust that we had not
named the fragments of the
compents. If the s"uare was the
basic unit and we could not thin'
of something smaller being an
element it is because we had not
learned to thin' of a fragment of
the s"uare as a component.
#or e$ample ta'e this s"uare as a
component that could be
multiplied +with different colors,
to ma'e up a comple$ composite/

(ut imagine that we learned to
see the only columns as ob*ects so
that we saw three ob*ects when
we saw the above s"uare 3 as we
might today if they were different

. . .
Perhaps we would do this if we
were used to building fences of
some sort so that we interpreted
all graphic s"uares/

in terms of fence slats. %t a
glance even if there were no
separating lines we might see it
as ? fence slats or three
components to a composite
#or e$ample in terms of slats
can't you imagine seeing that
wheresas the above s"uare was
composed of ? slats the one
below has B1


%phorism A23AF from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance commentary.,
Shawver commentary:
A2 In describing language3game
+@6, I said that the words -)- -(-
etc. corresponded to the colours of
the s"uares. (ut what does this
correspondence consist in5 in what
sense can one say that certain
colours of s"uares correspond to
these signs1 #or the account in +@6,
merely set up +sic, a conne$ion
between those signs and certain
words of our language +the names
of colours,.

What is the account in @61 It
is where LW says/

The s"uares form a
comple$ li'e a
chessboard. There are
red green white and
blac' s"uares. The
words of the language
are +correspondingly,
-)- -<- -W- -(-
and a sentence is a
series of these words.
They describe an
arrangement of
s"uares in the order/
Jsee +@6,K
=an you see how this sets up
what we are going to call the
components of the
chessboard1 We are told
specifically that -there are
red green... s"uares.- So we
have been told what we are
to consider the parts of the
33 Well it
was presupposed that the use of the
signs in the language3game would
be taught in a different way in
particular by pointing to
Our %ugustinian mythology
about language says that we
are taught how to use words
+signs, by pointing and
naming and here we are
being -taught- conte$tually
without our noticing.
7ery well5 but what does it mean to 9otice this phrase -certain
say that in the techni"ue of using
the language certain elements
correspond to the signs1 33Is it that
the person who is describing the
comple$es of coloured s"uares
always says -)- where there is a
red s"uare5 -(- when there is a
blac' one and so on1
elements correspond to the
signs.- It's a common way
of putting things but what
does it mean1 Is there a
universal meaning to this
(ut what if he goes
wrong in the description and
mista'enly says -)- where he sees
a blac' s"uare 33what is the
criterion by which this is a mista'e1
33Or does -)-s standing for a red
s"uare consist in this that when the
people whose language it is use the
sign -)- a red s"uare always comes
before their minds1
If someone mista'enly calls
a blac' s"uare -)- in what
sense is this a mista'e1 If
you have been drawn into the
language game of @6 by the
account and you recogni8e
that someone is mista'en in
calling a blac' s"uare -)-
how do you 'now this1 Is it
the case that a red s"uare
comes before your mind1
In order to see more clearly here
as in countless similar cases we
must focus on the details of what
goes on5 must loo' at them from
close to.
!ere LW is teaching us not
to accept the answer above
without e$amining what
happens in these situations.

A;. If I am inclined to suppose
that a mouse has come into
being by spontaneous
generation out of grey rags and
dust I shall do well to e$amine
those rags very closely to see
how a mouse may have hidden
in them how it may have got
there and so on. (ut if I am
convinced that a mouse cannot
come into being from these
things then this investigation
will perhaps be superfluous.
!ere LW is continuing with his
last comment from A4. 0ven if
we see that we have bought into
a certain cultural mythology that
distorts our vision this does not
mean that we can find our way
out of it. !ow do we do it1 if
we thin' that mice spontaneously
generate in gray rags and we're
convinced of this it might be
superfluous to e$amine the rags
(ut first we must learn to
understand what it is that
opposes such an e$amination of
The first thing we have to do is
understand what gets in our way
seeing what is happening.
details in

A?. Our language3game +@6,
has various possibilities5 there is
a variety of cases in which we
should say that a sign in the
game was the name of a s"uare
of such3and3such a colour. We
should say so if for instance we
'new that the people who used
the language were taught the use
of the signs in such3and3such a
way. Or if it were set down in
writing say in the form of a
table that this element
corresponded to this sign and if
the table were used in teaching
the language and were appealed
to in certain disputed cases.
!ow do we 'now that -)-
means that a particular s"uare
should be colored -red-1 We
can imagine it coming about
that -we 'now this- in a variety
of ways +other than the
insidious account we have
discovered above,. We might
say this on the basis of certain
%ugustinian language practices
that we had observed in the
tribe. That is we might have
noticed that the tribe points and
names s"uares -)- until the
children learn to do this. Or if it
were set down in writing that
red s"uares should be called
-).- Then this is how we would
'now that this is what they
should be called +imagine a
We can also imagine such a
table's being a tool in the use of
the language. :escribing a
comple$ is then done li'e this/
the person who describes the
comple$ has a table with him
and loo's up each element of the
comple$ in it and passes from
this to the sign +and the one who
is given the description may also
use a table to translate it into a
picture of coloured s"uares,.
The comple$ is li'e the grid we
say in @6 it is a cluster of
elements arranged in a
predefined way. !ow will one
describe the comple$ to another
who must arrange say a copy1
One might loo' at the comple$
and then loo' up each element
in a table.
This table might be said to ta'e
over here the role of memory
and association in other cases.
+We do not usually carry out the
order -(ring me a red flower- by
loo'ing up the colour red in a
Whereas ordinarily we rely on
our memories to recogni8e
simple colors li'e -red- we do
sometimes use a tool such as
this when we are trying to get
the e$act shade.
table of colours and then
bringing a flower of the colour
that we find in the table5 but
when it is a "uestion of choosing
or mi$ing a particular shade of
red we do sometimes ma'e use
of a sample or table.,

If we call such a table the
e$pression of a rule of the
language3game it can be said
that what we call a rule of a
language3game may have very
different roles in the game.
Wittgenstein is setting up this
table as a model of rule in a
language3game and he will use
this model in subse"uent te$t.

A@. Let us recall the 'inds of
case where we say that a game
is played according to a definite
% definite rule is one that is set
out e$plicitly that everyone
agrees on.
The rule may be an aid in
teaching the game. The learner
is told it and given practice in
applying it.

Say I e$plain before we begin
that the rule is that when you
type your comments you should
enclose them in brac'ets with
your initials. The rule is an aid I
devise in assisting our study but
it is not a part of the language3
game in the sense that we could
easily devise other devices that
would wor' *ust as well. It
would not change the playing of
the language game in any
important way if we used a color
code to 'eep trac' of who wrote
which comment.
33Or it is an instrument of the
game itself.

(ut an rule that is an instrument
of the game itself cannot be
changed without changing the
game. If the rule is that we can
as' each other "uestions and get
answers then it would change
our language game if we
changed the rule.
33Or a rule is employed neither
in the teaching nor in the game
itself5 nor is it set down in a list
of rules. One learns the game by
watching how others play. (ut
we say that it is played
according to such3and3such
rules because an observer can
read these rules off from the
practice of the game3li'e a
natural law governing the play.
33(ut how does the observer
distinguish in this case between
players' mista'es and correct
play1 33There are characteristic
signs of it in the players'
behaviour. Thin' of the
behaviour characteristic of
correcting a slip of the tongue. It
would be possible to recogni8e
that someone was doing so even
without 'nowing his language.
Imagine a new reader noticing
that everyone encloses their
comments within brac'ets that
contain their initials and
conforming to this implicit rule.
In that case too can we not say
that this is -playing according to
the rules-1
(ut in this case how do we
'now when people are playing
correctly according to the rules1
Perhaps by the way people
correct themselves or other such
recogni8eable signs that people
show they feel they have
violated the rules even the
implicit rules +apologies1,

AA. -What the names in language
signify must be indestructible5
for it must be possible to
describe the state of affairs in
which everything destructible is
destroyed. %nd this description
will contain words5 and what
corresponds to these cannot then
be destroyed for otherwiseQthe
words would have no meaning.-
I must not saw off the branch on
which I am sitting.
AA. !ere LW is spea'ing again
with his aporetic voice from
within the fly bottle. (ut there
is you can see +can you not1, a
certain distance from this
aporia. !e is listening to what
he is inclined to say here.
!e is inclined to say that there
must be ob*ects in the world
that are simple and
indestructible +which are either
true or false,. 0ven if I destroy
0$calibur it must be the case
that I at least have something
left that I can say is destroyed
fragments smo'e something.
If we do not have these simple
indestructible truths that we can
point to and name then how can
we continue1 Our entire logic
depends on this. Or so it seems
from within the fly bottle.
One might of course ob*ect at
once that this description would
have to e$cept itself from the
That is if we destroyed
everything and then described
the destruction we could not
destroy the description itself.
33(ut what corresponds to the
separate words of the description
and so cannot be destroyed if it
is true is what gives the words
their meaning 333 is that without
which they would have no
meaning. In a sense however
this man is surely what
corresponds to his name. (ut he
is destructible and his name
does not lose its meaning when
the bearer is destroyed
LW is still within his aporetic
voice e$pressing wonder at
these parado$es he is
entertaining. In this frame of
mind it seems that what
corresponds to the separate
words cannot be destroyed if
the words are true -The =hair
is in the corner.- If the words
are true then the chair cannot
have been crushed until it is no
longer a chair. Still and here's
the perple$ity a name still has
meaning once the ob*ect is
destroyed. !ow can this be1
33%n e$ample of something
corresponding to the name and
without which it would have no
meaning is a paradigm that is
used in conne$ion with the name
in the language3game.

The standard meter in Paris
gives us an e$ample of this
paradigm. Or a sample of
-sepia- that serves to define our
naming of colors. Samples li'e
this can give meaning to a
word. %s' yourself/ !ow long
as a griset1 If we had a sample
in Paris that told us that word
would have meaning.
AB. (ut what if no such sample
is part of the language and we
bear in mind the colour +for
instance, that a word stands for1
33-%nd if we bear it in mind then
it comes before our mind's eye
when we utter the word. +sic, So
if it is always supposed to be
possible for us to remember it it
This is LW's aporetic voice.
9otice that he often puts his
aporetic voice in "uotes but he
is inconsistent. I put a +sic,
after the -word- because I
believe it should have a
"uestion mar' there. This is
the cultural reasoning that puts
the indestructible simple in the
must be in itself indestructible.-
mind. It is what gives Plato his
essences or eternal ideas.
33(ut what do we regard as the
criterion for remembering it

!ere LW is "uestioning his
own aporetic voice. This is a
significant "uestion and he will
ma'e much of it in other
conte$ts. If we have a sample
of -red- say in our minds and
no e$ternal sample how do we
'now that we have remembered
the right color1 The color that
-red- is1 =an you see that this
would be problematic1 .ou
can hold the red sample up to
the apple and see that the apple
is the same color but that
wor's because the red sample
you are using is dependable.
What if you have gotten
confused and the red sample in
your mind is now distorted you
are thin'ing of it as -rust.-
!ow would you 'now1
33When we wor' with a sample
instead of our memory there are
circumstances in which we say
that the sample has changed
colour and we *udge of this by
memory. (ut can we not
sometimes spea' of a dar'ening
+for e$ample, of our memory3
image1 %ren't we as much at the
mercy of memory as of a sample1
+#or someone might feel li'e
saying/ -If we had no memory
we should be at the mercy of a
sample-., 33Or perhaps of some
chemical reaction. Imagine that
you were supposed to paint a
particular colour -=- which was
the colour that appeared when the
chemical substances E and .
combined.3Suppose that the
colour struc' you as brighter on
!ere he is further e$ploring the
"uestion of whether we can rely
on memory as if it were a
sample. We do sometimes
notices that colors have
changed he tells us but we do
not entirely trust our
observations. So if we rely on
memory as a sample we often
do not feel very secure about

one day than on another5 would
you not sometimes say/ -I must
be wrong the colour is certainly
the same as yesterday-1 This
shews that we do not always
resort to what memory tells us as
the verdict of the highest court of

AD. -Something red can be
destroyed% but red cannot be
destroyed and that is why the
meaning of the word 'red' is
independent of the e$istence of
a red thing.-
The aporetic voice. %gain the
emphasis is mine. This is a
paradigm +sample, case of the
Platonic3%ugustinian muddle.
What is it that cannot be
destroyed1 The color1 What
color1 In what way does the
color e$ist apart from things that
are so colored1
3=ertainly it ma'es no sense to
say that the colour red is torn up
or pounded to bits. (ut don't we
say -The red is vanishing-1 %nd
don't clutch at the idea of our
always being able to bring red
before our mind's eye even
when there is nothing red any
more. That is *ust as if you
chose to say that there would
still always be a chemical
reaction producing a red flame.3
#or suppose you cannot
remember the colour any
more.53When we forget which
colour this is the name of it
loses its meaning for us5 that is
we are no longer able to play a
particular language3game with
it. %nd the situation then is
comparable with that in which
we have lost a paradigm which
was an instrument of our
!ere's LW's clarifying voice.
!e is not really giving us an
answer here to the above
"uestion but he is directing our
attention. If we are inclined to
say +confusedly, that the red
would e$ist because it would
still e$ist in our minds +because
we could imagine a red s"uare
still, then this neglects the fact
that we sometimes cannot recall
the color. Suppose you suffered
brain damage and it did not
destroy your color vision but
you could no longer remember
which color was which. Would
red then still e$ist1

A6. -I want to restrict the term
'name' to what cannot occur in
the combination 'E e$ists'.
33Thus one cannot say ')ed
e$ists' because if there were no
red it could not be spo'en of at
%gain LW is using the "uotes to
indicate his aporetic voice. This
is the aporetic voice trying to
patch things up so that they
wor' as our cultural picture says
that they should. %ccording to
this patch up *ob we are going
to say that the word -red- will
lose its meaning when there are
no red ob*ects. Will this wor'1
33(etter/ If -E e$ists- is meant
simply to say/ -E- has a
In other words if the statement
-)ed e$ists- is true then this
means that -)ed- has a
3then it is not a proposition
which treats of E but a
proposition about our use of
language that is about the use
of the word -E-.
(ut notice this proposition does
not tal' about the e$istence of
-red-. It is a move in setting up
the language game. It has
nothing to do with the e$istence
of red apart from this new
language game.
It loo#s to us as if we were
saying something about the
nature of red in saying that the
words '0ed e+ists' do not
yield a sense 9amely that red
does e$ist 'in its own right'.

Important passage In 2;; LW
notices that our grammar is
lac'ing in a certain 'ind of
perspecuity that would enable us
to more easily see what is going
on. !ere it is. The phrase -)ed
e$ists- can be either a
negotiation of the meaning of
the term -)ed e$ists- or it can
be a statement about the world
33 but if it's a statement about
the world it has to be within a
particular language game.
We get confused however
when we see that the statement
-)ed e$ists- ma'es a 'ind of
sense to it. The sense it seems
to ma'e when we conflate the
two possible uses of this phrase
is that -)ed- e$ists apart from
any ob*ect that is red. Still this
seems perple$ing to us. It is
hard to imagine how red e$ists.
This is our aporia here.
The same idea 33that this is a
metaphysical statement about
red 33finds e$pression again
when we say such a thing as that
red is timeless and perhaps still
more strongly in the word
That is there are many ways to
e$press this metaphysical
thought that -red e$ists- beyond
red ob*ects and particular
language games. Sometimes we
say that it is -timeless- or
(ut what we really want is
simply to ta'e -)ed e$ists- as
the statement/ the word -red-
has a meaning. Or perhaps
better/ -)ed does not e$ist- as -
')ed' has no meaning-.
In other words if we are
tempted to say -red e$ists- then
we are pointing out that the
word red has a meaning. Or if
we say that -grue- does not
e$ist- this is a way of saying
that the word -grue- has no
Only we do not want to say that
that e$pression says this but that
this is what it would have to be
saying if it meant anything. (ut
that it contradicts itself in the
attempt to say it 33*ust because
red e$ists 'in its own right'.
Whereas the only contradiction
lies in something li'e this/ the
proposition loo#s as if it were
about the colour% while it is
supposed to be saying
something about the use of the
word 'red'.
(ut it seems as though the
statement -)ed e$ists- is
asserting a truth about red not
*ust giving us the rules of the
language +that the word 'red' has
meaning. The formulation fools
us because it is so similar to the
formulation we would use if we
were tal'ing about a thing and
not about meaning as if I would
say -The document you have
been loo'ing for I have found
out that it e$ists- it would be
clear that I am not tal'ing about
word definitions but about the
document e$isting. Still the
formulations seem so similar.
33In reality however we "uite
readily say that a particular
colour e$ists5 and that is as much
as to say that something e$ists
that has that colour. %nd the first
(ut our language does not ma'e
a distinction between these ways
of using the phrase -red e$ists.-
Within the rules of our
language both uses are e"ually
e$pression is no less accurate
than the second5 particularly
where 'what has the colour' is
not a physical ob*ect.
AF. -% name signifies only
what is an element of reality.
What cannot be destroyed5 what
remains the same in all
The %ugustine's voice again.
This voice tells us/ If -red e$ists-
it signifies something that cannot
be destroyed.
33 (ut what is that1 33Why it
swam before our minds as we
said the sentence> This was the
very e$pression of a "uite
particular image/ of a particular
picture which we want to use.
#or certainly e$perience does
not shew us these elements. We
see component parts of
something composite +of a
chair for instance,. We say that
the bac' is part of the chair but
is in turn itself composed of
several bits of wood5 while a
leg is a simple component part.
We also see a whole which
changes +is destroyed, while its
component parts remain
unchanged. These are the
materials from which we
construct that picture of reality.
This is the aporetic voice
spea'ing. It says /isn't there a
way in which this seems
compelling1 #rom within the fly
bottle1 :oesn't it sometimes
happen that when you say -chair-
you see something li'e a chair
flash before your mind's eye1
Well then maybe we should say
that this ghostly image is what
the word 'chair' refers to. It is the
idea perhaps that Plato had in
mind when he constructed his
theory of ideas.

%phorism B43B@ from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance
Shawver commentary:
B4. When I say/ -&y broom is in
the corner-3is this really a
statement about the broomstic'
and the brush1
What else could a statement li'e
this be1 )emember that in A2 LW
introduced the notion that we can
introduce the account into the
remar's so that this account
defines the terms to be used sets
up the language game rules.
Well it could at any
rate be replaced by a statement
giving the position of the stic' and
the position of the brush. %nd this
statement is surely a further
analysed form of the first one.
This is the voice of tradition
noticing that the word -broom-
could be replaced with something
li'e -brush plus stic'-1 This
phrase -brush plus stic'- it says
is an analy8ed form of -broom.-
3(ut why do I call it -further
The voice of aporia as's why this
is so.
33Well if the broom is there that
surely means that the stic' and
brush must be there and in a
particular relation to one another5
and this was as it were hidden in
the sense of the first sentence and
is e$pressed in the analysed
The voice of tradition answers and
gives its reasons. This T voice
says in effect -(room- and
-brush plus stic'- are the same
thing e$cept -brush plus stic'-
gives a more detailed listing of
what we actually have.
Then does someone who says that
the broom is in the corner really
mean/ the broomstic' is there and
so is the brush and the broomstic'
is fi$ed in the brush1
Perhaps this will remind you of an
earlier discussion of whether
-Slab>- in languge game ; really
means -(ring me the slab>- +cf
2F, It is in ways li'e this that
Wittgenstein teaches us going
over these points in one conte$t
and then in another using a
different versions of a basic model
to familari8e us with the problem
in a variety of cases.
3If we were to as' anyone if he
meant this he would probably say
that he had not thought specially
of the broomstic' or specially of
the brush at all. %nd that would be
the right answer for he meant to
spea' neither of the stic' nor of
the brush in particular. Suppose
that instead of saying -(ring me
the broom- you said -(ring me
the broomstic' and the brush
which is fitted on to it.->3Isn't the
answer/ -:o you want the broom1
Why do you put it so oddly1- Is he
going to understand the further
analysed sentence better1
The point is that the spea'er who
had as'ed for the broom was
as'ing for the gestalt whole not
the parts even if they were
attached to each other. .ou don't
see a person's face by noticing the
constellation of features. The
whole is more than the sum of its
individual parts.
Is he going to understand the
further analysed sentence better13
This sentence one might
say achieves the same as the
ordinary one but in a more
roundabout way.
%ctually it might be harder to
understand. Imagine it/ -Would
you had me the brush attached to
the broomstic'1-
3Imagine a language3game in
which someone is ordered to bring
certain ob*ects which are
composed of several parts to
move them about or something
else of the 'ind. %nd two ways of
playing it/ in one +a, the
composite ob*ects +brooms chairs
tables etc., have names as in
+2A,5 in the other +b, only the parts
are given names and the wholes
are described by means of them.3
In what sense is an order in the
second game an analysed form of
an order in the first1 :oes the
former lie concealed in the latter
and is it now brought out by
True the broom is ta'en to
pieces when one separates
broomstic' and brush5 but does it
Poof> There goes our great
distinction between names and
descriptions. If we call the ob*ect
a broom then it is a description to
say it is a brush with a broomstic'
attached because the composite
ob*ect has a name +i.e. -broom-,.
(ut if only the parts have names
then the whole must be described
by the means of the parts and each
of the parts become names.
So what loo'ed li'e a comment
about the unanaly8ability of the
broom +or the brush, is really a
comment about whether I can
further analy8e the language. If
invent ways to name more
infintesimal aspects of the ob*ect
then the ob*ect can be analy8ed
further. The s"uares can be
divided into triangles and then
that the order to bring the broom
also consists of corresponding

each s"uare is a composite of
B2. -(ut all the same you will
not deny that a particular order
in +a, means the same as one
in +b,5 and what would you call
the second one if not an analysed
form of the first1-
The %ugustinian voice again. =an
you see where he's coming from1
Practically spea'ing it seems that
as'ing for the 'brush' and the
'broomstic'' means the same thing
as as'ing for the broom. If the
instructions were followed in each
case the same ob*ect would be
3=ertainly I too
should say that an order in +a, had
the same meaning as one in +b,5
or as I e$pressed it earlier/ they
achieve the same. %nd this means
that if I were shewn an order in +a,
and as'ed/ -Which
order in +b, means the same as
this1- or again -Which order in +b,
does this contradict1- I should
give such3and3such an answer. (ut
that is not to say that we have
come to a general agreement
about the use of the e$pression -to
have the same meaning- or -to
achieve the same-. #or it can be
as'ed in what cases we say/
-These are merely two forms of
the same game.-
I have corrected the electronic
version of our te+t which has the
word 'strewn' when it should
have had -shewn- when in
%merican is -shown.-
(ut the "uestion is *ust because
they have the same practical effect
of resulting in the broom being
fetched doesn't mean that they are
the same game. I can get you to
turn around by saying -turn
around- perhaps but I can li'ely
achieve the same effect by saying
your name.
B;. Suppose for instance that the
person who is given the orders in
+a, and +b, has to loo' up a
table co3ordinating names and
pictures before bringing what is
Let this remind you of the table
discussion for the color of the grid
in A?3AB.
:oes he do the same when he
carries out an order in +a, and the
corresponding one in +b, and
no. .ou may say/ -The point of
the two orders is the same-. I
Why are we tempted to say
however that the point of a lamp
is that it gives light1 :on't you
thin' we are1 .et in a given case
in a particular situation the point
should say so too.3(ut it is not
everywhere clear what should be
called the 'point' of an order.
+Similarly one may say of certain
ob*ects that they have this or that
purpose. The essential thing is that
this is a lamp that it serves to give
light53that it is an ornament to the
room fills an empty space etc. is
not essential. (ut there is not
always a sharp distinction between
essential and inessential.,
may be entirely different. We are
inclined to thin' of a paradigm
case +as if the situation has been
set up for us, and ignore
alternative possibilities. We
recogni8e that they are there but
we let them slip under the rug to
'eep things simple +or for some
Why do we do this1
B?. To say however that a
sentence in +b, is an 'analysed'
form of one in +a, readily seduces
into thin'ing that the former is the
more fundamental form5 that it
alone shews what is meant by the
other and so on.
%h here it is again. The account
in the language set us up. It is the
same point he made in A2
#or e$ample we thin'/ If you
have only the unanalysed form
you miss the analysis5 but if you
'now the analysed form that gives
you everything.
The %ugustinian voice says that
the more minute the analysis the
more accurate things are.
3(ut can I not say that an aspect of
the matter is lost on you in the
latter case as well as the former1
(ut the level of description is *ust
different. Something may be
gained but something is also lost.
We lose the forest for the trees.

This relates to the point in 2F in
which we compared the language
game that said that in +;, -Slab>-
was not an abbreviated form of
-(ring me a slab>- anymore that
-(ring me a slab>- was a
lengthened form of -slab>-
9evertheless we are somehow
seduced into thin'ing that -Slab>-
is abbreviated.
(ut in each case we have a
different language game a
different -form of life.-
B@. Let us imagine language game
+@6, altered so that names signify
not monochrome s"uares but
rectangles each consisting of two
such s"uares. Let such a rectangle
which is half red half green be
called -C-5 a half green half white
one -7-5 and so on. =ould we not
imagine people who had names
for such combinations of colour
but not for the individual colours1
Thin' of the cases where we say/
-This arrangement of colours +say
the #rench tricolor, has a "uite
special character.-

Imagine it. a sentence li'e C 7
7 C would result in the grid being
colored in thusly/

. .
=ouldn't we imagine a culture
having such names1 Thin' of the
#rench flag or any flag and
imagine these rectangles loo'ing
li'e flags one flag on top of
In what sense do the symbols of
this language3game stand in need
of analysis1 !ow far is it even
possible to replace this language3
game by +@6,13It is *ust another
language3game5 even
though it is related to +@6,.

%h but you say it would be so
inconvenient> yes in 0nglish it
would be. (ut what if nothing
really mattered but the flags.
Women wore greenQwhite +or C
flags, and men wore greenQred or
some other division between
classes of people were designated
li'e this. %side from these flags
there was no concern with color.
.es it would be a different form
of life and the person who
thought that these different
statements were translatable to
statements that coded these flags
not as units +C or 7 but as s"uares
<reen White and )ed, would be
missing the forest for the trees.
%phorism BA3BF from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is
inserted by Shawver to
enhance commentary.,
Shawver commentary:
BA. !ere we come up
against the great
"uestion that lies
behind all these
someone might ob*ect
against me/ -.ou ta'e
the easy way out> .ou
tal' about all sorts of
language3games but
have nowhere said
what the essence of a
language3game and
hence of language is/
what is common to all
these activities and
what ma'es them into
language or parts of
language. So you let
yourself off the very
part of the
investigation that once
gave you yourself most
headache the part
about the general form
of propositions and of

We have now shifted to a new topic that he
announces straightforwardly. The topic is
presented in the form of an %ugustinian voce
or a -somone.- This someone wants
Wittgenstein to defie the essence of the
concept of a language game. 9otice within
the %ugustinian frame the 'essence- is e"ual
to -what is common to all these activities.-
This idea goes bac' to Plato who tal's of the
essence of various things or the
transcendental idea behind their various
sensual manifestations.
So the "uestion is/ What is the essence of a
language game1 and hence to all of
language1 What is the essence of language1
%lso notice that in the last part of this
passage the 7oice reminds LW that this
search for the essence was once something
that he tried very hard to do and it gave him
considerable trouble.
%nd this is true.3
Instead of producing
something common to
all that we call
It is true LW is saying that he hasn't yet
presented this essence that is common to all
language +or all language games,. !is answer
here in this passage is very famous and it is a
language I am saying
that these phenomena
have no one thing in
common which ma'es
us use the same word
for all3but that they
are related to one
another in many
different ways. %nd it
is because of this
relationship or these
relationships that we
call them all
-language-. I will try
to e$plain this.
powerful move in developing the
Wittgensteinian framewor'. (efore this
move it seems imperative that we define the
essence of what we are tal'ing about. 9ow
LW is going to show us another way to see
BB. =onsider for
e$ample the
proceedings that we
call -games-. I mean
board3games card3
games ball3games
Olympic games and so
on. What is common to
them all1 33 :on't say/
-There must be
something common or
they would not be
called 'games' -3but
loo' and see whether
there is anything
common to all. 33 #or
if you loo' at them you
will not see something
that is common to all
but similarities
relationships and a
whole series of them at
that. To repeat/ don't
thin' but loo'> 33

This aphorism has a little different structure
than some of the others that we are reading.
!ere LW is e$plicitly guiding our reading and
he does such a good *ob of it I am not going
to offer much commentary.
(ut a few notes/ 9ow notice your inclination
to say certain things has become the
Wittgensteinian voice. 9ow we can begin to
listen to this voice within ourselves. The
voice spea's within us when we want to say
-there must be something common among
-games.- There must be an essence if we
have a concept.
LW says in a manner of spea'ing -don't say
to yourself that this must be the case and then
give yourself a headache trying to see what is
not there. Let's loo' at specific 'ind of cases
and as' if the essence is there in those cases.
Loo' through these aphorisms while putting
the point that he is ma'ing out of mind. :on't
thin' so much or ponder what you're loo'ing
for *ust loo' at your memories and
understanding of games and detail what you
Loo' for e$ample at
board3games with
their multifarious
(oard games what are some1 =onsider
chess of course but thin' also of monopoly.
9ow pass to card3
games5 here you find
many correspondences
with the first group
but many common
features drop out and
others appear.

=ard games. What about po'er1 %nd what
about Old &aid. )emember that children's
card game1 !ow are these card games ali'e
and different from each other1 %nd how do
they compare with board games1 What about
the element of strategy1 Or how many
players can play and whether or not there is a
single winner or as in &onopoly +I believe,
there are different degrees of winning.
When we pass ne$t to
ball3games much that
is common is retained
but much is lost.33 %re
they all 'amusing'1
=ompare chess with
noughts and crosses.
Or is there always
winning and losing or
competition between
players1 Thin' of
patience. In ball games
there is winning and
losing5 but when a
child throws his ball at
the wall and catches it
again this feature has
disappeared. Loo' at
the parts played by
s'ill and luc'5 and at
the difference between
s'ill in chess and s'ill
in tennis.
Thin' of the way one wins or loses in tennis.
Winning is hierarchical. One can win a point
but lose the game. One can win the game but
lose the set. %nd one can win the set but lose
the match. One can win the match but lose
the tournament. =ompare this with baseball
+also hierarchical, or with chec'ers. %nd
howabout board games that revolve around a
throw of the dice1
Thin' now of games
li'e ring3a3ring3a3
roses5 here is the
element of amusement
but how many other
characteristic features
have disappeared>
sometimes similarities
of detail.

Then we have children's ritual games. :o
they have a winner1 What about drop the
han'erchief1 Or London (ridge is falling
down1 !ow about -spin the bottle.-1 %re
you winning or losing if the bottle stops
pointing to you1
What about *ac's1 Gac's is a girls' game that
was popular when I was a child and I was into
the game. .ou have 24 little ob*ects called
-*ac's- that you toss onto the ground as the
other girls sit in a circle. Then each girl has a
turn. She starts with a ball in her preferred
hand and she tosses the ball up and lets it
bounce and before it bounced again she pic's
up one *ac' and then catches the ball before it
bounces again. She does that with each *ac'.
Then she does -twosees- which means she
pic's up two *ac's in one sweep. She
continues that until she has done all ten
*ac's. Then if she completes that round
without difficulty she starts again with a
more difficult rule. Perhaps she doesn't let
the ball bounce at all or she not only pic's up
the *ac's but she puts them in a particular
place before she catches the ball. There are a
few of these rounds that are already invented
but it is common for the winning player to
invent the ne$t game.
!ow does -*ac's- compare with chess1 Or
with ring3a3ring3o3roses1 !ow are they
different1 !ow does it compare with tennis1
Or %merican football1
%nd we can go through
the many many other
groups of games in the
same way5 can see how
similarities crop up
and disappear.

:on't children invent games on the spot1 See
who can spit the furtherest1 Or see who can
solve a particular pu88le first1 Or who can
follow a rule the best +thin' of Simon Says,.
%nd the result of this
e$amination is/ we see
a complicated networ'
of similarities
overlapping and cries3
crossing/ sometimes
overall similarities.

%nd what you'll find I thin' if you go
through a careful study of these various types
of games is that there are similarities and
differences. Po'er is li'e chess in certain
ways. They both have clear rules and the
winner is li'ely to have practice and s'ill.
(ut they are different in some ways too and
if you loo' at how they are different you'll
find other games that are not different in these
ways but different in other ways.
BD. I can thin' of no !ere is the 'ey move and the new metaphor
better e$pression to
characteri8e these
similarities than
resemblances-5 for the
various resemblances
between members of a
family/ build features
colour of eyes gait
temperament etc. etc.
overlap and cries3cross
in the same way.3%nd I
shall say/ 'games' form
a family.

that LW e$tends to replace the old Platonic
metaphor of essence. The concept is one of
-family resemblance.-

9otice %l
and Gac'
have the
while 0lmer
and (ob
have the
same ears
and %l and
(ob have the
same smile.
There is no
among them
yet they all
each other.
Wittgenstein #amily

%nd for instance the
'inds of number form
a family in the same
way. Why do we call
something a
-number-1 Well
perhaps because it has
with several things that
have hitherto been
called number5 and this
can be said to give it
an indirect relationship
to other things we call
I suppose what LW means here is that we call
positive numbers negative numbers real
numbers or a se"uence of characters
+abc...8, numbers +see 6,. !ow are these
-numbers- li'e and unli'e a series of
characters that we would not consider
%lso consider phone numbers and the
numbers on football *erseys social security
numbers numbers that are ran's verus
numbers that can be added and subtracted.
Or let's ta'e an e$ample that re"uires less
the same name. %nd
we e$tend our concept
of number as in
spinning a thread we
twist fibre on fibre.
%nd the strength of the
thread does not reside
in the fact that some on
e fibre runs through its
whole length but in
the overlapping of
many fibres.

mathematical sophistication. Ta'e the word
-food.- Imagine a plate of food composed of
only vegetables or a food concoction made of
cheese and tomato sauce or food for the
dogs or for the goldfish. %lso imagine
spoiled food or raw food or petrified food. Is
there some single feature in these foods that
runs through all of them1 Thin' of artificial
food +li'e wa$ apples, and playfood +for
children's tea,. %nd don't say that the single
feature is that they are all related to eating
because that is a way we frame -wa$ food-
and -play food- but it is not a characteristic of
this -food.-
%nd what a closer e$amination shows is that
even if there isn't a single thread that runs
through everything +and there may be in some
cases of course, there is a family
resemblance between these different items.
Some are edible. Some are animal flesh.
Some are vegetable. (ut there need not be a
single aspect that is common to all the
=an you thin' of another e$ample that can be
analy8ed in this way1 Ta'e the concept of
-thought.- :o all the different acceptable
uses of this term have a common feature1 Or
ta'e the concept of -nothing.- Is the meaning
of -nothing- the same in these two sentences/
2. There is nothing in the bo$.
;. There is nothing for me to do.

(ut if someone wished
to say/ -There is
something common to
all these constructions3
namely the dis*unction
of all their common
properties- 33I should
reply/ 9ow you are
only playing with
words. One might as
This is an important passage too. It points to
the tric's we play to 'eep ourselves in the fly3
well say/ -Something
runs through the whole
thread3 namely the
overlapping of those
B6. -%ll right/ the
concept of number is
defined for you as the
logical sum of these
individual interrelated
concepts/ cardinal
numbers rational
numbers real numbers
etc.5 and in the same
the concept of a game
as the logical sum of a
corresponding set of
!ere's the %ugustinian voice again. It
always seems to have a combac'. To return
to the concept of -number- remember LW
had said that there need not be a single
common feature in all -number- systems.
--It need not be so
.or I can give the
concept 'number'
rigid limits in this
way% that is% use the
word 'number' for a
rigidly limited
concept% but I can
also use it so that the
e+tension of the
concept is not closed
by a frontier %nd this
is how we do use the
word -game-. #or how
is the concept of a
game bounded1

!ere is another important passage.
Wittgenstein is pointing to the way in which
we can locally and provisionally define a
concept. !ow do we do this1 In numerous
ways. Sometimes we set things up e$plicitly.
We say -I am using the word number here to
mean 'rational number.'- %nd sometimes this
slips in without our awareness. +We studied
this A23AF and see especially A2,.
What still counts as a
game and what no
longer does1
I thin' we can count this as the %ugustinian
=an you give the
boundary1 9o.
It is very hard to delineate what the
boundaries of a game are to define it so that
it includes both tic3tac3toe and )ugby.
.ou can draw one5 for (ut in a local and provisional conte$t you
none has so far been
might say -(y game I mean something in
which one 'eeps score and there is a definite
+(ut that never
troubled you before
when you used the
word -game-.,
(ut ordinarily you use the word -game-
without trying e$plicitly to define it locally
and provisionally. .ou *ust say -Is this some
'ind of a game1- and you ta'e it that people
will understand you.
-(ut then the use of
the word is
unregulated the 'game'
we play with it is
9ow the %ugustinian feels uncomfortable
with where we're going. It seems we need to
'eep things more tied down than this.
It is not everywhere
circumscribed by rules5
but no more are there
any rules for how high
one throws the ball in
tennis or how hard5
yet tennis is a game for
all that and has rules
The rules of the game can't control every last
detail of the action. There is always a
considerable amount action that is beyond the
rules of the game.
BF. !ow should we
e$plain to someone
what a game is1
If we don't have a common thread running
through everything we call a -game- it seems
very chaotic> !ow on earth do we teach
people to use this term -game-1
I imagine that we
should describe games
to him and we might
add/ -This and similar
things are called
'games' -. %nd do we
'now any more
about it ourselves1 Is it
only other people
whom we cannot tell
e$actly what a game
Still don't we teach this term -games- to
children1 %nd don't they learn it1 =an it
really be as diffficult as all that if we manage
to teach it so easily1
3(ut this is not
ignorance. We do not
'now the boundaries
because none have
been drawn. To repeat
we can draw a
boundary3for a special
The term -game- is not a difficult term for a
child to learn and the fact that it seems that it
should be is a flag for this being a confusion
left over from our %ugustinian muddle.
The situation is that we imagine that we have
one term here and the different senses are *ust
purpose. :oes it ta'e
that to ma'e the
concept usable1 9ot at
all> +0$cept for that
special purpose., 9o
more than it too' the
definition/ 2 pace R DA
cm. to ma'e the
measure of length 'one
pace' usable. %nd if
you want to say -(ut
still before that it
wasn't an e$act
measure- then I reply/
very well it was an
ine$act one.3Though
you still owe me a
definition of
variations on a common theme but in
practice we ta'e these vague concepts that are
loosely defined and we tie them down to
more particular definitions. It *ust ta'es a
moment to do this and the practice is all
around us. It is *ust that we fail to notice that
we do this. We have a theory of terms having
essential meanings +based on transcendental
essences, and this belief in the theory of
language is so strong we simply overloo' the
way in which we negotiate the language that
we use when other people do it and when we
do it ourselves.
LJSomeone says to me/
-Shew the children a
game.- I teach them
gaming with dice and
the other says -I didn't
mean that sort of
game.- &ust the
e$clusion of the game
with dice have come
before his mind when
he gave me the
This is a footnote in which LW reminds us
how we teach this ostensibly difficult concept
of -game.- 9otice how we have practices of
continuously clarifying our local and
provisional meanings.
%phorism D43DA from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance
Shawver commentary:
9ow we will dip into the reason
that our local negotiation of
language games +the setting up of
the accounts in +A2, through +BF,
do not always wor' and why we
have disagreements and
confusions. What is it about
langauge that ma'es it difficult for
us to accept any definition of things
at all1
D4. -(ut if the concept 'game' is
uncircumscribed li'e that you
don't really 'now what you mean
by a 'game'.-

!ere is the %ugustinian +actually
his positivist descendant,
spea'ing. The point is simple. .ou
need to define terms to be able to
use them. (ut Wittgenstein isn't
defining -language game- in any
clear way recall that captures the
essence of language games.
Language games form a family
resemblance. There is no essence
to tie them together.
33 When I give the description/
-The ground was "uite covered
with plants- 33do you want to say
I don't 'now what I am tal'ing
about until I can give a definition
of a plant1
(ut notice mostly we don't have
ready definitions for terms. 0ven
when we set up the language game
by giving accounts we don't
typically 'now that we are doing it.
We all learned to tal' "uite a bit
before we were even able to
generate definitions for the terms
we used.
&y meaning would be e$plained
by say a drawing and the words
-The ground loo'ed roughly li'e
Imagine it. I say -The ground
loo'ed roughly li'e this- as I point
to a front yard of someone's. (ut
this-. Perhaps I even say -it
loo'ed e$actly li'e this.-3Then
were *ust this grass and these
leaves there arranged *ust li'e
this1 9o that is not what it
means. %nd I should not accept
any picture as e$act in this
what does -this- mean. )ecall our
problem in defining -this- before.
Or pointing to an effort
to define it. What am I pointing to
here1 This is the whole problem
with teaching ostensive definitions
that we faced in 2324 and that
Wittgenstein elucidated in his
remar's ;6 and ;F.. Gust as it is
hard to tell if I am pointing to the
circle or the color of the circle so it
is hard to tell what I am pointing to
here. %nd I said that the similarity
beteween this front yard and the
one one I am describing is rough
but rough in what way1 =an I be
e$act in how it is rough1 Without
ma'ing this -rough- e$planation an
e$act one1
D2. One might say that the
concept 'game' is a concept with
blurred edges.3
!ere LW brea's his usual form and
he begins this aphorism in his own
voice. !e is suggesting a way to
thin' about things that will be
challenged in the ne$t passage.
-(ut is a blurred concept a
concept at all1-3
There's the challenge// The
imaginary interlocutor says in
effect -:on't I have to pin my
meaning down in order to be
Is an indistinct photograph a
picture of a person at all1 Is it
even always an advantage to
replace an indistinct picture by a
sharp one1 Isn't the indistinct one
often e$actly what we need1
The "uestion is whether you want
to call an indistinct picture a
-picture.- <enerally I thin' we do
unless it is more than *ust a little
indistinct. (ut with concepts don't
we often operate with -indistinct
meanings- of terms1 %nd in the
case of -language game- isn't that
what we need1
#rege compares a concept to an
area and says that an area with
vague boundaries cannot be
called an area at all. This
presumably means that we
cannot do anything with it.
Well here's a real case of the
positivist descedent who ma'es the
complaint that forms the problem
for this aphorism to handle.
3(ut is it senseless to say/ -Stand =learly we do this all the time. -I'll
roughly there-1
Suppose that I were standing
with someone in a city s"uare
and said that. %s I say it I do not
draw any 'ind of boundary but
perhaps point with my hand3as if
I were indicating a particular

be finished about noon- I might
tell someone. =an I call you after
that1 -Well- that person says -I
have to leave somewhere around
one o'cloc'. I'm not sure e$actly
but something around one. So try
to call before then.-
The communication seems sensible
and useful in a conte$t li'e that.
%nd this is *ust how one might
e$plain to someone what a game
is. One gives e$amples and
intends them to be ta'en in a
particular way.
Isn't this how we e$plain things
often enough1 There are
provisional e$planations that
prepare a place and then more a
more sophisticated understandings.
Imagine trying to e$plain -chess- to
a child. .ou say -It's the game that
you have seen :addy play with
Cncle Paul. .ou 'now the one
with those funny figures that ove
around a board that loo's li'e the
floor in our 'itchen1- Oh the child
says -the one that has soldiers1- 'ind of.- %nd that's the first
e$planation. Obviously the child
does not yet have a very solid
understanding of chess but this
initial rough e$planation lays a
groundwor' prepares a place.
33I do not however mean by
this that he is supposed to see in
those e$amples that common
thing which I 33for some reason33
was unable to e$press5 but that
he is now to employ those
e$amples in a particular way.
!ere giving e$amples is not an
indirect means of e$plaining 33 in
default of a better.

This is what he does not mean/ !e
does not mean that somehow this
e$planation of chess to the child
will give the child the essence of
chess or that I even 'new the
essence of chess at the time but
simply could not thin' of it. &y
e$planation to the child was not
merely a faulty e$planation either.
The child could not have
understood a fuller one. <iving
him the e$planation that I did will
however prepare a place for a fuller
e$planation. Over the ne$t year or
so imagine him watching his dad
and Cncle Paul playing chess and

learning a little at a time until
gradually he has wor'ing
definition but still does not 'now
"uite what a chec'3mate means
and after that he has a wor'ing
definition but does not 'now what
a Iueen's <ambit is and so forth.
33 Wittgenstein is showing us how
we can understand language being
learned in terms other than the
unambiguous pointing and naming
that %ugustine imagined in +2,
#or any general definition can
be misunderstood too.
9o matter how I point at the blue
circle and say -blue- you might
misunderstand me +cf. ;6,. %nd no
sentences either are so accurate
and so apt as to prevent all
The point is that this is how we
play the game. +I mean the
language3game with the word
What language game1 The
language game of showing others
what we mean. We introduce the
concept by preparing the place.
Listeners cannot understand our
language until a place is prepared
for it..
D;. Seeing what is common.
Suppose I shew someone various
multi3coloured pictures and say/
-The colour you see in all these
is called 'yellow ochre' -.3This is
a definition and the other will
to understand it by loo'ing for
and seeing what is common to
the pictures. Then he can loo' at
can point to the common thing.
This voice is persistent isn't it1
The voice that says we learn by
seeing what is common. Well we
sometimes seem to learn by seeing
what is common. The problem is
that we give this way of learning
language altogether too much
credit. There are other ways of
learning language and LW is
showing us a few.

=ompare with this a case in
which I shew him figures of
different shapes all painted the
same colour and say/ -What
these have in common is called
'yellow ochre' -.

This is the 'ind of e$ample the
%ugustinian in this passage was
pondering. .ou can imagine it.
There are various shapes and they
are all the same color. 0ven if the
person wasn't "uite sure about the
concept of 'color' +say didn't 'now
the difference between the concept

of 'color' and the concept of 'shade',
surely she would understand if she
could see the different shapes here
and be told -What these have in
common is called 'yellow ochre'-.
Isn't this how we learn to 'now
colors1 by seeing what is
%nd compare this case/ I shew
him samples of different shades
of blue and say/ -The colour that
is common to all these is what I
call 'blue' -.

(ut here things are a bit different.
:ifferent shades of blue might not
all be seen as -blue- especially if
one didn't 'now that ordinarily we
treat different levels of saturation as
the -same color- even though they
are different -shades.-
In other words some situations of
e$planation are easier to grasp
perhaps than others. If we imagine
the case of different ob*ects having
the same color as being useful to
teach people the concept of 'yellow
ochre' are we imagining that these
different ob*ects have precisely the
same shade of 'yellow ochre'1 (ut
don't we use the word in a rougher
'ind of way to individate a variety
of shades1 Ta'e the color blue and
notice the vast difference between
midnight blue ice blue robin's egg
blue babyblue and so forth.
In other words we can convince
ourselves that we detect the essence
of the concept by seeing e$amples
only by thin'ing of e$treme cases
in which the ambiguity of what we
are pointing to is minimi8ed. It is
hard to imagine what that e$treme
case would be in the case of
D?. When someone defines the
names of colours for me by
pointing to samples and saying
-This colour is called 'blue' this
'green' ..... - this case can be
Well this is a familiar e$ample.
Thin' of all of our tal' of the table
or the file cabinet in the mind. .et
it is true that we do teach these
words in situations that amount to
compared in many respects to
putting a table in my hands with
the words written under the
colour3samples.3Though this
comparison may mislead in
many ways.3
attaching labels to things it is *ust
that we have seen that this e$ample
as seductive as it seems to be is
misleading if it leads us to thin'
that such a table must be present in
the mind.+cf A@3A6,
One is now inclined to e$tend
the comparison/ to have
understood the definition means
to have in one's mind an idea of
the thing defined and that is a
sample or picture. So if I am
shewn various different leaves
and told -This is called a 'leaf' -
I get an idea of the shape of a
leaf a picture of it in my mind.3
(ut what does the picture of a
leaf loo' li'e when it does not
shew us any particular shape but
'what is common to all shapes of
leaf'1 Which shade is the 'sample
in my mind' of the colour green3
the sample of what is common to
all shades of green1

!e continues to show us the
problem with the idea that we
deduce the essence of the concept
from e$amples in which the one
thing held constant is the essential
feature of the concept +as in
differently shaped ob*ects all
having the color -yellow ochre- in
!e is countering this %ugustinian
presumption by referring to some
earlier discussions. In ?6 for
e$ample he tal'ed about our
tendency to solve the pu88le of how
we do things by presuming we do
things half3unconsciously +or even
unconsciously, in the mind that
correspond to what we might do
physically. If we can loo' up a
table to see what a color is we
imagine doing this in the mind
-(ut might there not be such
'general' samples1 Say a
schematic leaf or a sample of
pure green1-

This is the ne$t move after the
%ugustinian voice reali8es that we
do teach general concepts that
include considerable variation +and
families of variation, under their
rubric. -&aybe- the %uegustinian
says we have a 'ind of schematic
leaf in the mind roughly drawn.
Would that wor'1- That is 'ind of
li'e a table in the mind +cf. lwref
pictures before the mind.,
3=ertainly there might. (ut for
such a schema to be understood
as a schema and not as the shape
of a particular leaf and for a slip
of pure green to be understood as LW is saying there could be
such a schema but how would we
'now that it was such a schema and
not the shape of a particular leaf1-
%nd I might add how would we
a sample of all that is greenish
and not as a sample of pure
green3this in turn resides in the
way the samples are used.
'now how diverse a group of things
this schema would apply to1
%s' yourself/ what shape must
the sample of the colour green
be1 Should it be rectangular1 Or
would it then be the sample of a
green rectangle13So should it be
'irregular' in shape1 %nd what is
to prevent us then from regarding
it3that is from using it3only as a
sample of irregularity of shape1
Or let's reverse the e$ample here to
the earlier one/ What color would
the schematic leaf be1 %nd how
would we 'now that the term did
not apply to the color of the leaf1
D@. !ere also belongs the idea
that if you see this leaf as a
sample of 'leaf shape in general'
you see it differently from
someone who regards it as say a
sample of this particular shape.
9ow this might well be so 33
though it is not so 33 for it would
only be to say that as a matter of
e$perience if you see the leaf in
a particular way you use it in
such3and3such a way or
according to such3and3such
!ere I thin' LW confuses things a
bit. !e is using the phrase -see the
thing in a particular way- in one of
its possible senses. I see him as
saying you don't -see things
differently- unless it is something
li'e a gestalt picture of the duc'3
rabbit where it appears li'e a duc'
sometimes and li'e a rabbit at
others. I thin' we have a related
langauge game in which we say
that we -see things differently-
without this meaning that we
actually e$perience the visual
image differently. (e that as it
may Wittgenstein is I believe
tal'ing about -seeing things
differently- as seeing a different
aspect as in the case of the duc'3
rabbit. %t least to me this is the
interpretation that ma'es the most
Of course there is such a thing
as seeing in this way or that5 and
there are also cases where
whoever sees a sample li'e this
will in general use it in this way
and whoever sees it otherwise in
another way. #or e$ample if you
see the schematic drawing of a
cube as a plane figure consisting
%nd an important point. The world
around us has many aspects and
some of those aspects may be
noticeable if we see the world in a
certain way and not if we don't.
(oth ways may be e"ually correct
+as in the case of the duc'3rabbit,.
(ut how we see the world will have
an impact on what we do and on
our form of life.
of a s"uare and two rhombi you
will perhaps carry out the order
-(ring me something li'e this-
differently from someone who
sees the picture three3
DA. What does it mean to 'now
what a game is1 What does it
mean to 'now it and not be able
to say it1 Is this 'nowledge
somehow e"uivalent to an
unformulated definition1 So that
if it were
formulated I should be able to
recogni8e it as the e$pression of
my 'nowledge1 Isn't my
'nowledge my concept of a
game completely e$pressed in
the e$planations that I could
give1 That is in my describing
e$amples of various 'inds of
game5 shewing how all sorts of
other games can be constructed
on the analogy of these5 saying
that I should scarcely include this
or this among games5 and so on.
I understand this on the model of
people learning to ma'e *udgments
without 'nowing the criteria they
use to ma'e those *udgments and
even without there being
formulateable criteria. I learn to
drive steer a car turning the
steering wheel a little this way or
that in response to how the car
moves and I learn to ride a horse
by doing something similar even
balance on my feet as I'm standing
still by doing little corrections but
this doesn't mean that I would
recogni8e the rule or even that the
rule could be stated in a single
formula no matter how comple$.
This is especially clear to me if the
*udgment is obviously comple$ li'e
whether my boss is in a good mood
good enough to as' for a raise.
%phorism DB364 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance
Shawver commentary:
DB. If someone were to draw a
sharp boundary I could not
ac'nowledge it as the one that I
too always wanted to draw or had
drawn in my mind. #or I did not
want to draw one at all. !is
concept can then be said to be not
the same as mine but a'in to it.
The 'inship is that of two
pictures one of which consists of
colour patches with vague
contours and the other of patches
similarly shaped and distributed
but with clear contours. The
'inship is *ust as undeniable as the
=onsider again the concept of a
schematic leaf
In s'etching such a schema one
creates something that was not
initially there. I do not picture
such a schematic leaf in my mind
each time identify a leaf and if I
were to do so the one that I
pictured might not be e$actly li'e
yours. Still if we were each to
create such a schematic leaf
representing all leaves our
creativity would be constrained by
our similar understanding of what
counted as a leaf.
DD. %nd if we carry this
comparison still further it is clear
that the degree to which the sharp
picture can resemble the blurred
one depends on the latter's degree
of vagueness. #or imagine
having to s'etch a sharply defined
picture 'corresponding' to a
blurred one.
!ere is a schematic
leaf. Is that the one
you would have
drawn1 !ow
similar to a real leaf
must this leaf be in
order to be a
schematic leaf1 Will the point on
the right side be enough to ma'e it
serve for a maple leaf1 Or should
it be more pointed1 %nd if it were
more pointed would it it also
wor' for a smooth3sided leaf 1
!ow would you s'etch a sharply
defined picture corresponding to
this blurred one1
In the latter there is a blurred red
rectangle/ for it you put down a
sharply defined one. Of course3
several such sharply defined
rectangles can be drawn to
correspond to the indefinite one.3
(ut if the colours in the original
merge without a hint of any
outline won't it become a hopeless
tas' to draw a sharp picture
corresponding to the blurred one1
Won't you then have to say/ -!ere
I might *ust as well draw a circle
or heart as a rectangle for all the
colours merge. %nything3and
nothing3is right.- %nd this is the
position you are in if you loo' for
definitions corresponding to our
concepts in aesthetics or ethics.
%nd here is a
suppose your
tas' is to draw a
definite one that
with this indefinite one. %nd if
you imagined it even more
blurred1 %t some point wouldn't
the tas' become hopeless1
In such a difficulty always as'
yourself/ !ow did we learn the
meaning of this word +-good- for
instance,1 #rom what sort of
e$amples1 in what language3
games1 Then it will be easier for
you to see that the word must
have a family of meanings.
The situation is similar when we
try to envision the essential
features of a game or of any other
concept. To thin' in terms of
essences we must visuali8e a
blurred concept and yet when we
try to apply such a concept to a
case before us we will have the
same 'ind of difficulties we have
with the schematic leaf or
D6. =ompare 'nowing and
how many feet high &ont
how the word -game- is
how a clarinet sounds.
If you are surprised that one can
'now something and not be able
to say it you are perhaps thin'ing
of a case li'e the first. =ertainly
not of one li'e the third.
If one 'nows how high a mountain
is then one would surely 'now
how to say it. (ut isn't it possible
to 'now how a clarinet sounds or
how coffee smells without being
able to say what one 'nows1 %nd
isn't the case of 'nowing what a
game is rather li'e the case of
'nowing how a clarinet sounds1 It
is easy to 'now such things
without 'now how to say what one
DF. =onsider this e$ample. If one
says -&oses did not e$ist- this
The sentence -&oses did not
e$ist- has blurred boundaries
may mean various things. It may
mean/ the Israelites did not have a
single leader when they withdrew
from 0gypt or/ their leader was
not called &oses or there cannot
have been anyone who
accomplished all that the (ible
relates of &oses 33 or/ etc. etc.33
much li'e the blurred boundaries
of a schematic leaf or a blurred
rectangle. Gust as a number of
different leaf shapes could have
been ta'en from the blurred
schema so a number of different
meanings might be drafted onto
the statement -&oses did not
We may say following )ussell/
the name -&oses- can be defined
by means of various descriptions.
#or e$ample as -the man who led
the Israelites through the
wilderness- -the man who lived
at that time and place and was
then called '&oses' - -the man
who as a child was ta'en out of
the 9ile by Pharaoh's daughter-
and so on. %nd according as we
assume one definition or another
the proposition -&oses did not
e$ist- ac"uires a different sense
and so does every other
proposition about &oses.3%nd if
we are told -9 did not e$ist- we
do as'/
-What do you mean1 :o you want
to say ...... or ...... etc.1-
0ven the name -&oses- is not as
clearly defined as we are apt to
presume. What if someone not3
named &oses was still a person
who had done all that &oses is
repored to have done. Would that
be the same as &oses1 Or what if
he had done some of the ghings
but not all1 !ow much different
from the story of &oses could the
historical man have been in order
to *ustify the statement -&oses did
not e$ist1-

(ut when I ma'e a statement
about &oses33 am I always ready
to substitute some one of these
descriptions for -&oses-1 I shall
perhaps say 33 (y -&oses- I
understand the man who did what
(ible relates of &oses or at any
rate a good deal of it. (ut how
much1 !ave I decided how much
must be proved false for me to
give up my proposition as false1
!as the name -&oses- got a
and une"uivocal use for me in all
possible cases1 33
(ut if I were to ma'e a statement
about &oses all of these
considerations are not in my
mind. I haven't decided
beforehand which features of the
story of &oses are essential in
order for us to say that &oses
lived. (ut perhaps you want to
say that most of it must be true in
order to say that &oses e$isted.
(ut how much1
Is it not the case that I have so to
spea' a whole series of props in
readiness and am ready to lean on
one if another should be ta'en
from under me
and vice versa1
Suppose there were @4 stories of
&oses. If stories @ through ?;
were false would this be different
than if stories 23;6 were false1
%re there any essential stories1 Or
can I fall bac' on any1
=onsider another case. When I
say -9 is dead- then something
li'e the following may hold for
the meaning of the name -9-/ I
believe that a human being has
lived whom I +2, have seen in
such3and3such places who +;,
loo'ed li'e this +pictures, +?, has
done such3and3such things and
+@, bore the name -9- in social
life. 33%s'ed what I understand by
-9- I should enumerate all or
some of these points and different
ones on different occasions. So
my definition of -9- would
perhaps be -the man of whom all
this is true-.3(ut if some point
now proves false1 33Shall I be
prepared to declare the
proposition -9 is dead- false3even
if it is only something which
stri'es me as incidental that has
turned out false1 (ut where are
the bounds of the incidental133 If I
had given a definition of the name
in such a case I should now be
ready to alter it.
%lthough it may seem to us when
we spea' that our language is
unambiguous even the phrases
that at first seem without
ambiguiuty are on reflection very
e"uivocal that is sub*ect to
interpretation 33 much li'e the
blurred leaf that was to serve as a
schematic leaf. Is -9- dead1 #or
-9- to be dead -9- must have
lived but how will we decide that
the person I am referring to is a
specific person1 If someone lived
who had some of the features I
imagined for -9- but not all was
that -91-

%nd this can be e$pressed li'e
this/ I use the name -9- without a
fi$ed meaning. +(ut that detracts
as little from its usefulness as it
detracts from that of a table that it
stands on four legs instead of
three and so sometimes wobbles.,
So we are driven to notice that
words do not have fi$ed
meanings. %t first glance you may
thin' this would reduce their
usefulness to us. (ut it is not so.
Should it be said that I am using a
word whose meaning I don't
'now and so am tal'ing
nonsense1 3 3Say what you
choose so long as it does not
When we notice that language is
never unambiguous that is much
li'e the blurred leaf we might as'
-can I use a word JdorrectlyK
whose meaning I do not 'now1-
prevent you from seeing the facts.
+%nd when you see them there is a
good deal that you will not say.,

There is a sense in which our
understanding of the term is
limited. Shall we count this as a
case of not3'nowing1
The problem is that we can see
what is 'nown and what is not3
'nown. Our confusion comes not
from not3'nowing what the facts
are but rather from the fact that
the rule that would determine how
we should spea' is not definitive
enough to tell us how to answer.
It is the same as if I were to as'/
-Is it cold outside1- +since you
were standing outdoors, and you
might 'now it was B;S #ahrenheit
+imagine having a thermometer,
and yet not 'now whether to count
this as -cold- because the word
-cold- does not have such well
defined boundaries.
Still your understanding of the
temperature would limit how you
answered the "uestion +truthfully,.
+The fluctuation of scientific
definitions/ what to3day counts as
a observed concomitant of a
phenomenon will to3morrow be
used to define it.,

Scientific definitions reduce this
ambiguity somewha. What counts
as water in the vernacular is
different from what counts as
!;4. In he creation of the concept
of !;4 there has been the
systemtic e$clusion of seawater or
dishwater from the concept. Still
if there are a few molecules that
are not -!;4- shall we still
consider the vial to contain !;41
0ven here there is ambiguity that
tends to escape us.
64. I say -There is a chair-.
What if I go up to it meaning to
fetch it and it suddenly disappears
from sight.1 33-So it wasn't a
chair but some 'ind of illusion-.
33(ut in a few moments we see it
The rules that determine the right
way to use language in any given
language game are never defined
with absolute precision. We all
comfortably call the ob*ects we sit
on chairs but we have no rules to
again and are able to touch it and
so on. 33-So the chair was there
after all and its disappearance was
some 'ind of illusion-. 33(ut
suppose that after a time it
disappears again3or seems to
disappear. What are we to say
now1 !ave you rules ready for
such cases 333rules saying
whether one may use the word
-chair- to include this 'ind of
thing1 (ut do we miss them when
we use the word -chair-5 and are
we to say that we do not really
attach any meaning to this word
because we are not e"uipped with
rules for every possible
application of it1
label them if they stop behaving as
chairs. Language is simply not
that precise. There are blurred
boundaries that we fail to see and
that often do not bother us.

%phorism 62366 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is
inserted by Shawver
to enhance
Shawver commentary: This section is
concerned with rules and precision of rules
and suggests that precision is not always
better than imprecision.
62. #. P. )amsey
once emphasi8ed in
conversation with me
that logic was a
'normative science'. I
do not 'now e$actly
what he had in mind
but it was doubtless
closely related to
what only dawned on
me later/ namely that
in philosophy we
often compare the
use of words with
games and calculi
which have fi$ed
rules but cannot say
that someone who is
using language must
be playing such a
game. 33$ut if you
say that our
languages only
appro+imate to such
calculi you are
standing on the very
brin# of a
#or then it may loo'
as if what we were
tal'ing about were an
ideal language. %s if
our logic were so to
62. This is an important aphorism. 0arly
Wittgenstein the Wittgenstein of the
Tractatus thought of language as something
li'e a calculus. The idea was that if you
'new the rules of language you could apply
the calculus to understand it.
#or e$ample suppose you had the following
four sentences/
%. &ary went to the store
( Gac' went to the barber.
= &ary is tired
: Gac' earns lots of money
%nd suppose you also had four ways of
connecting those sentences/
v 3 meaning either or both
H 3 meaning -and-
T 3 meaning if 3then
L 3 meaning if and only if
%nd suppose you could also modify any
sentence by negating it and symboli8ing that
negation with a tilde li'e this/
%nd let's enrich this calculus. .ou can also
use parentheses. Csing the character names
above to name the four sentences couldn't
you figure out the following statement li'e
spea' a logic for a
vacuum. 33Whereas
logic does not treat of
language 33 or of
thought 33 in the
sense in which a
natural science treats
of a natural
phenomenon and the
most that can be said
is that we construct
ideal languages. (ut
here the word -ideal-
is liable to mislead
for it sounds as if
these languages were
better more perfect
than our everyday
language5 and as if it
too' the logician to
shew people at last
what a proper
sentence loo'ed li'e.

one would figure out a calculus1
+%H(, H P%
It would mean

While it is true that
-&ary went to the store
and Gac' went to the
(arber- is a true
statement it is not true
that &ary went to the
%nd as you can see this would not be
possible because it is not true for &ary to
have both gone to the store and not to have
gone to the store. So we can see that the
symbolic phrase
+%H(, H P%
is nonsense. because to be true it would
re"uires % to be both true and false.
9ow consider the following/
JP+%H(, v +(T=,K v +:L(,
=ould this statement be true1
.ou could figure this out using the same
process that we used above and it would feel
very much li'e performing a 'ind of
mathematical calculus.
This was the sort of vision of language that
inspired early Wittgenstein +and the logical
positivists, but now he is saying that it will
not wor'.
One might want to say that if it were a
misunderstanding that language wor'ed as a

calculus then it was because language is
defective in some way. (ut Wittgensein is
telling us that the failure of langauge to
conform to a calculus does not imply that it is

%ll this however can
only appear in the
right light when one
has attained greater
clarity about the
concepts of
meaning and
thin'ing. #or it will
then also become
clear what can lead
us +and did lead me,
to thin' that if
anyone utters a
sentence and means
or understands it he is
operating a calculus
according to definite
%nd these concepts of understanding
meaning and thin'ing are concepts
Wittgenstein will e$plicate.

6;. What do I call
'the rule by which he
proceeds'1 33The
hypothesis that
describes his use of
6;. Suppose you are playing chess and you
move your 'night. % child who does 'now
how to play chess as's you how you were
able to move the piece in such an odd way. If
you 'now chess the rule is probably clear in
your mind and you can state it
words which we
observe5 or the rule
which he loo's up
when he uses signs5
or the one which he
gives us in reply if
we as' him what his
rule is1 33(ut what if
observation does not
enable us to see any
clear rule and the
"uestion brings none
to light1 33#or he did
indeed give me a
definition when I
as'ed him what he
understood by -9-
but he was prepared
to withdraw and alter
it.3So how am I to
determine the rule
according to which
he is playing1 !e
does not 'now it
himself. 33Or to as' a
better "uestion/ What
meaning is the
e$pression -the rule
by which he
proceeds- supposed
to have left to it

unambiguously. .ou can say what the rule is
that guides and constrains the movement of
the bishop compared to the movement of the
'night. There is no ambiguity here.
(ut if you were as'ed the rule you used to
decide if a sentence were a well formed
sentence or grammatically flawed you
might find that you do not 'now the answer
immediately. .ou feel you have to thin'
about it a bit. It may be that you can choose
which sentence has a flaw but not 'now
immediately what the rule that this correct
useage obeys.
Similarly you might 'now how to use a word
in a sentence and use it regularly and
meaningfully yet still not be 'now its useage
well enough to give a definition
spontaneously and easily.
So as' yourself are you following a rule in
the cases in which you cannot easily and
spontaneously state the rule1 In what sense
are you following one1 %re you
subse"uently *ust trying to discover a stated
rule that wold capture the behavior you are
engaging in without any sense of trying to
conform to a defined rule1
6?. :oesn't the
analogy between
6?. See how far this new model of language
is from the model of language as a calculus1
language and games
throw light here1 We
can easily imagine
people amusing
themselves in a field
by playing with a ball
so as to start various
e$isting games but
playing many without
finishing them and in
between throwing the
ball aimlessly into
the air chasing one
another with the ball
and bombarding one
another for a *o'e and
so on. %nd now
someone says/ The
whole time they are
playing a ball3game
and following
definite rules at every
.es there are rules but the rules are not
binding in the same way that they are in
calculus. The rules of langauge do not
confine every movement that is made. In
languge one can stop metaphorically
spea'ing to toss the ball up into the air.

%nd is there not also
the case where we
play and3ma'e up the
rules as we go along1
%nd there is even one
where we alter them3
as we go along.

This is a particularly significant observation.
In language we will find ourselves ma'ing up
meanings for words as we go along. -What
do you mean by that1- someone as's you.
Then you say -I mean...- and you give the
word a definite sense not a sense that is "uite
what it is in the dictionary but a definite
sense. .ou are ma'ing up the rules of this
language game as you go along.
6@. I said that the
application of a word
is not everywhere
bounded by rules.
(ut what does a
game loo' li'e that is
everywhere bounded
by rules1 whose rules
never let a doubt
creep in but stop up
all the crac's where it
might1 33 =an't we
6@. %m I right that games are not completely
bounded by rules1 Sure there are gaps in the
stated rules. (ut can't we imagine some sort
of implicit rule that guides us in the spaces
between the rules1

imagine a rule
determining the
application of a rule
and a doubt which it
removes3and so on1
(ut that is not to
say that we are in
doubt because it is
possible for us to
imagine a doubt. I
can easily imagine
someone always
doubting before he
opened his front door
whether an abyss did
not yawn behind it
and ma'ing sure
about it before he
went through the
door +and he might
on some occasion
prove to be right,3but
that does not ma'e
me doubt in the same
Sure we can imagine such a thing but we
need not. It is not a re"uirement of games
that they be everywhere bounded by rules.

6A. % rule stands
there li'e a sign3
post.33:oes the sign3
post leave no doubt
open about the way I
have to go1 :oes it
shew which direction
I am to ta'e when I
have passed it5
whether along the
road or the footpath
or cross3country1 (ut
where is it said which
way I am to follow it5
whether in the
direction of its finger
or +e.g., in the
opposite one1 33%nd
if there were not a
single sign3post but a
6A. %nd even if we stated rules +li'e sign3
posts, every space this would not leave us
with some fle$ibility in how we played the
game. 0ven sign3posts have to be
interpreted. 0ven if a hand

points in a certain direction where is the rule
that says I must follow it in the direction of
the finger1

%nd even if we assume that the hand points
towards the flag is it pointing to the stripes
or the stars1 Or the flag as a whole1 Or to
chain of ad*acent
ones or of chal' mar

's on the ground33 is
there only one way of
interpreting them133
So I can say the sign3
post does after all
leave no room for
doubt. Or rather/ it
sometimes leaves
room for doubt and
sometimes not. %nd
now this is no longer
a philosophical
but an empirical one.
the colors1 Is there not room for
interpretation here1 Is everything completely
bound by rules1 %nd if we have this
fle$ibility in pointing is there not room for a
similar fle$ibility in how we interpret the
rules of a game1

6B. Imagine a
language3game li'e
+;, played with the
help of a table. The
signs given to ( by %
are now written ones.
( has a table in the
first column are the
signs used in the
game in the second
pictures of building
stones. % shews (
such a written sign5 (
loo's it up in the
table loo's at the
picture opposite and
so on. So the table is
a rule which he
follows in e$ecuting
orders.3One learns to
loo' the picture up in
the table by receiving
a training and part of
this training consists
perhaps in Pe pupil's
learning to pass with
Imagine the wor'ers in a language3game li'e
+;, having the following table to use to ma'e
their selection of stones.

his finger
hori8ontally from left
to right5 and so as it
were to draw a series
of hori8ontal lines on
the table.
Suppose different
ways of reading a
table were now
introduced5 one time
as above according
to the schema/

If we did include arrows in our own culture
it would li'ely loo' li'e this/

#or the most part however the action that
the arrows prompt is so common in our own
culture that the arrows are not needed. We all
approach such tables with our eyes already
trained to loo' in the way the arrows are
intended to guide us.
another time li'e

or in some other

#or this 'ind of loo'ing however we would
need arrows/

33Such a schema is
supplied with the
table as the rule for
its use.
(ut what Wittgensein had in mind for this
tribe is two tables. #or e$$ample imagine
one being up on a wall and the other being in
one's hand.

!owever if their mythology re"uired a more
comple$ the rule might be/
Or imagine things more comple$ still.
Perhaps this language game is not for the
purpose of building but for the purpose of
assuaging the temper of the gods and
supppose too that the paths the gods want
their servants to ta'e to read these tables
re"uires them to wor' through a ma8e of
arrows such as this/

in order to read a table li'e this/

=an we not now
imagine further rules
to e$plain this one1
9ow suppose the various rules in the
networ' of arrows was tied to a mythology so
that each arrow represented a sacred path that
must be followed e$actly. 9ot only did this
sacred path guide how one's eyes were to
move but also how one stood and the
e$pression one put on one's face/

%nd on the other
hand was that first
table incomplete
without the schema
of arrows1 %nd are
other tables
incomplete without
their schemata1
The initial table seemed easy to us/

(ut the ease we felt
surely reflected the
years of training we
had in reading such a
table. We no longer
needed guidance to
loo' from left to
right. The straight
arrows may have
such guidance to one uninitiated but only if
that person had already had training in how
to read arrows how to follow a line with the
eyes. .ears of reading ma'e it natural for
0nglish spea'ing people to follow the line
from left to right but of course there are
other traditions. We could complicate things
further by a re"uirement that the reader of the
table must *ump from line to line or move
the eyes bac' and forth or up and down the
line. There is no end of complicating
possibilities. .et these possibilities do not
confuse us. We have been trained to see and
read such tables so we do with ease *ust as
we have been trained to read the words on
this page and do that with ease 33 even
though it was not always so.
%ll these implicit rules seem to guide our
behavor and rules we can no longer state
that no longer guide us in a concscious way.
:o we want to say that the table needed to
include such rules in order to be complete1
If so would any table ever be complete1

6D. Suppose I give
this e$planation/

I ta'e '&oses' to
mean the I man
if there was such
a man who led
the Israelites out
of 0gypt
whatever he was
called then and
whatever he
may or may not
have done
besides.- 33

6D. LW is going to try to show us +or remind
us, how difficult it can be to tie down the
meaning of even an apparently simple
sentence. This may seem to you li'e a
change in sub*ect because we are no longer
tal'ing about tables and arrows but the
sub*ect is much the same. We are noticing
how many gaps there are in the rules we
might use to interpret things how much of
our understanding ta'es place without our
noticing how it all wor's.
(ut similar doubts to
those about -&oses-
are possible about
%s soon as you try to pin down these words
you can see how hard it is to ma'e sure the
person in history that we tal' about refers to
the words of this
e$planation +what are
you calling -0gypt-
whom the -Israelites-
etc.1,. 9or would
these "uestions come
to an end when we
got down to words
li'e -red- -dar'-
-&oses.- &aybe the real person had a
different name and maybe his story has been
modified through the years. !as it been so
modified that the person we thin' of as
-&oses- is no longer congruent with the
historical figure1 It is possible to doubt all of
these things.
-(ut then how does
an e$planation help
me to understand if
after all it is not the
final one1 In that case
the e$planation is
never completed5 so I
still don't understand
what he means and
never shall>- 33
%s though an
e$planation as it were
hung in the air unless
supported by another
This is Wittgenstein's "uestioning voice
voice of aporia wondering. If I can't tie
these things down with an e$planation I not
only fail to understand who &oses is but I
fail in all similar attempts. 0$aplanations
cannot help me understand> +or so it seems
to the aporetic voice,.
It seems +when in this apoetic mood, that we
must be able to use e$planations to tie down
all the ambiguities or else nothing will ever
be 'nown.

Whereas an
e$planation may
indeed rest on
another one that has
been given but none
stands in need of
another 33 3unless we
re"uire it to prevent a
That is we may be able to use one
e$planation to e$plain another 33 but no
additional e$planation is needed e$cept to
prevent misunderstanding. .ou do not need
an e$planation for the statement -The chair I
am sitting in is uncomfortable- unless you
don't understand it +and you might not for
e$ample if it loo'ed o you that I was not
sitting at all.,
One might say/ an
e$planation serves to
remove or to avert a
misunderstanding 33
one that is that
would occur but for
the e$planation not
every one that I can
The confusion comes about because we
imagine that e$planations contain a complete
rule that re"uire no training to interpret.
0$planations canavert misunderstandings but
only for those whose training is sufficient to
understand the e$planation. %nd we cannot
find sufficient e$plnations to replace that
history of training.
It may easily loo' as
if every doubt merely
revealed an e$isting
gap in the
foundations5 so that
secure understanding
is only possible
if we first doubt
everything that can
be doubted and then
remove all these

This is reminescent of :ecartes' -=artesian
doubt- or -methodological doubt.- !is idea
was you'll recall that he could doubt
everything e$cept that he was thin'ing +I
thin' therefore I am,. 0verything that came
after that point in the =artesian te$t was the
result of his reasoning things out and thus
proven ostensibly by the reasoning. It loo's
as though we must pin things down
completely. We must find a way to prove for
e$ample that historical person I call &oses is
indeed he -real &oses- or else it is all a
sham. (ut how many stories about &oses
would have to be a little wrong in order to
call it all a sham1 %nd even if the stories
were "uite a bit wrong would it be a sham1
What does it mean in fact for any name to
refer to an historical figure with accuracy1 It
seems it does not ma'e sense if we cannot
pin things li'e this down. On the other hand
the assertions do ma'e some 'ind of sense to
us given our training in this story of &oses
33 even though we cannot pin things li'e this
down without becoming aware of the
enormity of doubt about all the details.
The sign3post is in
order 33 if under
circumstances it
fulfills its purpose.

When does the e$planation the reasoning
come to an end1 )emember we may not
need an e$planation at all. If we are already
trained or practiced in how to interpret some
sign then the e$planation is no longer
re"uired. If there is a stop sign or some
strange sign it needs no e$planation if we
have been trained in its interpretation.
=onsider the e$planation I gave those reading
over our shoulder at the top of this note. I
e$plained our tradition that we put our
initials in brac'ets around the te$t. (ut I did
not e$plain that Gudy uses small characters
for her initials or that 9ic' sometimes uses
no brac'ets at all. Will they understand1 Is
it possible that they will simply ma'e sense
of it even if I don't e$plain these details1 %t
what point will the e$planation be more than
is necessary1
66. If I tell
someone -Stand
roughly here-33 may
not this e$planation
wor' perfectly1 %nd
cannot every other
one fail too1
66. If I want to ta'e a photograph of
someone I might say -stand roughly here-.
That means I am li'ely to be satisfied if the
person stands somewhere withina range of
places. Of course the person might stand
outside my preferred range or I might
discover that the range of places I thought
would wor' will not wor'. On the other
hand I cannot guard against this 'ind of
problem entirely by being more specific.
Imagine my saying -Stand on this blade of
grass.- =ouldn't I find this positioning not
"uite satisfactory too1
(ut isn't it an ine$act
e$planation1 3
This is really the voice of )amsey again or
perhaps the voice of early Wittgenstein or
one of the positivists who tried to reduce
language to a 'ind of calculus. This voice
has been very influential in the development
of modern psychology. This voice says
-=an't we find a way to represent our
thoughts and wishes very precisely so that
there can be no misunderstanding1 The
usefulness of phrases such as -stand roughly
here- helps disenchant us with this "uest by
showing us
.es5 why shouldn't
we call it -ine$act-1
Only let us
understand what
-ine$act- means. #or
it does not mean
that non3e$act e$planation can be very
useful indeed.
%nd let us consider
what we call an
-e$act- e$planation
in contrast with this
one. Perhaps
something li'e
drawing a chal' line
round an area1 !ere
it stri'es us at once
that the line has
breadth. So a colour3
edge would be more
In this paragraph Wittgenstein argues that
not only is non3precise language often useful
but that more precise statements +e.g. stand
precisely on this blade of grass, adds nothing
to the usefulness at times of less precise
statements. .ou could stand on a precise
blade of grass if I stand -stand roughly over
there- but it wouldn't be more useful than if
you stood in a slightly different place. This
is what Wittgenstein has in mind when he
suggests that this e$actness does not have a
function in some conte$ts.
e$act. (ut has this
e$actness still got a
function here/ isn't
the engine idling1
%nd remember too
that we have not yet
defined what is to
count as overstepping
this e$act boundary5
how with what
instruments it is to
be established. %nd
so on.
0ven when we try to be more precise +you
must step on this precise blade of grass, we
do not elminate all imprecision. 0ven here
we must define what will count as not doing
what the precise rule calls on us to do.
We understand
what it means to set a
poc'et watch to the
e$act time or to
regulate it to be
e$act. (ut what if it
were as'ed/ is this
e$actness ideal Of
course we can spea'
of measurements of
time in which there is
a different and as we
should say a greater
e$actness than in the
measurement of time
by a poc'et3watch5 in
which the words -to
set the cloc' to the
e$act time- have a
different though
related meaning and
'to tell the time' is a
different process and
so on.33
0ven when we tal' about precision it is not
clear how precise precision must be in order
to be precise enough.
9ow if I tell
someone/ -.ou
should come to
dinner more
punctually5 you 'now
it begins at one
o'cloc' e$actly-33
!ere's another e$ample to e$amine to help
us understand the function and limits of our
ideal of precision. !ow precise would such a
statement be1
is there really no
"uestion of e$actness
here1 because it is
possible to say/
-Thin' of the
determination of time
in the laboratory or
the observatory5 there
you see what
'e$actness' means-1
e$actness or how
nearly does it
approach the ideal13
The first thing to notice is that the 'ind of
-e$actness- that would be re"uired for
someone coming to dinner punctually is not
of the same level as that is re"uired in a
laboratory or an observatory. The conte$t
lets us 'now that different levels of precision
are involved in these different situations.
-Ine$act- is really
a reproach and
-e$act- is praise. %nd
that is to say that
what is ine$act
attains its goal less
perfectly than what is
more e$act. Thus the
point here is what we
call -the goal-. %m I
ine$act when I do not
give our distance
from the sun to the
nearest foot or tell a
*oiner the width of a
table to the nearest
thousandth of an
Language is often implicitly loaded to
convince us that something is good or bad.
People who call someone -youthful- for
e$ample are using loaded langauge to
communicate a positive "uality about
behavior or appearance that $ome might refer
to more negatively as -childish-. Sometimes
it is hard to find the negative loading for a
positive term +or the positive loading for a
negative term, and our languuge simply
doesn't appear to have the resources for
thin'ing about the ob*ect without this
particular evaluative loading.
That may be the case for the notion of
-e$actness- has that 'ind of positive loading.
+see my comments on -transvaluation-,
9o single ideal of
e$actness has been
laid down5 we do not
'now what we should
be supposed to
imagine under this
head 33 unless you
yourself lay down
what is to be so
called. (ut you will
'ind it difficult to hit
upon such a
convention5 at least
any that satisfies you.
What all of this study of exactness seems to
be teling us is that how much -e$actness- we
need depends upon the conte$t. Sometimes
demands for -e$actness- can *ust be a bother
with nothing useful to add. (ecause they
would be a -bother- this e$cessive precision
is not so universally ideal as our language
suggests to us.
%phorism 6F3244 from
Wittgenstein's Philosophical Investigations
with commentary on the right by
Lois Shawver

+0mphasis in bold is inserted by
Shawver to enhance
Shawver commentary
and supplementary notes:
6F. These considerations bring us
up to the problem/ In what sense is
logic something sublime1
#or there seemed to pertain to
logic a peculiar depth
3a universal significance. Logic
lay it seemed at the
bottom of all the sciences.33 #or
logical investigation
e$plores the nature of all things. It
see's to see to the bottom of
things and is not meant to concern
itself whether what actually
happens is this or that.

In 6F The "uestion is/ !ow did
we come to believe that logic is
sublime1 Why do we thin' that it
is sublime1
The people of our culture have
believed that logic is sublime for a
long long time. +SCPPL0&09T%).
%)TI=L0, Since %ristotle at least
philosophers have been inspired
with the idea that logic is
something something lofty and if
followed carefully can lead us to
a more accurate understanding. In
fact thin'ing this way it seems if
we could only get logic right
define things precisely enough
then we could ma'e sense of all

33It JlogicK ta'es its rise not from
an interest33 in the facts of nature
nor from a need to grasp cause
conne$ions/ but from an urge to
understand the basis or essence
of everything empirical. 9ot
however as if to this end we had
to hunt out new facts5 it is rather
of the essence of
our investigation that we do not
see' to learn anything new
by it. We want to understand
something that is already in plain
view. #or this is what we seem in
some sense not to understand.
This glorification of logic
emerges not from our need to
grasp particular connections
+such as what specifically causes
what, but a desire to find a 'ey
that will open up the secrets of the
world for us ma'e it all ma'e
sense. The "uest is not to uncover
something new detail but to
understand something that is
already before us but confuses us
because its mysteries are
somehow veiled.

%ugustine says in the =onfessions
-"uid est ergo tempus1 si nemo e$
me "uaerat scio5 si "uaerenti
e$plicare velim nescio-.

this translates as/ -What therefore
is time1 If you
don't as' me I 'now 3 if you as'
me I don't 'now.- In other words
the loftiness of logic is something
we understand until we are as'ed
about it. Then suddenly we see
how confusing it is to us.
3This could not be said about a
"uestion of natural science
+-What is the specific gravity of
hydrogen1- for instance,.
Something that we 'now when no
one as's us but no longer 'now
when we are supposed to give an
account of it is something that we
need to remind ourselves of. +%n
it is obviously something
of which for some reason it is
difficult to remind oneself.,
There are many scientific
problems that we either 'now the
answers to or we don't. (ut there
are other thngs we to undestand so
well we ta'e our 'nowledge for
granted until we are as'ed. Then
we are pu88led. It is as though we
'now the answer but can't "uite
remember what it is and need to
be reminded.

F4. We feel as if we had to
penetrate phenomena/ our
investigation however is directed
not towards phenomena but as
one might say towards the
'possibilities' of phenomena.

When we feel that logic is lofty
we feel as though we had to
penetrate the mysteries of what is
before us with the power of logic
but we do not actually loo' at
what we are studying in order to
try to do this. We simply thin'
about things or study them in our
"logical" reflection.
We might as' about our sub*ect
for e$ample in relationship to
certain possibilities. If time is the
sub*ect of our study we might
ponder for e$ample if time
would continue to e$ist if the
world stopped turning.

We remind ourselves that is to
say of the 'ind of statement that
we ma'e about phenomena.

Csing logic we try to recall things
about our sub*ect. We might say
to ourselves for e$ample that
-time seems to pass more "uic'ly
when you're busy.- %nd we

would as' ourselves -What does
that mean about time1- This 'ind
of logical reflection then is more
reflective than observational.
Thus %ugustine recalls to mind the
different statements that are made
about the duration past present or
future of events. +These are of
course not philosophical
statements about time the past
the present and the
Our investigation is therefore a
grammatical one. Such an
investigation sheds light on our
problem by clearing
misunderstandings away.

So our investigation is not based
on observations of new data.
Instead it is a study of the things
we say or have said about this
sub*ect. Our purpose is to clear
away certain misunderstandings
that seem to bloc' clarity about
whatever interests us. This means
that our study is a grammatical
one in the sense that we might
ponder the meaning of certain
terms or the connection between
different terms and remind
ourselves of the criteria for
different application of these
terms. If we wanted to 'now
what time is we might remind
ourselves of the way we name
time differently in different time
8ones for e$ample.
&isunderstandings concerning the
use of words caused among other
things by certain analogies
between the forms of e$pression
in different regions of language.

&any of our misunderstandings
result from the fact that there are
superficial similarities between
different regions of language. If I
say -love- when I am scoring
tennis this does not mean the
same thing as when I spea'
endearingly. These things
continuously confuse us.
supplementary note
3Some of them
JmisunderstandingsK can be
removed by substituting one form
of e$pression for another5 this may
be called an -analysis- of our
forms of e$pression for the
process is sometimes li'e one of
ta'ing a thing apart.

Some of this confusion can be
removed by replacing words with
other words that seem less
confusing. -Love- we might say
-means 8ero- so instead of saying
the score ?43love. We might say
that the score is ?438ero in order
to be less confused and
confusing. There are many
multiple uses of most terms that
get confused this way and we are
scarcely aware of them. When we

do study them unravel the
e"uivocations this we might call
F2. (ut now it may come to loo'
as if there were something li'e a
final analysis of our forms of
language and so a single
completely resolved form of every
e$pression. That is as if our usual
forms of e$pression were
essentially unanalysed5 as if there
were something hidden in them
that had to be brought to light.
When this is done the e$pression
is completely clarified and our
problem solved.
When we analy8e the
e"uivocations straighten things
out it sometimes begins to appear
as though we could finally get a
picture of the accurate meaning
that we could invent even ways
of tal'ing that allowed tus o spea'
in ways that are completely clear
so that the problem
at hand is solved.

It can also be put li'e this/ we
eliminate misunderstandings by
ma'ing our e$pressions more
e$act5 but now it may loo' as if
we were moving towards a
particular state a
state of complete e$actness5 and as
if this were the real goal of our

When we are mystified li'e this
we thin' we can find a way to put
things that will eliminate all
misunderstandings. It will *ust
re"uire so we thin' more
e$actness. It even seems that
e$actness not clarity is the real
goal of our investigation.
Somehow we have become
infatuated with the idea that
e$actness will bring us closer to a
final picture of the hidden
mysteries around us.
F;. This finds e$pression in
"uestions as to the essence of
language of propositions of
Our infatuation with e$actness
shows itself when philosophers
as' about the essence of language
in that they often strive for more
33#or if we too in these
investigations are trying to
understand the essence of
language 33 its function its
structure 33yet this is not what
those "uestions have in view.
It may seem that this is what we
in this boo' are trying to do as
well. (ut the "uestions we as'
are different.

#or they see in the essence not
something that already
lies open to view and that becomes
surveyable by a rearrangement
We need to use different
metaphors for their "uestions and
for ours. While they are see'ing
something deeper that will be
but something that lies beneath the
surface. Something that lies
within which we see when we
loo' into the thing and which an
analysis digs out.
unveiled as the mystery structure
of language we are see'ing
something that might be clear to
us by a certain rearrangement of
the details.
'The essence is hidden from us'/
this is the form our problem now
assumes. We as'/ -What is
-What is a proposition1- %nd the
answer to these "uestions is to be
given once for all5 and
of any future e$perience.

If we are in their frame of
reference and we as' "uestions
about the essence of things we
loo' for answers that can be given
now and for all time regardless of
what happens in the future. %fter
all the essence of language
cannot change. If langauge has an
essence so they thin' it e$ists
everywhere and whenever
langauge e$ists. 9ot so for us.
We will loo' at changeable
aspects of language that happen to
create patterns during our cultural
e$perience. #or e$ample
whereas they will loo' for what
-truth- really is apart from any
true statement we will be inspired
to notice the ways in which this
term is used in our culture and in
particular language games and
One person might say -%
proposition is the most
ordinary thing in the world- and
another/ -% proposition 3
that's something very "ueer>-
33%nd the latter is unable simply to
loo' and see how propositions
really wor'. The forms that we use
in e$pressing ourselves about
propositions and thought stand in
his way.
When they are loo'ing for
essences they do not loo' at the
way the statements actually wor'
and how we use them. They loo'
for something hidden from us.
We loo'
for something we can watch and

Why do we say a proposition is
something remar'able1 On the one
hand because of the enormous
importance attaching to it. +%nd
that is correct,. On the other hand
this together with a
misunderstanding of the logic of
When this logic of propositions
seems remar'able it is for two
reasons. One I endorse/ There is
much importance attaching to
language and why and how that is
so is worthy of our reflection.
The second reason we thin' logic
language seduces us into thin'ing
that something e$traordinary
something uni"ue must be
achieved by propositions.

is remar'able is that we are
seduced by certain illusions that
tell us that language is alien to
other things in the world. We will
find the distinction between
language and non3language "uite
blurry. Our culture tends to
polari8e the world mista'enly I
feel into language and not3
language failing to see that the
distinction is not so complete as
we at first thin'.
33 % misunderstanding ma'es it
loo' to us as if a propositions did
something "ueer.

Our recognition of the importance
of language plus our having been
seduced into seeing it as
something completely different
from non3language ma'es
language propositions
+statements, seem very odd
F@. '% proposition is a "ueer
thing>' !ere we have in germ
the subliming of our whole
account of logic.
This -subliming- of our logic is a
way of seducing ourselves into
this mystification that treats logic
as something "uite mystical.
The tendency to assume a pure
intermediary between the
propositional signs and the facts.
Or even to try to
purify to sublime the signs

When we sublime the logic of our
langauge in this way we turn it
into a 'ind of ghost which is
seems to wor' as an intermediary
between the statements we ma'e
and the words we say. We try to
get rid of the words +signs,
themself and stare at the essence
this linguistic ghost
so to spea' that connects our
words with the facts they are
meant to portray.
3#or our forms of e$pression
prevent us in all sorts of ways
from seeing that nothing out of the
ordinary is
involved by sending us in pursuit
of chimeras.

Seduced by the ghost of language
into seeing apparitions between
words and things +into seeing
-selves- -minds- -schi8ophrenia-
as things for e$ample, we are
distracted and do not notice the
ordinary that is involved.
FA. -Thought must be something FA begins with LW tal'ing
uni"ue-. When we say and mean
that such3and3such is the case we
33 and our meaning33 do not stop
anywhere short of the fact5
but we mean/ this3is3so. (ut this
parado$ +which has the form of a
truism, can also be e$pressed in
this way/ Thought can be of what
is not the case.

indirectly about the fly3bottleQ
That is he is e$ploring the
cultural thoughts that weave
together and bloc' our path out of
the fly3bottle. !ere at the source
of this impasse we find ourselves
saying things li'e -Thought must
be something uni"ue-. This is not
an innocent statement. It
represents our willingness to
imagine -thought- as something
mysterious and beyond
e$planation at the same time that
that we loo' for e$planation. This
is a path into thin'ing of language
as tied to metaphysical mysteries
such as Platonic forms.
supplemental article.
!ere is my paraphrase of the last
part of this aphorism/
When I say, "This is a cup." my
words seem to point directly to
this cup. I am pointing right to
it. y words don't fall short of
the cup and point !ust to a
concept. This is a cup, I say. It is
"ut words can only point to what
is true# Isn't this a truism# If I
say "This is a flower" and it is
really a cup before us, then my
words are not really pointing to
anything. That is fine. y words
are !ust pretending that there is a
flower there. I can't really point
to what is not here.
$r can I# If I loo% for my cup
and find a bare shelf and say, "y
cup is not here", aren't I pointing
to its absence# &nd how is this
different from loo%ing at the bare
shelf and saying, "The flower is
not here#" What would be
different about the shelf and what
I point to in the two cases# It
must be that there is something
else that I am pointing to other
than the cup itself.

FB. Other illusions come from
various "uarters to attach
themselves to the special one
spo'en of here. Thought
language now appear to us as the
uni"ue correlate picture of the
world. These concepts proposition
language thought world stand in
line one behind the other each
e"uivalent to each. +(ut what are
these words to be used for now1
The language3game in which they
are to be applied is missing.,

I point here to this bare shelf and
say -The cup is not here- but
what am I pointing to1 I might
say perhaps I am pointing to the
thought of the3cup3that3is3not3
here1 Or if not the thought then
to the proposition -This is a cup-
or to the web of language that
reflects this meaning or to the
-world- +as LW used the term in
the Tractatus when he said in the
beginning -The World is all that is
the case,. These are all more or
less synonyms. %s soon as you
'noc' one down I have a bac'up
concept that stands between the
word and the fact. These words
may loo' a little different to you
but they function in the same
way. They are place holders that I
use to tal' about these ghostly
Platonic images as i thin' about
my difficulties in e$plaining the
way langauge seems to me to
Is that any better1
(y having a string of abstract
concepts we construct in order to
have something to point to we
create a mysterious ob*ect of
meaning that language seems to
address. It suddenly appears
when we are pointing to that
thought whatever that should
mean.Then language begins to
appear to be something
remar'abe almost magical.

FD. Thought is surrounded by a
halo. 33Its essence logic present
an order in fact the a priori order
of the world/
that is the order of possibilities
which must be common to both
world and thought. (ut this order
it seems must
be utterly simple. It is prior to a
e$perience must run through all
e$perience5 no empirical
cloudiness or
uncertainty can be allowed to
affect it 33It must rather be of the
purest crystal. (ut this crystal
does not appear as an
abstraction5 but as something
concrete indeed as the most
concrete as it were the hardest
thing there is +Tractatus Logico3
Philosophicus 9o. A.AAB?,.

In this aporia it seems that
thought is surrounded by a 'ind of
halo. This halo of thought is
-essence- or -logic- and this
logical3essence3halo seems to
hold the world in some 'ind of
order to organi8e it. Without that
organi8ing halo the world would
appear chaotic. (ut this
organi8ing halo must be
completely simple perfect in
someway. It would not wor' for
this metaphysical3halo of essences
have something confused about it
something fu88y. %nd we must
have this organi8ing principle
prior to our being able to ma'e
sense of anything. Without this
organi8ing principle all if

We are under the illusion that what
is peculiar profound essential in
our investigation resides in its
trying to grasp the incomparable
essence of language. That is the
order e$isting between the
concepts of proposition word
proof truth e$perience and so on.
This order is a
super3order between 33so to
spea'33 super3concepts. Whereas
of course if the words -language-
-world- have a use it must be as
humble a one as that of the words
-table- -lamp- -door-.
%nd so in this state of
mystification we are under the
illusion that there is some essence
of langauge some magical
essence and that we are trying to
grasp this essence which is *ust
beyond our grasp. This essence
consists in the organi8ing
principles concrete almost
ghostli'e organi8ing principles.
%nd these appear to be permanent
fi$tures in the world. !ow can
they change we say in our
illusions they are the principles
that control the world of human
understanding1 See LF2

F6. On the one hand it is clear that
every sentence in our language is
in order as it is'. That is to say we
are not striving after an ideal as if
our ordinary vague sentences had
not yet got a "uite
une$ceptionable sense and a
(ut there is aporia while in this
mystification because for
e$ample we 'now that it is a bit
odd to say that we can point to
nothing and yet it seems we can.
It seems with my concepts I can
point to the fact that Gohn is not in
perfect language awaited
construction by us.33 On the other
hand it seems clear that where
there is sense there must be perfect
order. So there must be perfect
order even in the vaguest sentence.

his seat. I see the seat empty.
!ow can I do that1 Then
noticing this aporia and we thin'
that the problem is
that the language that we use is
not "uite perfect enough so we
want to ma'e it more perfect
more e$act. This perfect language
awaits our construction. What
will it be li'e1 Well it seems it
will be much li'e the one we
have only more e$act more
perfect. Thin'ing li'e this
we say to ourself that the
organi8ing principle that controls
everything is there even in the
fu88y imperfect principle but
still things do not "uite wor'
correctly. The organi8ing
principle is perfect we *ust have a
language that is an imperfect
picture of it. There are a few
flaws and we must figure them
out and fi$ them.
FF. The sense of a sentence 33one
would li'e to say33 may of course
leave this or that open but the
sentence must nevertheless have a
definite sense. %n indefinite
sense33 that would really not be a
sense at all. 33This is li'e/ %n
indefinite boundary is not really a
boundary at all. !ere
one thin's perhaps/ if I say -I have
loc'ed the man up fast in the room
33there is only one door left
open-33 then I simply haven't
loc'ed him in at all5 his being
loc'ed in is a sham. One would be
inclined to say here/ -.ou haven't
done anything at all-. %n
enclosure with a hole in it
is as good as none. 33(ut is that

In this perfect language that in
our mystification it seems we
must construct +if we are to gain
any clarity, we may of course
allow for a sentence to have some
fle$ibility. We might have a
structure li'e -The boo' is on the
table- that could be adapted to
-The pen is on the table.- (ut it
seems there must be something
"uite definite in the
boundaries of it all. We can't have
the basic rules be fle$ible. If I
leave any of the basic rules
fle$ible it seems I might as well
not have any rules at all. +Thin'
how this relates to Lyotard and his
notion that we negotiate the basic
rules of our language in paralogy.
We can say now in our
postmodernism -This is what I
mean by E- and sometimes
people can follow us.,
244. -(ut still it isn't a game if
there is some vagueness in the
rules-. 33 (ut does this prevent its
being a game1 33
-Perhaps you'll call it a game but
at any rate it certainly isn't a
perfect game.- This means/ it has
impurities and what I am
interested in at present is the pure
3(ut I want to say/ we
misunderstand the role of the ideal
in our language. That is to say/ we
too should call it a game only we
are da88led by the ideal and
therefore fail to see the actual use
of the word -game- clearly.
%nd so let me as' you must there
be e$act rules in order for us to
have a -game-1 Or is this *ust an
illusion of our logocentrism1 The
mystified voice responds well
you can call this game without
precise rules a game if you wish
but it is not a perfect game. (ut
now as I thin' through this
finding my way out of the fly
bottle Wittgenstein says I want
to say that we misunderstand the
nature of our tas' here. We are far
too da88led by the dream that
increased precision will show us
clarity to see any other prospects