Royal Institute of Philosophy

Wittgenstein on Language and Games

Author(s): J. F. M. Hunter
Source: Philosophy, Vol. 55, No. 213 (Jul., 1980), pp. 293-302
Published by: Cambridge University Press on behalf of Royal Institute of Philosophy
Stable URL: http://www.jstor.org/stable/3750812
Accessed: 02-09-2017 04:01 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

Royal Institute of Philosophy, Cambridge University Press are collaborating with JSTOR
to digitize, preserve and extend access to Philosophy

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
 Wittgenstein on Language
and Games
J. F. M. HUNTER

In reading Wittgenstein one can, and for the most part perhaps should,
treat the expression 'language-game' as a term of art, a more or less arbi-
trarily chosen item of terminology meaning something like 'an actual or
possible way of using words'. It would then be a fairly routine task to
work out answers to such questions as what features of the ways a word
is used are emphasized by this term of art, what philosophical purposes
are served by the description of primitive language-games or of variations
on actual language-games, or in exactly what way those purposes are
supposed to be served.
However, although Wittgenstein fell into the way of using the expression
'language-game' as routinely as anyone would use any other philosophical
term of art, and not in every case meaning, just by using it, to make a
philosophical point, it also struck him as an important philosophical
insight that 'we play games with words'. Yet neither he nor his interpreters
have shed much light on precisely why this should be an exciting thought.
That question is made particularly difficult by the fact that there are
so many differences between talking and playing games:
i. Games are mostly played for fun. 'It is just a game' is a reminder
that it is not serious, nothing turns on it; but while we sometimes talk
to amuse ourselves, most of our discourse, including most of the examples
Wittgenstein discusses, is not a form of sport, and something does turn
on it.
2. We propose games to our friends ('Tennis, anyone?'), but we do not
in any obvious way propose word-play ('Language-game, anyone?').
We do sometimes make something game-like out of a language-learning
session with a child, but while this may be of some importance, it is a
small point, and not a basis for treating very much of our discourse as a
form of play. Wittgenstein speaks of 'the language-game with the word.. .';
but even Wittgenstein initiates would not know what to do first if some-
one suggested they play the language-game with the word 'think'.
3. Most games are played to win, but it is by no means clear how one
might win or lose a language-game, except in special cases which there
is no reason to suppose Wittgenstein had particularly in mind. One can
win an argument, or get one's way by beguiling someone with words;
but Wittgenstein can hardly be taken to have thought that all, or even
very much, of our discourse is of those kinds.

Philosophy 55 i980 293

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
J. F. M. Hunter

4. Many games are played with equipment (pieces, balls, dice, mallets),
but words are in most ways unlike such equipment. One wields a bat,
but (except in special cases) not a word. The pieces in a game are very
much fewer in number than the words of a language, or even than the
words of a language-game. The pieces in chess play a role even when
they are not being moved, but it is difficult to see what role is played by
words we are not using. (Chess pieces are however like words in one way:
they have various and somewhat complex roles which, like the uses of
words, are both what is distinctive of them, and something not evident
from inspection of the pieces/words themselves. Chess pieces are rather
distinctive amongst game equipment in this regard, however; and it was
while watching football that Wittgenstein was struck by the game analogy.)
5. Many games have definite and well-recognized rules; but if there
are rules for the use of words, there is neither an official statement of
them nor much agreement as to what they are, and not many competent
speakers could specify more than half a dozen of them. Moreover, it is
not likely that Wittgenstein would be greatly excited by the old thought
that there are rules of language, even if he took it to be true; nor is it
likely that he should fasten on the game analogy as the only or the best
way of making this point. It is a point that can be very adequately expressed
without resort to analogy, and a sage practitioner, if he did employ an
analogy, would do so sparingly and without excitement. Furthermore,
in Philosophical Investigations ??8i-82 Wittgenstein expressed doubts as to
whether there is an interesting sense in which there are rules of language;
and he seemed as much interested in play that is not rule-governed as in
games having rules (ring-a-ring-a-roses in ?7; fooling around with a ball on
a field in ?83).
Of course, not all games are played for fun, involve winning or losing,
have rules, and so on; but the features that have been mentioned are at
least among the most striking characteristics of many games. An activity
can lack some of them and still be a game, or game-like, but can what
lacks all of them be a game, or game-iik-,?
If the answer to that question is to be atfirmative, there will have to be
less obvious features of games, in respect of which they can usefully be
likened to speech episodes. It may also be that the features of talking
to which they are likened are also not very obvious. We should remember
that Wittgenstein found this analogy illuminating. It is therefore to be
expected that pondering it should direct us to aspects of language that
we had not previously appreciated.
The following are nine possible resemblances, any of which could
be philosophically interesting. There is very little evidence as to whether
Wittgenstein was mindful of all or any of these, and hence the primary
claim that is made for them is that they are among the conclusions one
might reach if asked to make something of the thought that we play

294

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
 Wittgenstein on Language and Games

games with words. They are not presented in any particular order of
importance.
i. At the start of a game, it is often somebody's choice what the game
is to be. In poker, for example, the dealer can say whether it will be five
card stud, whether deuces will be wild, and so on; and what he says
settles the matter. He is not regarded as reporting anything, or as express-
ing a preference, in saying what the game shall be. Even if one knows
he does not care for five card stud, if that is what he announces, that
is what the game is.
This has an analogue in some kinds of speech episode. If I tell my wife
I sometimes think of quitting my job, it may not be clear what I am up to,
what my game is: to while away half an hour chatting about life's disap-
pointments and exchanging fancies as to how we might otherwise live,
or to engage in a serious discussion with a view to possibly taking a radical
decision. Her question 'Do you mean it?' is like the question 'What
game are we playing?'; and if I reply that I do, I have declared for the
serious game. It will then be a possible move for her, as it would not
have been had I declared differently, to argue vehemently against it, to
make dire threats, or equally to press me to go ahead and assure me of
her support.
In such incidents there is a possibility of games within games, that
scarcely exists in poker or cricket: I may have said I was serious because
I enjoy seeing my wife agitated or because I want to create a scene; but
still she has played correctly if she took me at my word, and indeed there
is nothing else she can do within the ostensible game. She may play her
own game within a game, perhaps seeing if she can make me back off
by encouraging me; but there is no use her asking over again whether I
mean it. I have already declared myself on that point.
How, in detail, the play will proceed is in no way settled by the dealer's
choice, but given this or that opener, some broad kinds of development
are in order, and others are ruled out. It is as if there were rules governing
this, as there are for the different kinds of poker the dealer may announce;
but while we know that threats and arguments are in order in the serious
game, but out of place in the light game, there are no rules telling us this.
There is neither a book of rules nor an oral tradition of explicit rules,
that are propounded to beginners, or cited when someone misplays.
2. It is a kind of good sense, rather than the application of rules, that
shows us what kinds of conversational developments will be appropriate
or otherwise. We have a model for this in some kinds of play. Children
learn the rudiments of hide and seek from watching it played or having
it briefly explained to them. Usually no one lays it down that one must
hide within a reasonable distance of home base, and not in the next block
or in a neighbour's attic, but the average child will see that such tactics
make the game unplayable. If some inventive and not very sporting child

295

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
J. F. M. Hunter

does hide in the next block, perhaps a rule against that will be instituted; but
for the most part good sense operates and such rules do not become necessary.
Similarly it makes for good fun, in playing with a ball, to throw it far
enough from another player that he will have to be nimble to catch it,
but it spoils the fun to throw it so far from him that he cannot possibly
catch it. Having caught a ball, one may run with it rather than throwing it,
and see if the other person can catch up, but one had best run in some
direction that leaves this possible; and if the other player proves slow,
one had best give that up and try something else-kicking the ball,
perhaps. Here one counts on the other player to see what is going on, and
play accordingly, and for the most part he will. It is a kind of sporting
sense that enables us to develop the play usefully.
There is an analogue of this in conversation. If someone says he often
thinks he should have been an accountant, not a philosopher, one may not
know whether he wishes to reflect sadly on the tribulations of his vocation,
or perhaps to make some amusing observations on the ways in which
his philosophical practice resembles accountancy, and so one may have
to improvise; but it is a dumb move to ask precisely how often he has
thought this. Not that there might not be an answer to that question, but
because there is virtually no chance of the answer contributing to any of
the ways in which the conversation might develop interestingly. It is
not as if he had said 'I often clean my spark plugs', where we do not
know just what he is recommending until we know how often.
Similarly if I am telling you about the curious dream I had last night,
it will generally be a feckless question what I was wearing, whether the
sun was low in the sky, or whether a dream person I mentioned was standing
or sitting-not again because there might not be answers to these questions
but because there is scarcely a prospect of the answers contributing to
the activity, to the game. I am engaged in sharing with you the fascination
or the terror of something I have experienced, or something I say I have
experienced (it often will not matter), and the only questions that are proper
parts of that activity are questions, the answers to which will help you
savour the fascination I am trying to share. You may indeed take a
psychoanalytic, rather than a story-fancier's interest in my dream, but then
you are not playing my game.
3. It is characteristic of some games, card and board games more than
others, that the moves we make affect the play independently of whether
we intend or appreciate their effect. I may not notice that my knight
move opened a file pinning your pawn, or that it left my rook unprotected,
but it did those things none the less. In making a move I let myself in
for all the risks, gains and losses that the move itself entails. I may, when
I notice some of its consequences, wish to retract my move, and may
sometimes be allowed to do so, but the consequences are inherent in the
move, not in my intentions.

a96

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
 Wittgenstein on Language and Games

Similarly what we say has meanings, independently of what we mean

by it. To re-use an earlier example somewhat differently, if I say I have
been thinking of quitting my job, this construction with the verb 'to think'
implies that, though still undecided, I am quite seriously inclined to do
this. Another person would have a right to complain that I had mis-
spoken if it turned out either that I had decided, or that I treated quitting
as out of the question, though I did often amuse myself thinking about
how dramatically I might proclaim my resignation, and so on. Yet a
learner of the language might easily make this mistake, extrapolating
from the fact that he could correctly say that he had thought a good deal
about it before deciding to quit, in one case, or in the other, that he very
often amused himself thinking about quitting. The thoughts one has need
not be any different depending on whether one is toying with an idea
or seriously considering it. There may be the same dissatisfactions brooded
over, the same risks reviewed. The attitude with which the thoughts are
pursued will be different, but an attitude is not something that happens,
a sort of quick-step in the dance of thought. I might, when toying with
the idea, have said to myself 'Of course, I am not serious about this',
but it would not be my having said this that showed me that I was not
serious. If I did not know whether I was serious, I need not believe
myself when I said it.
4. A further feature of some games suggests an answer to the question
how we know when to say 'I have been thinking of doing such and such',
if not by reviewing what has been happening. We work out what move
we wish to make, not by introspecting to see what our wish is, but by
reflecting on the possibilities contained in the present lay of the pieces.
Among the possible moves that are worth considering at all, one will
perhaps be safe but dull, another risky but very interesting. There is no
other test of what we want to do than whether, understanding the prospects
attending a certain move as compared with those attending other moves
we have considered, we can so move. If it remains unclear what we want
to do after various alternatives have been reviewed, there is no other way
of pursuing that question than by further reflection on the currently
available strategies and possibilities of the game.
The attendant prospects, if I consider saying I have been thinking
of quitting my job, include being met with encouragement, dissuasion,
threats, ridicule. As in games, there is often no telling what move another
person will in fact make, but these are the kinds of move that would not
be inept, and they are a different brand of move than would be appropriate
if I were to say I often amuse myself with thoughts of quitting my job.
If I am a competent speaker, which way I choose to express myself will
depend on my understanding of the sorts of prospect ensuing on this
or that choice of expression. Uncertainties can be resolved, not by examin-
ing oneself to see which wish exists, but by thinking further about the

297

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
J. F. M. Hunter

prospects. Nothing shows our wish better than that, understanding

what we will be getting into if we express ourselves in a certain way, we
can bring ourselves to take that move.
5. In games, explanations of why one moved thus and so are primarily
accounts of possibilities contained in, or believed to be contained in, the
present state of play. If I say 'I moved there to threaten his queen', and
explain the threat, I am primarily recording the fact that, as things now
stand, this move is a threat to his queen. I say 'primarily', because one
is no doubt also avowing that one did not take a move having that potential
unwittingly. We are inclined to think of this avowal as being an assurance
as to what went through the speaker's mind; but various things may
justify it. There may be a lay of the pieces so familiar to a practised player
that given it, he can make a threatening move knowing it to be such
without at all reflecting on it; and although this is rare, and perhaps
found mostly in cases of explaining the rationale of a standard opening
or end game, it is not rare in the analogous case of explaining why one
used a certain word. 'Why did you say you propose to go, rather than
that you intend to go?' 'Because I wanted to indicate that while I am
intending to go, it is something I am prepared to discuss.' It is a fact
about the use of the word 'propose' that it serves this purpose, just as it
was a fact about my move that it threatened his queen; but practised
speakers do not usually need to reflect on such facts in choosing their
words; and their explanations of a choice of word will reflect, not a thought
leading to that choice, but their day in, day out comprehension of
the language, and their general ability to communicate what they thus
understand.
6. In games there is sometimes provision (a) for adjusting a move,
if a piece is not placed squarely on one square or another, or (b) for with-
drawing a move, if it seems ill-advised. Another player will say 'Did you
intend to move to this square, or this one?', in the former kind of case,
or 'Did you really want to move there?', in the latter; and while these
forms of expression may suggest that it is understood that if such is the
move one intended or wanted to make, that is the move one must make,
all that is in fact sought in the former case is the removal of an am-
biguity, while in the latter case a player is being given the opportunity to
reconsider. It may, however, with some players, be the convention that
if he declines, he has lost his right to retract that move, and must bear
with the consequences.
We can see these move-adjusting affairs as providing at least a ground
plan for a philosophical treatment of some questions about professions
of belief, intention, and some comparable speech acts. The fact that we
say 'He said he believed it, but he didn't' leads us to suppose that as well
as the saying, there is normally also believing, and hence to press the
question what this believing is; but if, when we query a profession of

298

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
 Wittgenstein on Language and Games

belief, we are giving a person an opportunity to retract, and also fixing

it so that the belief can be re-affirmed only at a price, we may be able
to eliminate the supposition that there is something or other called 'believ-
ing', and with it the vexed question what that something is.
The question 'Do you really want to make that move?' turned out to
be, not a question of whether a certain wish existed, but an invitation to
reconsider; and similarly 'Do you really believe that?' may be, not a
question of whether a certain belief exists, but of whether, on due reflection
a person will still care to affirm something he has said. There are many
complexities to be sifted and sorted in a comprehensive working out of
this idea, and for present purposes it can be regarded only as an interesting
possibility.
7. In games we may have opinions as to how another person should
move or will move, but only the player himself can say what his move is.
This is not because he alone knows what it is, but because it is his move.
It is the way the game is played, that each person is responsible for his
own moves.
Similarly, while we have beliefs, suspicions and doubts about what
another person is thinking, believes or intends, we treat him as the best
authority on those subjects. We are much inclined to suppose that the
reason for this is that the thinker, believer or intender has access to some-
thing that is hidden from us; but on the game analogy it is possible to
regard his authoritative status as being due to the fact that it is up to him
to make these moves. He is performing an act having various social conse-
quences, and only if he does it is it fair to hold him to these consequences.
8. Some games endure, and others have a brief period of popularity
and then are played no more. The reasons for a game's longevity no doubt
differ a great deal from chess to poker to cricket, but the general reason
why a game survives is that it suits human beings, whether by providing
a controlled outlet for hostility, a variegated test of wit and ingenuity,
or a form of relaxation in which personalities interplay interestingly.
We are much inclined to think that the word 'intend' survives and is
used because there are intentions, and they play an important role in our
lives, and the same for the words 'think', 'believe', 'hope' and so on;
but a different view of the matter emerges from the game analogy: the view
that the use of a word like 'intend' is a human invention like a game, which
survives because it has proved to suit human beings to have at their disposal
the mechanism it provides for authorizing expectations as to what we will
be doing, while allowing for life's contingencies and for human incon-
stancy. Saying we intend to do something gives another person some
right to count on our doing it, rather more right than if we had said
we 'plan' to do it, and not as much right as if we had promised. It is
useful to have at our disposal these various ways of authorizing
expectations.

299

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
J. F. M. Hunter

In a different world we might have no use for a word like 'intend',

but the difference would be, not that in that world no one had something
called an intention, but perhaps that people were all very stoical, and
never changed their minds or were irresolute in the pursuit of their
projects. In such a world a cautious word like 'intend' might never gain a
foothold and be used.
9. It is characteristic of games that they create the possibility of the
activities peculiar to them. One could not get a home run except in baseball
or some similar game, pin a queen except in chess or some similar game,
or finesse a high card except in bridge or some similar game. Chess was
not invented as an outlet for our desire to pin queens or straddle rooks,
but we acquire such desires through learning chess.
It is possible to see our use of some words as being analogous to this.
When we have learned English we can truthfully say we hope; but could
a person who learned neither English nor any similar language hope, and
have the desire to say so, but not yet the means?
The use of 'hope', one could suggest, is too complicated for that. We
say we hope roughly when (a) it is neither unlikely nor certain that the
hoped for event will occur, and (b) if it were to happen we would be
pleased. If the event is either very unlikely or not expected at all, we use
instead the word 'wish', while if we have no doubt that it will occur,
but do not expect it soon, we use words like 'yearn' or 'long'. On the other
hand if an event is neither unlikely nor certain to happen and we shall
be displeased if it occurs, we use the word 'fear', and if it is quite unlikely
and we would be displeased, we say we hope it will not happen. In short
there is a somewhat complex nest of distinctions we master when we learn
this part of the language, and we have the distinctions, not because they
are obvious in the way the difference between two colours is obvious,
but because it has proved socially useful to draw them. It is possible
that some sensitive souls might make the distinctions spontaneously
before learning the words, and thus have the concepts of wishing, hoping
and yearning, but not yet the words, but at least normally there will be
no hoping before the word 'hope' is learned. It is not as if hope were a
distinctive feeling occurring naturally, to which we had given a name.
It is part of hoping to recognize that an event is neither certain nor unlikely
to happen, and that recognition is a sophisticated capacity which is not
itself a feeling or capable of being discerned in a feeling.
An obvious objection to this is that we attribute hope to dogs and young
children, although they have not the beginninos gf linguistic competence,
and one would not want to argue that this was a mistake. Yet there is
clearly enough going for the thesis that there must be some way around
this difficulty; and indeed there is.
We attribute wishes and hope to mature language-users, and by extension
to other living beings. If a child is clearly eager for his dinner and it will

300

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
 Wittgenstein on Language and Games

probably be ready soon, we may say he hopes it will soon be ready; but
we lose our grip on the question whether we might better have said he
wishes it were ready now, he wants it, or he fears it may be some time
before he eats. This is not because, being unable to interrogate the child,
we lack the evidence on which a decision might be made, but because
we have no reason to suppose that the child has either made the assessment
of the prospects of supper being ready, or based an attitude on such
assessment. In an adult, 'I hope supper will be ready soon' expresses a
restrained, balanced attitude, a weighing off of one's eagerness against
an assessment of the prospects; but if a child shows only a moderate
interest in his prospective supper, then the younger he is the less we can
get any grip on the question whether he has the balanced attitude an adult
might be expressing if he said he hoped supper would be ready soon.
Hence although we do quite routinely say of young children that they
hope, we can be taken to mean only that there is some analogy between
the child's case and a typical case of an adult hoping. The analogy may
for example be that (a) the child is keen for supper, and (b) we think it
possible but not certain that supper will soon be ready. Although normally
it is the hoper's assessment of the prospects that is involved when we
attribute hope, in secondary uses it has come to be allowable to use the
word 'hope' as an expression of the attributer's assessment of them.
I have now sketched nine different ways in which it could be philo-
sophically interesting to draw analogies between features of some games,
and possible theses about some uses of words. I have provided only
sketches, and in every case a great deal more would need to be said,
fully to articulate and defend the suggestion I have made. I hope however
that in each case I have said enough to show that there is at least an inter-
esting possibility there, which might profitably be pursued further.
Although, in the case of some of my suggestions, evidence could be
cited indicating that Wittgenstein may have been thinking along similar
lines, there is more often no such evidence, and the evidence there is
is very far from conclusive. I will give some examples:
(i) Wittgenstein sometimes (e.g. PI ??36o, 421, 569) suggests that
we look on words (concepts, sentences) as instruments; and my suggestion
in section 3 that, like moves in some games, what we say has consequences
whether or not we intend them, is one way of taking this suggestion;
but since it is not very clear whether this is what Wittgenstein had in
mind, and also since he did not mention games in this context, this evidence
must be reckoned slight.
(ii) In section 7 I suggested that just as, in games, it is up to each player
to make his own moves, so similarly we might regard the authoritative
position we accord to people as to what they believe, intend or expect,
as being due just to the fact that it is their move. In PI ?247 XVittgenstein
says that 'Only you can know if you had that intention' is a remark about

30I

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms
J. F. M. Hunter

how we use the word 'intention'. He might have said that it is the way
the game is played, that everyone must make his own professions of
intention; but he did not put it that way, and hence it remains quite
conjectural whether he had a game analogy in mind.
(iii) In PI ?492 Wittgenstein suggests that inventing a language could
be compared to inventing a game; but since he does not indicate what
basis of comparison he has in mind, the passage is not strong evidence
that he would make the same point as I made in my section 8.
(iv) At PI p. 174 Wittgenstein expresses some views about hoping
similar to what I was saying in section 9, but again since he does not
connect what he says there in any way with games, his remarks count at
best as slight evidence.
Hence my explorations of this analogy must be presented only as
conjectures; but it does seem to me that I have shown the game analogy
to be fertile; and even if none of the connections I have drawn were what
excited Wittgenstein, perhaps I have shown that the question what did
excite him about this analogy could well have a more interesting answer
than that most often given-that language, like games, has rules.

University of Toronto

302

This content downloaded from 108.61.242.12 on Sat, 02 Sep 2017 04:01:30 UTC
All use subject to http://about.jstor.org/terms