Wittgenstein's language games

Lois Shawver (rathbone@crl.com)
Mon, 3 Jun 1996 21:34:36 -0700 (PDT)

A number of people have asked for my hand-out on language games.
Actually, I thought I already posted it but realized I sent it to one
individual instead. I took this opportunity however, to go over the
handout and improve it, so maybe it is just as well that I delayed
sending it.

If you're new to the concept of language games, the important thing to
keep in mind as you start to learn the concept is that it does NOT mean
"word games" as we ordinarily use that phrase. A "language game" in
Wittgenstein's sense has no hint of the implication of playing tricks on
people's minds. The word "game" in "language game" is, I think, a bit
misleading in this respect so be careful not to bring these associations
into your study of the term.

The following is an introduction. I'll be happy to answer questions (even
though I may be a bit slow). The concept is introduced in Wittgenstein's
most important "later" text, "The Philosophical Investigations", which
Wittgensteinians refer to affectionately as "The Investigations".


Wittgenstein begins the Philosophical Investigations with a description
and critique of a certain model of learning language that was contained in
the writing of St. Augustine. According to this model, a person learns
language by 'ostensive definition' of the terms. A term is 'ostensively
defined' if you name it and point to what it is your naming. So, you say
to a child, "cup" as you point to a cup, and the child learns to use the
word cup.

This is a picture of language acquisition has some validity, of course. We
do this with children, but it focus on this practice hides the fact that
all the functional language we learn is learned in 'language games.' That
is the child learns words to make things happen. The child says
"cookie" not to name the cookie, but to get a cookie, or at least to get
an enthusiastic response from someone listening.

Wittgenstein uses the term "language game" in two related ways. One way
is to introduce a thought experiment to help explore how language works.
He usually refers to these thought experiments as studies of "primitive
language games", but the language practices he describes are imaginary.
The other way he uses the term is to help shed light on existing language

Early on in this text he presents a simple example of a primitive language
game. It takes place in an imaginary primitive society and it goes sort
of like this: A leader stands on a building site with a several workers.
There are stacks of building objects, beams, rods, etc. The leader
says the word 'beam' and the worker knows to go and get the beam and put
it in a certain pile. The word 'beam' does not just label the object. It
makes something happen. It has a function in this language game. The
point is that even though the leader uses the term to name the object, it
is part of a practice designed to get a response to that name. This is
not idle naming as Augustine's theory suggested.

A little later there is another example of a language game. It is the
game of counting apples. You go to the grocer and ask for 6 apples. He
then says the number one, two...six by heart as he reaches for an apple
with each count and puts them in a sack. This is a language game. A
person might learn to do this without learning the broader language just
as we can imagine a person learning to ask for beams and rods in the
example above without learning all of English, or even anything more of
English than this one game. The point is that we are not using the
phrase "apple" or "one" just to name, but as part of a practice that has
its place at the grocer. The language practice would not make sense if
people just went to the grocery store to count and name apples.

These serve as examples of how we learn our basic language, not by
labeling objects, but by doing things with language. Often we do not
understand the meaning fo the words initially. We simply learn to use
the words in these contexts, like the child who learns to recite the
'pledge of allegiance' without knowing what 'pledge' means or
'allegiance' means It is simply a language game, something one does when
a teacher says, "Now, we'll all stand and say the pledge of allegiance."

Wittgenstein uses the word 'game' because of its connection with 'board
games' or, particularly, chess. There are implicit and explicit rules as
to how to 'play the game' but within those rules there is a lot of
freedom. 'Games' does not imply playfulness, and the connotation of
playfulness is very confusing for many people.

Wittgenstein does not have a well developed theory of language
development. It is more of a set of insights and observations that, when
read together, suggest a theory of language development. I think it is
easiest to explain him in terms of a theory of language development,
however. So if you'll permit me to expand on him a little to present it
that way, please keep in mind that this is partly my extention of him.
You can ask me, if you like, what part is me and what part is him.
The terms I am using are his, but I am organizing it developmentally in a
way that he only more or less implies. The important thing is
understanding how we move from a primative use of language (as
illustrated by primitive language games) and a parasitic use of
language. (A parasitic use of language is a language practice that
can only be learned after a prior mastery of a more primitive language
practice. Long division is, in this sense, a parasitic on
multiplication, addition and subtraction since one needs to learn these
more elementary practices before one can learn long division.)

Basically, Wittgenstein's philosophy reminds us that our childhood basic
language games are learned without familiarity with words and so we cannot
learn language through definition. The child learns to say "Cookie!" as a
request for a cookie. Then the child is able to move into a more advanced
form of understanding.

Then, some words that are learned in these basic childhood language games
are passed metaphorically into a more parasitc language game. For
example, take the word "catch". A child learns the word in the game of
"catch",perhaps, as someone tosses a ball. But later the term is used
paratically. Perhaps someone tells a joke and says, "Did you 'catch'
that?" The idea is that we carry some of the meaning from the primitive
use of language to the parasitic use of language and understand without a
definition. Notice a "parasitic" use of language is to use language
metaphorically but in a way that passes implicitly without our noticing.

The enormity of this observation is only striking when one's attention is
awakened to the wealth of implicit metaphors that fill our ordinary
speech. Take that last sentence, for example, and let me retype it with
the metaphors capitalized.

when one's attention is AWAKENED to the WEALTH of implicit
metaphors that FILL our ordinary speech.

Words seem to be passed from primary or primitive uses of language into
this more metaphorical or parastic use of language with little awareness
on our part. These hidden metaphors lead us into language games that we
have learned in other contexts without our awareness by a process that
the postmoderns call "metaphorical structuring".

The concept of a parasitic language game in Wittgenstein fits in
remarkably, to my way of thinking, with Derrida's work, especially his
ideas as elaborated in the essay White Mythology, even though Derrida is
apparently unaware of this. Levi-Strauss' concept of bricolage is also

The concept figures highly in Wittgenstein's argument that there can be no
private language. A paper that I wrote many years ago is devoted to
explaining this argument. If you read this paper, however, know that
while I still endorse my analysis of Wittgenstein's argument here, my
recommendations as to how to proceed in research in areas that appear to
have to do with private langauge (comments that I attached at the end of
this paper) were, I now feel, based on my research immaturity.
Still, it is a good introduction to Wittgenstein's critique of the
private language argument, especially if you have read the above handout
that can provide you with a bit of framework that I believe that paper
needed to have and did not.

Shawver, Lois and Dokecki, Paul. "A Wittgensteininan analysis of the
role of self-reports in psychology." Psychological Records, 1970,
20, 289-296.

..Lois Shawver