Tuesday, August 16, 2005
Primitive Language Games
What follows below is a slightly modified version of a passage from an article I published in 1995:
A great deal has been written about Wittgenstein, but there are curious gaps still open. Relatively little has been written about Wittgenstein as a literary artist, and it seems to me that Wittgenstein’s affiliations to Hegel as well as to Augustine and spiritual autobiography deserve more attention. As far as I can tell, neither Hegel nor Augustine shows up in the writings of Ayer, Cavell, or Kripke on Wittgenstein.
Wittgenstein begins the Philosophical Investigations with Augustine’s account of how, as a baby, he began to acquire language. When Augustine’s elders named an object, at the same time they indicated the object “verbis naturalibus omnium gentium” [with the natural language of all peoples]--with facial expressions and movements of the body. Augustine learned to associate names with objects. This ostensive account of language acquisition, Wittgenstein proceeds to demonstrate, can account for only a subset of what we call language. He goes on to show how complex and mysterious that process of pointing out an object or a quality of an object is, and Augustine’s account of language acquisition is soon left behind. Wittgenstein, though, was one of the great prose artists of this century, and one whose favorite rhetorical strategy was one of implication. The meaning of the Investigations, in a sense, emerges in the blank spaces between paragraphs. The quotation from Augustine is particularly pregnant with hidden meaning. It is not an accident that the quotation comes from the Confessions. Like the Confessions, the Investigations look back over the life of their narrator as if he had died and become someone else, someone who understands the errors of that former way of life. Wittgenstein, who generally had little interest in publishing his work, wanted to publish his first book, the Tractatus Logico-Philosophicus, together with the Investigations, in a single volume. Between the writing of the Tractatus and the Investigations, Wittgenstein experienced a turn away from his earlier philosophy that amounted to a conversion.
Wittgenstein ends his quotation from Augustine with the phrase “After I had trained my mouth to form these signs, I used them to express my own desires.” Wittgenstein omits the final sentence of this paragraph: “In this way I made my needs known to my family and they made theirs known to me, and I took a further step into the stormy life of human society, although I was still subject to the authority of my parents and the will of my elders.” One might say that these last two sentences foreshadow Wittgenstein’s argument against the possibility of a private language. In a sense, Wittgenstein repeats the elision of the crucial last sentence throughout the Investigations: questions about the role of desire and authority in language acquisition shadow or halo Wittgenstein’s examples without ever reaching full articulation.
Wittgenstein begins his interrogation of Augustine’s notion of language by describing two primitive language games which, despite their simplicity, cannot be accounted for by Augustine’s account of language acquisition. By language game, Wittgenstein tells us, he means games of the sort by which children acquire language; all of the language games in the Investigations examine a practice through the lens of the question of how that practice might be learned. In the first language game, Wittgenstein sends someone to the store with a slip marked “five red apples” (Paragraph 1). The shop-keeper looks up the word “red” in a table of color samples and, counting to five, takes out an apple that matches the color sample with each numeral. “But how does he know where and how he is to look up the word “red” and what he is to do with the word ‘five’?” asks an imaginary questioner. The fundamental point of the example--or is it the fundamental point?--is by now familiar: learning a language involves not only having things named but being trained to follow rules. A reader without philosophical training, a naive reader, may have other questions about this passage. What sort of philosophical work is filled with colloquies of ghostly voices, voices that ask naive questions, voices that come from nowhere to answer questions, and voices that may or may not be the true narrator, the consciousness that takes up all the other perspectives and frames them within a larger understanding? Shouldn’t one be able to locate the authority of a philosophical work in a single voice?
Wittgenstein’s second primitive language enables a builder, “A,” to give orders to an assistant named “B.” The language consists of the words “block,” “pillar,” “slab,” and “beam.” A calls out one of these names and B brings a building block of the appropriate type. The peculiar fascination of Wittgenstein’s examples, their pathos and their austere beauty, comes from what one might describe as a form of preterition. Wittgenstein raises but never answers the question of what binds these agents together into communities. Are they parent and child, lover and beloved, jailor and prisoner, Hegelian master and slave? Later in the Investigations, Wittgenstein tells us not to be troubled by the fact that the builder’s language consists entirely of orders. Like Augustine, Wittgenstein believes that the games through which we enter language as children have this lop-sided quality, these asymmetries of authority and desire. Furthermore, similar primitive languages survive embedded in the language of adults:
It is easy to imagine a language consisting only of orders and reports in battle.--Or of a language consisting only of questions and expressions for answering yes and no. And innumerable others.--And to imagine a language means to imagine a form of life. (19)
If this is reassurance, it is the sort of reassurance one associates with Kafka. We would not want to speak these languages, the language of an examination or interrogation where one can only answer “yes” or “no,” the language of the soldier’s reports or of the orders followed by the mute builder’s assistant. Unfortunately, these primitive languages seem to be part of our own language. Wittgenstein’s absurdly simple examples unfold into structures that implicate us. It is not clear that we can avoid playing these games, inhabiting the role of A or B. When we read a philosophical text like Wittgenstein’s, are we playing B? When we read as literary critics, are we playing A? Part of the difficulty of the Investigations--the difficulty, that is, of paraphrasing Wittgenstein or abstracting principles from his work--may come from Wittgenstein’s reluctance to play the role of either A or B.
It is easy to imagine a language consisting only of orders and reports in battle.
I actually have a hard time imagining that, if the reports are to be any good.
Sorry, Matt. I hope that didn’t sound snarky since I didn’t mean to be. I just know doodly about Wittgenstein and so my reaction is no doubt sophomoric. But I just don’t know how to parse the famous last sentence from the passage you quote or really the whole thing. Is it really possible to imagine a wholly distinct language that would have its own form of life that would consists of orders and reports in battle--something on the order, to use John’s example, of chess notation? That doesn’t seem likely because actual battles aren’t like chess and require the communication of so much information that a completely specialized language wouldn’t do. In other words, while you can imagine special codes used in war, it doesn’t seem to me possible to imagine a separate language. The implication of that seems to me different than the one you draw, if I understand right: not that commands and authority, etc. are built into language, but that where human beings are concerned, language can’t be commands alone.
Don’t worry, Sean. I know you weren’t being snarky, and I am happy just to know that the post inspired a thought or a question.
I actually can imagine a language that would consist entirely of commands and reports in battle. Ants and even slime molds coordinate complex activities using an extremely simple set of algorithms and messages, as Steven Johnson explained in his wonderful book Emergence: The Connected Lives of Ants, Brains, Cities, and Software. And armies of ants can be surprisingly effective on the strategic level. But I think you are entirely right to find the idea disconcerting. Like many of Wittgenstein’s counter-factuals, this one is designed to be strange and disconcerting, to produce a powerful esthetic and emotional response. Wittgenstein’s philosophical thought-experiments, in other words, are also literary experiments. I think the same can be said of Foucault, whom I disagree with on almost everything but I whom I nonetheless see as a literary artist of a very high order. Foucault’s anti-humanist story of people without interior selves ("The soul is the effect and instrument of a political anatomy; the soul is the prison of the body") interestingly resembles Wittgenstein’s argument against the possibility of a private language.
To turn to something very different and largely irrelevant but possibly amusing with regard to the idea of a simple language consisting strictly of battlefield reports and commands, do you know Tolstoy’s historiographical meditation in War and Peace, which I mentioned in a previous post? Tolstoy points out that orders issued to Napoleon and reports brought to him had almost nothing to do with his success or failure in battle. At the battle of Borodino, “These instructions - which strike one as exceedingly confused and obscure, if one ventures to throw off the superstitious awe for Napoleon’s genius in treating of his disposition of his troops - may be condensed into four points - four commands. Not one of these was possible to carry out” (892). Moreover, “during the whole battle Napoleon was so far from the scene of action that (as it turned out later) he knew nothing of the course of the battle, and not a single instruction given by him during the fight could possibly be executed” (895).
Similarly, Napoleon intended to invade England rather than Russia: “Napoleon was giving commands all through his reign for an expedition to England. On no one of his undertakings did he waste so much time and so much effort, and yet not once during his reign was an attempt made to carry out his design. Yet he made an expedition against Russia, with which, according to his repeatedly expressed conviction, it was to his advantage to be in alliance . . . .”
“Our false conception that the command that precedes an event is the cause of that event is due to the fact that when the event has taken place and those few out of thousands of commands, which happen to be consistent with the course of events, are carried out, we forget those which were not, because they could not be carried out. Apart from that, the chief source of our error arises from the fact that in the historical account a whole series of innumerable, various, and most minute events, as, for instance, all that led the French soldiers to Russia, are generalized into a single event, in accordance with the result produced by that series of events; and by a corresponding generalization a whole series of commands too is summed up into a single expression of will” (1362). Ants do it better.
Whoops! I meant “orders issued BY Napoleon,” not TO him. And “but whom I nonetheless see as a literary artist of a very high order,” not “but I whom I nonetheless see as a literary artist of a very high order.” Sorry for the hasty typing.
Oh, I see Matt. I wouldn’t have taken ant and slime-mold coordination to be language and am still intuitively doubtful about the thought that they are. But I’m out of my depth here. I think the Tolstoy example fits with my original intuition. Trying to communicated about anything that complicated would require I think a language that would have to be able to serve the purpose of more than commands and reports alone.
The very fact that Augustine’s argument is considered worthy of being taken seriously and refuted is significant to me. Very interesting—I wish you’d done the work on Hegel, too.
Alas, I don’t understand Hegel’s work well enough to do the job. Perhaps some other member of the Valve community does.
Nietzsche was also a big influence, but a more diffuse one, and difficult to talk about. And that which one cannot speak of, one must pass over in silence.
Matt has noted how Wittgenstein truncates Augustine’s story—perhaps oddly, perhaps with reason, given the arguments that follow. Even within the quotation itself, however, I think we find some noteworthy items—items elided somewhat by Wittgenstein’s analysis.
Here’s an offhand list, with no claims to thoroughness:
There is more to Augustinian language-learning than ostention and “pointing.” There is an entire host of bodily moments – gestures, expressions, eyeplay, and tone of voice.
This web of “body language” is meaningful because it is intentional – it communicates and embodies the elders’ intentions.
This body language seems to communicate more than just the names of objects. Actions, feelings, attitudes, and (social) relations are communicated as well (or at least Augustine implies as much).
These forms of language and language-learning have a “natural” basis – a core of shared human aptitudes, at the least, and something like innate ideas, at the most. Regardless, the ability to learn language is presented as innate, and such innate capacities are based in bodily predispositions and interactions.
Children, to continue, are not taught language. They are born language-learners.
Children learn language, not word by word, but by listening to language in use – in “sentences” that express intentions and reflect emotions.
And, finally, looking against at Matt’s snipped ending, we see that language-learning corresponds with entry into a world social roles, systems, and rules.
Maybe this is all oldie-moldy stuff for Wittgenstein scholars. And there is no reason to argue that the Philosophical Investigations should have addressed all these aspects in full. Still it is interesting that the Wittgenstein’s opening gambit reduces the complexity of Augustine’s vision so drastically.
There seems to be a lot that is still alive and interesting in Augustine’s original text – a lot that even gestures towards the “forms of life” (biological and social) to which language is committed.
Strong misreading, anyone?
Wonderful comment, Peter. You have anticipated my next Wittgenstein post, which will be about Wittgenstein and biology (i.e. the language bioprogram hypothesis of the Chomskyites). In a succinct and elegant way, you have explained what is wrong with Wittgenstein as a philosopher. I would love to hear more of what you think about him as a literary artist, though.
Sean, I can offer only a somewhat frivolous response to your serious question. I think that by Wittgenstein’s definition, communication among ants would probably have to count as a language game.
Sorry if I jumped the gun, Matt. I am still thinking about what we gain from labeling Wittgenstein a literary artist — trying to formulate that move so it doesn’t sound too much like a dodge. “Oh, so you think his thought experiments are abstruse or ambiguous or absurd. Well, that’s part of his art — part of his Modernist project!” In other words, I meant to do that!
Still, there is definitely an aphoristic (even an absurdist’s) art — an ability to turn simple constructions back upon themselves, both prompting and confounding our reactions. In this regard Wittgenstein’s words are reminiscent of Kafka and Nietzsche, but also Wilde and possibly Dickinson:
Pain has an element of Blank
I like a look of Agony / Because I know it’s true.
I’m Nobody! Who are you?
I cannot live with You – / It would be Life – / And Life is over there –
I heard a Fly buzz – when I died –
Of course, Wittgenstein isn’t quite this directly, painfully enigmatic. He would have preferred the quasi-rhetorical question, somewhat modest and mostly bullying: Why can we not say that we heard a Fly buzz when we died? Does faith lack a grammar?
The PI is in many senses about primitive forms of language: workmen’s commands, conversation, ostention, “performatives.” Whereas the TLP was somewhat concerned with semantics and logical form, the Witt. of the PI is about syntax; it’s more anthropological, or at least implicitly so, than philosophical. The anti-idealism hinted at in the TLP is fleshed out, and it’s not entirely improper to perceive a behaviorist inclination in many passages of the PI. No, it’s not Pavlov S-R theory or Skinner’s “operants”, but the emphasis on language games as public, of meaning defined by use, etc. (in addition to criticizing the essentialism of Augustine, W. also denies Russellian descriptions and much of the logic of the TLP), all moves in the direction of a sort of materialist psychology where subjective language or subjective mental states are negated; thus literature, certainly subjective, is still in a precarious state, though less so than with the analytical requirements of the TLP. (Solipsism is still a problem for Witt. in the PI and his resolution is not clearly indicated). I do not believe Witt. is jake with Hegelian idealism either.
Given the abstruse nature of the PI and of the occasionally ironic interlocutor/speaker interaction, especially in the later sections, different and perhaps conflicting interpretations are not unlikely. Theists may find some writing not hostile to a religious perspective; Austinian phil. of language types may find sections congenial to their thinking; yet more materialist, empirical types also seem to have at least some support as well: “The human body is the best picture of the Human soul.” So, asked on a date (platonic, of course) by Chomsky or by Skinner or better, Konrad Lorenz, who does Wittgenstein choose? I think W. is closer to the biologists and behaviorists than to neo-idealists such as Chomsky, though many of the issues raised in the PI seem relevant to cognitivism as well, though not of the platonic sort that Chomsky often suggests.