29 March, 2007


At the beginning of his Philosophical Investigations (1945-49), Wittgenstein quotes a passage from the Confessions (398) of Saint Augustine:
When they (my elders) named some object, and accordingly moved towards something, I saw this and I grasped 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 peoples: the expression of the face, the play of the eyes, the movement of other parts of the body, and the tone of voice which expresses our state of mind in seeking, having, rejecting, or avoiding something. Thus, as I heard words repeatedly used in their proper places in various sentences, I gradually learnt to understand what objects they signified; and after I had trained my mouth to form these signs, I used them to express my own desires.
This empiricist account is remarkably similar, in fact, to how Locke described the acquisition of language in his 1690 Essay on Human Understanding. It was an influential account; Wittgenstein, however, rejects it as simplistic: 'Augustine does not speak of there being any difference between kinds of word. . . Augustine, we might say, does describe a system of communication; only not everything that we call language is this system'. It is an account suited to a conception of language consisting of concrete nouns—'apple', 'chair', and so forth—but cannot deal with more sophisticated words—from 'if' and 'but' to 'exasperation' and 'piety'. For Wittgenstein, the acquisition of language comes not from 'explanation', but rather from 'training':
An important part of the training will consist in the teacher's pointing to the objects, directing the child's attention to them, and at the same time uttering a word; for instance, the word "slab" as he points to that shape. . . This ostensive teaching of words can be said to establish an association between the word and the thing. . . But if the ostensive teaching has this effect, —am I to say that it effects an understanding of the word? Don't you understand the call "Slab!" if you act upon it in such-and-such a way? —Doubtless the ostensive teaching helped to bring this about; but only together with a particular training. With different training the same ostensive teaching of these words would have effected a quite different understanding.
The learning of words—the establishment of associations 'between the word and the thing'—can only operate in a pre-determined context, such that the learner understands what it is he is to learn. In this context, which Wittgenstein calls a 'language game', the child knows what it means when a teacher points to a slab and says, 'Slab'.


Isn't it ironic, then, that Wittgenstein's great predecessor in critiquing the notion of 'ostensive definition' was, in fact, Saint Augustine? The Bishop of Hippo had already written his dialogue De Magistro nine years prior to his Confessions; it was the first work he wrote after his conversion to the faith in 387, and still in the mould of his pre-Christian works, such as Contra Academicos. For my money, Augustine's early dialogues are the most sophisticated philosophical works since Aristotle—a stylistic return to Plato, after the dialogue form had been diluted by Cicero's bland moderateness.

De Magistro concerns the role of signs in the acquisition of language and knowledge. It begins with Augustine asking his disciple Adeodatus the question, 'When we speak, what does it seem to you we want to accomplish?' The reply is 'So far as it now strikes me, either to teach or to learn'. The rest of the work is given to countering this thesis. Of things, words can only give ostensive definitions:
When a question is raised about things that aren't signs, these things can be exhibited either by doing them after the query, if they can be done, or by giving signs with which they may be brought to one's attention.
But Augustine is just as aware as Wittgenstein about the problems with this:
If anyone should ask me what it is to walk while I was resting or doing something else, and I should attempt to teach him what he asked about without a sign, by immediately walking, how shall I guard against his thinking that it's just the amount of walking I have done?
In other words, ostensive definition can only teach when the learner knows how to interpret it correctly; he needs a context for the demonstration. We notice not only a similarity of reasoning, but also of style—Augustine, like Wittgenstein, constantly returns to concrete examples and thought-experiments to demonstrate his points. Here's another:
Suppose that someone unfamiliar with how to trick birds (which is done with reeds and birdlime) should run into a birdcatcher outfitted with his tools, not birdcatching but on his way to do so. On seeing this birdcatcher, he follows closely in his footsteps, and, as it happens, he reflects and asks himself in his astonishment what exactly the man's equipment means. Now the birdcatcher, wanting to show off after seeing the attention focused on him, prepares his reeds and with his birdcall and his hawk intercepts, subdues, and captures some little bird he has noticed nearby. I ask you: wouldn't he then teach the man watching him what he wanted to know by the things itself rather than by anything that signifies?
The difference here is that the man watching already knows what birdcatching is. The demonstration only 'jogs his memory' of the craft as a whole. Signs cannot teach the knowledge of things: 'When a sign is given to me, it can teach me nothing if it finds me ignorant of the thing of which it is the sign; but if I'm not ignorant, what do I learn through the sign?' Signs can only remind us of things we already know, or teach us other signs:
Words have force only to the extent that they remind us to look for things; they don't display them for us to know. . . From words, then, we learn only words.
Augustine is essentially a rationalist, like his master, Plato—the only source of knowledge is a spiritual faculty, a prior awareness, of which external evidence can only remind us:
Regarding each of the things we understand, however, we don't consult a speaker who makes sounds outside us, but the Truth that presides within over the mind itself, though perhaps words prompt us to consult Him.
Plato brought in the doctrine of anamnesis—recollection of the world of Forms, experienced by the soul before mortal birth. Augustine instead invokes God; in his account, which is much more rigorous than Plato's Meno, he anticipates the objections made by Wittgenstein to the empiricist account of learning, although he draws from them very different conclusions. If Augustine retreats to an even more internalist epistemology—the criterion of truth being in one's spirit—the Austrian philosopher advocates almost pure externalism: language as a group activity, an acquired behaviour.

Update 28/09/07: Alasdair MacIntyre, in his 1984 essay 'The Relationship of Philosophy to its Past', writes: 'Augustine’s account of the place of ostensive definition in language learning points towards the divine illumination of the mind; Wittgenstein’s very similar account—that Wittgenstein erroneously took his account to be at odds with Augustine’s reinforces my central thesis—points towards the concept of a form of life.' So it's not just me.


Anonymous said...

Have read it several times... I'm not entirely sure I understand the distinction to which you are so beautifully pointing.

Could you elaborate?

Anonymous said...

"For Wittgenstein, the acquisition of language comes not from 'explanation', but rather from 'training'..."

It's poetic that now, with present strengths in the science that Augustine in his way lay so much of the foundation for, we uncover that in-dwelling spirit, or knowledge, residing...in genes. How else to explain it? The tacit-yet-impossibly abstract? The understanding is built directly into the machines.

John B. said...

Signs cannot teach the knowledge of things: 'When a sign is given to me, it can teach me nothing if it finds me ignorant of the thing of which it is the sign; but if I'm not ignorant, what do I learn through the sign?' Signs can only remind us of things we already know, or teach us other signs.

Just the other day I was talking about my daughters' learning the names for colors and how I think it's in that circumstance that we can most easily see the rather arbitrary nature of language. For my younger daughter, every color was "red" for quite some time.

Malone said...

Adeodatus was Augustine's son, was he not?

This post struck a chord with me, since I'm the parent of a bright five-year-old; I have lived on the front lines of language acquisition for some time. In the past few days, I have been obliged to answer the following "what is" questions: "what is infinity? the president? government? war? prosody? copyright law?" Interestingly, my explanation of "infinity" (an entirely abstract concept) and prosody (with reference to "twinkle, twinkle little star") were accepted immediately. We are still talking about "president," "government," "war," and "copyright law." Is that due to the inadequacy of my explanations or my child's innate interests and ability to connect concepts?

Augustine's empiricism and Wittgenstein's reliance on social context both seem inadequate to explain the phenomenon of language acquisition - neither theory recognizes the agency and creativity of the individual in manipulating a symbolic system of communication.

Andrew W. said...

One of Wittgenstein's tricks is that, despite the fact that his philosophical writings rarely mention other philosophers, he was in fact deeply engaged with the philosophical "problems" of his day.

So it's a wondrous thing that Wittgenstein scholar P.M.S Hacker can show that Locke makes a picture-perfect adversary for Wittgenstein's understanding of language, even though it doesn't appear Wittgenstein ever read Locke.

It just so happened that he and Russell had found a new cloak for Locke's ideas.

And as yet another commenter with a small child, despite my chomskian tendencies to see my son's language acquisition as a kind of swtich that turns on certain elements of syntax, I have to say that Wittgenstein's views do indeed, at least to my limited faculties, and pace Proserpine, recognize the agency and creativity of the language user much more than the Lockean conception.

Just take those language games at the start of the Philosophical Investigations - those are complete languages in and of themselves, not primitive languages. They require something of both speaker and hearer that goes beyond ostention.

Wittgenstein's account is very much informed by the peculiar fact that people seem to know when to use words correctly, which can often mean using them in ways they've never been used before, yet nonetheless correctly.

So when people say Joyce writes nonsense (and they often do, at least where I'm from), it's a Wittgensteinian account that helps us to understand why Joyce is a great writer, and not the pretentious hack he's taken for.

Thanks Conrad, for showing us how Wittgenstein used the wrong straw man - perhaps he knew Augustine had a more sophisticated view of language, but I suspect not.

And bravo for marking a more fruitful path of exploration with respect to Augustine's philosophy of language!

Conrad H. Roth said...

What a feast! I feel a bit left out now, not having a kid...

Hank: Are you referring to my final sentences? If so, what I meant was that for Augustine, the signs displayed by others in teaching (eg. words, pointing, etc.), can only push us towards consulting our own, inner intellectual faculty (or God)--the latter are the source of truth and understanding, not external sensations. Language plays a very limited part. Whereas for Wittgenstein, there is no 'inner truth' or 'inner language'--language consists rather of a series of learned behaviours which are conducted in a group setting.

Steve: yes, the rationalists are gaining day by day. And 2 centuries ago they thought they were beaten!

John: A good illustration!

Proserpine: yes he was. I had forgotten that rather important fact. I'm amazed your five-year-old could ask you about prosody and copyright law--must be pretty precocious!

Otto: It is an intriguing aspect of W, indeed; you feel like you're reading him in a vacuum. W's position, like Chomsky's, does push language to the front of thought--Augustine's, on the other hand, devalorises language and separates it from the spirit, which he owes more to the divine. Words--breath, spirit--become almost a useless parody of the Spirit. I think Augustine was reacting against his own language-heavy training in rhetoric.

Anonymous said...

Pardon my spelling. You've gotten me excited and agitated. This is wonderful. I confess, I'd never thought about many of these things in these ways.

Of course I'm familiar with the Bishop of Hippo and his City of God/City of Man comparisons. I'd never before heard of Wittgenstein (thank you).

(Does everyone know this stuff but me?)

John's comment was especially poignant for me, as they so often are. It is APPARENT that he is a teacher.

"Signs cannot teach the knowledge of things...."

But this discussion shakes me. Rather, it shakes a long and firmly held belief of mine. (And that's never fun.) I'd always found some real joy in the following quotation (really! joy!).

"The quality of our thoughts is bordered on all sides by our facility with language.” J. Michael Straczynski

(I think J. Michael Straczynski was/is some kind of Star Trek guy or something... or so I hear. I originally heard the (mis)quote as Oscar Wilde (who is always the safest attribution I find).)

I have always believed that knowing 100 words (labels/categories) for "red" is DIRECTLY linked to being able to SEE and UNDERSTAND 100 hues of "red." For me, it has always been language that leads to ideas. I know the difference between Cobalt Blue and Ultramarine and when I see a painting, I can SEE the two colors. AND the gradients in between.

But I think I'm wrong. I think I've BEEN wrong.

I wonder if there are discrete phases of learning… like acquisition of basic skills and then refinement and then something else… like, maybe we begin to THINK DIFFERENTLY at different points. I'd always thought of it as ONE continuum, but maybe it is several in some kind of sequence?

I'm going to have to look into this.

And I just wanted to play some golf today. Damn.

Anonymous said...

Thanks, Conrad; I wish you had written this a year ago, when I was knee-deep in Wittgenstein.

Thanks Conrad, for showing us how Wittgenstein used the wrong straw man - perhaps he knew Augustine had a more sophisticated view of language, but I suspect not.

It's worth pointing out that Aug.'s view of language is not more sophisticated than this in the Confessions, though; and his later chapters on time and memory are also dealt with in PI; and that choosing Augustine as his straw man brings a quiet religious context for the PI which reverberates in interesting ways; but I am too feverish and sickly right now to elaborate.

Still, I haven't heard about this part of Aug.'s thinking before; De Magistro is on my to-read list, but it has now been bumped up a few notches.

Conrad H. Roth said...

Hank: I'm flattered to have had such an impact. I wouldn't want you to think that I actually agree with Augustine, however. I have more sympathy for W. The issues of the actual psychological / linguistic development of children has been of great interest to linguists for the last 50 years, and has become quite scientific--thus putting it somewhat beyond my ken.

Chris: I am sorry to hear you are sickly, but nonetheless I hope at some point you might enlighten us with your insights into W's relationship to Augustine.

Anonymous said...

The seed of the idea, which I have not let germinate properly, goes something like: The religious seems to fall for Wittg under "that which one can say nothing about", although it's also very important to him; Augustine also seems to feel something like this, but is determined to try to say something about it anyway. And this leads him down all sorts of incorrect thinking (about language, time, memory) -- but he has used these in the Confessions to build some sort of proof (or just a suggestion?) of God! So Wittg pulls the rug out from underneath Augustine -- but, perhaps, without making him fall.

Conrad H. Roth said...

"The religious seems to fall for Wittg under "that which one can say nothing about""

Have you read, by any chance, this?

"Augustine also seems to feel something like this."

You really think so? I can see a comparison to Augustine's famous remarks about time; but surely Augustine thinks quite the reverse about religion! He does, after all, say a hell of a lot about it; and there is no sense of grim determination in his writing. There is a wonderful sense in works like The City of God, On the Trinity, and On Christian Doctrine that Christianity is something reasonable, philosophical and communicable. His main critique of the religious currents of his time--both Gnostic theurgy and the decayed remnants of Roman state religion (which he cites in Varro)--is that they are confused and unreasonable.

Anonymous said...

Ah it's been a few days. And reading Auerbach's "Literary Language and Its Public" I found a quote that sideways gets at my thoughts.

From De Trinitate:

...when you hear, "He is truth". Do not ask what truth is. For mists of corporeal images and clouds of phantasm will rise forthwith and confuse the clarity that flared up in you in a first impulse when I said: "Truth." When the word truth is spooken, remain if you can in that first impulse which struck you as a flash of lightning. But you cannot; you fall back into this world of familiar, earthly things...

Now, this is at least Wittgenstein in its belief in the inability to come up with a total definition for "truth". But it also seems not unlike his conversion experience, with its understanding in a flash of lightning. We can, Plato-style, try to search for that exact definition of "truth" -- or of what that religious conversion is -- and it might be interesting and useful (to prepare the mind for that moment -- I believe the term "grace" should appear around here?) and certainly someone with Augustine's sensability isn't going to be able to prevent himself from thinking very easily. But all that cogitation is, at best, an approximation of the actual definition, of the actual shape of faith. I haven't read enough Augustine to know if he thinks that there's a chance his thinkings might in some way be misleading, even -- I suspect so, and can only assume his mind, which was certainly clever enough, would have gotten him to that point eventually. Certainly he doesn't seem to think you need or should even want his intellectual abilities and curiosities to be a good Christian or to "understand" faith.

So yes; certainly it didn't stop him from trying, but I think he didn't feel like he was ever going to be able to talk about the thing itself, just maybe outline the implications of the thing.

Conrad H. Roth said...

I see what you're saying, although I would want to draw quite a sharp distinction between the Neoplatonic idea of God as being essentially unknowable (which Augustine inherited from his pagan training), and the Wittgensteinian notion that the whole field of religion is excluded from the realm of discourse, on the basis of something like a verifiability criterion.

Augustine wants in the quoted passage to retain something of the 'numinous' feel of the Christian revelation; but in the rest of this text he repeatedly offers his reader a way to understand one of the two central paradoxes of Christianity (ie. the Triune Godhead) by the philosophical method of analogia. This strikes me as a thoroughgoing rationalism, even if it is (as it was in Plato) in the service of an ulterior mysticism.

Books 10 and 14 of De Trinitate, incidentally, contain accounts of language-learning that also conflict with the Confessions, though an exposition of this (if it is wanted) will have to wait till another time.