Augustine on semantic indeterminacy

St. Augustine’s dialogue The Teacher is concerned with the nature of language.  There are several passages in it which address what twentieth-century philosophers call semantic indeterminacy -- the way that utterances, behavior, and other phenomena associated with the use of language are inherently indeterminate or ambiguous between different possible interpretations.  Let’s take a look.  (I will be quoting from the Peter King translation, in Arthur Hyman, James J. Walsh, and Thomas Williams, eds., Philosophy in the Middle Ages, Third edition .)The dialogue is a discussion between Augustine and his son Adeodatus.  Several pages into the dialogue (at pp. 12-13 of the text I’m quoting from) the question arises whether someone can teach another person the meaning of a term without using words or other signs such as pointing one’s finger at a thing, but instead just by way of one’s actions:

Augustine: What if I should ask you what walking is, and you were then to get up and do it? Wouldn't you be using the thing itself to teach me, rather than using words or any other signs?
Adeodatus: I admit that this is the case.  I'm embarrassed not to have seen a point so obvious.  On this basis, too, thousands of things now occur to me that can be exhibited through themselves rather than through signs: for example, eating, drinking, sitting, standing, shouting, and countless others.
Augustine: Now do this: tell me-- if I were completely ignorant of the meaning of the word ['walking'] and were to ask you what walking is while you were walking, how would you teach me?
Adeodatus: I would do it a little bit more quickly, so that after your question you would be prompted by something novel [in my behavior], and yet nothing would take place other than what was to be shown.
Augustine: Don’t you know that walking is one thing and hurryinganother?  A person who is walking doesn't necessarily hurry, and a person who is hurrying doesn't necessarily walk.  We speak of 'hurrying' in writing and in reading and in countless other matters.  Hence given that after my question you kept on doing what you were doing, [only] faster, I might have thought walking was precisely hurrying -- for you added that as something new -- and for that reason I would have been misled.
End quote.  Augustine’s point is that the behavior Adeodatus was proposing as a means by which one may teach the meaning of the word “walking” is ambiguous or indeterminate between the meaning walking and the meaning hurrying.  Nothing in the behavior considered by itself could determine one or the other interpretation, nor could it rule out yet some other possible interpretation (such as jogging or being chased).  Hence exhibiting that behavior could not by itself teach the meaning of “walking.” 
Later on in the discussion (at p. 27), Adeodatus himself reinforces the point with a related but slightly different example:
Adeodatus: … For example, if anyone should ask me what it is to walk while I was resting or doing something else, as was said, 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?  He'll be mistaken if he thinks this.  He'll think that anyone who walks farther than I have, or not as far, hasn't walked at all.
Here the idea is that by walking six feet (say), you will have done something the meaning of which is indeterminate between the meaning walking and the meaning walking six feet.  Hence if someone asked you what “walking” means and you carried out that behavior in response, he could come away thinking “Oh, ‘walking’ means moving in that manner” but he could also come away thinking “Oh, ‘walking’ means moving six feet in that manner.”  Again, since nothing in the behavior considered by itself could determine either of these meanings or some other meaning altogether, the behavior by itselfcould not suffice to explain the meaning.
Now, you might think that further behavior that provides a larger context for the walking, or gestures, or explanatory utterances, or other elements of the overall communicative environment, will suffice to determine which meaning is intended.  Augustine himself doesn’t pursue the issue much further, but in fact the indeterminacy would afflict all of these other aspects of the situation as well.  This is the lesson of examples like W. V. Quine’s “gavagai” example in Word and Object, and Saul Kripke’s “quus” example in Wittgenstein on Rules and Private Language.  Any collectionof behaviors, gestures, and even utterances, will be ambiguous or indeterminate between different possible interpretations.  Even if you add to the story mental pictures and other images, including inner “utterances” -- as when you call before your mind the way that the sentence “By this action I mean walking!” sounds, or the way the sentence looks when written out -- that will not solve the problem, because those images are also susceptible of different possible interpretations. 
So what does determine what is meant?  Here different philosophers offer different answers.  Quine famously held that there simply is no fact of the matter about what one means by an utterance.  Meaning is not merely indeterminate from behaviorand the like, but indeterminate full stop.  But Augustine would not agree with that.  (Which is a good thing, since, as I have argued many times, the idea that there is no determinate meaning full stop is incoherent.  See e.g. my article “Kripke, Ross, and the Immaterial Aspects of Thought,” reprinted in Neo-Scholastic Essays .) 
But again, what does determinate what is meant?  Augustine doesn’t say much about that -- the indeterminacy of semantic content is not his main topic, after all -- other than to note (at p. 28) that someone who is intelligent will be able to figure out the significance of behavior, a judgment with which his son concurs:
Adeodatus: … If he is sufficiently intelligent, he’ll know the whole of what it is to walk, once walking has been illustrated by a few steps.
Of course this is, in one sense, not terribly informative or helpful, even though it is perfectly true that we typically are able to figure out what is meant by different behaviors.  For we want to know exactly how an intelligent person figures out the meaning, given that the behavior is inherently ambiguous or indeterminate in its significance. 
In another way, though, Augustine’s point is a deep one, even if this is best seen by reading it as an answer to a question that is not exactly the one he was addressing.  Materialist or naturalist accounts of thought and its content typically suppose that they can be explained in terms of causal relationsof some kind.  The idea is that a thought will have the content that (say) the cat is on the mat if it bears the right sort of causal relation to the state of affairs of the cat’s being on the mat.  Spelling out what the “right” sort of causal relation would be is where things get very complicated.  And the main issue is that indeterminacy problems afflict every attempt to spell out the analysis.  For the state of affairs we call the cat’s being on the matcan also be described as a state of affairs involving a domesticated mammal’s being on the mat.  So why does the fact that this state of affairs causes the thought entail that the thought has the content the cat is on the mat as opposed to the content a domesticated mammal is on the mat?  You can add details to the description of the causal relation to get around this problem, but the revised account of the causal relation will in turn face indeterminacy problems of its own.  (An example would be Fred Dretske’s account of semantic content, which I discussed in a post a few years ago.)
At the end of the day, the indeterminacy can only be eliminated by simply conceptualizing the relevant causal relata in this specific way rather than that way.  That is to say, it can be eliminated only when there is an intellect present which can do the needed conceptualizing.  Yet the whole point of the causal theory of content was to explain where thoughts having a certain conceptual content come from.  So any such theory must fail.  It inevitably must presuppose the very thing it was supposed to be explaining.  (This is a point which has been made in different ways by Karl Popper and Hilary Putnam, and which I develop in “Hayek, Popper, and the Causal Theory of the Mind,” also reprinted in Neo-Scholastic Essays.)
The deep point implicit in what Augustine says, then -- though again, this isn’t really the set of issues he was addressing -- is that the intellect’s grasp of meanings is more fundamental than any behavior, gestures, utterances, aspects of the communicative context, etc. that might be used to teach or express meanings.  Hence you are not going to be able explain the former in terms of the latter.  You are not going to be able to reduce intelligence to patterns of behavior or dispositions to behavior (as the behaviorist holds), or explain it in terms of causal relations between the human organism and aspects of its environment (as causal theories of content hold), etc., because the behavior, causal relations, etc. have whatever semantic associations they have only by reference to an intellect which grasps those associations.  The intellect is itself the central and irreducible element of the semantic situation.  (It is irreducible to inner “utterances” and other mental imagery too.  When I entertain the thought that the cat is on the mat, I might “hear” in my mind the English sentence “The cat is on the mat,” but that auditory image is not itselfthe thought.  See “Kripke, Ross, and the Immaterial Aspects of Thought” for more on this subject.)
In this vein, Augustine also notes (at p. 29) that it is a mistake to think that gestures like pointing are the key to understanding meaning.  Pointing one’s finger is, after all, itself just another piece of behavior susceptible of alternative interpretations, and is not in the first place fundamentally about the thing pointed to at all:
Augustine: … I don’t much care about aiming with the finger, because it seems to me to be a sign of the pointing-out itself rather than of any things that are pointed out.  It's like the exclamation ‘look!’ -- we typically also aim the finger along with this exclamation, in case one sign of the pointing-out isn’t enough.
In other words, what pointing primarily does is to call attention to the fact that the one pointing is trying to call our attention to something, and only secondarilydoes it indicate the thing that is being pointed to.  This reflects the fact that the presence of an intellect is fundamental to the semantic situation, and the significance of gestures, utterances, actions, etc. is only derivative.  Imagine a garden hose lying on the ground in such a way that it seems to “point” to a certain tree.  We don’t regard this as genuine “pointing” -- in the sense that deliberately aiming your finger at someone is genuine pointing -- because we know that the hose does not have an intellect and thus cannot be trying to call our attention to something.  We would regard it as genuine pointing only if we supposed some person had come along and arranged the hose that way in order to get us to notice the tree.
It would be absurd, then, to try to explain how intellect gets into the picture by starting with meaningless physical elements and their behaviors, then supposing that some kind of “pointing” arises in sufficiently complex systems -- say, by means of causal relations of some sort -- and then in turn supposing that intellects arise in some subset of these systems which cross some yet higher threshold of complexity.  All of this would get things precisely backwards.  For “pointing” of the relevant sort could arise only where there is already an intellect present, which intends by the “pointing” to call attention to something.
Related posts:
Augustine on the immateriality of the mind
Augustine and Heraclitus on the present moment
Kripke contra computationalism
Oerter and the indeterminacy of the physical
Oerter on indeterminacy and the unknown
Do machines compute functions?
Can machines beg the question?
Da Ya Think I’m Sphexy?