April 12, 2015

What is understanding a sentence?



Michael Johnson gives the following argument for compositionality in his article on the Internet Encyclopedia of Philosophy:

We are capable of understanding a very large number—perhaps an infinite number—of sentences that we have never heard before. …
If we understand the meaning of a new sentence whose meaning we haven’t been specifically taught before, it must be that we can work out its meaning from information available to us when we hear that sentence and other things that we have already learned.
Suppose for a moment that English is a compositional language, in the sense that the meaning of a sentence of English can be computed (worked out) from its syntactic structure and the meanings of its morphemes. This would explain how one could understand a novel utterance [shouldn’t this be “sentence”?] such as There is a cauliflower-shaped spacecraft from Saturn on television. English speakers who have never learned the meaning of this sentence specifically have nevertheless learned the meanings of each of the words in it: cauliflower, shape, the past tense morpheme -ed, spacecraft, and so forth. Furthermore, part of mastering a language involves acquiring the ability to parse sentences of that language, that is, to figure out their syntactic structure—for example, figuring out that cauliflower-shaped modifies spacecraft, but on television doesn’t modify Saturn. Thus if English is compositional, English speakers have all they need to understand novel English sentences they have never encountered before—provided those sentences don’t contain unfamiliar words. (Michael Johnson, “Compositionality” , Internet Encyclopedia of Philosophy¸ accessed 11 April, 2015)

This is one version of a familiar line of argument. However, it doesn’t seem clear to me what precisely it is supposed to prove. I had never heard the sentence, “There is a cauliflower-shaped spacecraft from Saturn on television” until I read this article. I immediately recognized it as an English sentence, and I suppose I can say I understood it: it didn’t bewilder me, I understood how it was meant to be taken. (This was partly due, no doubt, to the context in which I encountered the sentence, as an instance of “sentences-understood-though-never-before-heard”.)
I could, as it were, identify the place of the sentence in the language calculus. Or we might say, I can state its truth-conditions (“the sentence p is true if and only if p”). But considered as an achievement, it seems to me, this is tautologous. Mastery of the system gives me mastery of the system; the question, however, is what enables us to reach beyond the language cage. After all, compositionality I take it is supposed to explain how we are able to communicate by means of combinations of words we have never encountered before, and all its seems to explain is how we are able to recognize combinations of words as the combinations of words they are. Suggesting that understanding a sentence takes us beyond mere recognition because it enables us to tell what the truth-conditions of the sentence are is no help: for one thing, this is true only if truth-conditions are understood as tautologically conveyed by the sentence itself. (“’Snow is white’ is true if and only if snow is white.” ) Second, if I don’t understand what the speaker is up to I don’t know how he means for his words to be related to the question of truth: is he trying to tell me something, musing on a formulation, telling a joke, composing a line for a science fiction story, planning the interior design of a room (“On top of the television set there will be this funny-shaped spacecraft-looking thing.”)?
            In fact, the condition that I must know the words in the sentence in order to understand it – in this sense – seems gratuitous, as long as I know what grammatical categories they belong to. Thus, to cite a famous example, while I have no idea what a runcible is, I have no difficulty understanding that the sentence “Smith kept a runcible at Abbotsford” is true if and only if Smith did indeed keep a runcible at Abbotsford. Why should my familiarity with the words have a bearing on my ability to extract the truth-conditions from the sentence?
            The idea seems to be that compositionality can explain the mysterious leap from simply hearing a sequence of words to a state called “understanding the sentence …”, a state which, although it does not yet tell me what I am to do with the words spoken, yet puts me in a position to deal with them wherever I might encounter them. And I doubt whether any such state needs to be postulated.

6 comments:

  1. "The idea seems to be that compositionality can explain the mysterious leap from simply hearing a sequence of words to a state called “understanding the sentence …”, a state which, although it does not yet tell me what I am to do with the words spoken, yet puts me in a position to deal with them wherever I might encounter them. And I doubt whether any such state needs to be postulated."

    I agree with this, and would add that reflecting on the complexity of what it means to understand something may show how problematic a search for "a state of understanding" may be:

    If someone says that “Smith kept a runcible at Abbotsford”, the first problem may not be to understand the truth conditions. If that is the case, you would have to understand that he, for instance, wants you to go get the runcible. In that case you could say that you need some "truth conditions" that are not tautological, i.e. some way of identifying what is talked about (I would call these then truth conditions). But as the sentence stands, it is not obvious that anybody wants you to get the runcible. Maybe you can understand from the context that a runcible is something very fancy, and the person telling you is informing about it to you out of envy. Last, it might be something like a code that a spy tells you mistaking you for an underorcover agent. Even this could, perhaps, be understood from the context. In this case you would not have much to go on from, and could not perhaps answer the person much more than "you must have mistaken me for someone else". Still, this would show a kind of understanding of the sentence.

    So if the theory is supposed to explain a "state" of understanding, then which state is in question? I believe that as soon as this question is realized then the inklination of making theories of understanding dissipates. It seems that there endless amount of "states" of understanding, which all enables a person to respond in different ways. It would seem that if we were to ask "when has someone understood a sentence", this would be a moral question. I.e. in different situation we have different expectations for an understanding; this expectations would then be an expression of what kind of relationship we want to have with another person, or what kind of relationship we ought to have.

    I happen to have a McIntyre book here beside me, so i come to think of a point of criticism of contemporary culture that would perhaps be in line of his critisism of emotivism: sometimes it feels that the level of expectation of an understanding is fairly low in social situations. That the most important factor is a kind of like/dislike category. The most important thing in many situations (im thinking facebook and youtube posts) is to figure out whether one likes something or not (hence, the like/dislike buttons). Understanding what someone is saying as a way of forming relationships to one another is not of importance (yes, I count youtube as a form of social interraction). If this is in any sense correct, then the risk is that this way of understanding what other people say can dull our senses, and make it more difficult to understand phenomenon or other people when a more complex understanding is called for. It is sometimes said that entertainment works to dull us. I do not know whether entertainment is the cause or the effect of this, but some internal connection seems to be there.

    This is, i feel, an example of the social and moral dimension of the question of understanding. The moral and social dimension, i feel, is obscured when the meaning of a state of understanding is simply taken for granted, and the main question is instead to explain how an assumed understanding happens.

    -MK

    ReplyDelete
    Replies
    1. Thank you very much for your comment! Indeed, I think it serves to deepen the point I was making. What I take you to be saying is this: not only does the understanding of one and the same range of words vary with the context of speaking, but even in a single case what will count as having understood what was said may be indeterminate. What real understanding amounts to may even be seen as a moral issue.

      Yes, there are contexts, such as social media, which encourage a superficial view of what is involved in understanding. I don’t suppose this is a new phenomenon, though: haven’t there always been various kinds of social gatherings in which one is expected to say, “I see what you mean” and not mean very much by it.

      I hope my understanding of your comment has not been superficial. In any case, it reminds me that there is much more to be said along these lines.

      Delete
  2. I basically agree. But as far as I know what has been taken to be compositional after Grice, is not the whole "meaning" of a sentence, but something more narrow, perhaps something close to what Grice called "what is said". Of course there is no agreement on how to understand this technical term.

    The minimalists claim that the utterance content (i.e. "what is said") does not include any pragmatically determined elements, except disambiguation and reference fixing for indexicals, that are not triggered by grammar.

    The contextualists disagree, and claim that the semantic content is determined by a number of pragmatic processes. (See the work done by the relevance theorists and Recanati etc.)

    So in some sense perhaps the contextualists are closer to a broadly Wittgensteinian perspective, but not very close because they still accept a Gricean mentalistic picture of meaning and communication, which I think is a grave error and leads to nowhere.

    But I do think that something like compositionality is a problem for Wittgensteinians. We need better theories or at least descriptions of how humans become competent members of a community of speakers. It seems to me that there is in humans some kind of initial ability that is then used as a bootstrap to gain more complex habits and abilities, which is again used to gain more competence and so on (this is all compatible with Wittgenstein, I think). We need research in AI and models that make Wittgensteinian ideas of language more practical. The "truth conditionalists" after all have things like Montague grammar and things that seem to work (at least to some extent). And there is a lot of work done on the syntax-semantics-pragmatics interface that apparently many people think is productive.

    So unless we have an alternative research program I don't think the Wittgensteinians will be properly heard in these debates. I think there has recently been interesting things happening in this area more broadly (in AI etc.) so I'm hopeful, but it's not my area of expertise as a student.

    Johannes
    (Sorry if I multi posted, I'm not sure if the earlier posts came through.)

    ReplyDelete
    Replies
    1. “… a Gricean mentalistic picture of meaning and communication, which I think is a grave error and leads to nowhere”

      - Important point; but all the same one should keep in mind the degree to which what we think of as communication varies with the situation of speaking. There is (or ought to be) a lot of psychology – a very special kind of attention - involved, say, when a criminal investigator interrogates a five-year-old victim of sexual abuse, when you’re consoling a heart-broken teenager, or when you’re out on a first date. The situation is quite different when you’re trying to interpret the clauses in a contract or read a user’s manual for putting together an IKEA bed. Less verbal nitpicking, more reading between the lines in the first type of case. The lesson seems to be: beware of any unified conception of what constitutes communication.

      “So unless we have an alternative research program I don't think the Wittgensteinians will be properly heard in these debates.”

      - I’m not an expert either, but like you I would be surprised if people who do research on language learning pay much attention to Wittgensteinian ideas these days – it would be interesting to hear of any counter-examples. (For a case of non-attention, see my earlier blog http://languageisthingswedo.blogspot.se/2012/04/st-augustine-rides-again.html.) But then I doubt whether we are to think that the best Wittgensteinian contribution would be an alternative theory of language learning. Psycho-linguistics, Wittgenstein would have said, is not an a priori science, it has to be carried out through experiment and observation. The contribution, rather, would be in the form of a dialogue with researchers, pointing out that certain widely held assumptions may be questioned, suggesting interesting avenues for new research, etc.

      “It seems to me that there is in humans some kind of initial ability that is then used as a bootstrap to gain more complex habits and abilities, which is again used to gain more competence and so on (this is all compatible with Wittgenstein, I think).”

      - Yes, this sounds like common sense on the whole. But in this context there is often a danger of dressing up conceptual (internal, logical) connections as though they were empirical.

      Delete
  3. "...I would be surprised if people who do research on language learning pay much attention to Wittgensteinian ideas these days – it would be interesting to hear of any counter-examples."

    What I was thinking by the "interesting things happening" recently, was work by researchers like Michael Tomasello on language acquisition etc. and the embodied cognition stuff.

    And also research in AI using "embodied" robots that have sensory-motor apparatus. (See for example https://ai.vub.ac.be/research/topics/evolutionary-linguistics , an entertaining TEDx video relating to this: https://www.youtube.com/watch?v=YmcOCDVsBlo , or this paper: https://www.csl.sony.fr/downloads/papers/2012/steels-12c.pdf .)

    Yes you can find things to criticize in these things, but they are steps in the right direction I think.

    "The contribution, rather, would be in the form of a dialogue with researchers, pointing out that certain widely held assumptions may be questioned, suggesting interesting avenues for new research, etc."

    Yes that's a standard Wittgensteinian line to take, I'm ambivalent about it.

    ReplyDelete
  4. Thank you for the references. (As for Wittgenstein-inspired research on language acquisition, this time in the case of bonobos, I'd like to mention Segerdahl, Fields & Savage-Rumbaugh, Kanzi's Primal Language.)

    It may be the standard line, but there are good reasons why it is. Keeping track of what's empirical and what is not is never a bad idea.

    ReplyDelete