Prescriptivity, Normativity, "Oughtiness," etc. in Classical Chinese

I can’t tell whether this is an easy question to answer or a hard one. How is normativity expressed in Classical Chinese, if it is clearly distinguishable at all from “descriptivity,” for lack of better term (in my mind right now at any rate)?

(I don’t intend to engage in any future “lexical fallacy” arguments, here. I’m not sure yet where this line of questioning is heading in general. It just occurred to me…)

14 replies on “Prescriptivity, Normativity, "Oughtiness," etc. in Classical Chinese”

  1. That a certain asteroid didn’t slam into the earth a million years ago is (a) good, but not (b) normative, nor (c) morally right, nor (d) morally required, nor (e) obligatory, nor (f) in accord with duty. I think these are six different things.

    Classical Chinese has a bunch of things too: li (ritual), yi (right), dao (way), yi (appropriate), de (virtue), shan (excellent), hao (good), etc. The first three seem closer to “normativity” than the others. Classical Chinese seems to differ from English in stressing this other kind of thinnish term: sheng (sage), xian (worthy), junzi.

    Maybe to have “normativity” is to be a norm or be explicitly enjoined by a norm. Maybe norms are verbal imperatives meant for explicit use in practical reasoning. Normativity would then be very different from goodness.

    Manyul, maybe your question springs from the thought that we some one idea or English term, such as ‘ought to’, is fundamental to all those things, and is the proper complement to the descriptive or facts or truth or “is”. I guess the main candidates for the fundamental would be ‘ought/should’, ‘moral’, and ‘good’. So we could ask about each of these three whether Classical Chinese has some very closely parallel form of expression. A fourth candidate might be ‘reason’.

    Manyul, I wonder whether your question has something to do with the “pincers” of naturalism you mentioned earlier.

  2. Hi Bill,

    I really have a few different, specific things in mind, after having thought about your questions.

    (N1) I’ve been thinking a lot about Hansen’s distinction between a “discourse dao” and a “performance dao.” The main difference he seems to be talking about is between a way of doing something that is presented as normative (a “discourse dao” ) and a way in which something is *actually* done (a “performance dao” ). If there is this distinction, then some actual dao can be criticized for not according with a normative dao. One question that comes up naturally (philosophically?) then, Hansen suggests, is how to to adjudicate from among normative daos that are put forth by various schools. But the thing that’s been bothering me about this is how someone would distinguish between a dao, or way, of doing something that is an actual way that someone does or did it (e.g. the actual way that the sage-kings acted; the actual way that a junzi acts/feels; the actual ways in which the Zhou rituals were carried out), which would be a “performance dao,” from some account of doing the thing that is a normative account. Because it seems to me like the only way to express a dao/way of doing something in Classical Chinese is to state it as some description of how someone actually does/did/could do/would do it. There’s no nice(?) “ought” construction available.

    (N2) When someone, in early China, says that an action is ritual (li ye 禮也) or is “right” (yi ye 義也), etc., it is assumed, by people like us, that some kind of normative recommendation or prescription is being presented to the listener/reader. But on further reflection, that seems hazy to me.

    (N3) The term gu 故 seems like a logical operator, meaning “therefore…” Or, maybe it means something that is not normative but indicates a causal relation, “as a result of this cause…” I’m not sure we can take it for granted that the early Chinese distinguished as we do between inferential and psychologistic relationships among our beliefs.

    Maybe there are other related issues. Let’s start there.

  3. Hi Manyul, those are great questions. I don’t have much in the way of answers.


    If, as is sometimes suggested, early Chinese thinkers were mainly concerned not with finding out what to do but with getting oneself and others to do what one already supposes one ought to do, then we should be surprised to find the central vocabulary slighting the distinction between accepting a norm and acting on it.

    Here’s another way to ask part of your question: Is there in fact any case where the phrase “X 之道” definitely refers to X’s intended or announced way that is unfulfilled by X?

    You say Hansen’s “discourse dao” is a kind of action presented as prescribed. In the page you linked, it looks to me as though what Hansen has in mind is the prescription or evaluation itself: the linguistic act or its linguistic product. He writes: “Throughout classical texts, daos are spoken, heard, forgotten, …. I have some mild reservations about ‘speak’ as a translation. It is in some ways too narrow and in others too broad.” But maybe your version fits the texts better. Dao as a verb can mean something like “speak” or “speak of,” but can dao as a noun refer to a linguistic act or object: what is spoken as distinct from what is spoken of?

    Classical Chinese has the imperative mode, sometimes used without any definite issuer, as in the middle of DDJ 4: “Blunt the sharpness; untangle the knots; soften the glare; let your wheels move only along old ruts” (Lau 1963). Pithy Confucian maxims often take this mode.

    Maybe gai 蓋 sometimes has the force of “ought to”? I’m thinking especially of the use in Mengzi 1A7, marked here in red:

    Gai sometimes just means “probably,” but then so does “ought to.” A fair coin ought to come up heads about half the time.


    I suppose you have in mind the fact that Confucius once says something is ritual (li ye 禮也) as a preface to recommending against it (9.3). I don’t know if anyone ever uses ‘yi’ that way. I suppose one might say in English that sometimes one is constrained not to fulfill an obligation (e.g. an obligation incurred by way of a promise), by more pressing moral considerations. But that sort of thing seems easier to say about li. I wonder whether the “hazy” worry you mention applies also to the idea that calling something “just” or “part of being moral” amounts to prescribing it.


    Is inference a non-causal relation among beliefs? A person infers one proposition from another: I think that’s a causal relation between her beliefs. Then there’s the implication or entailment relation between propositions, which I guess is what you had in mind.

    I think you mean gu 故. I agree that we shouldn’t take for granted that early Chinese thinkers distinguished between implication and causation. We shouldn’t take for granted that they actually noticed such a distinction, and we shouldn’t take for granted that they never went wrong by confusing the two. But I’d add that English too seems very friendly to such confusion.

  4. I think my cavilling above about inferential relations was wrong. ‘Beliefs’ can be used to refer to propositions, and an inferential relation between such beliefs is that one of them may properly be inferred from the other. Sorry.

  5. Maybe that gai 蓋 is really he 盍 “why not?” – a rhetorical question implying “There is reason.” Would that be a rough equivalent of “ought to”?

  6. Hi Bill,

    Doesn’t gai 蓋 have more of a “suggestion” meaning, as in “Why not…” or “Perhaps you may…”(e.g. “Why not return to the roots?”/”Perhaps you may return to the roots.” for 蓋亦反其本矣)? But maybe that is enough like “ought to.” I’m not so sure.

    The point about li ye 禮也, yi ye 義也, and the like is, as you suggest, like when someone says something “is just” or “is moral,” but more specifically if the speaker is thinking of contingent, specific institutions as “justice” or “morality”–something like *the current system of justice* or *conventional morality*. There seems to me to be something like that going on when something is declared to be yi 義. Or, perhaps not; the point would be that in contemporary discussions (or back in Plato’s dialogues), we can disambiguate by talking about “justice itself” or “Morality” with a capital ‘M’ or something like that. With li 禮, maybe that is clearer since there does seem to be an actual, contingent institution to which it most often refers. With yi 義, at least some of the discussion in Mencius 6A4-5 suggests to me that there are some pretty specific conventions involving age and kin-relatedness that are the topic of discussion under the rubric of yi. I would concede without much fight that yi is broader than that, but I’m not sure the generality of “right” or “moral” is appropriate as an interpretation.

    Still, even with conventional systems, e.g. etiquette, I suppose saying “That’s the polite thing to do” does seem to endorse. So, maybe N2 isn’t really problematic? (Am I my own formidable opponent?)

  7. I’ve become an epiphenomenon!
    Maybe gai/he just doesn’t say much, as you suggest. Or maybe there’s a kind of etiquette such that one says “Why not?” to suggest “I don’t see any reason against” to suggest “There is adequate reason for.” What would the principle of etiquette be, to ground such a practice? Do the characters in early Chinese writings tend to shy away from saying directly “You’re wrong”?

  8. To take a brief stab at the question I would say that terms such as 禮,規矩,則, etc., generally speaking, express normativity (at least for Confucians). In Hansen’s terms these would probably fall under the category of “right performance dao”. The emphasis in the previous sentence, however, would be on generally speaking , because there seem to be circumstances where things such as 禮 are not meant to be normative. In these cases normativity seems to be determined by recourse to 權, 時,義, dao (Hansen’s discourse dao?), 天, etc.

    In this light perhaps the questions worth pursuing would be: What are the sources of normativity? And how does one recognize when terms such as 禮 lack normativity?

  9. Agui, those are good questions! The first is huge, and I don’t have an answer to the second.

    Manyul writes, “I suppose saying ‘That’s the polite thing to do’ does seem to endorse.” Maybe we can generalize the point: commonly saying anything that tends to count in favor of X tends to endorse and can be a way of endorsing. For example: “It’s in your interest to do X.” How about that?

    Terms that can suggest duty are zhi 職,ren 任,and ming 命.

    Another kind of normative expression is to use a word ‘X’ to mean “excellent, true, or proper X.” Sometimes such a usage will be a clearly separate sense of the word, as in Mencius’ use of 王 wang to mean “true king.”

    Another kind of expression:
    ke 可 = permissible,
    bu ke = impermissible,
    bu ke bu = required.

  10. From the Mencius 2B8:

    Legge: “Zi Kuai had no right to give Yan to another man, and Zi Zhi had no right to receive Yan from Zi Kuai.”
    Lau’s translation is virtually identical.

  11. Hey Bill; that’s a really interesting example. That translation of ‘bu de’ 不得 raises all kinds of flags in me. “No right to…” seems to involve too much anachronistic baggage. Would “does not get to…” (preserving the “attain” meaning of 得) seem less so, or does it have the same baggage? How about “has not achieved the status to…”?

  12. ‘Right’ is indeed alarming, but I think it shouldn’t be alarming here because the rights in question are concomitants of one’s role or position, not universal rights. It does seem that ‘de’ here is being used mainly in the sense of ‘attain’. ‘Does not get to’ seems most literally accurate, though it sounds to me too informal. ‘Status’ seems too concretely specific. How about ‘was not in a position to’?

    One would want to look at comparable uses of ‘de’, and offhand I don’t know of any! I’ll keep an eye out.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.