Saturday, April 11, 2009

Misleading Philosophical Jargon

Which philosophical terms do you think do more to obscure than clarify?

Some terms are ambiguous in ways that invite (unintentional) equivocation, e.g. 'autonomy' and 'intrinsic value'. While safely usable in some contexts, I think it's often preferable to replace these vague terms with whatever more precise understanding you have in mind.

(I should add: not all polysemous words are problematic. There are a bazillion different "internalisms" and "subjectivisms" in common use, but equivocation is rare enough, and -- as I note here -- there may not be any better terms available.)

Other terms are simply badly chosen, or have misleading connotations, e.g.:
(1) "Possible worlds" are really possible states of the world, or world-descriptions. Relatedly, talk of truth "at" a world might be better stated as truth "according to" a world-description.

(2) Moore's use of "naturalistic fallacy" to describe normative reductionism is, I think, a doubly intolerable usage. First, it usurps a far more intuitive alternative usage, namely, denoting the fallacious move from "X is natural" to "X is good". Second, Moore's usage isn't even sensible in its own right, since his target isn't really naturalism per se. You could just as well have a supernaturalist version of normative reductionism, as per (certain versions of) Divine Command Theory. (Besides, it's awfully tendentious to call the very statement of your opponent's position a "fallacy".)

Can you think of any further suggestions (for either list)?


  1. But then, if it's a property of our world '@' that it is actual, then it seems that "@ is the actual world" comes out as a necessary truth.

    How does that follow? Being actual is a contingent property of our world. What comes out necessary is that our world is actual at itself. But every world is actual at itself, and necessarily so.

    "Possible worlds" are really possible states of the world, or world-descriptions. Relatedly, talk of truth "at" a world might be better stated as truth "according to" a world-description.

    This is not a matter of jargon. Whether worlds are identical to world descriptions or are (Plantinga style) abstract simples or are (Lewis style) concrete individuals, depends on all sorts of other commitments and arguments. There is an important debate, too, about the distinction between truth-at-a-world and truth-in-a-world, since some argue (Williamson for instance) that not everything that is true at a world is true in it. So, as far as I can see, there is no simple 'just a matter of jargon' way of handling these distinctions.

  2. "World-descriptions" seems as bad, probably worse, than "possible worlds" since it seems to presuppose certain substantive views about the topic. I seem to remember Lewis arguing that there may be inexpressible propositions. Presumably, whether there are or aren't is a substantive question.

    Suppose there are such propositions. Call one "P." If P is inexpressible, then presumably there is no world description according to which P is true, but nevertheless there are possible worlds in which P is true.

    The worry is that descriptions are the sorts of things that you can express in language, and it's a substantive question whether for each possible world there is some possible language in which we can describe the way things are in that world--such a question shouldn't be swept under the rug by our choice of vocabulary.

    (I'm inclined to agree that "possible states of the world" is better than "possible worlds," and it doesn't have the problem mentioned above)

  3. I find the word "rational" pretty dodgy, since people use it to run together all kinds of normativity from practical means-end rationality (you shouldn't accept a set of bets that you know will result in a net loss for you no matter what) to logical consistency (you shouldn't believe contradictions) to maxims of theory choice (you should adopt strong, simple theories over the ugly disjunctive one that would be politically convenient if true) to whatever the author wants to bludgeon you into accepting (you should behave morally even if it doesn't serve your ends, and you should acknowledge that various things the author happens to like are good).

  4. Mike - "How does that [quote from another thread] follow?"

    It only follows if we adopt the static, space-like picture that is misleadingly evoked if we use 'world' talk rather than 'world-state' talk. See my first comment in that thread (responding to Brandon) for further explanation.

    (+ Daniel) But fair point that "world descriptions" might likewise fail as a neutral term here. "Possible world-states" might stick nearer to the pre-theoretic notion of 'ways the world could be'.

    I thing I don't like about "truth at" talk is that it risks taking the metaphor of "modal space" too literally. Possibilities aren't places. There aren't some truths here and others over there. Strictly speaking, there's just the actual truths. Then there's what would have been true if things had turned out differently. Speaking of truth "according to" a possible world-state serves to evoke connotations of fictionality ("according to the fiction..."), which seems a healthy precaution if we are to take the conceptual risk of talking about counterfactuals as in any sense "true". "Truth at" talk seems a much more radical departure from the pretheoretic notion it's meant to be getting at. Worse: it might confuse people into thinking that counterfactual "truths" are a kind of truth. This is like thinking that decoy ducks are a kind of duck.

  5. Wizard - are those really "run together" often in practice, or is it just that we have a single term with broad application? Put another way: isn't everything you said about 'rationality' just as true of, say, 'normativity'? But that's a helpful word; there's a reason why we apply it to all these things, for they have something important in common. Banning these terms would seem to impoverish rather than clarify our discourse.

    We should want to at least be able to verbalize and consider the possibility that someone really ought to do something other than what they are inclined to do. (It seems rather tendentious to call this "bludgeoning", or to assume that it's always merely a matter of imposing one's arbitrary preferences -- what one merely "happens to like"!)

  6. I second "naturalistic fallacy."

    My candidate is "person," which is used in discussions of both personal identity and ethics (as a 'member of the moral community'). The two usages are related enough to be easily conflated.

  7. I think that quite a large number of very common philosophical terms fall into the first list: proposition, belief, desire, intuition, and so forth. They aren't always misleading, since enough careful context can compensate for the problem, but they are apparently simple terms that are not really simple at all. As I've noted before, Scottish common sense philosophers in the eighteenth and nineteenth century appealed to at least eight distinct things to do the work some modern philosophers use the one word 'intuition' to do. And similar things can be said for the other terms.

    I think with the terms in the first list, however, we should distinguish between those that are better replaced by some more precise term and those that are sometimes perfectly serviceable if we are well aware of the danger of equivocation and take adequate precautions. It's not at all clear to me that 'intuition' is of any real use, but 'desire' (for instance) obviously is, if we take the trouble to compensate for the fact that on its own it obscures rather than clarifies, in part because our vocabulary for the things captured by the word 'desire' is not very extensive, and most other words we use would also have the same problem. Equivocation involving it is not at all rare; but the trouble of avoiding the word is almost always greater than the trouble of most confusions and difficulties it might cause.

    I remember Peter Geach somewhere puts 'coherence' (and 'incoherent' etc.) on the first list (not in those terms, of course): his argument being that, as it is often used, it suggests consistency, but without holding you to the rather strict standards you can be held to with the term 'consistency'.

  8. Sean - 'person' is an interesting suggestion. I'm inclined to think that it means, simply, 'agent', though we can ask two very different questions: what it takes to be the same person now as I was before, and what it takes to be a person at all. Could you say a bit more about the different meanings that you have in mind?

    (I guess there's some potential for ambiguity if you think, e.g., that you were once a fetus, although fetuses aren't agents. One could then use 'person' either (i) as a temporary predicate that applies to beings only while they possess rational capacities, or (ii) as a kind-term that applies permanently to any being that is an agent sometime. Is something like this what you had in mind?)

  9. I have to echo Mike Almeida's initial comments about how talk of possible worlds is substantive. There is a healthy debate about whether actualism can really account for modality. Surely you're right that sometimes people use 'truth at a world' talk even though they aren't committed to some version of possibilism, and perhaps in some of those cases they are using misleading jargon (although, I think we usually can easily parse what they're claiming). Nevertheless, there is a real substantive debate about which views can account for modality. Your reply to Mike just shows which side of the debate you fall into.

  10. "consequentialism" is terrifically confusing. This should be a general term for a family of views that emphasize consequences of various sorts, but is usually stupidly and confusingly used as a synonym for utilitarianism (not a very illuminating term itself). The way philosophers use it, "morality" is a profoundly confusing term. It turns out that morality, as most people understand it, is quite contrary to philosophers' moral theories(like "consequentialism"). Actually, most philosophic jargon is misleading, so this may be a tediously unending subject.

  11. Errol - I don't mean to be begging any substantive questions in the metaphysics of modality. Maybe David Lewis is right and counterfactuals really are genuine truths at another place. My point is just that our terminology shouldn't commit us to any such picture (especially since very few philosophers would explicitly endorse it). All else equal, and prior to engaging in any of those specialized debates, most philosophers should prefer to use terminology that is closer to ordinary modal thought. This would better communicate what they have in mind, and help avoid certain common confusions. (Kripke explains an important one.)

  12. Will - I've always heard 'consequentialism' used in the way it ought to be (encompassing, e.g., ethical egoism) -- utilitarianism is merely the paradigm. Though I have heard reports that some people use it to refer only to impartial consequentialism (still broader than utilitarianism, since one might plug in other values besides wellbeing) -- I'd agree that that's bad usage.

    "It turns out that morality, as most people understand it, is quite contrary to philosophers' moral theories"

    This is a puzzling remark. Compare: physics, as most people understand it, is quite contrary to scientists' physical theories. This is not because specialists are changing the subject. It's because ordinary people have false views. The disagreement is substantive, not terminological.

  13. Great discussion. The point about "autonomy" is particularly appreciated, and I have always been a little baffled by Moore's use of "naturalistic fallacy." I suppose the question is how to balance an equivocated nomenclature with an overpopulated nomenclature (e.g., recognizing the different types of "autonomy" but also realizing their root similarities).

  14. Richard, I thought inrtinsic value was pretty straight forward. Something has intrinsic value if it is valuable in and of itself. By contrast, something has instrumental value if it is valuable because it leads to something else which is valuable.

    Oh.. and by the way, Ive got an ongoing series of posts on my Blog: A Singaporean Renaissance on consequentialism. Hope you will comment.

  15. Murali - see my linked post for the explanation how 'intrinsic value' is ambiguous. (Sometimes it contrasts with other things besides instrumental value. E.g. some people use "intrinsic value" to mean absolute value, in contrast to agent-relative value. Consider, e.g. a family heirloom that I have reason to care about non-instrumentally but you don't. Others use 'intrinsic value' even more strictly, as meaning value that things have "in themselves", i.e. in virtue of their intrinsic properties, rather than depending in any way on their relations to other objects. Consider, for example, Abe Lincoln's hat, which some might consider non-instrumentally valuable -- it sure isn't a means to anything else -- but the hat's value is in some sense due to its relation to the great man.)

    For an example of the confusion this can cause, see the recent discussion at 'Go Grue' -- where different commenters were using each of these different senses of the term.

  16. "Though I have heard reports that some people use it [i.e. "consequentialism"] to refer only to impartial consequentialism"

    That's correct, some people take consequentialism to be what they call teleology, plus impartiality. I'm not familiar with the literature, but I assume that this is because consequentialism is normally cast as a view about the relationship between rightness and goodness, and goodness is normally taken to be agent-neutral.

    "Moore's use of "naturalistic fallacy" to describe normative reductionism is, I think, a doubly intolerable usage. First, it usurps a far more intuitive alternative usage, namely, denoting the fallacious move from "X is natural" to "X is good"."

    I'm not really convinced that it "usurps" this more intuitive usage. Isn't there a very intimate link between the two errors?

    "(Besides, it's awfully tendentious to call the very statement of your opponent's position a "fallacy".)"

    I'm not sure that this is right. "The naturalistic fallacy" does not refer to moral naturalism. It refers to a certain problem with moral naturalism. And I'm not sure that that is tendentious, since many objections and problems for numerous views are called "The ___ problem", or "The ____ objection".

  17. "Legal realism" is not just confusing, it actually describes the opposite of the view to which it refers. That is, "legal realism" is really a species of anti-realism (for some kinds of law some of the time).

  18. I'm teaching the problem of evil right now. The term "moral evil" is quite unhelpful. "Agent evil" or something would make more sense. On that note, the term "evil" is quite unhelpful as well, since it's a technical term even though it looks like an ordinary one.

    And, as we all know, "valid" has its issues. It's already an ordinary term that just courts confusion about what validity is a property of. The ordinary term applies to statements or propositions, not arguments.

    "Skepticism" is also a bad one, I think. It's an ordinary term that just means doubtfulness. Yet philosophers tend to you use it to refer to a position about not having any knowledge about something. And still others use it just to mean external world skepticism.

  19. Mention of Moore's naturalistic fallacy has had me picking up my copy of 'Principia Ethica' wondering if I could tell what Moore is up to. It is the matter that Moore mentions in his preface which I find a little contentious. Moore says philosophers should try to discover what question they are asking and I think there is a single important question for ethics and that is the question as to what is at stake. Moore probably does regard more at stake than the use of a word. It has also occurred to me that it is too easy to see the fallacy in Moore's reasoning, which prompts the question as to what is it to expose a fallacy?

  20. Richard, I do think they're run together in practice. There are Dutch books, which are often presented as arguments that rationality (in the sense of logical consistency) requires you to have beliefs that conform to the probability calculus, because rationality (in some pragmatic sense) requires you not to accept sets of bets that jointly result in a net loss no matter what. Then you get people like Patrick Maher arguing that rationality (in what I think is a *different* pragmatic sense) requires you not to accept books from Dutch guys in the first place.

    I gave a talk on probabilistic coherence and partial belief recently, and got a lot of comments of the form, "It's better to have slightly incoherent but non-insane partial beliefs than to have coherent partial beliefs that assign a high credence to the proposition that the Flying Spaghetti Monster is going to fight Yog Sothoth tomorrow, thereby causing the apocolypse. So probabilistic coherence must not be a requirement of rationality." That looks to me like people expecting "rationality" to mean one thing in a case where it actually means several.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)

Note: only a member of this blog may post a comment.