Because they are incapable of wanting anything.
Earlier I remarked that Geoffrey Pullum, Ph.D., professional linguist, ardent descriptivist, and language blagger extraordinaire surely didn’t really think this but wrote it in Language Log posts for rhetorical effect: that languages want to change and love ambiguity and show no interest in avoiding polysemy or ambiguity. Perhaps he really does, without the purpose of irony, cuteness, rhetorical effect, or simple convenience/time-saving in informal blag posts, anthropomorphose languages to such a great extent. We’ve all heard about real-life and fictional animal researchers becoming attached to their research subjects the way they would a pet or loved one, artistic creators feeling more love and devotion to their creations than to real people, entrepreneurs exhibiting more devotion to their businesses than their families, car or boat lovers thinking of their cars or boats almost as people, etc. I guess some linguists can become so immersed in and attached to the subject of their life’s work that they ascribe to it human qualities that it doesn’t deserve.
Geoffrey Pullum has repeatedly, over and over and over again, written that languages love ambiguous meanings and show no interest in avoiding ambiguity or polysemy. No, of course they don’t, because they are abstract things, basically concepts, that are incapable of thought or emotion. People, on the other hand, very obviously are interested in avoiding ambiguity. At least, some of us are. I doubt very many people, even those dreaded prescriptivists whom Dr. Pullum basically makes into a caricature of intolerant, unthinking, dogmatic unreasonableness, which in itself makes me wary of intellectually associating myself too closely with descriptivists of his type, care too much about the polysemy that is so typical of short, one-syllable words that we’ve inherited from ages past. But I find myself sympathetic to prescriptivists, if that’s what it makes them (me), who object to a rarer word with an established meaning acquiring a new one that it doesn’t need and doesn’t help anyone communicate.
The point of those posts by Dr. Pullum was to point out that polysemy does not mean ambiguity and thereby prove that prescriptivists are wrong in wanting words to maintain constant meanings, or even in believing that they can.
I would say the archetypal ongoing definition shift that divides descriptivists from prescriptivists is the adjective “disinterested”. It means impartial or not having a stake in either side. In contrast, “uninterested” means bored or not engaged. I don’t know if it’s descriptivist, prescriptivist, or neither to note that “impartial” is basically a synonym of “disinterested”, so do we really need the latter at all?, and that it is kind of stupid to have two negation-of-”interested” words mean different things. Then again, the word “interest” means different things, so I have no problem with “disinterested” and “uninterested” meaning different things. The sentences, “I’m not interested in that,” and, “I don’t have an interest in that,” can have distinct meanings from each other—the first would mean “uninterested”, and the second would typically (or at least occasionally) mean “disinterested”. Ah, but “interested” isn’t the same word as “interest”; maybe that influenced the meanings of the sentences more than just the polysemy did. How about, “I have no interest in that,” vs., “I don’t have an interest in that”? Phrased and juxtaposed thusly, more ambiguity certainly seems to creep in. Normally, when we use “interest” to mean stake in a matter, we use “matter” or other contextual clues to clarify our meaning (“I don’t have an interest in that matter”).
The “interest” examples provide yet more support for my point that in contrast to languages, people are quite obviously interested in avoiding ambiguity and avoiding problems caused by polysemy. For example, if we wanted to write a sentence with a structure like one of those in the previous paragraph but were worried about the ambiguity of the word “interest” (meaning fascination/attraction/intrigue vs. stake/bias/potential benefit or loss), we would use a different word. This is because we are interested in avoiding ambiguity and the problems that polysemy causes. The languages we’ve inherited and were able to evolve on our own have plenty of polysemy, occasional ambiguity, and also plenty of ways to avoid both, so we can further influence our language through usage to enable as much precision, clarity, and anti-ambiguity as possible.
One way to achieve this in the “interest(ed)” examples is to choose a different (coincidentally, also polysemic) word that could, despite its polysemy, have only one meaning in the sentence at hand: “I have no stake in the matter,” or, “I don’t have a stake in the matter.” Or, you know, something completely different but still constructed in the same way: the colloquial, “I don’t have a horse in that race.” Another way to achieve clarity in a short sentence like that is to use a word with a specific meaning for this purpose, “disinterested”. We wouldn’t be compelled by Language to choose the word it wants us to use and has devised for this specific purpose; we ourselves would choose a different word, possibly the polysemic “stake”, which people themselves have come to use for certain meanings because of the benefit of doing so, not caring whether it were polysemic or what it meant in other, irrelevant contexts.
On this point I both agree and disagree with Dr. Pullum: polysemy doesn’t (necessarily) cause ambiguity, certainly not enough to paralyze our use of language, but at the same time, we are not slaves to a disconnected, extra-societal, self-evolving Language that dictates what words mean and how they must be used. We have words with multiple meanings or one meaning because of how our ancestors used them, and we can continue to influence our usage of our language by conscious choice to achieve or at least maintain as much precision and clarity as is reasonable. What is reasonable can be debated extensively but also largely agreed upon, by either the linguistics community specifically or by society at large, and can include the maintenance of a word’s definition when it seems beneficial to do so and, especially, when it is obviously not necessary and not helpful for its meaning to change.
Language shows no interest in having “disinterested” and “uninterested” mean the same thing or different things or doing anything else. People, on the other hand—at least some people—seem to have an interest in keeping both “disinterested” and “uninterested” at one meaning each. Descriptivists are quick to remind us that “disinterested” used to mean what “uninterested” means today. This either occurred because the abstract concept of Language wanted and willed its meaning to change, or it happened because people came up with two different, non-overlapping meanings for these two words, which became standard and popular because it was helpful. Maybe it was the latter, and maybe people can choose, by way of usage, to keep it that way. Maybe it’s worth considering that people who use “disinterested” incorrectly are only doing so out of ignorance of its definition or a simple lapse of prefix distinction, not because Language wants oh-so-badly for it be polysemic, and that we could actually influence usage in a good way by simply reminding people of its definition while sparing ourselves half of the effort we expend arguing over whether this polysemic shift is a good, bad, or completely neutral thing.
Language Log commenter Jeff Rembetikoff said,
I don’t think the more reasonable breeds of prescriptivists object to each and every word with multiple meanings. Instead, they object to a rarer word losing a distinction between it and a more common word due to perceived sloppiness. Their frequent objections to certain uses of, for example, collide and comprise seemed reasonable enough to have influenced my own usage.
I agree. I don’t want words to have a certain, static meaning per se; rather, I want their meanings and distinctions to be clear and precise, and I want people separated by time and space to mean the same thing when they write, say, hear, and read the same words. Ardent descriptivists, in contrast, seem to consider any change a good and desirable thing per se because it’s what language does and is therefore by definition good. They spend hours and hours and millions of words defending their complete lack of a position, which seems to border on pointless, to me, except in such cases as people’s communication or understanding is impaired by undue prescriptivism or other strictness. I don’t really care about the polysemic words that we have, despite their meanings being broad and ambiguous, because the phrases and sentences they are used in are clear, unambiguous, and easily distinguishable from their other uses.
Going back to Dr. Pullum’s examples of completely unambiguous, unconfusing polysemy, I was particularly interested in his example “see”. Here are the definitions he lists for “see”, which exclude the original vision-related one:
1. understanding: I see what you’re saying.
2. judging: I see honesty as the fundamental prerequisite.
3. experiencing: Our business saw some hard times last year.
4. finding out: I’ll see whether he’s available.
5. dating: I heard that she’s seeing someone.
6. consulting: You need to see a doctor.
7. visiting: I’d be go and see my aunt for a while.
8. ensuring: I’ll see that this is done immediately.
9. escorting: Let me see you to your car.
10. sending away: I’ll come to the airport and see you off.
The thing that struck me about all of these uses is that in my world of scientific writing and editing, and possibly in the primary literature of most academic fields, every single one of these uses of the word “see” would be inappropriate. We would never use “see” to convey any one of these meanings (maaaayyyybe “see a physician”). Using “see” to mean any of those things in a biological or medical paper would be imprecise and inexact because we have better words with more specific, specialized meanings that would more precisely convey the meaning of the sentence. In everyday speech, such pomposity wouldn’t be necessary, but in higher-level writing, it’s often beneficial and even necessary, my objections to bombasticness in general notwithstanding. In the case of “see” as used with any of the meanings listed above, I would consider a replacement necessary, for clarity and precision. The reason “see” would be inappropriate in primary literature is not because Language wants and compels us to write differently for different situations, nor would it be (only) because scientists and scientific editors are bad writers who conflate “more syllables” with “better writing”. We would use more specific words because we want to and we see the benefits of doing so.
I see the benefit of keeping many words’ meanings distinct and constant, the best example of which is “disinterested” vs. “uninterested”. Many more I don’t care about. I think it’s reasonable to agree with me on any or all of them (if I ever compile a list, I’ll let you know), and it doesn’t make me ignorant, intolerant, pompous, unrealistic, or prescriptivist to advocate lexicographical constancy for some words when there are clear reasons to do so and unclear benefits of change.
My desire is not to avoid polysemy or change per se, but rather to promote things that make sense and discourage things that don’t. All of those different uses of “set”, “draft/draught”, “charge”, and “put down” make sense and don’t cause ambiguity, so there is no problem with them. Advocating the use of “disinterested” to mean “uninterested” probably doesn’t cause much lack of clarity, either, but it’s stupid, so I oppose it. It’s stupid because it already means something else and there is already a word that means “uninterested”. It’s not a short, simple, common word with lots of everyday uses, especially uses that are distinguished in prepositional and adverbial clauses. It’s a long, uncommon adjective whose meaning it makes more sense to preserve, so I advocate its preservation and constancy. This position is simple, beneficial, common-sensical, and very nearly unobjectionable, as far as I can tell.