Note to a recent commenter on the issue of Lexicographical Prescriptivism: I don’t think we can limit our usage surveys to “educated people.” That’s a slippery category.
I could, however, add that usage should ideally be “unselfconscious”: as soon as you ask people a usage question they’re liable to fall back into Lexicographical Prescriptivism!
Also, there are different levels of formality. Usage agrees—right now—that “whole nother” is appropriate for casual speech but not for formal discourse or writing. But that could change.
I agree that usage plays a very important role in determining what something means in certain contexts, but I’m not sure I that I agree with you when you say that “words mean whatever most people mean when they use them. If enough people mean something that is technically wrong, then that something is no longer wrong.”
Several questions come to mind:
1. Why is 50%+ usage the magic number? (I assume that’s what you mean by “most people” and “enough people.”) Why not 25% or 75%? What’s your rationale for this?
2. Is it possible, then, to speak of what a word means without lots of qualification (e.g., time, location, etc.)? For example, were the surveys an accurate reflection of usage? How long before the surveys are outdated and unreliable? A year? A month? Less? Does “most people” refer to most people in the world, a country, a state, a city? Can a word mean one thing in one state, but not mean that same thing in another? In two neighboring cities? On two ends of the same street? What about groups spread across cities, states, and even countries? May we speak about the meaning of a word with reference to Caucasian, middle-class, 17-year-old, bisexual males? When we speak of what a word means, which group should we have in mind? Why? And, to ask my original question, do we need to qualify what we have in mind?
3. Does the reverse hold true? Can non-usage invalidate a meaning? In other words, if only 5% of the population uses a word in a certain sense—even if that sense is in all of the dictionaries—can we safely say that the word doesn’t mean that? What if that meaning has never been held by more than 50% of the population at any time in history?
4. Is it right to oppose a misuse of a word? If so, at what point does it become wrong to oppose the misuse? 50.01%? Is the battle officially lost once the majority tips in the direction of the misuse?
Excellent, Phil! My blog doesn’t get many enriching comment sessions. Your questions sharpened my thinking. But I’m sticking to my guns! I don’t think any other basis than usage is sufficient to determine meaning, and I’m willing to live with the ensuing mess.
1. You caught me in a bit of imprecision. Actually, I was very aware that I had done so, because I don’t think anybody can pin the number down. If a full 40% of English users typically use the second sense of peruse, I’m willing to call that a sufficient quorum. But here I tread on territory that is properly not my own. I’m not sure what percentages from usage panels constitutes a quorum for any dictionaries. Perhaps they don’t say because no one can! At what point was “let” last used to mean “prevent”? When did it become a second sense? When was the balance tipped past 50%? No one knows. A dictionary just needs to let us know the empirical facts.
2. Dictionaries already do qualify their time and location. That’s why dictionaries have multiple editions (time) and why we have the AHD and OED (location). Many American dictionaries will also say “chiefly Brit.” with regard to a given usage. There are also regional dictionaries, though they save a lot of work by focusing only on what usages set the given regions apart. I’d say it’s entirely possible to have two people on the same street using two opposite senses of “peruse.” Every nation is composed of a wild mess of overlapping groups and sub-groups. The need to understand each other works as a conservative force keeping us from utterly scattering. But there are times when I can’t understand anything the kids in my evangelistic outreach are saying. When they do not want to communicate to me, they have their own dialect of sorts.
3. I touched on this under number 1. We have to qualify what we mean by “meaning.” “Let” still means “prevent” within the KJV. But no one in common English speech means that. Dictionaries need to tell us this, even if it’s a messy situation while the change process is going on. They do tell us this by frequently saying that a given sense is “obsolete.”
4. It is wrong to oppose the misuse of a word on the grounds of Lexicographical Prescriptivism. Aesthetics, precision—I’m fine with those reasons. Wes Hedrick offered both with regard to “peruse.” But you can’t say, “Peruse means to read carefully!” If people don’t mean that, “peruse” doesn’t mean that.