In thinking about your question, we might recall the conversation between Humpty Dumpty and Alice in Through the Looking Glass. At one point, Humpty Dumpty exclaims "There's glory for you." Alice protests that she doesn't know what he means.

Humpty Dumpty smiled contemptuously. "Of course you don't -- till I tell you. I meant 'there's a nice knock-down argument for you!'" "But 'glory' doesn't mean 'a nice knock-down argument'," Alice objected. "When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean -- neither more nor less." "The question is," said Alice, "whether you can make words mean so many different things." "The question is," said Humpty Dumpty, "which is to be master -- that's all."

Philosophers often distinguish between the semantic meaning of a word -- its assigned meaning in language -- and the speaker meaning of a word --what the speaker intends to mean by the word. Most of the time, if we want to communicate successfully, we need the speaker meaning of what we say to match up with the semantic meaning. If we behave like Humpty Dumpty, and our speaker meaning diverges from the semantic meaning, then our listeners will be like Alice, puzzled and uncomprehending of what we are trying to communicate. But over time, if speakers consistently use a word in a way that differs from its semantic meaning, then it does seem as if the word can thereby acquire a new semantic meaning. Language--meaning--does change with use.

This website lists many examples of semantic change, at least one of which I found disturbing:

"Philosophy is originally a science concerned with the use of reasoning and argument in the pursuit of truth and greater understanding of reality and the metaphysical. Now it has come to mean little more than ‘policy’ in a sentence like The company's philosophy is to be aggressively competitive."

Somehow, the semantic changes wrought by persistent Humpty Dumptyism seem especially troubling when it concerns one's own livelihood! (And for what its worth, those of us on AskPhilosophers are not simply policy wonks -- we here are still concerned with reasoning and argument in the pursuit of truth and greater understanding of reality.)

Nice try. But the answer is yes, it is still ambiguous. The river's bank is presumably much bigger than the financial bank, so "Judy went to the bank" could mean she is fishing on the river's bank outside the financial bank or she could be cashing a check inside the bank on the bank.

Now, if you had said that the only river in town is located entirely within the financial institution and the only place to be in the bank is on the bank ...

Suppose I ask, pointing out into empty space: What color is that apple? I take it that the question cannot be answered, that, to borrow a phrase from Sir Peter Strawson, "the question just doesn't arise". This is because the word "that" is not, in this utterance, used to refer to anything, so there isn't anything of which it has been asked what its color is. It is like asking: What color is OobaDooba?

The ruler case is the same: If there is no ruler, than you can't ask how long "it" is, because there is no "it" about which to ask. I take it that this means the answer, in your terms, is "N/A". I can see why you might think it could also be 0. After all, couldn't there be a ruler with no length? In principle (though not in fact), I suppose there could be: But it would have to be a ruler nonetheless, which might mean that it had width and breadth, but no length. (Think of a plane stood on end.)

These are all good philosophical questions (as well as practical ones for translators!). I strongly recommend that you read a wonderful book on the subject by Douglas Hofstadter: Le Ton Beau de Marot: In Praise of the Music of Language

Thanks for your nice question. Even if a language such as a sign language has a sign that is made just in the way that I move my hand when I point at something, that does not mean that when I am pointing at something I am using a word. Similarly, even if a baby happens to utter a series of sounds that sound exactly like the word 'hyperbole', this does not mean that the infant is referring to anyone's overstatement. Accordingly, when I point, particularly because I don't speak any sign language, I'm not using a word. The most that can be said is that I am making a hand movement that is identical to a word in some other language. You can, I think, see this even more dramatically if you think about gaze: we often use gaze, particularly overt gaze, as a form of ostension. However, even if there were a language that used overt gazing as a word (meaning perhaps, "look at that"), that doesn't mean that I am using a word in overtly gazing at something.

Mitch Green

Words have meanings, but they also have histories. The term "Paddy Wagon" is defined in the Merriam-Webster Dictionary as "an enclosed motortruck used by police to carry prisoners." That definition is adequate, if all you want to know is what people are talking about when they use that term. But if this definition is all you ever know about the term, you will be baffled by the fact that people of Irish descent usually take offense when they hear that term used. To understand why they have that reaction, you need to know where that expression came from. You need to know that "paddy" is a derogatory way of referring to an Irish person or a person of Irish descent, and you need to know that there used to be (and maybe still is) a stereotyped belief that people of Irish descent are more likely than others to drink heavily and to become rowdy and disorderly -- so disorderly that police must come and pick them up by the "wagon"-load. Originally, bigoted people who used the expression "paddy wagon" to refer to a police truck were expressing their contempt for people of Irish descent, without having to explicitly assert it. An Irish or Irish-American person who heard someone using the term "paddy wagon" would have been absolutely right to be offended by it -- even though the term purported to be nothing more than a way of referring to a particular kind of vehicle, it was really conveying an insult.

Now many people who today use that expression have no idea about its history. They do not mean to be expressing contempt for people of Irish ancestry -- how could they mean this, if they don't even know that the term has such connotations? Nonetheless, if someone -- say, me -- uses the term, the term's sordid history is going to be evoked for people who do know about it -- especially for people of Irish descent. And since that history is a painful one -- like many of the people immigrating to the United States over the years, people coming from Ireland suffered terribly discrimination and ill-treatment -- simply hearing the word becomes itself a painful experience. They may therefore mount a kind of campaign to take the word out of common usage, explaining to people who don't know why they find the word hurtful. Once I learn all this, it seems to me that I ought to simply stop using it.

Words or phrases like "paddy wagon" have become for many of us dead metaphors; that is, they do contain some descriptive or connotative material, but we don't pay any attention to it. If we did attend to this, we might realize that these terms and phrases allude to negative stereotypes. To "jew someone down" -- to cut a particularly shrewd deal -- alludes to the stereotype of Jewish people as money-grubbers, operating at the margins of honesty. (It's not quite to "gyp" them -- to actually cheat them -- as Gypsies do, according to bigots.) I don't know the etymology of "Dutch treat", but I would not be surprised to learn that it invokes some negative stereotype of people of Dutch descent.

But there are other words where no amount of reflection on their descriptive content can reveal why they are a source of offense to some people. Consider the derogatory word "kike". The etymology is uncertain (see Wikipedia for some speculations: http://en.wikipedia.org/wiki/Kike ), but it doesn't seem to attribute any negative property or allude to any stereotypical picture of Jewish people. In this case, what is offensive to people of Jewish descent is that the word was used specifically to convey contempt for them. The term "Jew", a term that Jewish people use to refer to themselves, was rejected in favor of a term invented by others so as to signal hatred or contempt. The story is the same for the term "nigger". That word exists in the language in order for bigots to express their bigotry in an efficient way.

What about terms like "handicapped" or "retarded"? These represent yet a third kind of case. Here the illustrative parallel is with the terms "Negro" and "black". Even if "Negro" did not function, as "nigger" did, to specifically express contempt for people of African ancestry, still the term was alien -- that is, it is not a term that people of African ancestry chose to designate themselves. Also, the term was associated with a period of history in which people of African ancestry were subject to brutal subordination. The process of rejecting the alien term in favor of a freely chosen self-designation was part of the process of rejecting the whole system of segregation and the ideology of racism with which the term "Negro" was historically associated.

If people with cognitive disabilities tell me that they find the term "retarded" alien, or that it evokes for them a painful history of discrimination and ill treatment, I ought to listen. Ditto for "insane," and "lame." After all, what, in the end, does it cost me? A little self-consciousness until I form a new linguistic habit. Big deal.

As you say, even those of us not trained in reading lips can often make out what is being said. So all that is censored is the actual sound of the word(s) being spoken. Don't ask me why people find this comforting or morally improving. In my view, most allegedly bad language is simply in bad taste. Etiquette is not the same as morality, no matter how vigilantly it is someetimes enforced. Of course, not all bad language is simply bad taste--hate speech actually is immoral, I think. And I also would allow that the border between hate speech and other "bad" language (e.g. certain obscenities) can be somewhat blurry. So *+#!~+%$* that!

Thanks for your perceptive comment/question. Derogatory words do identify groups. So both 'Wop' and 'Italian' apply to all and only Italians. However, the derogatory words convey more than this. You describe this further things by saying they kind of state that the persons are bad, etc. Philosophers of language have spent some time trying to get clear on this notion of "kind of stating", and so I'll try to give you the flavor of that effort here.

In addition to what we literally say with our words, we now know that we can communicate a great deal more. Sometimes what I leave unsaid carries a lot of meaning: I ask you how you like the painting I just completed and you reply, "Well, it's very colorful." Here you don't actually say or state that the painting is not so great, but your meaning is clear. Philosophers often call this type of meaning, "implicature." Some implicatures, such as this one, depend heavily on context. In other cases, however, our implicatures depend on the conventional meaning of our words. So for instance, the word 'but' suggests a contrast or tension between the two conjunctions it conjoins. My describing a person with the words, "She was poor but honest." would be taken as conveying the idea that there's some contrast or conflict between poverty and honesty. You can see that by noticing that many people might take issue with my speaking that way. For a more obvious case, consider, "Bob was from Newfoundland, but intelligent." Here I didn't *say* or *state* that being a Newfie tends to make you dumb, but I did suggest as much, and this is due to the meaning of the word "but."

This approach sheds light on epithets such as those you mention. The word 'wop' carries as part of its meaning a derogatory attitude toward Italians. In that respect it differs from 'Italian', even though the words apply to the same class of individuals. So too for many other derogatory words.

Is this phenomenon, as you suggest, strange? Note that you said the word 'boche' is applied to only one person, but it's a general term, just like 'German'. So both words have what's called an 'extension' (the set of individuals to which they apply). What I'd suggest is striking, and in need of explanation, is how a word, by virtue of its conventional meaning, manages to express an attitude. But I suspect to understand this, we need to stand back and try to figure out what it is to express anything, not just attitudes. I've tried to do this in a book called '_Self-Expression_'. (http://ukcatalogue.oup.com/product/9780199283781.do)

Mitch Green

Thanks for your message. The philosophy of language is not exclusively interested in how humans actually communicate; it is also interested in the various ways in which we could communicate, where 'we' is not limited to members of our species. Nonetheless, the field is not entirely divorced from empirical considerations. For instance, theories concerned with what it is to *mean* something are sensitive to the cognitive requirements that have to be met for one to do so, and some such theories are at risk because they make empirically implausible predictions about what, for instance, a child would have to know in order to mean anything.

By contrast, some areas of the philosophy of language, such as those that share a border with metaphysics (for instance, theories of propositions) are relatively un-empirical. Most theorizing in such areas can be done "in the armchair." The result is that some areas of the philosophy of language are relatively empirical, while others are not.