Is there any synchronic difference between homonymy and polysemy?

As a literate English speaker, I can usually tell when words that are pronounced identically have different etymologies thanks to our wretched but etymologically informative spelling system. I know that "here" and "hear" come from different roots, for example, just as I know that "run" as in "I want to run away" and "run" as in "Dad runs the company now" come from the same root.

But, as most of us know, children acquiring language don't have access to etymological information. So, spelling aside, is there any synchronic difference between homonymy and polysemy?

One possible answer is that morphologically different words are synchronic homonyms rather than polysemous words. Consider "lie" (repose horizontally) and "lie" (use language to deceive). A naive person like me could argue that the two "lies" are synchronic homonyms and not one polysemous word because they inflect differently in past tense (lied, lay).

But couldn't critic counter that "lie" meaning "present tense of lie.deceive" and "lie" meaning "present tense of lie.repose" are still examples of polysemy?

What am I missing?


In short: no, you need a diachronic definition.

If mouse "device" had the plural mouses, and mouse "animal" had mice, would you say it was still polysemic? I think the answer has to be yes: the paradigm/inflexion is in the end not crucial. Mouse and mouse are still "the same word". So we cannot use that as a criterion to distinguish between true homonymy and polysemy.

Another potential criterion could be, "is the meaning of one sense of the word close to that of the other sense?". But you can see how this is impracticable: polysemy can lead to wildly different senses, and homonymy can lead to similar senses, such as through folk etymology, where words converge because they are perceived as etymologically related.

Then what should we use? I think the only criterion that can be rigorously applied to distinguish between the two kinds of same-sounding/written words is this:

In the history of this word, were the two senses ever marked as written or pronounced differently?

If yes, it is a true homonym; is not, it is merely polysemy. Incidentally, one could argue that every word is polysemic, since all words can be used metaphorically.

So the answer is, no, only a diachronic criterion works to distinguish between the two categories that our intuition wants to put different labels on.


I disagree with the accepted answer.

If there is a difference in inflection between mouse "device" and mouse "animal", the natural way to model this is with two different lexemes. Anything else seems like a poor model to me.

Historical categories don't remain the same. Just because the etymology of "mouse" is still transparent, it doesn't mean it's still both the same lexeme.

Of course, concepts such as "polysemy" and "homonymy" are made up by linguists; I don't expect them to have psychological reality. But that doesn't mean we can't use them to model semantics, just as we use terms such as antonyms, meronyms, etc.

Be that as it may, there are certainly several competing defintions of homonymy and polysemy, so it's kind of wrong to argue which one is "right".


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.