¿Are semantic boundaries good or no beuno?


So I'm writing an article right now that I've basically had finished for almost a year and a half! Ah! Why can't I just finish it?? Well, there's this one (stinking) paragraph that I'm having trouble wording. I don't want to say too much and I don't want to be unclear. So I thought maybe I'll just throw some thoughts down in this venue... and see what others have to say.


When doing lexicological work—that is, research geared towards investigating the semantic "stuff" that's typically ascribed to lexemes—it's important to come up with boundaries that separate certain uses from other uses.

Now, even though this is not reflective of how meaning actually flows through various forms, it's certainly helpful when trying to pen down a lexeme's range of meaning in something like a dictionary.

For in all reality, meaning—which is coded into various forms and collocations—is a very fluid beast. Different senses are merged and stretched with others all the time. It's just that we, with a linguistic competency of whatever language we are speaking, pick up on the nuances effortlessly and constantly get the gist of what's being said—and if we don't, we ask for clarification.

Studying a dead language (or "ancient" if you want me to be PC) we don't have this luxury. All we have is what's been left in script.

I don't even know what I'm trying to say now. I think it all goes back to the argument that boundaries are helpful—superficial though they may be. Otherwise, we're left with the task of detecting the varying degrees of Sense C involved in Sense A—and oh, is there a touch of D and F in there as well?

I ain't got time for that. Much less, the brains!

So I'll content myself with utilizing a structured set of criteria to draw boundaries so as not to let my awesome powers of intuition run unbridled. Which is what, I think, many of the great lexicographers of the past have done. But let me be clear: these guys were studs. They were probably right more than wrong. But this still doesn't mean that they didn't overanalyze some uses of a word and under-appreciate others. And more importantly, it does not exonerate them from the shortcomings of failing to educate their readers (i.e. lexicon-users) on how and why they came to the conclusions that they did. This, I think is the real shortcoming.

How many times in your first years of study did you pick up one of these lexical tomes only to be dumbfounded with which use was active in the text you were translating? All you basically get are a list of glosses after all, and these—in the words of Barr (1973:120)—"can hardly be dignified with the term 'meanings'".

So if we can agree that boundaries are important, I think it's just important to realize that even though boundaries (i.e. semantic categories) may be drawn up, these lines are fuzzy. It's ok to be convinced that some uses you're looking at are a little of this and a little of that. Blurriness can be admitted. It's not the scarlet letter of the categorical community anymore.


The dark ages of Necessary and Sufficient Conditions no longer reign. A new dawn has emerged, and its name is Prototypes and Fuzz.

Well, I think I'm done rambling. Any thoughts?