1

Is there a word for this?

I'll use an example to show what I mean:

Let's say you don't know what sepsis means, which is bacterial infection of blood. So, you start thinking. You break the word up into what is most likely its components. In this case, it's likely you'd recognize its stem, seps, and its nominal suffix, is. Then, looking at seps, you remember the Seps, a snake whose venom caused putrefaction in a Greek bestiary. Then, you remember your house's septic tank, and the theme of putrefaction continues. With some more medical context, adding a non-linguistic logical element to your analysis (which in other examples might not be needed), you infer that sepsis is a bacterial infection of the blood. Other parts of the analysis might include looking for clues to find out what language the word stems from, and what languages it might have been transferred to, which further informs the analysis.

So, is there a word for this kind of analysis? Morpho-semantic analysis perhaps?

Note: I asked this question on Lingusitic.SE as well.

4
  • 2
    I suppose this approach might be the best you could do if you didn't have access to dictionaries, the Internet, and other authoritative resources. But that situation wouldn't often arise for many people today, and "guessing" at the meaning of a word like this would give you the wrong answer so often I can't really see the point. Words actually mean whatever the majority today think they mean, which is often not the same as what all this background detail might suggest (there's no "intrinsic" meaning beyond that established by custom and practice). – FumbleFingers May 19 at 17:17
  • 1
    On the other hand, it's often very helpful to take note of "associated" meanings and etymology, once you have actually looked up the meaning of some previously unknown term. Knowing what septic means won't help you if you're puzzling over the meaning of, say, septuagenarian (and it's worth noting there are very few English words related to septic). But if you know septuagenarian, you might well be able to guess the meaning of septillion - or at least, be able to remember the meaning more easil;y for when you meet it again in the future. – FumbleFingers May 19 at 17:27
  • Okay, Mr. Portokalos. How about the word kimono? – Jim May 19 at 18:32
  • @Jim I haven't seen My Big Fat Greek Wedding, so I don't get the reference. I know pretty much no Japanese either, and given how kimono is loanword from Japanese, I'm would not be able to estimate its meaning based on its morphology. However, I probably would have been able to see that it Japanese, because I think I've read that Japanese words usually have alternating consonants and vowels, and the word kimono intuitively seems Japanese to me (though that might be an effect of me knowing that it is). So, an analysis would not give any semantical info, but maybe some etymological info. – A. Kvåle May 19 at 19:53
2

In fact the term "Morphosemantic analysis" has been coined, and it corresponds to your feeling of what it should mean, although this new branch of knowledge is certain to be greatly limited in the importance of the results to be expected.

From Morphosemantics and their limits: three Inuit examples - Sedyl (PDF)

Morphosemantics may be defined as the semantic analysis of words through their constituent morphemes (Dorais 1984a: 3; 2010: 137). In polysynthetic languages, where lexemes generally result from the aggregation of several morphemes, morphosemantics can offer a particularly interesting and, hopefully, useful tool for getting access to the underlying meaning of words, often yielding significations that go beyond the surface meaning of the lexeme.

Here is a definition of the term "morphosemantics" (ref.).

Morphosemantics is generally a knowledge in linguistics, pertaining to morphological analysis combined with a semantic interpretation of words.

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.