Are “go” and “went” part of the same lexeme, i.e. the same set of inflected forms?

Consider this brief Glottopedia entry. The entry defines its subject matter as follows:

“A lexeme is usually defined as a set of inflected word-forms that differ only in their inflectional properties.”

The entry also provides these examples:

“The English word-forms boy and boys make up the lexeme BOY. The Latin word-forms habeo 'I have', habes 'you have', habet ' s/he has', habemus 'we have', habebam 'I had', habebunt 'they will have', and so on make up the lexeme HABERE.”

But can “went” really be described as part of a set of inflected word forms in the lexeme that also includes “go,” “goes,” and “going”? If so, how is “went” inflected?

Or am I confusing “inflections” with “inflectional properties”? (I don’t know whether there is a distinction between these terms or not). Could “went” have “inflectional properties” that make it part of the lexeme “GO” even if "went" can’t be described morphologically as being inflected?


Yes, they are. In the sentence John goes to school, goes can be replaced by went and the two sentences have the same meaning (semantic content) except for tense.

The base form is usually called lemma in formal and applied linguistics (in this case go) and it stands for the meaning of the inflected word.

| improve this answer | |
  • If semantic content is the criterion, would that mean that if you had two exact synonyms, they would be the same lemma? – dainichi Sep 27 '13 at 10:21
  • It depends. I said "lemma" but maybe I should have said "sememe". If you think that "exact synonyms" exist, then they would have the same sememe. As for lemmas and their inflected forms, they build up paradigms so the main criterion is combinatorial (not purely semantic). The form "went" belongs to the collection of inflected forms of the verb "go" because there is no form like "*goed" ("went" occupies its place). A better example is, I think, the Catalan verb "anar" (to go) which has two inflected form for the 1st person plural ("anem" & "vam"), that exclud each other in different contexts. – Atamiri Sep 27 '13 at 14:19
  • Here (Pacific Northwest), "went" seems to cannibalize various other forms, instead of just being relegated to *goed. "have went", "had went", etc. strike when you least expect them! – Mark D Oct 10 '13 at 22:46

Yes. Several frameworks consider words such as go/went or I/me to be two representations of a single lexical item. Suppletion is the term for when the inflected forms are etymologically unrelated, while the term allolex can be used for the individual forms.

(I think allolex originated in the Natural Semantic Metalanguage framework, originally only for semantic primes, but I can't see any reason why it can't be applied to any other words.)

| improve this answer | |

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.