lexeme

A lexical item corresponding to the set of all words (or of all multi-word expressions) that are semantically related through inflection of a particular shared basic form.

A lexeme is a unit of lexical meaning, which exists regardless of any inflectional endings it may have or the number of words it may contain. Thus, fibrillate, rain cats and dogs, and come in are all lexemes, as are elephant, jog, cholesterol, happiness, put up with, face the music, and hundreds of thousands of other meaningful items in English.

In a typical lexicalist approach (e.g. Koontz-Garboden 2006), the unmarked lexeme is taken as lexically listed, even if its meaning (as it often does) includes templatic entailments, and the derivational morphology is taken to operate on the underived form to yield the derived form. This is the case not only morphologically, but also semantically.

A lexical item corresponding to the set of all words (or of all multi-word expressions) that are semantically related through inflection of a particular shared basic form.

A lexical item corresponding to the set of all words (or of all multi-word expressions) that are semantically related through inflection of a particular shared basic form.

For the second sense, where “word” means “item that should have its own dictionary entry,” lexicographers sometimes use the term “lemma,” but that has other meanings too, so among linguists the term lexeme is now standard, and I’ll use it. For the different forms or shapes that belong to a lexeme we can use the term word-form. ¶ And as a typographical convention, from now on I’ll always put lexeme names in bold italics with a capital letter. So I’ll say there is a lexeme called Pamper [emphasis in original]. It’s the name of a small collection of word-forms with particular spellings – four of them: pamper, pampered, pampering, and pampers. I will always put word-forms (and all the phrases and clauses I mention as examples) in italics.

An individual instance of a continuous character sequence without spaces, used in lexical analysis.

Near-synonym: token