Formof

Formof is a term used in linguistic annotation to designate the surface form of a lexical item as it appears in text, distinguishing it from the canonical lemma or base form. In this usage, the formof captures inflected, derived, or orthographic variants that a word may take in different contexts, such as tense, number, mood, or clitics.

It is often contrasted with the lemma (the canonical dictionary form). Some annotation schemes adopt formof to preserve exact orthography during processing while lemmas are used for semantic analysis.

Examples help illustrate the concept. For the token "cats" in a corpus, the formof would be "cats" while the lemma is "cat," with a morphological feature indicating number: plural. For the token "ran," the formof would be "ran" and the lemma would be "run," with a feature indicating past tense. In each case, formof records the precise surface realization independent of the underlying lemma.

Applications of formof labeling include corpus annotation, natural language processing pipelines, and the documentation of language variation across dialects or genres. By separating surface form from lemma, researchers can analyze inflectional patterns, orthographic variation, and historical spellings without conflating them with lexical meaning.

See also: lemma, inflection, morphology, lemmatization, orthography.