Explanation
Word does not identify an automatically identical unit at every analytical level. Orthographic words depend on writing conventions, phonological words on sound organisation and grammatical words on morphological or syntactic criteria. Their boundaries can diverge. Clitics make this particularly visible when an element has grammatical functions of its own but is prosodically attached to a neighbour. A lexeme also differs from a particular word form. Anyone counting or comparing words must therefore state the unit used rather than treating spaces, meanings and grammatical units as interchangeable.
Movement of thought
1. Specify the analytical level. 2. State boundary criteria. 3. Examine actual forms. 4. Compare alternative segmentations. 5. Document the counting decision.
Scientific use case
Corpus work can report tokenisation separately from linguistic segmentation.
Example
English “can't” often appears as one orthographic unit but contains an analysable combination of can and negation.
Counterexample
One concept can be expressed by several words, and one word can have several meanings.
Distinction
Distinguish word from concept, lexeme, token and a universally valid space-delimited unit.
Effect on thought and perception
The concept makes criteria for linguistic units themselves objects of investigation.
Resonance
A space on paper is not every boundary in language.
Provenance
Parallel corpus 98f0950771ae4662f291d2f15f5c51b6797f3c20ba458a68615cdb177002ea18; case-bounded source review accepted; human specialist review not claimed.
Cite this entry
Reich, Michael (Hrsg./ed.): „Word“. Atlas der Sprache, AT-NAV-G05-045. ATLAS 259, data model 1.6.0. https://atlasoflanguage.com/concept/AT-NAV-G05-045The address remains stable; editorial content may evolve. Editorial reading state: 2026-10-15. Include your access date for precise references.