Also called: function words, stopword stylometry
Function word analysis measures how often an author uses grammatical words such as the, and, of and but, and it is the backbone of stylometric authorship attribution because those choices are made unconsciously and stay stable across topics.
Content words tell you the topic; function words tell you the author.
The counter-intuitive part is that the most informative words are the ones carrying the least meaning. Change the subject of an essay and its nouns change completely while its function-word distribution barely moves. That stability is exactly what an authorship signal needs.
The method's limit is its dependence on the list. Function words are language-specific and not translatable one to one, so a multilingual system needs a curated list per language, built from that language's own frequency data rather than from an English list run through a translator.
| Property | Content words | Function words |
|---|---|---|
| Varies with topic | Strongly | Barely |
| Chosen consciously | Yes | Rarely |
| Easy to imitate deliberately | Yes | Difficult |
| Frequency in text | Low per word type | High, so estimates are stable |
| Transfers across languages | Partly, via translation | No — needs a per-language list |
Cite this page
MyWritingTwin. “Function word analysis.” Last updated 2026-09-15. https://www.mywritingtwin.com/glossary/function-word-analysis