- A subsense of a subsense.
- 1986, International Federation of Translators, Babel, page 44:
- To simplify the discussion, I will assume that the lexeme has no subsenses, subsubsenses, etc.—although this is seldom the case in actuality.
- 1998, Department of Asian and Middle Eastern Studies, University of Pennsylvania, Reviews, volume 8, →ISBN, page 82:
- The Oxford English Dictionary gives scores of different senses, subsenses, and subsubsenses for "eat" whereas the Hanyu Da Cidian offers about a score of different senses for chi; most of the extended, figurative, and slang usages of chi ("eat") are very recent – within the last century or two.
- 2005, Sampson, Geoffrey; McCarthy, Diana, editors, Corpus Linguistics: Readings in a Widening Discipline, A & C Black, →ISBN, pages 365–367:
- The paired sense data can be classified as one of four levels of similarity: the Roman-numeralled homograph level (band-I (group) vs band-II (ring)), the major sense level (band-I.1 (music group) vs band-I.2 (other group)), the subsense level (which we arbitrarily use to refer to the distance between a general sense number such as I.1 and its specialization (I.1.2)), and finally the subsubsense level (such as between I.1.1 and I.1.2). […] In contrast, at the finer subsubsense level only 52 per cent of the given pairs were translated differently. This suggests that homograph-level distinctions are broadly salient and tend to be treated consistently as separate words across languages, while subsubsense distinctions appear to be less salient in that separate lexicalizations for these similar concepts have not evolved in the majority of the languages studied. […] However, for all but the finest subsubsense level, these two different measuring strategies tend to yield results that are closely correlated. […] Yet these induced trees precisely mirror the sense hierarchy given by the HECTOR lexicographers, at not only the homograph level but down to the subsubsense level as well.