While “entrenchment” and “conventionalisation” are often used interchangeably in historical linguistics, I would like to suggest that the two are separate and indeed only indirectly related. I take entrenchment as a process pertaining to an individual’s mental representation of linguistic (or other) categories, and conventionalisation as an inherently social process establishing or changing structures in the speech community, where these can be in the language itself (convention of language) or in discourse (convention of language use, cf. Morgan 1978). Consequently, any process of entrenchment is part of language acquisition (in the broad sense of changes in an individual’s representation of language over their lifespan), whereas only conventionalisation is in the remit of diachronic change.
The history of negation in French and some other Romance languages offers a striking example for the interplay of conventionalisation and entrenchment. In Old French pas ‘step’ had acquired by metonymy the grammatical function of marking optional emphasis in negation, in contexts of strong counter-expectation on the part of the hearer (Detges 2001, Detges / Waltereit 2002):
(1) Climborins, ki pas ne fut produme.. ‘Climborin, who was not at all a gentleman…’
This form contrasted with the simple negation without pas:
(2) Mais ço ne set ‘He does not know this’
In Modern French, pas has become obligatory part of sentence negation (ne…pas) and is not associated with any particular counter-expectation at all.
Recent work by Hansen / Visconti (forthcoming) suggests that this change was very much a gradual one, moving from explicit denial of a previous utterance in discourse via more subtle or implicit forms of counter-expectation to complete obligatorification. As a side-effect, pas became much more frequent in discourse and hence arguably more entrenched in the process.
Reviewing the evidence, I will take up the earlier (Detges / Waltereit 2002) claim that the change was driven by speakers taking rhetorical advantage of the greater noteworthiness that emphatic pas afforded at each stage of the process and preferring it over negation without pas whenever possible, thereby extending the realm of pas further and further. The original figurative use of pas was of little importance at the later stages.
It will be argued though that pas could travel this continuum with its use always being in perfect compliance with the conventions of the language at the respective stage. In other words, I will defend the view that gradual changes such as the one under consideration need not imply an increase in conventionality.
More broadly, adopting a Construction Grammar view on a language’s form-function inventory, it can be claimed that innovations in general need not go against existing conventions, and that in fact only a presumably small proportion of them actually do, paradoxical as this may seem at first sight.
Detges, Ulrich (2001): Grammatikalisierung. Eine kognitiv-pragmatische Theorie. Tübingen.
Detges, Ulrich / Waltereit, Richard (2002) : Grammaticalization vs. Reanalysis : A Semantic-Pragmatic Account of Functional Change in Grammar. Zeitschrift für Sprachwissenschaft 21, 151-195.
Hansen, Maj-Britt Mosegaard/ Jacqueline Visconti. Forthcoming. “On the diachrony of reinforced negation in French and Italian”. In: Rossari, Corinne et al., eds., Grammaticalization and pragmatics: facts, approaches, theoretical issues. Oxford: Elsevier.
Morgan, Jerry L. (1978). Two types of convention in indirect speech acts. In Peter Cole, ed., Pragmatics [Syntax and Semantics 9], New York: Academic Press, 261-280.