Séminaire : Johan van der Auwera, Cycles dans “Théories et données linguistiques : La grammaticalisation”, 7 Mars, 14h Inalco, Paris

Nous avons le plaisir d’annoncer la prochaine séance du séminaire “Théories et données linguistiques – thématique spéciale : la grammaticalisation” du SEDYL. Attention, les salles des prochaines séances changent ;  cf. aussi le programme en-dessous de ce message.

Vendredi 7 mars 2025, 14h à 17h (max)

INALCO, 65 rue des Grands Moulins, Amphi 6

(Lien Zoom : https://zoom.us/j/3276584955, Meeting ID: 327 658 4955)

Johan van der Auwera: Cycles

The study of linguistic cycles occupies a prominent place in diachronic linguistics, both in formal and functional theories. But the notion of cycle has also been criticized, in a mild way, e.g. by deploring that ‘cycle’ pushed back the Gabelentzian notion of ‘spiral’  (e.g. Hansen 2018), or in a harsher way, as when Givón (2016: 253) considered the cycle an ‘illusory epiphenomenon’, the real phenomenon being grammaticalisation. This presentation will survey some aspects of the scholarship on cycles.

In the first part I will focus on the most famous of all cycles, the so-called ‘Jespersen Cycle’, and on its most prominent illustration from French. In simple terms, French once expressed clausal negation with ne, then it expanded ne to ne … pas, and in spoken informal French one now often just uses pas. I will show that this kind of phenomenon is by no means unique to French and that there are different types. Illustrations come from all continents.

In the second part I will focus more closely on the question of what it means to say that some process is a cycle.  I will argue that it is important to distinguish between defining a cycle either (α) as a process in which a first stage is similar in a non-trivial way to a last stage or (β) as a process that is similar in a non-trivial way to another process. I will argue that the better concept is that of the α cycle and I will bring in cycles other than the Jespersen Cycle, viz. the so-called ‘Jespersen Argument Cycle’, the ‘Free Choice Cycle’, and the ‘Quantifier Cycle’.

The third part is a plea for caution. One should be aware that a cycle is often (?always) a simplification of a more complicated diachrony. For instance, however widely accepted the Jespersen Cycle for French pasmay be, it does not reflect that the first stage arguably had both non and ne nor that the intermediate stage (or, better, ‘stages’) had more strengtheners than just pas, one of which, viz. point, survives to this day, as does ne on its own.