David Willis proposed that changes in systems of indefinites often lead to items becoming more negative (e.g. French personne ‘person’ > ‘anyone’ > ‘no one’), yet this does not seem to be a completely unidirectional development and reverse cases have been proposed. His paper investigates such questions of directionality in the historical development of indefinites against the background of work in first language acquisition.
Using Jäger’s recently proposed feature system characterizing indefinites as [AFFECTIVE] and/or [NEGATIVE], he argues that, while the emergence of [AFFECTIVE] (negative polarity restrictions) and [NEGATIVE] (development of inherently negative distribution) and the loss of [AFFECTIVE] are to be expected, it is not clear how a restriction to negative contexts could be lost in ordinary cases. Standard examples of such cases in Celtic and Slavonic were also examined and argued to be at best inconclusive.
Read the seminar conclusions.
Main University navigation links
