Cyclical change (review)

Language change occasionally shows reoccurring developments or cycles: an item is gradually reduced and disappears, being replaced by a new item that undergoes the same fate. A prototypical and well-studied case of this type of change is the development of negative markers known as Jespersen's cycle (Jespersen 1917), but cyclical change is found in many other domains of syntax as well, as the volume Cyclical change demonstrates. It contains twelve papers from a workshop on cyclical change held at Arizona State University in April 2008.

The introduction by Elly van Gelderen presents a synthesis of all of the papers in relation to central theoretical questions on cyclical change, including typical steps in a cycle, differences in speed, the role of the structural position of the item to be reanalyzed, and the relation of cyclical change to her theory of feature economy (van Gelderen 2004).

Jack Hoeksema's 'Jespersen recycled' forms the first of four papers on Jespersen's cycle in Part 1, 'Negatives'. Drawing on English, German, and Dutch data, Hoeksema investigates negative polarity items, notably minimizers, as a source for new negators. He suggests that these turn into negative markers only if they mostly cooccur with the negative particle. Johan van der Auwera's 'The Jespersen cycles' is a meta-paper on the negative cycle, comparing different theoretical accounts and data from the literature. Van der Auwera concludes that instead of one uniform development, there are several Jespersen cycles: they basically differ as to which element loses its emphatic meaning at which stage. In 'The negative cycle in Early and Modern Russian', Olena Tsurska investigates feature changes underlying the development of Russian from a nonstrict to a strict negative concord language, that is, from only postverbal negative indefinites always cooccurring with the negative particle ne to obligatory cooccurrence of all negative indefinites with ne. Theresa Biberauer discusses developments of negation beyond Jespersen's cycle in 'Jespersen off course? The case of contemporary Afrikaans negation'. While Afrikaans employs two negative particles, no development toward replacement of nie₁ by clause-final nie₂ can be observed due to the origin of nie₂ as a tag-negator. Independent developments are underway in Colloquial Afrikaans in clauses containing negative indefinites.

Part 2, 'Pronouns, agreement and topic markers', begins with Diana Vedovato's 'Weak pronouns in Italian: Instances of a broken cycle'. She argues that the remains of weak subject pronouns in Italian, viz. 3rd person egli, esso, and dative loro, are due to prescriptive regulation that breaks the cycle of pronouns developing from strong to weak to clitic to agreement markers. Kyongjoon Kwon investigates 'The subject cycle of pronominal auxiliaries in Old North Russian', during which the former 'be'-auxiliary was reanalyzed as a strong pronoun. He contends that this kind of development forms an important step to be added in van Gelderen's (2004) subject cycle. In 'Two instances of a broken cycle: Sentential particles in Old Italian', Cecilia Poletto explains that, as a consequence of the loss of verb movement into the focus domain of a split CP, the Italian sentential particles e and sì became restricted to cases where a full CP is still projected, and changed in function by adjusting to the changed syntactic structure around them.

In Part 3, 'Copulas, auxiliaries and adpositions', Terje Lohndal's chapter, 'The copula cycle', explores the grammaticalization of copulas from demonstrative/personal pronouns or verbs (e.g. 'to remain/stay/stand') in various non-Indo-European languages, adopting van Gelderen's (2004) economy principles. These lead to changes not only in narrow syntax but also at LF, as Remus Gergel demonstrates in 'Rather: On a modal cycle'. As the implicature from 'earlier/faster' to 'preferred/more likely' becomes part of the lexical semantics of rather, part of the LF is erased, and the high position that rather used to reach by LF movement (quantifier raising (QR)) is...