Cyclical change (review)

The collection of articles included in this volume is based on papers presented at the Workshop on the Linguistic Cycle, which took place at Arizona State University in April 2008. This workshop was spearheaded by Elly van Gelderen, who sought to bring together linguists working on cyclical (or cyclic) change within different frameworks. Cyclic change is a diachronic phenomenon whereby an element grammaticalizes to a new functional item and is concurrently or subsequently replaced by a new element which can then renew the cycle by undergoing the same grammaticalization process. Van Gelderen’s research has invigorated the field of diachronic linguistics by providing a testable generalized theory of cyclic change within the Minimalist Program (van Gelderen 2004; cf. Roberts and Roussou 2003). Her theory of linguistic cycles is explanatory in its aims. She argues that syntactic economy principles (Chomsky 1995) are active during first language acquisition just as they are in adult grammar. These economy principles, such as an acquisition version of the preference for Merge over Move, called the Late Merge Principle, are argued to be responsible for the directional bias evidenced by cyclic changes; for further details see van Gelderen (2004, 2011). Her proposals continue the tradition of explaining syntactic change via L1 learning (.g., Lightfoot 1979).

Van Gelderen begins the volume with the sketch of the big-picture questions addressed: In which domains do cycles exist? How uniform are cycles across different syntactic domains? What are the sources of renewal (i.e., once an item or expression has semantically bleached, what replaces it)? And finally, why does cyclic change exist and how can we explain it? The articles in this book focus on answering different questions but share the goal of furthering our understanding of the grammar by analyzing data and processes which become comprehensible only through a diachronic lens. The book is divided into four sections, the first three addressing different linguistic domains: (i) negatives; (ii) pronouns, agreement and topic markers; and (iii) copulas, auxiliaries and adpositions. The fourth section includes only one paper, which outlines an innovative artificial-language based approach for simulating cyclic change. In total this volume consists of 13 papers.

The first section of the book covers Jespersen’s Cycle, which concerns the development of negative operators and negative reinforcers. In the first article, Hoeksma uses evidence from West Germanic languages to argue that negative polarity items (NPIs) are only candidates to renew the cycle (by becoming negative quantifiers) when the likelihood (measured with probability statistics) of these items co-occurring with negation is very high. He also articulates a lesser known negative cycle whereby a DP-like polarity item, e.g., English naught, becomes an adverbial negative. Neutralized contexts are thought to be the source of reinterpretation, a recurring theme throughout the volume. Next, van der Auwera argues that Jespersen’s Cycle is better understood as a collection of related cycles with eight possible trajectories. For example, the tripling of negative items is an optionally pursued third stage of negation (evidence from Lewo (Vanuatuan)). Ultimately, he argues that the phenomenon of the negative cycle is an oversimplified one.

The next two papers focus on the negative concord aspect of Jespersen’s Cycle. Tsurska presents a case study in Russian sentential negation. She argues that Russian sentential negation has changed from Non-Strict to Strict Negative Concord (Ziejlstra 2004). This change involves the loss of [iNEG] on the preverbal negative marker and is thus in accordance with van Gelderen’s proposal that cyclic change involves reanalysis of interpretable features as uninterpretable features. Biberauer compares negation strategies in Standard Afrikaans and Colloquial Afrikaans to those of English, German, Dutch, and French. She argues that, while Afrikaans appears to be “off-course” in terms of the negative cycle, there are good syntactic reasons for the supposed unresponsiveness of Afrikaans nie for participation in the negative cycle. Essentially, Biberauer argues that the last step of the cycle can only be taken when candidate elements are low enough structurally, and bear appropriate substantive...