In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Null pronouns
  • Henrik Rosenkvist
Null pronouns. Ed. by Melani Wratil and Peter Gallman. (Studies in generative grammar 106.) Berlin: De Gruyter Mouton, 2011. Pp. 270. ISBN 9783110238709. $140 (Hb).

The possibility of omitting referential subject pronouns from a finite clause has been a popular topic for generative grammarians for quite some time. Following a couple of influential works by Rizzi (1982, 1986), a substantial part of the research has been concerned with null subjects in the Romance languages and in languages such as Greek, Mandarin, and Finnish, while the Germanic languages have been absent from the scholarly debate. Indeed, since all standard Germanic languages require referential subject pronouns to be overt, some linguists have claimed that referential null subjects as such are incompatible with the V2-property. As shown by the articles in Null pronouns, however, that conjecture is erroneous. Referential null subjects in both Old and Modern Germanic languages are discussed in depth in the book, and it is therefore an important addition to this field of research.

Null pronouns is a collection of five articles from a 2006 workshop in Jena, plus an introduction by the editors. In the introduction, Melani Wratil and Peter Gallman provide a background to the research on null subjects in the framework of principles and parameters, discussing PRO, pro, the pro-drop parameter, and the much-debated relation between null subjects and verbal inflection. Throughout the introduction, problems with previous approaches to null subjects are highlighted, including the notion of a particular pro-drop parameter, null subjects and V2, null subjects and rich inflection, the identification of null subjects, and the historical development of null subjects. Although the introduction is relatively brief, it provides a useful outline of generative research about null subjects. In some places, however, one might have wished for a more complete set of references. For instance, the notion of I-subject (i.e. the idea that the inflectional affix on the finite verb may function as the subject of a clause) goes back to Borer (1986), and as shown by Roberts and Holmberg (2010:3), the correlation between rich agreement and null subjects was probably first observed by Apollonius Dyscolus, a second-century grammarian in Alexandria.

Following the introduction, Katrin Axel and Helmut Weiß explore the emergence of null subjects in some Modern Germanic vernaculars in ‘Pro-drop in the history of German—from Old High German to the modern dialects’. In early Old High German (OHG), null referential subjects varied quite freely with overt subjects, but null subjects only appeared in clauses where the finite verb moved to C, that is, main clauses and embedded clauses with V-to-C movement. They claim that this requirement indicates that OHG null subjects needed to be c-commanded by agreement features in C. While OHG verb inflection was distinct in parts of the paradigm, however, the forms for first- and third-person singular show no agreement at all, and hence one would expect that such subject pronouns should always be overt in OHG. That is not the case; third-person null subjects were very frequent. To deal with this problem, Axel and Weiß evoke Sigurðsson’s (1993) concept of free discourse coindexing, without going into details. In a number of Modern Germanic vernaculars, such as Bavarian and Frisian, first- and second-person (but not third-person) null subjects can also be found, in main clauses as well as in embedded clauses of all types. Another conspicuous, and typologically rare, feature of these language varieties is complementizer agreement. The complementizer wenn ‘if’ in Bavarian, for instance, may be inflected for second-person singular. In the clause wennst willst ‘if you.sg want to’, there are two instances of -st, the affix for second-person singular (but no overt subject). The analysis presented by Axel and Weiß is that the OHG licensing condition still applies, and that null subjects in Modern Germanic vernaculars may appear in nonroot contexts because complementizer agreement, not present in OHG, can c-command and identify a null subject also in a verb-final embedded clause. This idea, which is also presented in Axel & Weiß 2010, is appealing, but other Germanic...

pdf

Share