Word Order Change in Icelandic: From OV to VO (review)

In this revision of her doctoral dissertation at the University of Tromsø, Hróarsdóttir combines sophisticated discussion of theoretical syntax with valuable new textual data on the rigidification of VO syntax in the history of Icelandic. H notes in passing that Icelandic has received widest attention as a case in which OV syntax has been lost in favor of VO syntax while maintaining a fully functioning nominal case system (in contrast to English, for example, where a similar word order shift was accompanied by, and has often been attributed to, the decline of its case system). However, her overriding concern is with characterizing what happened in the history of Icelandic word order and not with why it happened.

After three general theoretical background chapters, H’s original contribution begins with Ch. 4 (the decline of OV word order, 53–78). Here she introduces her historical database (personal letters, when available), beginning with the fourteenth century and continuing up to 1870. She analyzes the data for each successive century, with the additional detail of separate analyses for each of seven successive age groups between 1730 and 1870, the critical period during which OV syntax was completely eliminated in most contexts, contexts in which OV syntax is judged unacceptable/ungrammatical by current speakers. The many tables presenting the statistical results of her analyses show that OV and VO syntax were both present from the earliest documented period (fourteenth century), with OV syntax gradually declining until the critical period (1730–1870) when it was rapidly eliminated. The tables also indicate that the word order variation was subject to consistent grammatical conditioning until the demise of OV syntax, most notably conditioning by the nature of the objects, for example, whether definite or indefinite, light or heavy, and so on.

Discussion of representative examples and of tables of statistical data, and the historical trends they exhibit, continues through Ch. 7 (word order patterns, 177–223). In Ch. 8 (object positions, 227–55) H turns to applying generative syntactic models to the historical trends discussed in the preceding chapters. She rejects the less conventional model of alternating bases (proposed by Anthony Kroch and Susan Pintzuk for early English), arguing that generation from two separate bases is implausible for single clauses in which word order is ‘split’ between a preposed and postposed object (e.g. Indirect Object-Verb-Direct Object). Instead, she adopts a version of principles and parameters syntax (proposed by Richard Kayne) in which all constituent movements are leftward. Here, universal implications of the Kaynian model aside, H proposes, contrary to the book’s subtitle, that Icelandic has been a basic VO language throughout its recorded history and that the changes in word order reflect the loss of optional leftward movements of postverbal constituents. Implications for the history of English and the other Scandinavian languages are sometimes alluded to in passing. In Ch. 9 (parameter change, 257–343) she refines her model to unite into a single parameter the two optional movements that were simultaneously lost during the nineteenth century. One of the lost options allowed OV order; the other allowed Main V-Aux V order. She labels the single parameter ‘remnant VP-preposing’. Thus, H illustrates the interesting theme that sometimes historical considerations can be useful in selecting among competing synchronic descriptions, in this case one in which a single rule is preferable to one in which object preposing and auxiliary ‘postposing’ are analyzed as independent movements.

H’s book advances knowledge of the Icelandic word order change, and it is clearly written and well-organized. It is worth further scrutiny by anyone interested in the details of word order change, even for its data if not for its particular theoretical conclusions.