This paper reports the discovery of a syntactic dialect difference between northern and southern Middle English in the grammatical implementation of the so-called constraint and argues that this difference is most likely a linguistic contact effect of the Viking invasions of northern and eastern England in the eighth and ninth centuries. In the South, the Middle English V2 constraint behaves as it had in Old English; that is, as a variant of the IP-V2 type, the type found in modern Yiddish and Icelandic. In the North, however, the constraint is of the CP-V2 type, as found in modern Mainland Scandinavian and in German or Dutch. It is argued that the northern form of V2 arises as a consequence of the collapse of agreement inflection in northern Middle English, which leads to a loss of V-to-I movement, just as happened in mainland Scandinavian when agreement inflection was lost there. Once, V-to-I movement is lost, IP can no longer host the topic-verb configuration that defines the verb-second phenomenon. The collapse of agreement is further argued to result from imperfect second language learning by the Scandinavian invaders, who formed a large enough proportion of the population of the North to impose substratum effects on the northern dialect. Investigation of the available documentary evidence is shown to support dating both the collapse of endings and the appearance of CP-V2 word order to the period of the Viking invasions. It is suggested that the loss of V2 word order in English, which has made it the only Germanic language that does not respect the constraint, may have resulted from the extensive contact between northern and southern speakers that lead to the mixture of northern and southern features that characterize the language of the Midlands (and London) in the late Middle English period.
Stored only when both papers are in the corpus — 144 citations exist upstream.