Published online by Cambridge University Press: 19 February 2026
This article looks at a hitherto unnoticed series of parallels between Middle English (ME) and Medieval Spanish (MS). In the first place, while both appear to have been essentially VO languages, they allow identical classes of object to move leftward over the verb. In ME, movement is normally to a position between the auxiliary and the main verb, whereas in MS the movement is normally to the left of the auxiliary, if one is present. Second, both languages allowed the fronting of nonfinite verb forms. We argue that the parallels in question are the result of analogous processes of reanalysis operating in the two languages at earlier periods in their history. Both MS and ME descend from languages that tended toward O-V(-Aux) order. According to our analysis, when the structural mechanism giving rise to the latter order was lost, certain instances of OV ordering were nevertheless able to survive in reanalyzed form. The ‘low’ pattern of reanalysis in ME versus the ‘high’ pattern observable in MS can then be attributed to the blocking effect of the subject in Spec-TP in English and its absence in the Spanish case (where the subject is either null or not in spec-TP). As regards nonfinite fronting, this too can be regarded as a relic of the older word order. For MS we posit an analogous reanalysis to that observed for relic OV ordering, whereas for ME the evidence points toward an undisturbed continuation of the earlier V-Aux pattern. This difference between MS and ME can again be attributed to the absence versus presence of a spec-TP subject.
The research on which this article is based was made possible by a grant from the Centre for Research in Linguistics and Language Sciences at Newcastle University. Parts of the article were presented at the 7th Symposium on the History of English Syntax, Nijmegen, 6-7 June 2009, and the Linguistics Association of Great Britain’s annual conference, Edinburgh, 6-9 September 2009. In addition to members of the audience at those gatherings, we are grateful to Michelle Sheehan and Richard Waltereit for helpful comments on a draft version of the article. We would also like to thank Greg Carlson and three anonymous Language referees, whose detailed and perceptive comments led to a substantially revised and much improved final version. All of the usual disclaimers apply.