The Constant Rate Hypothesis is the claim that a syntactic change affects all grammatical contexts (like questions, negative sentences etc.) at the same rate of change. In this paper, I test this claim with a large dataset of possessive have structures and find that it is plausible to think the prediction is correct.

I report a regression model for the change from OV to VO in Middle English. It focuses on genre (prose versus poetry) as a predictor by including data from a recently published corpus, the PCMEP, along with other independent variables. I highlight the importance of the genre variable, poetry being considerably more conservative than prose, and recommended that poetry texts should be considered in studies on early Middle English syntax more generally.

Middle English loses the negative particle 'ne' in its function as a sentential negator as well as in negative concord structures. This paper measures 'ne'-drop in both contexts with quantitative data from three syntactically parsed Middle English corpora. Surprisingly, the results show that the rate of 'ne'-loss with negative prepositional phrases (PPs) seems to be significantly slower than elsewhere. Different explanations for this observation are explored.

Middle English changes the realization of temporal subordinators from a th-form (then) to a wh-form (when). The innovation is quantified with data from four syntactically parsed corpora. The change may have had an antecedent cause in the loss of subject-verb inversion after adverbial then.

Word2Vec models are used to study the semantic chain shift FOOD>MEAT>FLESH in the history of English, c. 1425-1925. The development stretches out over a long time, starting before 1500, and may possibly be continuing to this day. The semantic changes likely proceeded as a push chain.

In my PhD dissertation, I discuss general mechanisms of syntactic change on the basis of three case studies from the history of English.

I use a probabilistic Bayesian classifier trained on fourteen syntactic features to ascertain the date of composition of undated Old English texts.

Old English main clauses and main clauses introduced by a conjunction pattern differently in terms of verb placement and topicalization. This paper propses an LFG model to capture these distributional differences.

This paper claims that Old English se þe relative clauses are the result of two independent, overlapping rules. The overlap is made possible by the fact that the rules are not conditioned on some contextual factor.

Early English develops from a verb-final to a subject-verb language. But in order to measure this change accurately, diagnostic elements must be identified that cannot postpose. In this paper, I show that self is such a diagnostic element.