The historical development of the it-cleft
A comparison of two different approaches
Amanda L. Patten
This chapter compares two approaches to a particular grammatical change. While Ball (1991, 1994) investigates the development of the it-cleft configuration from within the generative tradition of the 1990s, I have recently re-examined the historical it-cleft data from a constructional perspective (see Patten 2010, 2012). In this chapter, I show how our different theoretical assumptions lead us to categorize and analyse the data differently. I conclude that a constructional approach is better at interpreting the diachronic facts.
Construction grammar was developed as an alternative to the model of grammatical knowledge assumed in generative theories. As a result, we have access to a large literature which outlines the crucial differences between these theories of grammar and which, to a lesser extent, examines how these differences manifest themselves in our understanding of (or rather our perspective on) language change. What is less clear is how these different perspectives (or expectations) actually affect the way that we undertake historical language research and whether our findings and conclusions are to a large extent dependent upon those theoretical assumptions.
In this chapter, I compare two approaches to a particular grammatical change: one largely in the generative tradition of the 1990s, the other in the more recent tradition of usage-based construction grammar. The change I examine is the development of the English it-cleft. While there has been much debate about how best to analyse the structure of sentences such as (1), the history of the it-cleft configuration is a relatively under-researched area.
(1) It's Howard who plays the bassoon
Until recently, Ball's (1991, 1994) work dominated the literature on this topic, as the only substantial diachronic account to result from empirical research. Ball claims that the it-cleft originated in Early Middle English. She finds that the it-cleft subsequently underwent two major changes in its development, occurring with a new range of foci and permitting hearer-new information into the cleft clause. Ball explains these developments as resulting from two separate mergers in Late Middle English, involving the it-cleft, an OE impersonal construction, and the reverse pseudo cleft. However, in Patten (2012), I re-examine the historical it-cleft data from a constructional perspective. Contra Ball (1991), I find that the it-cleft can be traced back to Old English. I go on to argue that the later development of it-clefts with non-NP foci and it-clefts which express hearer-new information in the cleft clause results from the gradual expansion of the existing it-cleft schema, involving grammatical constructionalization (see also Patten 2010).
In this chapter, I explore the extent to which these different diachronic accounts reflect and are influenced by two different sets of theoretical assumptions. While I adopt a constructional framework, Ball's study draws from within the generative tradition of the early 1990s. I suggest that this has implications both for our treatment of the historical data and our perspectives on language change.
The structure of the chapter runs as follows. In 2, I briefly outline generative and constructional models of language structure and language change. In 3, I explain how these two theories of grammar lend themselves to different and competing analyses of the PDE it-cleft configuration. From the outset then, Ball and I approach the historical data from different perspectives and with a different understanding of the proper object of study. As I explain in 4, this means that we apply a different set of criteria to the data, influencing how we sort through and factor out (what we consider to be) the relevant and irrelevant examples. I show that this alone leads Ball (1991) and Patten (2012) to propose divergent dates for the origin of the it-cleft. In 5, I compare Ball's (1994) and Patten's (2010) accounts of the subsequent development of the it-cleft configuration. I show how we are able to interpret the diachronic data in ways which are consistent with, and which reinforce, our diverse theoretical assumptions. I conclude, in 6, that the diachronic facts of the it-cleft case study are more amenable to a constructional explanation. I then go on to discuss the implications for historical language research (and the role of language theory) more generally.
-  I would like to thank Willem Hollmann, Elizabeth Traugott, and two anonymous reviewers for their valuable comments on an earlier version of this chapter.
-  The early historical cleft literature comprises a short paper by Matsunami (1961) and brief discussions in more general works on historical syntax (including Curme 1931, Traugott 1972, and Visser 1963). See Filppula (2009), Johansson (2008), Patten (2010), and Pérez-Guerra (1999, 2012) for more recent contributions.