A Corpus-Based Approach to Non-Concord in Present Day English Existential There-Constructions

Abstract

Although number agreement between the verbal form of the existential thereconstructions (hereafter TCs) and the post-verbal NP (hereafter PVNP) seems to constitute the norm, the degree of non-concord in this respect is a noticeable phenomenon worthy of being studied and characterised.From a corpus-based approach, this paper seeks to investigate whether or not there is a direct and/or causative connection between non-concord and factors such as the medium of expression, the type of polarity, the length and structural complexity of the PVNP (notional subject), the presence of co-ordination in the post-verbal sequence or the existence of some intervening elements between the verb and the PVNP.It is hypothesised that TCs containing long, structurally complicated postverbal sequences or those with some intervening element between verb and NP will present a higher frequency of non-concord. Speech, with its faster and somehow unplanned character and its higher degree of interaction between participants, is also assumed to foster this lack of agreement. *

Existential there and catenative concord. Evidence from the British National Corpus

“…(4) ... there seems to be some kind of disturbance going on ... (EFP 843) The entire corpus was needed to provide a sufficient number of instances. In contrast, for Martinez Insua & Palacios Martinez (2003) and Breivik & Martinez Insua (2008) subcorpora were sufficient for their investigations of the much more frequent constructions with be. 9 Seem is the best attested catenative verb in the corpus (about 30,000 of the 60,735 instances of the lemma seem).…”

“…(1) There is a book on the table (2) There are some books on the table 1 A similar analysis based on CobuildDirect and limited to seem is to be found in Olofsson (2007). For an analysis of the concord patterns of the basic there + be construction in contemporary spoken British English, see Martinez Insua & Palacios Martinez (2003) and Breivik & Martinez-Insua (2008). Spoken Canadian English has been similarly analysed by Meechan & Foley (1994).…”

“…Apart from the formal/informal dimension, the situation with be is fairly clear-cut in the sense that it is possible to claim that the postponed element is -really‖ the subject of the clause in which it is positioned. 3 Concord in this type of construction has been extensively studied and reported on, for instance by Meechan & Foley (1994), Martinez Insua & Palacios Martinez (2003) and Breivik & Martinez-Insua (2008).…”

Two Types of there-sentences and Feature Specification

“…The second assumption in (12b) is also a famous assumption in literature, of third person singular BE-verb is frequently used by colloquial standard English speakers. In addition, many corpus studies have found that sentences corresponding to (10a) are frequently used in both spoken and written English (Martinez Insua and Palacios Martinez 2003). Although Sobin's work provides an important insight that "default" case needs to be appealed to account for the unexpected morphological case realization in there-sentences, there still remain many unclear issues.…”

“…However, the system proposed here is well compatible with the asymmetry in number mismatch between there-BE and there-V, as in (32). Adopting a corpus-driven approach, Martinez Insua and Palacios Martinez (2003) investigate variations in number agreement in there-constructions.…”