Left and Right Dislocation of the G-Topic

LEFT AND RIGHT DISLOCATION OF THE G-TOPIC

Ilaria Bacolini

September 2010

Abstract

In English, the lowest position of the Topic in CP does not appear to be available for left realizations of the G-Topic. The reason for such unavailability remains unexplained. What is known, however, is that there are no recognized cases of the G-Topic dislocated on the left. In spite of this, the present analysis will lead us to maintain that, even in this language, left dislocation of the G-Topic is present, but depends on “particular structural conditions of the CP” and on the uniqueness of such a Topic as the only element in CP that can occupy the functional projection of TopP. This paper is divided into three main sections: in § 2, we will present an introduction to the notion of Topic and to other notions considered important to understand subjects in question. In § 3, we will explain in detail our aforementioned hypothesis; finally, in § 4, we will discuss right dislocations of the G-Topic and, in particular, the so-called ‘accented’ Topic.

1.Introduction

The present study aims to investigate different aspects of languages, that is to say, what is called the Parameter of linguistic variation in Generative Grammar (GG). According to Simone (2002), it is more appropriate in primis to discuss language and then linguistic theories, starting from the ‘facts’, that is to say from statements that can be drawn from several languages and which define the world’s reality, in addition to linguistics (Dummet 2008). Thus, the focus of the present analysis tends towards the syntactic structures used by speakers to reach a communicative goal. Specifically, I will investigate the properties and functions of the Familiar Topic (according to the analysis proposed in Frascarelli & Hinterhӧlzl 2007) in languages such as English, where the realization of the Familiar Topic Left Dislocation does not often occur, and in Italian, where the Familiar Topic can appear in any type of sentence (cf. Bianchi & Frascarelli 2009). This crosslinguistic difference has induced us to investigate this linguistic occurrence in order to verify our working hypothesis, that is to say the possibility of realizations - in English – of the Familiar Topic Left Dislocation (LD), which is also realized as if it was a Topicalization. A further aim research concerns the Right Dislocated (RD) Topic which is often analysed together with e marginalization even if they do not have any elements in common. As we will see, the right dislocated Topic, in some cases, functions differently to those described in the literature, which classifies this type of Topic like as “accented” Topic.

2. The Topic

In the literature, there are several works concerning the Topic and, in particular, its left dislocations in the sentence (cf. Rizzi 1997; Frascarelli 1997; Prince 1998; Cecchetto 1999; Frascarelli & Hinterhölzl 2007 among others); however, little research has focused on the lowest Topic of the left periphery and, in particular, the RDs of such a Topic. For this reason, I propose to proceed step by step, starting in primis from the definition of the Topic and then introducing the necessary concepts in order to arrive eventually at the aim of the present study (the Familiar Topic). In Reinhart (1981) an excellent definition of Topic is offered (What the sentence is about); however, in the International literature, the notion of Topic is not as homogeneous as it might seem. Indeed, confusion regarding the definition is common especially when the Given/New dichotomy for distinguishing the Topic from the Focus is adopted. In order to avoid such a mistake, we have adopted the definition suggested by Krifka (2006: 31), since the author overcomes the Topic/Given Information association, proving us with a suitable definition: “The topic constituent identifies the entity or a set entities under which the information expressed in the comment should be stored in the CG content”. For a complete understanding of such a statement, one needs some clarification: Krifka divides the Common Ground (CG) into two distinct areas, that is to say, the CG content and the CG management. The first is connected to those aspects of the Information Structure (IS) which contribute to the information content at a specific moment of the conversation; the second provides the tools to develop the CG content, that is to say the conversational moves made by the participants in the conversation which determine the development of the CG content (cf. Bianchi & Frascarelli 2009, from now onwards B&F 2009). In the light above, we can now understand that the entity evoked by the author is simply the theme of the discourse which must be part of the common knowledge. This can happen in two ways. In the first, everything that is said regarding the aforementioned entity is stored in the CG content, becoming given information. Let us assume that such an entity is already part of the common knowledge; in this case, it is recovered from the CG that is be updated: the information regarding it in the comment is added to the CG content, that is to say, to the related file card^[1]. Also for Reinhart (1981) the CG is not simply a messy set of entities, but rather a set of propositions that provide instructions for the hearer, or more specifically which file card it is better to open (CG management) in order to store the related information (CG content). In other words, the CG is like a ‘tower’ where the file cards are stored; as the conversation advances, the stored file cards are covered by new file cards, but if the speaker decides to ‘reopen’ one of the previous file cards, it returns to the top of the pile. In this way, the aforementioned authors have been able to define the Topic, without recurring to the overused Given/New dichotomy.

[...]

^[1] Metaphor appeared for the first time in the study carried out in Hein (1982), and then adopted by several other linguistics.