Obviously, in the process of preparing and writing a thesis, smaller and larger frustrations and excitements crop up which are not shared with one's supervising professor, but rather with fellow students and friends who are kind enough to put up with all sorts of talk about the thesis. Such experiences are shared with a lot of 'colleagues', of course, but I feel compelled to single out a 'happy few' of friends who have regularly and – luckily – tactfully inquired into the state of proceedings. They are Marian Belsack, Dany Buts, Servaas Carbonez, Peter De Gryse, Helder De Schutter, Wouter Deswarte, Luc Naert, Tine Olivier, Marisa Peeters, Wim Remysen, Stefaan Simoens, Annemie Vandaele, Jeff Van Meir and Peter Willemse. If there is, perhaps, no God to reward them in afterlife, there is at least me to repay their kindness with my gratitude.

On a more practical level, I would like to thank Stefaan Simoens for helping me out on that cursed April day my computer decided to take a vacation and turn to wilful disobedience. To Wim Remysen I say thanks for okaying the additions or amendments I made to a number of French examples so as to bring out contrasts with examples taken from the literature. Finally, even though I got myself a decent printer at last, I have not forgotten the various occasions on which Johan Lenaerts was kind enough to let me use his.

On a more general level, I would like to see this thesis as a modest tribute to all the people that have had a formative influence on me over the years – teachers in nursery, primary and secondary school, professors and research assistants, and of course my parents, other relatives, and friends. Each in their own way, they have helped to unlock some of the mysteries of this world, but also to stand in uncomprehending awe for many others; not to give up in face of difficulties ahead, but also to recognize the limits of any human being's possibilities; to enjoy the fullness of everyday existence, but also to look for ways of transcending the obvious.

The single most formative environment for me, as, I believe, for anyone, was of course my home, which having been and continuing to be a home filled with books proved to be a very stimulating place in which to grow up – our large garden always just a doorstep away for whenever the bookishness had to be counterbalanced. I thank my mother, Denise Hoedt, for countless practicalities that have helped me to deliver this work in time, but also, more fundamentally, for her continual support and belief in me, offering me the opportunity to study as much as I want. And finally, I would like to dedicate this thesis to the memory of my father, Guido Vandelanotte. As a teacher of English, quality time was no airy catchphrase for him; he would have the blackboard written on even before class if he could, and even though he probably placed a larger stress on grammar than is customary in high school teaching, he never lost sight of music, poetry and prose, through which he tried to pass on more than knowledge. It is his profound love for the English language, instilled in me, that is my source of inspiration.

Science for beginners
 
 

If linguists do not decide to quit their jobs, they should at least be aware of the fact that whatever answers they provide to descriptive problems, these will always remain partial, temporary and imprecise. Language, like nature, is just 'there', in its infinite complexity and versatility, and it has little concern for the mind-bending effects it has on those people trying to get a hold on it. It will never be grasped, as a matter of fact: language is always well ahead of linguists, continually adapting to new circumstances or undergoing changes. If the professional lives of linguists are spent on "imprecisions", to borrow Larkin's word, one may rightly ask why they bother. Why do linguists – or physicists for that matter – probe into the 'things that are'? It is their job, for sure, they get paid to do so, or, as in my case, they have to in order to obtain their degree. Surely that cannot be the whole story?

I certainly hope not. The ignorance of Larkin's poem I take not to be a downright negative one: there is genuine wonderment and admiration in the perception of "the way things work: / Their skill at finding what they need, / Their sense of shape, and punctual spread of seed, / And willingness to change". From such a position of wonderment stems reflection and research. From such wonderment, too, may originate art – Larkin's poem, for instance. I would suggest that it is not only in the shared starting position of wonderment that art and science – to use the big words – are conjoined rather than opposed. Both, in their separate ways, are fundamental expressions of the human being's need for transcendence – the 'more than sense' which separates our lives from the unthinking lives of animals. Without science and without art, and other forms of transcendence such as religion, forgiveness, and love, life would be no more than a struggle for life, an endeavour to retain the condition of living, and that is all. That would surely not be much. So linguists are right, after all, in keeping their jobs.

My 'job' in this thesis is to offer a functional approach to free indirect speech or thought. Before explaining a bit further what that amounts to, I concede at the outset that I am just an 'ignoramus' starting out with wonderment, enthusiasm and "a hunger in himself to be more serious", to borrow a phrase from Larkin's "Church Going". That the answers I will provide to specific descriptive problems can only be partial, imprecise and temporary should be evident. All the same, I hope they may transmit some of my wonderment and enthusiasm, and provoke further thought.

Preparing a thesis – plans and outcome
 
 

The original title for this thesis was "A corpus-based study of free indirect speech". As appears from the title – or at least, the subtitle – in the final draft of this book, the corpus has silently withdrawn. The idea of a corpus-based study was, in fact, to offer a detailed account of the concrete realizations of the category of free indirect speech or thought in (literary) texts. Such an endeavour can, however, only be undertaken successfully if there is a more or less generally accepted understanding of what free indirect speech or thought is, about. My exploratory reading on free indirect speech or thought soon revealed that such was not the case. The impression I was left with, was that free indirect speech or thought was certainly a much debated issue, but at the same time, that this discussion always revolved around the same issues in the same (vague) terms. I am referring specifically to the 'voice' question: is there, in free indirect speech or thought, a superimposition of two voices – narrative and characterial – or is there just the character's voice?

The descriptive gaps are not only caused by an insistence on one specific issue, it seems to me. Free indirect speech or thought has been widely studied in many literary texts, but its linguistic features have not been given the attention they deserve. Speech and thought representation in general, and free indirect speech or thought more specifically, are exceedingly complex areas of grammar which give ample scope for detailed observations on the one hand, and 'functional' generalizations on the other hand.

It is, then, the nature of the topic, combined with the state and amplitude of the linguistic literature available on this subject, that caused my work to take a different course. In terms borrowed from logical semantics, one might say that rather than developing an 'extensional' description of the category of free indirect speech or thought, I concentrated on the 'intensional' level. In other words, instead of pursuing the original plan of drawing up a prototypical model of different concrete realizations on the side of the 'referents' or concrete tokens of free indirect speech or thought, I took the logically prior step, viz. defining and describing the forms and functions of the more general and abstract 'category' of free indirect speech or thought. An internal description of the category of free indirect speech or thought cannot be complete, however, without proposed delimitations vis-à-vis related categories. Added to the 'internal-intensional' description, there is an attempt at a better 'external' demarcation of the category. These two taken together should, in the end, offer a useful descriptive tool on the basis of which a thesis of the title "A corpus-based study of free indirect speech or thought" could indeed be conducted systematically.

Outcome

This thesis is organized in five parts. In an introductory part 0, the general framework which I will be using throughout this book is sketched. Apart from locating this work within the field of 'functional' approaches to language phenomena, it outlines the central concepts from which I will work in my argument, and gives a first, tentative characterization of the topic of this thesis, free indirect speech or thought. Part 1 offers a selective survey of literature, organized thematically and focusing mainly on the 'dual voice controversy' mentioned above. Its aim is to come to a status questionis which provides challenges to be taken up from the functional perspective described in part 0.

Starting from the gaps in existing descriptions of free indirect speech or thought – notably the vagueness surrounding the 'dual voice controversy' – I set out to offer a comprehensive and consistent 'internal-intensional' description, linking features and functions of specific components of clause grammar up with the general functional categories distinguished in the framework of part 0. It is also in part 2 that an initial description of a fourth type of speech and thought representation, besides the traditionally distinguished direct, indirect and free indirect, will be included. I will try to demonstrate that this fourth type is semiotically distinct: it can be set apart from the other modes of speech and thought representation not only in semantic, but also in correlating grammatical terms.

The internal description of part 2 is complemented in part 3 by an external delimitation of free indirect speech or thought vis-à-vis a number of related sentence types which may sometimes be confused with it. Towards the end of that part, a more systematic classification of a number of sentence types rendering the 'mental' processes (such as saying, thinking, wanting, liking and perceiving) of a character in a novel will be proposed.

Finally, part 4 summarizes the main contributions of this thesis in terms of what I consider to be its three main contributions: the enhanced functional description of the internal make-up of the category of free indirect speech or thought, the description of the fourth type of speech and thought representation called disclaiming indirect speech or thought, and the systematization of various narrative modes of rendering the mental processes of characters. Part 4 is rounded off with some suggestions for further research and some concluding thoughts.