Like any other well structured text a preface should have a beginning,
a middle and an end and generally display a modicum of textual cohesion. The end, for example, is often a statement to the effect that the
author shoulders full responsibility for any mistakes still left in the
book, thus routinely exonerating from guilt all those to whom gratitude has been expressed. Since this preface tends not to abide by the
acceptable canon of preface-cohesion, I shall begin by accepting total
responsibility for everything that appears in this book with the exception of the data extracts, which are transcriptions of verbal interactions
among the members of my extended family. To the extent that transcribing real live interaction after the event cannot do justice to what
was going on, the responsibility for the transcriptions is indeed that of
the researcher, but what was said is genuine and not of the researcher's making. I have simply tried to represent it as faithfully as
possible, and I am deeply grateful to the members of my family who
gave this research substance by being willing to cooperate. Their natural spontaneous response to being recorded has been of inestimable
value.
The book itself arose from a profound concern that some of the
central concepts in use to describe and analyse verbal interaction in
spoken English did not tally with my own commonsense understanding of those notions. In particular, the term "interruption", it seemed to
me, was being used almost indiscriminately to refer to occurrences of
simultaneous speech caused by incoming speakers not waiting for current speakers to finish what they were saying before beginning their
turns at talk. I was not particularly worried about such occurrences
being defined as interruptions, but rather about the tendency of almost
all researchers to evaluate them as speech dysfluency and unacceptable
social behaviour and to use them as a variable with which to identify
and quantify asocial speech behaviour. After all, as long as a term
within an empirically grounded model of discourse is adequately defined, it does not really matter what label we give it. My lay understanding of interrupting has always been that it is a negative activity,
and in this respect (and only in this respect) I found myself in agreement with the majority of researchers