the bends at an early age, and we have never gotten rid of them" (p. 9).
In this book I've tried to get rid of some of my bends. I hope others will
find it "unbending" too, but not in the conventional meaning of the term
as dogmatic and inflexible. I do not expect all readers to agree with me!

Some of the material I include here comes from a series of lectures and
practical classes I gave on the topic of language and gender at Oxford in
Michaelmas Term 1993 and Hilary Term 1994. In preparing these sessions
I was greatly aided by consulting the Language and Gender Syllabi Col-
lection edited by Elizabeth Hume and Bonnie McElhinny and published by
the Linguistic Society of America's Committee on the Status of Women.
The collection contains 26 syllabi from courses on language and gender
offered in a variety of departments, including linguistics, anthropology,
English, and education. From them I got many references, ideas, and ma-
terial for exercises.

I am also grateful to a number of colleagues such as Bernd Heine for
the Ewe examples in chapter 4, and to Lou Burnard of the Oxford University
Computing Services for help with the British National Corpus, an early
pilot version of which I have used to collect some of the material in chapters
4 and 5. I would also like to thank Judi Amsel for her assistance and her
comments on an early version of chapter 9.

To facilitate the use of the book as a textbook, I have tried to avoid
cluttering the text itself with detailed citations and I have avoided footnotes
altogether. I have mentioned some additional sources of reading at the end
of each chapter and given full bibliographical details in the references at the
end of the book. In the text itself, I have used specific dates after the names
of scholars for whom I have cited more than one reference in the bibliography
in order to make clear which particular work provided the source for what
I say. Where I have quoted directly or paraphrased a particular point, I
have given the date and page number in the text. Each chapter ends with
discussion questions and exercises, many of which involve readers either
in collecting their own data for analysis or analyzing data.

-- Suzanne Romaine

-xiv-