In Gender shifts in the history of English, Anne Curzan sets out to analyze how and why the English language came to change its original grammatical gender system, which it shared with all other Germanic languages, into a natural gender system. She also addresses a number of related questions, such as the relationship between this shift and sexist language. The study is based primarily on the Helsinki Corpus, which comprises texts from Old, Middle, and Early Modern English, but the author also makes use of a number of other sources. Curzan's research methodology may be termed eclectic; it includes corpus linguistics, prototype theory, historical syntax, sociolinguistics, and feminist theory, as well as references to a number of other theories and schools. This eclectic mix is, however, composed with wit, elegance, and great learning, which makes it very interesting to readers of any theoretical persuasion.

In her introductory chapter Curzan points out that the very word gender is often ill defined in linguistics, and that in modern usage the word in its social meaning is often wrongly applied to biological designations of sex for reasons of “political correctness.” She also makes the important observation, “The gender constructs in the English language reflect social constructs of gender in the world of its speakers.”

In chap. 1, “Defining English gender,” Curzan gives an overview of the various uses of the term gender in English and other languages. She claims quite correctly that the natural gender system of Modern English “[s]tands as the exception, not the rule among the world's languages.” She discusses the distinction between grammatical and natural gender and the ways in which various languages with grammatical gender systems handle conflicts between the two categories syntactically. She points out that the natural gender system of English is not without its exceptions, and that it must still be regarded as a grammatical category that requires a systematic analysis of the use of anaphora in the language. In the conclusion of the chapter she defends her use of the term natural by stating that it does not rest solely on biological sex but on social concepts of sex and gender.

In chap. 2, “The gender shift in histories of English,” Curzan presents a critical survey of the various ways in which the disappearance of grammatical gender and the development of natural gender have been described in histories of English. She describes earlier grammarians' attempts to find reasons for the gender of individual nouns such as ship (feminine) and ocean (masculine). She also quotes a number of 18th- and 19th-century grammarians who regarded grammatical gender systems as “unreasonable” and “irrational” and hailed the gender system of Modern English as an enormous gain in simplicity and clearness. Curzan goes on to discuss morphological explanations for the loss of grammatical gender. She claims that while the loss of inflections in late Old and early Middle English made an important contribution to this development, that it is not a sufficient explanation of the cause of the shift. She finds a number of sociolinguistic factors to be of equal importance. She further discusses whether Middle English can be described as a creole language, a language mix that may have facilitated the loss of the grammatical gender system. Her answer is that, although there may have been significant Old Norse influence on English syntax, there is no need to look upon this as a disruptive creole situation.

In chap. 3, “A history of gender, people, and pronouns: The story of generic he,” Curzan describes the various views on so-called generic he that have been advanced over time, from the traditional assertions that he is the natural or correct anaphoric referent to mixed-gender expressions to today's view that it is sexist. She traces the use of he from Old English onward. Of special interest is her description of prescriptive recommendations written in the past two centuries. While most of these advocated he, a significant change is now noticeable in recent handbooks, which recommend singular generic they.

Chap. 4, “Third-person pronouns in the gender shift: Why is that ship a she?,” focuses on when and how English lost grammatical gender in the use of the singular third person pronouns he, she, and it. Curzan claims that, while grammatical gender remained stable in the Old English period, “the seeds of grammatical gender decay had already been planted in the Old English anaphoric pronouns.” There was even then a clear discrepancy between grammatical and natural gender in the use of anaphoric referents. Thus, Curzan claims, natural gender was not invented in Middle English; it was rather a survival from Old English. The further development of this system was, according to Curzan's sources, strengthened above all by reanalysis of certain masculine anaphoric pronouns that came to be used with inanimate reference. This is, in her view, an good example of the principle that “syntactic change is often triggered by structural ambiguity and, it moves in a direction to avoid ambiguity.” The ambiguity of the masculine forms caused them eventually to be replaced by unambiguous forms. (This seems to be an explanation very much like the one usually offered for the replacement of the ambiguous Old English third person plural forms with Old Norse they, their, and them.) Curzan also argues that grammatical and discursive factors may have interacted in the use of anaphoric pronouns in the development of the gender shift. In her conclusion she claims that the present-day use of masculine or feminine anaphoric reference to inanimate nouns is not just a matter of personification. It may also represent residues from old patterns of diffusion of the gender shift, as well as foreign influence on certain sets of English words.

Chap. 5, “Gender and asymmetrical word histories: When boys could be girls,” is of particular interest to the present reviewer because of my own humble efforts in this area of research. Curzan traces the historical development of a large group of words belonging to the central domain of humanity: terms for the concepts ‘woman’, ‘girl’, ‘man’, and ‘boy’. Many of these have obscure and uncertain semantic and etymological histories. Some major trends are, however, discernible, according to Curzan. One is that non-denotational or evaluative meanings (I would prefer to use the expression “axiological meanings”) may have a marked influence on the semantic development of words and eventually lead to new senses, a position to which I would like to give my whole-hearted support. Quoting a number of other sources, Curzan shows, for example, how a large number of initially neutral female terms have developed negative axiological senses that have often ousted the original sense. She also shows that derogatory terms for women and men are targeted on different aspects of their respective behaviors and personalities: Invectives about women involve sexuality, while those about men focus on their mental and/or physical competence. Curzan also traces and discusses the obscure origins and subsequent developments of words such as girl and boy and the potential pejorative senses of such terms. Another interesting paragraph deals with the use of animal metaphors for men and women, where those referring to women, especially those based on terms for domestic animals, are generally much more derogatory than those referring to men. Curzan also takes up the asymmetry between seemingly equivalent antonymous pairs such as boy-girl, bachelor-spinster and lady-gentleman. She points out that symmetry is difficult to define, and that pairs of this kind may become asymmetrical for a number of different reasons, since gender is just one of the factors that may affect their meanings.

Curzan argues convincingly that history does matter in the sense that old beliefs and cultural patterns that affected words in the past may also affect the usage of new words in the same domain. She claims that linguists have paid too much attention to language-internal changes while ignoring language-external ones, and that “speakers' attitudes about words can affect their development.”

In her final chapter, “Implications for non-sexist language reform,” Curzan points out that, in spite of general resistance to all kinds of language reform, the feminist movement has been highly successful in enforcing changes in the domain of male and female terms in a relatively short period of time. She argues that language is a social instrument, and that speakers' linguistic choices – whether conscious or unconscious – have political implications in the field of human relationships.

It is not really possible to do full justice to this excellent book in this short review. Let me just say that I wish to give it the highest possible marks for its scholarship, convincing argumentation, admirable historical insights, and exactitude. I am sure it will be a valuable textbook in a number of academic disciplines such as English, sociolinguistics, and women's studies.