A Dictionary of English Normative Grammar, 1700-1800 (review)

REVIEWS A Dictionary of English Normative Grammar, 1700-1800. 1991. Bertil Sundby, Anne Kari Bjorge, and Kari E. Haugland. Studies in the History of the Language Sciences, 63. Amsterdam: John Benjamins, ? + 486 pp. $95.00. It is always a pleasure to see the completion of projects that were described in Dictionaries in their nascent state, and this work (henceforth DENG) is no exception to that principle. In his promissory essay of 1985, Sundby estimated that the completed work would total "approximately two hundred printed pages" (1985, 151). Wisely, he did not project a completion date—he had begun in 1980—but one suspects that as the length doubled, so too did the time. The first steps involved identifying an audience and determining the scope of work. Though many histories of English allude to 18th-century grammarians as more or less authoritarian cranks, Sundby recognized that their comments on the nuances of usage provide useful clues for the "historiography of grammar" and for the social stratification expressed in English variously in bodi Britain and Nordi America. In 1985, Sundby detected growing interest in die insights of the grammarians (182), but in his introduction to DENG he is obliged to declare diat even now "scholars have been slow to recognize die linguistic potential of die old grammar-books" (1). One reason for such slow progress, of course, is diat it has not been easy to gain an overview of opinions about usages. It is precisely diis difficulty that DENG sets out to remedy. The second step required Sundby to define die corpus of works to be covered. R. C. Alston's bibliographical work and die subsequent publication of facsimile reprints, English Linguistics, 1500-1800 (1967-72), provided the essential ground work. Even this valuable series did not yield the complete corpus, diough it was essential to it. Thus, Robert Baker's Remarks on the English Language first appeared in 1770 and that edition is die one Alston reproduced in facsimile; Sundby and his colleagues have also employed die second edition of 1779 with Baker's additional remarks. In addition, Sundby surveyed Nordi American works not included by Alston—for instance, Caleb Bingham's The Young Lady's Accidence (1792) and Benjamin Dearborn's The Columbian Grammar (1795). The result is a bibliography on an international scale with 150 tides published in England, 14 in America, 14 in Scotland, 7 in Ireland, and 2 in "other countries" (Antigua and Belgium). In the end, 136 titles were investigated in addition to die 57 reproduced in facsimile by Alston. Having assembled the books to be studied, Sundby and his colleagues were faced with organizing a scheme by which observations might be classified . Naturally enough, die scheme was variously revised, and diere is little point in rehearsing its evolution here. Some of die difficulties faced, and overcome , are described in the introduction. The category "Idiom," it may be 1 72Reviews said, was deleted, and a few competing constructions (for instance, prevail upon/over), promised in 1985, were not ultimately treated. In the end, DENG presents 14 categories of "error," each with an introductory explanation and overview. For a user interested in a particular expression, however, a detailed index allows immediate access. Thus, Thomas Gray wrote to his friend Horace Walpole that his poem "must be, Elegy, wrote in a Country Churchyard ." When it first appeared in print, it was titled "STANZAS written in a Country Church-yard." The status of these competing usages can be ascertained simply enough by reference to write in the index and from there to a full treatment under the category "Differentiation" (233-34). As the categories were refined, Sundby and his associates found themselves with an increasing mass of material. At last, the number of excerpts totaled nearly 18,000. (The work was compiled by developing a data base, though nothing is said about the software beyond the fact that Ivar Utne created it. Computers themselves promote a fascination with counting things that might not otherwise have been so easily indulged.) The most censorious grammarians were John Knowles and Louis Brittain, citing, respectively, 722 and 531 types of error; the least censorious are low on the list mainly because their focus...