Grammar and Usage. A review of Modern English Usage, by H. W. Fowler; The Philosophy of Grammar, by Otto Jespersen; A Grammar of Late Modern English, by H. Poutsma; and Le Langage, by J. Vendryes
In lieu of an abstract, here is a brief excerpt of the content:

8 ] Grammar and Usage A review of Modern English Usage, by H. W. Fowler Oxford: Clarendon, 1926. Pp. viii + 742. The Philosophy of Grammar, by Otto Jespersen London: Allen & Unwin. Pp. xvii + 347. A Grammar of Late Modern English, for the Use of Continental, Especially Dutch, Students, Parts I and II, by H. Poutsma Groningen: P. Noordhoff, 1904-1926. Pp. xi + 812. Le Langage. Introduction linguistique à l'histoire [Language: A Linguistic Introduction to History], by J. Vendryes Paris: La Renaissance du Livre, 1921. Pp. xxviii + 439. The New Criterion: A Quarterly Review, 5 (Jan 1927) 121-24 Aquestionraisedanddebatedfromtimetotime,andalwaysdroppedwithout any conclusion having been reached, is the question of the kind of education necessary or desirable for the acquisition of a good English style. Appeal is usually made to the evidence of “great writers,” but such appeal is never satisfactory . One or another distinguished writer may be cited to prove that Latin and Greek are essential, or that Latin alone is enough, or that neither is necessary , or that a scientific training is as good or better, or that no training whatever ,eventhestudyofEnglishmodels,isrequired.Greatwritersaredeceptive: their virtues may blind us to their defects, or their virtues may be so singular as to make trivial in them vices which would be unpardonable in others. Again, nostyle,especiallynoEnglishstyle,istheperfectmediumforeverycontent.La pensée, says Gourmont, est l’homme même:1 nobody thinks about everything, orthinksineverywaypossible;andthesubjectmatterandthemodeofthought determine the style. We can study the styles of great writers, certainly, but we cannot educate ourselves on their model. We educate ourselves largely on instinct; we can educate others only by the humble light of reason. [ 9 Grammar and Usage Reason, if we formulate our problem in a plain and practical way, seems to me to recommend certain general tips and precepts. It recommends in general a classical education. True, some of our best prose is the prose of scientists; but this statement is reserved by several restrictions. One is that in a scientific training superiority of mind counts for everything: the great scientist writes well, the mediocre scientist indifferently, and the hack scientist badly. Another is that the scientist, when he writes on some subject other than that in which he has been trained, and which he has helped to create, may write, and think, in a very slovenly fashion indeed. We are concernednotwiththeexceptionalstylebutwiththemoderatestyle :ahumane training should teach a man to write well when he writes at all; it should teach him not to write at all of anything about which he has not thought well. The process of learning to write is the process of learning what we are competent to think about and write about. A literary or humane education is certainly that most suitable for all but those who expect to occupy themselves with one of the more exact sciences . The more vague and dubious a “science,” the worse its practitioners write, and the more they depend upon jargon. Physicists, mathematicians, chemists, may write extremely well; psychologists, economists, and especially the American pundits of that American science known as “sociology,” more often write badly. The greatest variation occurs perhaps in the study of history, which is usually written well or badly according as the historian considers his study an art or a science. There is no worse writer than the pseudo-scientist: he believes that his “science” will take care of the thought and the style. The real scientists know better. We may say then, that in general a classical training is the best foundation . Balance and proportion are important: Gibbon and Johnson are great writers, but are their styles not impaired by an excess of Latinity, and might they not have been the better for an infusion of the agile conversational language of Plato? But Latin and Greek, though they require many years of study, are not enough. The English language owes much to good classicists, but it owes something also to writers who, because they were not so saturated with Latin and Greek, have elicited and exploited the peculiar virtues of our Saxon-Danish-Norman inheritance. The particular virtue of the best Tudor prose – as that of Lancelot Andrewes – is its combination of Latinity with the native idiom.2 But this style could not persist; its Latinity is too heavy, its nativeness too crude, for our later culture; we demand a moreperfectassimilationoftheelements.Andtheeighteenthcenturyshows 1927 10 ] that there is always some danger – greater for English than for the more natively Latin tongue of France – of excessive Latinity. If we had Gibbon at...