be diminished by ceasing to write, but by writing others which, like Aaron's serpent, shall swallow up the spurious.[5]
19. I have said that some grammars have too much originality, and others too little. It may be added, that not a few are chargeable with both these faults at once. They are original, or at least anonymous, where there should have been given other authority than that of the compiler's name; and they are copies, or, at best, poor imitations, where the author should have shown himself capable of writing in a good style of his own. What then is the middle ground for the true grammarian? What is the kind, and what the degree, of originality, which are to be commended in works of this sort? In the first place, a grammarian must be a writer, an author, a man who observes and thinks for himself; and not a mere compiler, abridger, modifier, copyist, or plagiarist. Grammar is not the only subject upon which we allow no man to innovate in doctrine; why, then, should it be the only one upon which a man may make it a merit, to work up silently into a book of his own, the best materials found among the instructions of his predecessors and rivals? Some definitions and rules, which in the lapse of time and by frequency of use have become a sort of public property, the grammarian may perhaps be allowed to use at his pleasure; yet even upon these a man of any genius will be apt to set some impress peculiar to himself. But the doctrines of his work ought, in general, to be expressed in his own language, and illustrated by that of others. With respect to quotation, he has all the liberty of other writers, and no more; for, if a grammarian makes "use of his predecessors' labours," why should any one think with Murray, "it is scarcely necessary to apologize for" this, "or for omitting to insert their names?"—Introd. to L. Murray's Gram., 8vo, p. 7.
20. The author of this volume would here take the liberty briefly to refer to his own procedure. His knowledge of what is technical in grammar, was of course chiefly derived from the writings of other grammarians; and to their concurrent opinions and practices, he has always had great respect; yet, in truth, not a line has he ever copied from any of them with a design to save the labour of composition. For, not to compile an English grammar from others already extant, but to compose one more directly from the sources of the art, was the task which he at first proposed to himself. Nor is there in all the present volume a single sentence, not regularly quoted, the authorship of which he supposes may now be ascribed to an other more properly than to himself. Where either authority or acknowledgement was requisite, names have been inserted. In the doctrinal parts of the volume, not only quotations from others, but most examples made for the occasion, are marked with guillemets, to distinguish them from the main text; while, to almost every thing which is really taken from any other known writer, a name or reference is added. For those citations, however, which there was occasion to repeat in different parts of the work, a single reference has sometimes been thought sufficient. This remark refers chiefly to the corrections in the Key, the references being given in the Exercises.
21. Though the theme is not one on which a man may hope to write well with little reflection, it is true that the parts of this treatise which have cost the author the most labour, are those which "consist chiefly of materials selected from the writings of others." These, however, are not the didactical portions of the book, but the proofs and examples; which, according to the custom of the ancient grammarians, ought to be taken from other authors. But so much have the makers of our modern grammars been allowed to presume upon the respect and acquiescence of their readers, that the ancient exactness on this point would often appear pedantic. Many phrases and sentences, either original with the writer, or common to everybody, will therefore be found among the illustrations of the following work; for it was not supposed that any reader would demand for every thing of this kind the authority of some great name. Anonymous examples are sufficient to elucidate
