There was thus a simultaneous development in language studies on both sides of the Atlantic, with neither side in the early days knowing much about what the other was doing. However, it is usual to try to date the beginning of a science by referring to the publication date of some pioneering work; and those who have tried to do this for linguistics generally give the honours to a European. Despite the tendency these days to see the origins of linguistics in the work of almost every scholar since Plato, it is generally accepted in more sober mood that the work of the Swiss scholar Ferdinand de Saussure holds pride of place as the first real essay into linguistic theory as we understand it now. To call him the 'founder' of the subject, as is sometimes done, is perhaps a bit extreme, in view of the American work taking place at the same time. Moreover, there were other strands to the early history of linguistics which contributed to its foundation - for example, the general reaction against the principles and practices of traditional grammars, which had developed in the late nineteenth century in the context of a fresh pedagogical interest in language teaching, and associated with such names as Henry Sweet (satirized by Shaw as Henry Higgins in Pygmalion), Harold Palmer, and the Danish linguist Otto Jespersen. But there is no doubt that Saussure's pioneer thinking on theoretical issues had a fundamental and lasting effect on language study - and a very specific one too, in view of the fact that his was a dominant formative influence on at least three schools of linguistics later (those of Geneva, Prague, and Copenhagen). A number of his theoretical distinctions are striking anticipations of current issues - his distinction between langue and parole, for instance, which I shall shortly discuss appears with very little difference in the competence/performance distinction of generative grammar.
But it is not possible to understand what Saussure did without seeing him in his own time, and especially against the intellectual linguistic background of the nineteenth century, against which so much of his work is a reaction. Let us, then, begin near the beginning, with an excursus into the nature of nineteenth-century 'comparative philology'.
The easiest way to identify comparative philology is by saying that it is what most people coming across linguistics for the first time expect the subject to be about - the history of language and languages, and the study of the origins and development of words and their meanings ('etymology'). In fact, as we have seen in the first part of this book, this would be a highly misleading interpretation, for the history of language comprises but a small component of the discipline as a whole. It is, moreover, a component which many people have come to disparage, in view of the melodramatic approach to much historical study in earlier centuries (in connection with the origins of language), and the pedagogical tendency to confuse matters of history with matters of current relevance in language structure. The psychological gap between linguistics and philology has indeed been very great - and it still is, in some parts of the world, particularly on the continent of Europe. Linguists would get very emotional if they were referred to as philologists by mistake; and many philologists would look rather pityingly at the new upstart discipline which they would feel lacked the decades of painstaking textual analysis on which their approach was based. These days, however, there are many signs that the old opposition between the two fields is coming to an end. From the point of view of linguistics, at any rate, it is beginning to be realized that any opposition was due more to the use of different procedures in the analysis of data than to any radical difference of opinion as to the intrinsic interest of historical vs non-historical data. Nowadays, the problems which historical linguistics raises and the facts which its methods bring to light are seen as highly relevant to the development of linguistic theory as a whole. It has been recognized that a linguistic theory will be of very limited value unless it can provide an account of the mechanisms underlying language change - either as seen in the individual (as when a child learns a language, this being sometimes referred to as 'linguistic ontogeny'), or in the community as a whole (as when a language changes from one distinct form into another, e.g. Latin becoming French - 'linguistic phylogeny'). And with an increasing number of linguists becoming interested in historical matters and using modern techniques for their analysis, it is likely that an integrated approach to historical phenomena within linguistics will not be long in being formalized. There is, however, considerably less ecumenical spirit in many schools of traditional philology, and any attempt to identify philology with linguistics would still be premature. Even with the same subject-matter, and the use of similar techniques, it will be a long time before the pejorative connotations the two labels have acquired are eliminated. Accordingly, it is still prudent even these days to keep the terms 'historical linguist' and 'philologist' distinct, the former referring to someone trained in linguistics who is applying this knowledge to the study of the older states of language, the latter to a follower of the older, nineteenth-century traditions of study.
The contribution of the nineteenth century towards the development of a scientific approach to language cannot be underestimated, even though the preoccupation throughout this period was almost totally historical. Earlier study of language history, as we have seen, was largely haphazard and vague. There was little objective, systematic analysis of the similarities and differences between language forms, or of the chronological changes in a language. If similarities were noted, it was often to dismiss them as coincidental; differences were dismissed as unimportant, or reinterpreted to suit the presuppositions of a particular (e.g. original language) theory. If the changing nature of language was considered at all, it was as part of a natural process of corruption, measured against the changeless status of Latin. Above all, no one, with the possible exception of some early Jewish scholars, had noticed anything systematic about either resemblances or differences. The first to point out objectively the fact of a systematic language similarity was a French Jesuit missionary named Coeurdoux, who showed in 1767, with many examples, that Latin and Sanskrit had definite grammatical and lexical correspondences; but his suggestion was not published until much later, and by that time, Sir William Jones had said the same thing more emphatically, and included Greek and Celtic in his observations. He had had an opportunity of studying Sanskrit in detail while Chief Justice in Bengal, and in a speech to the Asiatic Society in February 1786, largely to do with matters other than language, he made a statement which was to inspire the basic principle of comparative linguistics:
The Sanskrit language, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger affinity, both in the roots of verbs and in the forms of grammar, than could possibly have been produced by accident; so strong, indeed, that no philologer could examine them all three, without believing them to have sprung from some common source, which, perhaps, no longer exists. (My italics.)
Even though this was unsupported in detail, Jones's impressions were in print, and thus circulated widely. Within the following thirty years, the effects of the stimulus became apparent, and the reverberations of the theory took over a century to settle.