It is important to be aware that the comparative method was largely empirical, and (under German stimulation) thorough; it was based on textual evidence, information about speech being deduced from an examination of writing; again, it was primarily concerned with comparison of individual sounds (rather than words, or meanings); and it tried to show the systematicness behind the linguistic variation which it noted - an aim which was not always successful, in view of the fact that insufficient attention was paid to the structural character of the language stares being compared before comparative decisions were made. Rut such an ordering of priorities (non-historical description preceding historical comparison) was not seen as important until Saussure. Meanwhile, as the century progressed, techniques were clarified, principles were more precisely stated, and a scientific atmosphere became more normal.
The remainder of the century saw the accumulation of a great deal of information on the history of languages, and, in particular, on the details of Proto-Indo-European. Sanskrit was of the utmost relevance in such work, and was hailed as such by many linguists; Max Miiller, for example, in 1868, said, 'A comparative philologist without a knowledge of Sanskrit is like an astronomer without a knowledge of mathematics.' An important reflex of this detailed study, however, was an increasing theoretical linguistic interest which took the newly discovered facts about language into consideration. Scholars began to meditate on the underlying principles which the facts of sound-change and related developments suggested. In particular, there was the growth of an evolutionary attitude to language, stimulated by Darwin's work. If plants and animals have a birth, development and death, then why not language too? Early on, W. von Humboldt had emphasized the fact of linguistic flux, in an attempt to explain the phenomenon in terms of the changing mental power of the users of language; and certain aspects of his thinking have been commended by Chomsky as striking anticipations of important features of current linguistic theory. August Schleicher first tried to develop the theory systematically, making Hegelian philosophy take account of the Darwinian theory of natural selection; to him, the typological classification of languages as isolating, agglutinative and inflectional was an example of a Hegelian triad of thesis-antithesis-synthesis. Synthesis being the climax of development, the standard of excellence in language was thus tied to the amount of inflection it possessed: Latin, Greek, and Sanskrit were indubitably best on this count, Chinese least of all, and all languages since Proto-Indo-European were in the stages of a slow decay, as the comparative method showed quite clearly that this parent language (Ursprache) was probably more inflected than any of the attested languages.
A fundamental objection to this approach, emphasized by many scholars, is that language has little in common with an organism, such as a plant. It has no separate physical existence, therefore it has no separate life or death. Language change resides primarily in the users of the language, and only indirectly, via these speakers, can language be seen as an abstracted whole. Language is but one aspect of an organism's behaviour, an activity which is continually changing; it is no more than a set of useful conventions. There is a further, more specific objection, that if languages like French and English are biologically distinct, on different 'branches' of the family tree, then once they have split up, how could the one influence the other in any direct way? Yet this has often been the case, as is shown by the number of words borrowed from French in recent centuries by the English. But despite these objections, such theories had a great influence on the development of linguistics during this part of the nineteenth century. The emphasis till then had been on philology in its more widely accepted sense, i.e. language study as an end to understanding a nation's culture (in particular, its literature). But with the stimulus of natural science, linguistics came to be studied as a more autonomous discipline, with the suggested status of a physical science. It began to be studied for its own sake. Otto Jespersen, in his book Language: Its Nature, Development and Origin, published in 1922, calls this the 'emancipation' of the subject. It was supported at the linguistic level by further developments among contemporary scholars.
The work of Jacob Grimm and others had already produced a more 'mechanical' outlook on linguistic data, with more and more sound-Maws' being formulated. But there was still a large amount of material that could not be accounted for, and sound-changes which seemed to be exceptions to the otherwise readily perceivable patterns of development. But when Karl Verner proved in 1875 that one set of unsatisfactorily explained sound-changes could be shown to fit a regular pattern by formulating a new phonetic principle hitherto ignored, a new attitude in linguistic scholarship became apparent; it was supported in other publications appearing at the same time (by Saussure, for example) that showed the relationship between Sanskrit and Indo-European more clearly. Certain scholars thus began to assume that all exceptions were explicable in the same way, that is, that they only remained exceptions because insufficient study had been made of the material to determine the underlying principles of development which could be formulated as laws. Sound-changes were not haphazard, it seemed: a comprehensive, objective examination of the data, paying careful attention to the mutual influence exerted by sounds, could produce a satisfactory explanation of a regularity behind all sound changes. 'Sound laws have no exceptions' became a canon of the new attitude, held by men who were called by their older contemporaries, a trifle sarcastically, 'neogrammarians' (Junggrammatiker). Of major importance in their doctrine was the concept of analogy (cf. p. 70), as this was seen as the linguistic force which tended to normalize differences in language.
This approach thus focused attention on the physical side of language; but its methodological rigidity naturally evoked some heated criticism. It was too mechanistic an approach, it was said, which left the human being out. Language had two sides, not one: there was form, but there was also function (or usage), and this social (or pragmatic) province provided an indispensable perspective for language study. But this criticism the neogrammarians largely ignored.
The criticisms were largely valid; the social basis of language had yet to be thoroughly expounded. However, the result of the movement was to inject a greater scientific precision and awareness into linguistics, and this supported the tendency to see the subject as a kind of natural science. The perspective was still evolutionary – all explanations continued to be historical in the following years - but there was a more rational, empirical approach to language, especially in its contemporary, living forms, which was first developed in theoretical detail in the work of Saussure. The old, fanciful, vague theorizing was gone; reliable major work, synthesizing and codifying the results of widespread scholarship, was becoming available. The comparative method had been proved to be of great use in historical linguistics, and a number of important points had been raised and clarified.
For example, such philological procedures firmly dissolved all the old theories that one of the spoken languages of the world was the oldest. Moreover, it caused further confusion in the
anthropological camp among those who maintained that whichever language it was that Adam and Eve spoke in Eden, it was sure to have been a simple language; for the comparative method indicated that the further back one went in reconstruction, the more complex the inflections of language appeared to be. Indo-European was much more inflected than either Greek or Sanskrit; and there was no evidence that Indo-European was anywhere near the starting-point of mankind's language. There was a geographic coincidence between the linguistic judgement and the historical, in so far as both sets of evidence pointed to a place of origin for civilization to the north of the Indian sub-continent, but it was all very hypothetical, and how long a variety of Indo-European was being spoken in that place was indeterminable.