Old and New Views on Language History and Language Relationships
It is regrettable that linguists, especially dialectologists and historical-comparative linguists, should have tried to force their work into static (i.e. minilectal) frameworks. For these are wholly recreant to analyses which are necessarily cross-lectal. As such, they cannot logically be made to fit into an idiolectal strait-jacket, whether the structuralist one or the transformationalist one, without becoming something other than what they are. Dialectology and comparative analysis obviously require the linguistic analogy of Bach's well-tempered instrument. And the polylectal descriptive models of the dynamic framework are precisely this. Who, having once enjoyed a well-tempered instrument, could wish to return to one in which only tunes in one key (idiolect) can be played? Scholars in historical linguistics ought to seek their inspiration further in the past--in J. Schmidt and H. Schuchardt. We should be looking for models to implement the insights of those scholars, not aping single-key instruments totally unsuited to our needs.1
I maintain that, so far from its being feasible for historical investigations to employ static descriptive models, what is needed is for descriptive linguistics to adopt dynamic, or time-based, models of change. Trying to do linguistics without a satisfactory understanding of the history of the language under analysis is like trying to fly a dirigible with gasoline in the engine but no gas in the bag: The thing is started and all sounds fine; only she will not get off the ground. In addition to the study of languages, there is the study of the history of theories. But this study can never approach in importance the study of the history of languages and should never be permitted to replace it in linguistic curriculums.
Current interest in the theoretical importance of language-hybridization has led some historical linguists to a revived interest in the work of those few nineteenth-century figures who appreciated the relevance of mixture and gradience____________________