Methodology in Cognitive Linguistics
What needs to be done
The argument developed in the previous pages may be summarized in three steps. First, a growing increase in the empirical demands on linguistics is to be expected: corpus research, experimental designs, surveys, and the accompanying forms of quantitative data analysis are likely to occupy a more important position in linguistics than they have so far done. Even if they will not penetrate massively in linguistics in the short run, they will probably determine what is methodological cutting-edge research in languages studies. Second, the theoretical make-up of Cognitive Linguistics provides an excellent basis for participating in this empirical revolution: its theoretical principles are generally advantageous for the incorporation of empirical methods, even though the primary interest of Cognitive Linguistics in semantics might favor an opposite methodological orientation. Third, the actual situation in Cognitive Linguistics is a mixed one, with a clearly growing interest in empirical methods on the one hand, but a lingering dominance of traditionally analytic approaches on the other.
What needs to be done, then? The primary step will be for Cognitive Linguistics to make its mind up: in which direction is it going to develop? Will it wholeheartedly embrace the empirical turn, or will it stay with a predominantly traditional approach? Assuming that only the former option is fruitful in the long run, a number of recommendations for the future development of Cognitive Linguistics may be spelled out.
First, if it is correct that the spread of empirical methods is hampered by insufficient knowledge of approaches and techniques, didactic efforts are necessary to make cognitive linguists acquainted with corpus analysis, experimental design, survey techniques, and statistical processing of data. A program in Cognitive Linguistics worthy of that name should resolutely opt for advanced training in empirical linguistics. In the process, it will have to be made clear what the essence of empirical research really is, so that the misconceptions that now sometimes mar the debate may be avoided. No, empirical research does not involve abandoning theory formation in favor of purely descriptive research: rather, it involves trying to provide proof for theories, and from there, refining the theories. No, empirical research does not imply that intuition and interpretation have no role to play in linguistic research: rather, it implies that interpretation is but one step in the empirical cycle of successful research. And no, empirical research is not about restricting the investigation to one kind of method or technique: rather, it is about using experimental, and corpus-based, and other empirical approaches in combination to achieve maximally reliable results.
Second, the standards of scholarship in Cognitive Linguistics will gradually have to be raised to such a level that unsubstantiated claims will become unfashionable. This is not to say that the bold and inspired analytical and theoretical work that sparked off Cognitive Linguistics to begin with, should be ruled out: that would be a counterproductive check on the creativity of the community of cognitive linguists (and if there is one thing you cannot deny them, it’s creativity). Nor is this to say that the desired level should be achieved in one go and simultaneously by all: what we are talking about is a medium-term development that leaves room for different paces of development, just like it is in theory development. But even if the speed with which Cognitive Linguistics will get there may be moderate and variable according to the circumstances, the direction of the development should be clear, and all those who have responsible positions in Cognitive Linguistics– leading authors, editors of journals and books, referees, conference organizers– should consciously and consistently pay more attention to empirical grounding, to keep that development going.
And third, Cognitive Linguistics (and linguistics at large, for that matter) will have to learn how to live with a slower pace of development. Empirical thoroughness and reproducibility come at a price: that of painstaking and time-consuming effort. While the late 20th century in linguistics may have had a tendency to prefer conceptual audacity and the construction of grand theories, these are values that we associate primarily with 19th century philological accuracy and meticulousness. Perhaps, then, the early 21st century may profit from a return of some the 19th century spirit …
