pressing me to address the matter of practice creates a position somewhat akin to that taken by some contemporary scholars—e.g., Rampton—to the effect that “all is practice”; that is, there is no theory, no reason, but rather only lots of reasons, uncoordinated, disjointed, the extreme of postmodernism In such a solution, there is no role for linguists or applied linguists, and the role of educators consists of dealing with a disconnected group of individual teachers working in some vague sense of collaboration You see, if it is not possible to trace the development of the field both synchronously and diachronically then there is nothing left to talk about; all that remains is a disjointed kind of practice unhinged from the sort of structure that supports all social science and leaves odd bits that can exist only within teacher education THE AUTHOR Robert B Kaplan, emeritus professor of applied linguistics at the University of Southern California, Los Angeles, United States, was the founding editor-in-chief of the Annual Review of Applied Linguistics, a member of the editorial board of the Oxford International Encyclopedia of Linguistics (1992, 2002), and editor of the Oxford Handbook of Applied Linguistics He has served as president of NAFSA, TESOL, and AAAL TESOL, Applied Linguistics, and the Butterfly Effect ALAN DAVIES University of Edinburgh Edinburgh, Scotland What interests me in this discussion, initiated by Alister Cumming, is the use of the term theory Let me start with the butterfly effect This term is often attributed to Ray Bradbury in his science fiction story “A Sound of Thunder” (Bradbury, 2005, pp 203–215), which first appeared in 1952, but the idea goes back to Duhem (1906/1954) It was taken up by Edward Lorenz (1972) in his talk to the American Association for the Advancement of Science, which was given the title: “Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas?” Is this a theoretical question or, rather, does it make a theoretical claim? Obviously not in the strict sense because there is no way of reversing time so as to experiment with and without the flap of the butterfly’s wings At the same time, in a different—wider—sense of theory, 296 TESOL QUARTERLY the butterfly does answer to our profound desire to understand and attempt to explain the relation between cause and effect, ourselves and our world And the butterfly is but a colourful illustration of the very serious—and scientific—field known as chaos theory, which uses computer simulation to investigate the notion of sensitive dependence in initial conditions There seem to be two main uses of the term theory The first is the scientific use: This use requires a testable model which, following Popper (1934/1959), is falsifiable Empirical observation and evidence are emphasised, the purpose being to generalise about natural phenomena so as to predict future behaviour of those phenomena The second use of theory is the narrative sense in which a story is told What narrative does is help us make sense of our world Narrative, writes Bruner (1990), is “one of the most ubiquitous and powerful discourse forms in human communication” (p 77) It has had an impact on philosophy, literature, cultural studies, anthropology, and the social sciences Just as in the scientific category there are harder (for example, physics) and softer (the butterfly effect) investigations, so in the narrative use of theory there are harder and softer studies A good example of the harder is ethnomethodology, Garfinkel’s (1967) “tell-a-story-aboutable, analyzable—in short, accountable” (p 33) An example of the softer use of theory in narrative is the theory of creationism and its partner intelligent design If theory in science is modernist, appealing to universal truth, theory in narrative is postmodern Scientific theory is, paradoxically, weak, in the sense that it is prepared to be disproved, can be falsified by new evidence Narrative theory of the softer kind is strong: For example, intelligent design is religious, a matter of belief and by definition not falsifiable But there is a hard edge too to narrative which allows for evaluation for good reasons through narrative fidelity (the external criterion of audience judgement) and narrative possibility (the coherence of the story being told) The question we are concerned with is the place of theory in TESOL To me it is the wrong question A more pertinent question would be to enquire about the place of theory in applied linguistics Applied linguistics was established institutionally in the 1940s and 1950s, first in Michigan, then in Edinburgh (Davies, 2007), later expanding widely with national associations, such as the American Association of Applied Linguistics (AAAL), the British Association of Applied Linguistics (BAAL), and the international umbrella organisation Association Internationale de Linguistique Appliquée (AILA), as the theoretical cover for TESOL But if TESOL after World War II was the trigger, the occasion which made SYMPOSIUM: THEORY IN TESOL 297 applied linguistics necessary, it was never intended to be the sole concern of applied linguistics True, TESOL is still the largest stakeholder in applied linguistics: Scratch an applied linguist and you are likely to find someone who is now—or has in the past been—involved in TESOL But the intention, as the work of the Center for Applied Linguistics (CAL) illustrates, was always to relate to the learning and teaching (and increasingly use) of all languages I say all languages rather than language in order to differentiate the interest of linguistics, which is in language, from that of applied linguistics, which is in languages There has, of course, been severe resistance, especially by long-established professionals in languages other than English, in French, German, Spanish, and other languages, resistance both to the claims (and ambitions) of applied linguistics to be relevant across languages and to the perceived hegemony of English That suggests, perhaps, that had applied linguistics been fuelled up not by English but by, say, Swedish, there would have been less resistance Maybe, but I doubt it Entrenched interests take no prisoners In terms of my representation of the two extreme uses of theory— science and narrative—applied linguistics falls toward the narrative end It certainly tells a story and, like ethnomethodology, is prepared in its curricula and its assessments and its planning, to be accountable And it predicts: If this, then In doing so, it relies on borrowed theories, of learning, of pedagogy, and of social behaviour What applied linguistics adds is both the bringing together of these models and an experimental, research-based approach which proceeds by moving from problem and practice to theory, that is, by theorising practice Is TESOL therefore only (part of) applied linguistics? Yes But is it any different from other second language teaching operations? Again, yes Although TESOL does indeed depend on the same theories as other language teaching branches of education, it is also different and for two reasons The first—again—is applied linguistics, because so much of the discussion and exemplification in applied linguistics is of TESOL The second is that in the early 21st century, English is just different from other languages, different because of its (no doubt temporary) spread and take-up This may be imperialist, it may be regretted, but it is nevertheless true Does it have to be English? Research appealing to chaos theory could be timely: Applied linguistics meets the butterfly effect? THE AUTHOR Alan Davies is emeritus professor of applied linguistics at the University of Edinburgh He has taught English and applied linguistics in the United Kingdom, Kenya, Nepal, Hong Kong, and Australia His more recent publication is Assessing Academic English: Testing English Proficiency 1950–1989 (Cambridge University Press, 2008) 298 TESOL QUARTERLY