Some Blind Spots
There is no doubt, if one surveys the history of applied linguistic endeavor, that sophisticated solutions have emerged from its interventions into language problems. There are, however, also areas of neglect within applied linguistics. Most notable in this regard, perhaps, is the failure to develop a generally accepted way of assessing itself, i.e. evaluating the success of applied linguistic solutions. Because there are so many variables that cannot be controlled, the consistent and credible assessment of the effectiveness of different teaching techniques, for example, still confounds experts in the field. What is more, the early and promising debate on language program evaluation (cf. Alderson 1984, 1992; Alderson and Baretta 1992; Baretta 1986, 1990; Baretta and Davies 1985; Kroes 1991a, b; Lynch 1996; Mackay 1994; Rea-Dickens and Germaine 1992; Weir and Roberts 1994) has given way to a para- digmatically varied investigation of components of applied linguistic designs, as is evident from the facets of applied linguistic artefacts regarded as researchable in later work, as those mentioned by Paltridge and Phakiti (2010). Such paradigmati- cally inspired variation is, of course, not wholly negative: it emphasizes that applied linguistic work is never neutral, and can be undertaken from many angles and points of view. As the discussion below of the systematic themes or styles of doing applied linguistic work during its short disciplinary history (Chap. 8) will also show, there is great variation in how we arrive at responsible applied linguistic designs. But the lesser attention after 1990 to ambitious language intervention program evaluation is without doubt ascribable to the postmodernist redirection of applied linguistics at about that time.
In the early history of applied linguistics, there was a realization that to achieve status as a discipline, this area, viz. the validation of the solution to teaching problems by the development of measurement techniques, must receive attention (Corder 1972; Qvistgaard et al. 1972). It must also, preferably, be an evaluation of the proposed design that can be done empirically in the classroom itself. In another applied linguistic sub-discipline, that of language testing, this has of course been a central problem: the validation of tests has been uppermost in the minds of all serious designers of such assessments since the earliest work in language assessment (cf. Weideman 2011, 2012 for a conceptual overview). Remarkably, course designers have not foregrounded the validity of their designed interventions to the same degree as test developers.
From the angle of the disciplinary inputs into applied linguistic work, though, the most neglected area is to be found in the lack of pedagogical information. With a few exceptions of recent date, there is scant knowledge within the field of pedagogical trends, specifically as these influence classroom didactics. The only possible exception here is where the designs are supported by or delivered through new technological means, as is evident in the renewed interest in computer-assisted language learning and assessment, now that electronic means of instructional delivery have become more readily available. Put differently: most applied linguists today would be open to information from new insights in linguistics, or may be able to relate psychological research (specifically as regards learning theory) to their work, but very few would be able to make explicit the pedagogical assumptions that underlie their efforts. The pedagogical naivety underlying even some quite recent work (e.g. Lea 2004) in the teaching of the ability to handle academic discourse that aims to address exactly this angle is a continuing sign of this inability. Those who are able to articulate coherently the pedagogical information underlying their work would mostly have such assumptions through the mediation of learning theory, which is more conventionally understood as a branch of psychology.
Though we have considered some exceptions above (Chap. 3, in the section on “Applied linguistics as pedagogical engineering”), pedagogy is not usually conceived, within applied linguistics, as an independent discipline that has its own unique input to make into the inter-disciplinary matrix of the field. As far as learning theory itself is concerned, investigations have not been limited to how second or additional languages are acquired in general terms, but have also attended to individual learning strategies, varying with age, motivation, attitude and so forth (cf. Roulet 1975: 80; Clement et al. 1977; Hosenfeld 1974,1976; Wesche 1979; Wenden and Rubin 1987; Horwitz 1987; Wenden 1991; Cohen 1998; Lepota and Weideman 2002).
One last area of neglect that must be mentioned here is that of mother tongue or first language instruction. Although this problem has been addressed by language planners and scholars working in bilingual education (since, very often, “mother tongue” instruction is, as Rosen 1982: 49, remarks, simply a euphemism for the teaching of the standard, national language, and not the actual home dialect of the student) there has been very little innovation inspired by applied linguistics in this regard. In South Africa, specifically, one is often asked, when discussing the didactic techniques and implications of the mainly communicative syllabi now adopted nationally for English and Afrikaans (as additional languages), whether such techniques have nothing to offer for first language instruction. They do of course, but there has been little to offer, by way of a designed solution, from the side of applied linguistics, since first language instruction is not perceived as a problem, specifically a problem of a disadvantaged individual or group. Neither can there justifiably be any applied linguistic intervention before this kind of instruction is perceived as a problem in its own right by the teachers and educators concerned.
Not until home language instruction becomes a problem, therefore, as is currently the case in South Africa, is an applied linguistic intervention possible. In a report to the South African Council for Quality Assurance in General and Further Education and Training (Umalusi), Du Plessis et al. (2013) outline a possible construct for the assessment of home languages, in order to lay the foundations for a plan that is aimed to make the examinations across the 11 home languages offered at secondary school level equivalent. The report was commissioned by Umalusi after they had become increasingly concerned about the lack of equivalence among these examinations, a lack that is all the more concerning because these are high stakes assessments: on the basis of their results, decisions are taken to allow students entry into higher education. In their report, Du Plessis et al. (2013) outline how a blueprint to use as a design basis for examinations might be derived from the common curriculum, and how the curriculum relies on a number of sociolinguistic starting points for its theoretical justification .
If applied linguistics finds its limits in the nature of the concrete problem that is being addressed, as has been argued here, there is, finally, some explanation for the views of the field outlined above as the ‘intra-disciplinary’ and ‘inter-disciplinary’ perspectives. Both of these, as has been pointed out, see some kind of conceptual continuity between linguistic, applied linguistic and other activity. Such views may be taking their cue from the fact that linguistics itself progresses from a consideration of elementary concepts to elementary ideas (Weideman 2009a), so that these ideas form a bridge to an investigation of complex linguistic concepts. The complex linguistic concepts in turn link linguistic theory to the factual complexity of language acquisition, the notion of language ‘authority’ and so forth. This progression may explain how, for lack of a foundational perspective that honors the uniqueness of theoretical, technical and a-theoretical activities, applied linguistic concerns with concrete problems come to be confused with the analytically qualified theoretical investigations that support them.
It is to rule out this confusion that Santema (1978) distinguishes between models of knowledge and models of manufacture. The latter have as their function to serve as technical blueprints in the execution of a task, the former are analytically qualified constructs that enable one to identify the crucial factors of a phenomenon, state, event, process, or factual arrangement. They also, of course, serve as the theoretical justification or defensibility of the designs for manufacture that are imaginatively conceived. The same distinction is implied by Corder (1972: 4) when he sees the task of the ‘pure’ scientist as “improving the picture we have of the world”, while the ‘applied’ scientist sets out “to improve the efficiency of some practical task”.