The question of how to determine a learner's proficiency in a language he or she is currently learning has been a subject of research for many years now. One measure that is widely used in studies on first language acquisition is the mean Length of Utterance (MLU), introduced by Brown (1973) in his pivotal work on the development of English in three children aged 18 to 44 months. However, this index cannot be applied to second language investigations, not only because the learner has already passed through the early phases of learning his or her first language2 and thus has a prior, more or less conscious knowledge of a language system, but also due to the fact that the skills the learner has acquired in the first language open up space for exploring the world and gaining new experiences. A journey into the world of another, new language also entails entering into a wealth of new systems: the language system, the learning environment, or a new culture – and all this at a time when the learner is already equipped with his or her first-language background and prior experiences. Therefore, by the 1970s L2-researchers were already calling for another, objective (Hakuta, 1975; Larsen-Freeman, 1978) developmental index for measuring second language proficiency.

Around a half century ago Lado proposed a “skills-and-element” model of second language proficiency in which three elements of language knowledge (phonology, structure, lexicon) could be assessed separately in the context of four language skills: listening, reading, writing, and speaking (Lado, 1961). The model was expanded and fine-tuned by Carroll, who claimed that even more elements should be measured within the framework of skills, i.e. phonology and orthography, morphology and syntax, and lexis (Carroll, 1968). About a decade later, Canale and Swain called for another, multi-component model of second language proficiency, with the focus on communicative competence, including grammatical, sociolinguistic, discourse and strategic competence (Canale & Swain, 1980). Ten years later Bachman (1990) pointed out that the earlier skills and competence models should be further enriched through research on how language is used to achieve communicative goals. Furthermore, he postulated that language use needs to be viewed as a dynamic process.