2.3.3 Behavioristic CALL

Technology use tends to mirror existing practices and "whenever a new medium comes in," says McLuhan, "it takes its initial content from the old" (Sprecher, 1987, p.14). McLuhan's (1964) "rearview mirror" phenomenon helps to explain why the first uses of computers in language learning were designed on the textbook and language lab paradigm. Chiquito, Meskill and Renjilian-Burgy (1997) describe the early phase of CALL as an attempt to "transfer existing foreign language textbooks to computer-based applications. Students could then essentially use the computer to turn pages of the textbook, fill in the blanks in workbook drills, and choose multiple-choice answers to questions" (p.72). As was the case with the language lab, this phase of CALL relied on its ability to simply do more efficiently many of the teacher's classroom tasks such as drilling and patterning. According to the typology of Maddux, Johnson, and Willis (1997) this type of computer use would be characterized as a Type I application. Unlike Type II applications which aim to make available new ways of teaching, Type I applications simply aim to render existing practices more efficient.

The influence of the language lab on the initial uses of CALL included a reliance on a behavioral philosophy of learning. Thus the first phase of use of computers in language teaching (the sixties and seventies) can be referred to as Behavioristic CALL (Warschauer, 1996a) or Instructional CALL (Wyatt, 1987). Backer (1995) provides a summary of this approach to computer-assisted language learning:

Chronologically, the first trend in CALL (originally called CAI -Computer Assisted Instruction) was an electronic extension of "programed learning" or "programed instruction" (PL and PI, respectively) based on the behaviorist theories of Skinner and Bloomfield. According to these theories, all learning could be broken down into small "frames" and the learner could be drilled and evaluated in each frame until mastery. The teacher then brought the student to the next frame. In the computerized version, the progress of the student could be monitored and guided through "branching". Proficient students could automatically be sent ahead, while slower students could be routed to remedial lessons. According to Audio-Lingual and Cognitive Code methodologies, a major focus of language teaching was grammatical structures through drill and practice. Thus, the earliest attempts at computer-assisted language instruction, first appearing at some large universities in the late 1950's, stressed learning grammatical structures through electronic PI. (p.5)

This first phase was conceived in the fifties and implemented in the sixties and seventies (Warschauer, 1996a). Numerous theorists have provided typologies of the phases of and approaches to CALL. Wyatt (1987), in reference to the teaching of English as a second language, proposes a typology that distinguishes three approaches to or phases of CALL: instructional, collaborative and facilitative. According to Wyatt, Instructional CALL software corresponds to the first phase and can be identified with the Audio-Lingual Method. Thus early Instructional CALL shared many features with behaviourist teaching methodologies (Backer, 1995). Wyatt describes Instructional CALL as follows:

• Materials are presented in a highly-structured, predetermined manner.
• Repetitive language drills and practice are the main substance.
• Students are passive responders, not initiators.
• The computer functions as an authoritative instructor.
• A detailed set of high- and low-level learning objectives is provided.
• Learning paths are predetermined.
• The computer instructs the student; students learn from the computer.

Hubbard (1987) describes the behaviorist approach to CALL as one which presents vocabulary and structure appropriate to the learner's level through pattern reinforcement. It aims to maintain the learner's attention to the task and provides sufficient material for mastery and over-learning to occur. An essential premise underlying the approach is that of positive and negative reinforcement. Behavioristic CALL is designed to promote student mastery of a body of rules by indicating to the learner whether or not the language they produced matched that stored in the computer's memory (Garrett, 1987). Unlike communicative approaches which clearly downplay explicit error correction (Schulz, 1996), Behavioristic CALL relies on it. The "wrong-try-again" model thus requires the learner to input the correct answer before proceeding, provides the learner with positive feedback for correct answers and does not accept errors as the correct answer (Hubbard, 1987).

Picard and Braun's (1987) descriptors of the didactic approach echo many of the characteristics of Behavioristic CALL. Such an approach typically is teacher-centered; determines the learning path; verifies what the student has learned and proposes reinforcing exercises; transmits knowledge and corrects errors. A didactic approach placed the computer in the role of electronic drill master (Backer, 1995), computer-as-tutor (Taylor, 1980) or computer-as-magister (Higgens, 1986). In this role, the computer initiates and controls procedures and judges performance. "The computer-magister knows the truth, intervenes to guide the student toward that truth, and then judges the student's performance" (Backer, 1995, p.3). "The computer asks the questions and has the answers", automates "routine correction" thus eliminating "the arithmetic burden imposed on language teachers" boasts Hope, Taylor and Pusack's (1984) description of CALL. They describe further the computer in the role of the teacher:

Programmed imaginatively, the machine embodies the best strategies and insights of the experienced language teacher, multiplying the teacher's contacts with students for certain kinds of language practice. Good programs can offer in this way, individualized attention and can allow students to work at their own pace. Students can work in privacy without fear of reprisal or ridicule regardless of how slow they might be or how often they give incorrect answers. Immediate diagnosis saves time and frustration and helps students weed out their errors. Computers possess the quality of infinite patience. They treat each student in the same way without favoritism. They are also very consistent in their responses, regardless of how many hours they have been working. Even the best of teachers cannot show the same level of enthusiasm, interest, and energy, day in and day out. (p.16)

In spite of the perceived advantages of Behavioural CALL, its detractors grew particularly as interest grew in CLT. The focus on form rather than on meaning, sketchy and vague help or feedback, the computer as "evaluative task-master that asks all the questions and judges all the answers" and "discrete points of grammar or vocabulary, mostly out of context and devoid of any real meaning": these are some of the criticisms cited as the defects of Behavioristic CALL (Underwood, 1984). Backer (1995) notes that a further problem with CALL was its underlying "assumption that one student would work at each computerized work station". This assumption required "computer hardware well beyond the financial means of most language learning facilities" and resulted in physical and psychological isolation of students (Backer, 1995, p.6).