|
|
|
|
|
|
|
|
WITH the development of affordable computer hardware and software in every field, the potential benefits of computer-aided instruction (CAI) in foreign languages and literatures now seem limited only by our imaginations. Although any new pedagogical technique can expect to meet some resistance within the profession, it is now clear that the integration of CAI materials into foreign language curricula has become indispensable if not inevitable.
CAI adapts itself particularly well to foreign language study because of its potential to address all aspects of the discipline at all levels of instruction. In addition to facilitating elementary instruction, CAI can enrich the advanced student's learning experience through programs on geography and culture, on the syntactical problems of writing in a foreign language, and on the tools of literary criticism. CAI enables students to work at a level commensurate with their educational and instructional needs. CAI programs also allow teachers to make the best use of their classroom hours, because it releases them from time-consuming repetition and drill exercises. Programs that provide immediate feedback and explanation free up valuable classroom time for the more creative and motivating features of foreign language learning. Easy access to interesting and stimulating programs in the more inviting areas of the language arts can produce more highly motivated and intellectually active students.
In 1985 funds were put aside at the University of New Hampshire to support and encourage the systemwide use of microcomputers in the educational process. The project, named DISCovery and headed by Betty le Compagnon, set out to establish academic computing as an integral part of all scholarship at UNH. On the recommendation of a joint faculty-administration committee, DISCovery funds were divided into two portions. One funded the building and staffing of three microcomputer clusters that would be open to students and faculty members for general academic computing, CAI assignments, and manuscript preparation. The clusters housed approximately one hundred AT&T 6300 (IBM-compatible) and Apple Macintosh Plus microcomputers. The remaining funds supported faculty activity in the development of CAI programs for specific courses. A call for proposals went out to the faculty in December 1985 requesting project descriptions, which could include various combinations of hardware, software, programming assistance, and faculty stipends.
The project funded for CAI in foreign languages was developed by a committee of ten foreign language faculty members working in five different languages. Three areas of inquiry interested us in particular: (1) how the unique educational characteristics of the computer could be used most effectively to facilitate drill and practice; (2) how a more positive experience could be created for the first- and second-year language student faced with learning materials best taught through drill and practice; and (3) how new educational strategies should be developed to enhance our second-year curriculum.
With these three objectives in mind, we divided our proposal into two parts: phase 1, an experimental phase during which we set out to determine what software the publishing industry had to offer at the elementary level; and phase 2, a developmental phase in which we proposed to write our own intermediate-level software. By beginning with a pilot program at the elementary level, we hoped to generate essential feedback on those practical and instructional difficulties that would need to be addressed in the construction of a more ambitious CAI program for the intermediate level. These intermediate courses carry our highest enrollments and function as the primary seed courses for foreign language majors.
Our short-term objectives in phase 1 included the institution of CAI programs in as many languages as possible for the academic year 1986–87. We were able to introduce programs in French, German, Spanish, Russian, and Latin, five of the nine languages then taught at UNH. Each pilot program initially used commercially prepared software. Though the software functioned differently in each language, it was always integrated into an overall program of classroom instruction and language-laboratory assignments in an effort to provide a flexible and individualized elementary curriculum. The use of centrally located hardware in the university's computer clusters provided easy access for the students and ensured wide participation by the various language departments. The use of commercially prepared software allowed us to discover and deal with the many practical problems inherent in establishing a new educational strategy without the added difficulties involved in actually creating programs. It also ensured that we could have our programs in place for the fall 1986 semester.
Phase 1 was designed to help us gain expertise and experience and overcome our admitted reticence in undertaking the tasks outlined for phase 2. Our knowledge of computers generally extended only to word processing and a little experimentation with data bases. The prospect of creating in-house materials intimidated most of us. We hoped that a careful evaluation of student and faculty responses to commercially prepared elementary software would provide important insights into the design and development of intermediate programs that would enhance, not duplicate, the existing intermediate curriculum. Our aim was to attract a larger number of serious language students from which majors and minors could be drawn.
With these goals in mind, we began surveying commercial software and were immediately struck by the meager offerings of the publishing industry. Our university clusters contained IBM-compatible AT&T and Macintosh Plus computers; publishers wanted to sell us Apple II software. After a frustrating summer of trying to locate satisfactory products, we decided that each department would make independent curricular decisions about software based on departmental policy and the availability of programs in each field.
In designing elementary French, faculty members chose a direct-method textbook and a generic software program called CLEF . We could not consider the integrated first-year CAI programs, because they all used Apple II software. CLEF consists of fifteen disks with thirty sets of exercises on the basic elements of French grammar, such as definite articles and -er verbs. CLEF, however, fails to cover many essential grammar points. Similarly, there was only one elementary Latin CAI program, Latin Flash Drill (Latovsek), that met curricular needs and was compatible with our hardware. Because of the inadequacies of existing software, we began to give serious consideration to authoring our own.
The use of CAI promised to be especially valuable in a highly inflected language like Russian. The recent introduction of programmable non-Roman character sets opened up many new possibilities. Instructional programs had already been developed at Bryn Mawr for use on IBM-compatible computers: one correlated Cyrillic letters and the basic sounds of the Russian language, and the other dealt with the Russian verb system ( BMCAI ). Since Bryn Mawr planned to convert to the Macintosh format in the fall of 1986 and both programs were in the public domain, they appeared to be ideal candidates for our Russian CAI efforts. The basic-sounds program promised to be particularly helpful to beginning elementary students trying to master the correlation between the Cyrillic alphabet and the Russian sound system. These concepts also function as important building blocks for learning the Russian nominal and verbal systems since they give insight into the rules underlying the existing surface forms. Both these Russian programs would have applications not only for initial learning but also for reinforcement and review at the upper levels.
Along with the advantages of CAI for Russian language learning, however, came a set of problems. Although the Bryn Mawr software could produce Cyrillic characters on an IBM computer, our AT&T computers would all require the installation of a special logic chip to allow for the on-screen production of Russian characters. While the chips were not expensive, they did have the disadvantage of scrambling important graphic characters in any programs not using Russian. This severely limited the number of machines to which we could give our students access. To make matters worse, the projected Macintosh conversion scheduled for 1986 never materialized, and we found ourselves with two excellent programs that we could run on only a small number of machines. Since even fewer commercial programs were available in Russian than in the Romance languages, the prospect of in-house authoring became more and more attractive.
Our German Department was able to choose Sevin, Sevin, and Bean's Wie Geht's , which is accompanied by a fully integrated, text-specific computer program. This software presented a careful, cross-referenced version of all the materials covered in the text, complete with scoring capabilities, page references for error checking, and a variety of types of exercises for the chapters. However, while it added greatly to the practice and review capabilities of our German students, this program limited computer review to the concepts and vocabulary presented in the Wie Geht's text. Moreover, some troublesome bugs in the program disconcerted beginning students. Our German faculty members wanted a more flexible approach. They wanted to be able to focus on specific grammar problems as they became apparent in the classroom and to be able to introduce and drill new vocabulary. Again, authoring our own programs seemed to be the only alternative.
After a frustrating investigation of textbooks and software, our Spanish faculty determined that their most feasible course lay in generic software. They purchased three programs: one on vocabulary drill ( Matchmaker: Spanish Vocabulary Drills ) for use in all Spanish 1 courses, another on stem-changing verbs and present-tense verb conjugations ( Radical Changing Verbs ), and the last on pronouns ( Direct and Indirect Object Pronouns ). These generally well-written programs immersed the student in an entirely Spanish environment, providing comprehension drills based on running dialogues. Unfortunately, the program on vocabulary drills was aimed primarily at the high school level, and the other two programs proved too advanced for first-year students, even at the college level. Our Spanish faculty decided that CAl could not become an integral part of their curriculum unless more flexible and appropriate alternatives were found.
Our efforts in exploring the commercial software market, then, were plagued from the very beginning by difficulties—some deriving from decisions made by the publishing industry and some generated by our own lack of expertise in a rapidly developing field. As we struggled to establish phase 1 of our project, it became increasingly apparent that the products offered commercially did not meet our needs. In their first column reviewing CAl in Foreign Language Annals, Nina Garrett and Robert S. Hart warned, “Teachers who expect to purchase one neatly integrated package to fill all their courseware needs are likely to be disappointed” (59). We found that even a combination of commercially produced programs fell far short of fulfilling our requirements.
At the same time, our students were demanding more and more CAI programs, encouraging us to find viable alternatives to commercial software. They enjoyed using the computers and found the immediate feedback in drill and practice extremely helpful. Although we had agreed from the start that the complex variables present in our project would render it immune to actual statistical analysis, questionnaires that were circulated to students at the end of the first semester revealed that the response to CAl in foreign languages was overwhelmingly positive. Despite the problems that had been encountered, most students felt that the programs had improved their skills and enhanced their learning experience. In fact, both faculty members and students wanted to broaden the scope of the project. Initially, only two sections, on average, of each course had been targeted for participation in the CAl project. These limitations reflected practical considerations in the newly operative clusters as well as a desire to maintain an informal control group for evaluation purposes. Our students' questionnaires came back complete with suggestions for new types of exercises and more course-specific materials, and faculty members who were not part of the original committee began showing an active interest in participating in the project.
The prospect of authoring programs still intimidated most of us, although we now had a highly motivated faculty and an enthusiastic student population. We looked for a vehicle that would carry us through the complex process of authoring, and we found it in an authoring system called MacLang , developed at Harvard University by Judith Frommer.
MacLang is an authoring program specifically designed for foreign languages, although it has broad applications in almost any discipline. It contains features like nonstandard character sets, on-screen display of accents, as well as many other elements essential to language instruction. MacLang runs on the Apple Macintosh or Mac Plus computers, both extremely user-friendly and thereby attractive to both students and faculty members. Some project participants already had Macs of their own, and their previous experience with this particular computer proved beneficial to us all.
In November 1986, Frommer introduced MacLang to a group of interested language faculty members at a special CAI workshop. This group far exceeded the original ten project participants. We immediately realized that MacLang authoring could potentially provide us with the tools we needed to create effective CAI programs for our students. It seemed an accessible and unintimidating means of producing highly focused, course-specific software carefully tailored to our curricular needs. Moreover, MacLang was available for a nominal fee from Harvard University.
MacLang is an extremely flexible program. The current version 3.0 can produce drills using one of five formats: vocabulary, fill-ins, paragraphs, multiple choice, and jumbles. Vocabulary exercises require translation in and out of the target language. Multiple choice or fill-ins provide a substitute method for vocabulary drills in direct-method courses as well as a means of drilling a wide variety of grammar points. Paragraph exercises allow for the insertion of multiple blanks in a single sentence for the drilling of compound structures. “Jumbles” presents the student with scrambled sentences appropriate for practicing word order.
Creating programs on MacLang follows a simple menu-driven procedure. The instructor makes initial choices regarding exercise format and language. Choosing the target language for the exercise automatically loads the special characters needed for that language into the computer. The instructor then follows four simple steps: writing an introduction, choosing “System Error Messages,” adding specific “Standard Error Messages,” and composing the questions and answers. The introduction presents the directions for the exercise with models as needed, along with any reminders or explanations to which the student can then refer at any time during the exercise. “System Error Messages” are general messages that the student receives after answering any given question. In French, for example, typical “System Error Messages” include “Bravo” (for a correct answer), “Recommencez,” “Ce n'est pas tout à fait juste,” and “La bonne réponse est … ,” which would appear after a predetermined number of attempts specified by the instructor in the authoring process. “Standard Error Messages” are messages specific to frequent errors in a given exercise. The instructor can number these messages and enter them along with the actual questions and answers if they are repetitive or add them individually with each question if they are not.
MacLang's authoring process itself consists of three parts: writing the questions, entering the correct answers, and entering close answers or anticipated errors. In a fill-in exercise on relative pronouns in German, a typical question might be: “Der Mantel, ———ich kaufen möchte, ist sehr schick.” Here the correct answer is “den.” A typical close answer would be “der.” An appropriate explanation might be: “The case of the relative pronoun is determined by the pronoun's function in its own clause.” If the error corresponds to a “Standard Error Message” entered earlier, typing the number of the appropriate error message calls the entire message automatically to the screen when the student responds with that particular anticipated error. In addition, an appropriate “System Error Message” also appears. For an unanticipated error, only the generic or “System Error Message” will redirect the student to the correct response. After typing in all questions and correct and close answers, the author chooses how many attempts the student will have at each question, in what order the questions will be presented (including random order), and whether individual scoring is desired. The latter option automatically records all student errors so that the instructor can retrieve these data and add them to the anticipated-errors file to provide the widest possible range of close answers. After completing the actual questions and answers, the instructor then transfers the exercise to a student disk and sends it to the computer cluster for student use.
At UNH, clusters function on a library basis. The student checks out disks in the cluster and completes exercises as assigned by the instructor. In MacLang, the student moves immediately to an opening menu that offers a choice of exercises. The introduction for the exercise appears automatically on the monitor and starts up the exercise. The student then proceeds through the questions, typing responses. A wrong answer produces either a “System Error Message” and a “Standard Error Message” or a “System Error Message” alone if no “Standard Error Message” has been entered. If the student continues to answer incorrectly, the correct answer appears after the number of tries predetermined by the instructor. Only then can the student move on to the next item. At the end of the exercise, each student receives a final tally of right and wrong answers.
MacLang gave us the tools we needed to respond to our students' demands for highly focused, interesting programs to supplement their classroom work. It has allowed us to begin authoring our own software with a maximum of flexibility and ease. By March 1987, we had MacLang programs in place in the five original language courses, were developing intermediate materials in French and Spanish, had begun work on elementary programs for Italian, and had upper-level students designing projects for methodology courses using MacLang. In addition, MacLang interfaces well with a number of different Macintosh programs like MacPaint , MacWrite, and MacDraw, and it can be linked to graphics generators like the Apple Thunderscan Digitizer to produce interesting and challenging programs. It can greatly facilitate learning in any discipline where basic concepts require reinforcement through drill and practice.
When our grant period drew to a close in spring 1987, we had fulfilled our original intention of exploring the potential value of CAl in foreign languages. Student evaluations spoke overwhelmingly in favor of CAI. Faculty interest in program development continued to increase despite the amount of time required to author programs. At least one publisher and several other universities had expressed interest in our in-house programs. We had, however, failed to gain the institutional support necessary to ensure the continuation of the project. Without stipends, released time, and general funding, continued commitments of time and resources on the scale required became difficult and professionally risky. Issues of tenure and publication began to make themselves felt at the departmental level, and requests for continued support at the college level met with a general skepticism about the advisability of dedicating financial or human resources to what many perceived as a problematic field.
Several critical issues presently stand in the way of CAI at the college level, not only in foreign languages at UNH but also nationwide across all disciplines. At the forefront stands the all-important issue of promotion and tenure and the way efforts in program development will be viewed in professional evaluation procedures. Will the programs generated by faculty members be accepted as legitimate professional activities worthy of advancement and compensation rewards? Will departmental interest in CAI prove compelling enough to make software development projects a consideration in discussions of released time, course loads, summer stipends, and sabbatical leaves? In turn, will broad and permanent commitments to CAI at the college level persuade hesitant publishers to produce college-level texts with adequate and affordable accompanying software?
Until the academic community as a whole confronts these issues and shows justifiably cautious publishers that it is committed to this new technology, the future of CAI in colleges and universities will continue to rest, as it does now in foreign languages at UNH, on the personal commitment of individual faculty members.
The senior authors are, respectively, Director of the Parker Learning Laboratory and Assistant Professor of French at the University of New Hampshire. Other contributing authors are Claire-Lise Malarte (French); Arna Bronstein, Edward Larkin, Mary Beth Rhiel, Deborah Roney, and James Roney (German and Russian); and Angela Shipman and Alan Stegmayer (Spanish and Classics).
We would like to acknowledge the support of the following persons who were essential to the success of our project: Betty Le Compagnon, the director of Project DISCovery; Judith Frommer, author of MacLang; Stuart Palmer, Dean of the College of Liberal Arts at UNH; and Mary Walkup, senior Russian major at UNH and all-around expert in MacLang authoring.
BMCAI Russian Verb Program, BMCAI Russian Basic Sounds. Computer software. Bryn Mawr Coll., 1986. IBM, disk.
CLEF: Computer-Assisted Learning Exercises for French. Computer software. Gessler, 1985. IBM, disk.
Direct and Indirect Object Pronouns. Computer software, Spanish. Concordia U, 1986. IBM, disk.
Frommer, Judith. MacLang. Computer software. Harvard Coll., 1986. Macintosh Plus, microdisk.
Garrett, Nina, and Robert S. Hart. “Foreign Language Teaching and the Computer.” Foreign Language Annals 18.1(1985): 59–63.
Latovsek, Robert B., Jr. Latin Flash Drill. Computer software. Centaur Systems, 1985. IBM, disk.
Matchmaker: Spanish Vocabulary Drills. Computer software. American Educational Computer, 1984. IBM, disk.
Radical Changing Verbs. Computer software, Spanish. Concordia U, 1986. IBM, disk.
Sevin, Sevin, and Bean. Wie Geht's? 2nd ed. Computer software. Holt, 1984. IBM, disk.
© 1988 by the Association of Departments of Foreign Languages. All Rights Reserved.
|
|---|
|
|
|
