CALL It's Scope and Limits

The setting

Frank Berberich stood large over the console of the wonderful CALL lab put together by Kaz Nozawa at Toyohashi Gikadai as he led a curiosity-filled group through his current thinking about CALL software. The setting for the chapter meeting of Toyohashi JALT was well-chosen: the lab is designed for teaching both language and CALL software construction.

Berberich's projected scenarios for CALL

Frank's work at Tsukuba's University of Library and Information Science has led him to some grounded and broad attempts to characterize what is being done and what might optimally be done in CALL in general and in Japan in particular.

First Frank laid out his three imagined scenarios for CALL, the Star Trek scenario, where the target language is instantly integrated into one's mind; the 2001 scenario, where the machine is a fully human conversationalist and tutor; and the Now scenario, the current state of the art (in in the non- ideal sense of the term) scenario, where interaction is via keyboard/screen and audio/visual multimedia are basic.

Berberich's dimensions of CALL

Reaching far beyond other researchers' difficult-to-use qualitative lists of CALL activity types, Frank has created a proposed list of dimensions of CALL, a set of continua which make characterizing individual CALL objects (e.g. software) simple and revealing. A couple of examples of these continua: User Memory Load, the degree to which the user's memory is exercised; Data Access, the extent of the system's database; System Layering, the complexity of the system in terms of how much it is doing with the user data. These terms were demystified as Frank led the audience through a series of demonstrations of the design of a number of pieces of CALL software, both programs and CDROMS, and showed how each would rate on his dimensions.

The current state of CALL software

Some basic views underlying Frank's talk: current CALL software is limited, perhaps disappointing, since most items are either slick programmer productions which miss much of the wisdom that educators have to offer, or are educator produced and lack the stimulating interface that a programmer could provide.

Computer Adaptive Testing

All this suggests a need for "layered systems that can deal with flexible input and output, freely branch within and access a large base of tasks and data, depending onuser inputs, and can collect and process multiple user inputs for ongoing refinement of the system." For a hint of such a system, readers are directed to Frank's article,

Berberich, F. (1995). Computer Adaptive Testing and its extension to a teaching model in CALL. "CAELL Journal" 6 (2), 11-18.

Nozawa home page

So much information, so many sources, all laid out in a brief meeting. And there were distractions: the audience sat at computers linked to the web, and many had a good browse of Kaz Nozawa's home page, an excellent springboard for those seeking some map of what, where and how in CALL. The wise observer would merely note Kaz' URLhttp://www.lc.tut.ac.jp/nozawa/nozawa.html and browse it later, but warm and generous environment proved irresistible for many of us.

Toyohashi JALT are to be commended for putting together an event worthy of giving up a beautiful sunny Sunday.