The first prerequisite for the ability to write at a high professional level is of course reading, massive reading of the kind of writing one is expected to produce. As Smith (1988) has noted, "To learn to write for newspapers, you must read newspapers; textbooks about them will not suffice. For magazines, brose through magazines rather than through correspondence courses on magazine writing. To write poetry, read it ..." (p. 68).

General reading of the kind recommended earlier will help bring the future writer to the point where specialized texts are comprehensible. Moreover, reading anything will help all writing, thanks to similarities among prose styles (Biber, 1986), but real professionals must master all the subtleties of writing in their own field, and this comes only after a great deal of professional reading.

It is clearly impossible to include professional reading as part of an EFL program, and it is inefficient to try. This kind of reading needs to come later, once the future writer has begun specific education in the chosen profession.

Research has not yet begun to be done in this area, but I suspect that when it is done it will reveal that advanced EFL writers have acquired an impressive amount of the approved writing style of their professions. In fact, I think it is safe to predict that their competence in many areas of expository prose exceeds those of non-professional native speakers. They often, however, have gaps in late-acquired peripheral aspects of writing, conventions of writing that contribute little or nothing to actual communication, but have only a cosmetic effect.

Someday, because of the influence of nonnative writers of English (recall that they are now a majority on the internet), English expository prose might evolve; some of what is considered ungrammatical today may, in a few decades, be grammatical, and more aspects of institutionalized non-native varieties of English (e.g. Indian English, Nigerian English; Kachru, 1992) may be considered acceptable for international written communication. But until that happens, EFL writers must obey all current conventions and supply all the "cosmetic" aspects of language. Just how to close this gap is not yet clear.

One thing we do know: Formal instruction is not the answer. In addition to the demonstrated limited effects of formal instruction (Krashen, 2002), many advanced writers of EFL are obviously fully conversant with the known rules of writing; in fact, many are teachers of advanced writing in English as a foreign language and even do research in the field. What is clear is that these problems can and should be dealt with long after the formal EFL program is finished.

The composing process

Developing formal competence is only part of the story with writing, however. Competent writers also need to master the composing process, strategies that help them use writing to solve problems, stimulate cognitive development, and avoid writing blocks. Good writers, it has been reported, begin with a plan but are willing to change the plan, are willing to revise, pause to reread what they have written, delay editing until their ideas are clear on the page, and engage in regular daily writing rather than "binge writing" (Krashen, 1984; Rose, 1984; Boice, 1994).

The scant evidence available so far (Lee and Krashen, 2002) supports the hypothesis that at least some aspects of the composing process are universal in written languages, and that there is transfer from the first to the second language. If this is true, writers only need to develop the composing process once, in the primary language: Those who develop an efficient composing process in the first language will also have one in their second language.

8