After many unsuccessful attempts, I finally got a letter published in the NY Times [sentence removed by editors in brackets]:

Marlene Heath's experience with grade promotion based on accomplishment rather than age ("A Failure Policy that Succeeds," 3/30) makes me think that there is hope. Perhaps if more districts, or even individual teachers, adopted a policy of holding back students who had not mastered grade-appropriate skills, we would be able to abandon the punitive and stifling No Child Left Behind tests. [Perhaps if grade inflation were not commonplace, university accreditation agencies wouldn't be pushing for the college version:"student learning assessment."]

The vast majority of teachers know their material and know how to teach it to prepared students, but administrative fiats like social promotion and tools like student evaluations of teachers have obliterated the concept of meaningful grades and meaningful progress.

We can turn this around, but teachers cannot do it in isolation or without administrative support.

But, as they say, no good deed goes unpunished. I recently received the following e-mail:

My 8th Grade English class has a debate ... the topic I was assigned was:"Should there be a proficiency test to enter High School?", with me defending the affirmative position. In my search for sources, I came across a letter to the New York Times you submitted regarding Social Promotion. If you would take a moment to state your position on the subject, I believe it could help in my project. Thank you for your time.

Homework? Yuck.... and there's something a little perverse about asking 8th graders to debate whether they should be tested before they advance to the next level.

OK, seriously, though, this is an interesting question, I took debate in high school, so I'm quite sympathetic to the problem of finding sources. And as a scholar of modern Japan, where school entrance examinations are a very important component of social structure and life course, I've had some time to think about these things. So here goes, but you might not like my answer:

First, a basic description of the Japanese educational entrance exams. They come at important decision points: between elementary and middle school, between middle school and high school, between high school and college; elite elementary schools have entrance exams, as do elite preschools. These exams are not graduate tests, looking at basic skills from the previous level, but rather they are sorting mechanisms: the results of the exam determine the level of school into which one may advance. For example, the range of high schools available in Japan include vocational schools (agriculture, business, engineering/technical), terminal schools (no expectation of advancement to college), public schools (usually some students do advance to college, but not all; there's a range of public schools, too), and elite high schools (high rates of college matriculation; sometimes to specific elite colleges). A student's performance on the high school exams (and private schools often have their own entrance exams) determines which of these institutions they are suited for.

These exams are, therefore, very important, and very competitive. There is, as a result, a booming exam-preparation industry in Japan, ranging from private tutors (usually college students making money on the side, but sometimes people make a living doing this) to weekly supplemental classes (for someone who needs a little help in calculus, for example) to full-scale, five- or six-day-a-week (Japanese students still go to school for a half-day every other Saturday) after-school programs known as" cram schools," whose tuition can rival that of private high schools or colleges, and whose faculty, often with advanced degrees, command salaries at or above the level of college instructors. There are routine denunciations of this system within Japan: the loss of leisure time, the pressure on young students (though it is not actually a factor in suicide rates, as once believed; gender and age, certain life transitions, are more important than the exams), the cost, the limited nature of examined knowledge and the importance of"teaching to the exam" are all regularly discussed in Japanese public life. There is a clear acknowledgement that the system produces competence, but rarely brilliance, and the flexibility of US college admissions are seen as a source of strength and quality. But the Japanese system has so far been resistant to change; even the idea that non-exam factors should be considered in college entrance is rarely discussed seriously (and by seriously, I mean by college officials or Ministry of Education functionaries whose opinion might matter).

The exams serve a useful function: they provide a neutral, broad, even-handed score by which to judge students from a variety of schools and backgrounds. It is a powerful form of fairness, to use the exams in this way. It is educationally unsound and socially pathological, but it's"fair" and students and teachers are driven to high levels of competency by the high stakes of the test.

Part of the problem with the idea of a"proficiency" test, as I see it, is that it sets a bar too low. Not that there aren't students who would fail; I get some students in my university classes that should go back and repeat high school, and more who haven't successfully made the transition from high school to college (perhaps because they started high school at too low a level). If the test is a minimal skills test, though, many students (and perhaps teachers) will see a passing grade as sufficient, and will not strive for more. On the other hand, if the test is more than minimal skills, as these tests often are, then it is actually raising the minimum requirements for graduation without actually involving teachers, and penalizing students who do achieve the basic skills but who may not be capable of much more in the short term.

Minimal skills should be reflected in grades, not in graduation tests. Ninth-grade teachers should be able to trust eighth-grade teachers to give grades that are meaningful, and eighth-grade teachers need to be able to trust that seventh-grade teachers give grades that are meaningful. Conversely, eighth-grade teachers need to be given a curriculum (goals and standards) that will bring students from the seventh-grade level to the ninth-grade level; if the eighth-grade teacher is teaching sixth-grade level skills and knowledge, then the"passing" grades are not meaningful. For all the pathology of the Japanese examination system, the coordinated national curriculum does provide a consistency of standards and knowledge, and the group-oriented classroom culture of Japanese primary and secondary education does help the vast majority of Japan's students to achieve those goals, in stark contrast to the highly competitive exam junctures. Ironically, perhaps, the idea of"social promotion" which is so problematic in the US is not at all considered a problem in Japan: the vast majority of students advance with their classmates (and since they always stay part of the same classroom group, they get to know each other really well) and repeating a year is very, very rare; nonetheless, because the group as an integral component of classroom culture works together to bring slower learners up to speed, there is little to no concern that grade promotion is not consonant with skill achievement.

Of course, there are problems with Japan's educational system, and it is closely integrated with family and economic systems, some of which are in trouble. I'm not arguing for our copying it, and there are good reasons to think that some of its stronger components would be problematic when applied to an American classroom. But I am suggesting that there are lessons which can be learned from it: two in particular. First, testing has unintended consequences. The way in which No Child Left Behind mandates are drawing resources away from history and science education to boost reading and mathematics, for example, or the reversal of classroom mainstreaming (integration of the physically or cognitively disabled) in order to preserve NCLB score levels, or the increased drop-out rate, which school administrators don't mind because it raises their NCLB scores. Second, and the first is really an example of this, education is part of a social, economic and institutional system, and you can't just change one component of that system. If achievement tests are to be required, they must come along with curricular reform, they must be supported by parents and employers, they must come with both rewards and consequences for students as well as rewards and consequences for teachers and for administrators but those cannot be purely quantitative measures, and it should be diagnostic and graduated rather than simply pass/fail.

To sum up: I agree in principle with the idea that students should master minimal skills before advancing academically, but high-stakes standardized proficiency testing is not the way to solve the problem.