Kevin Carey responds to my skepticism about the possibility of "closing the achievement gap" in public education by saying "it's all a matter of how you define 'close the achievement gap.'" And so it is. Carey says under one definition ("erase all academic differences between students of different economic backgrounds") this is impossible, whereas under a different definition ("bring all students, including low-income students, up to defined minimum levels of proficiency") it's realistic. I suppose I agree with that.

On some level then, this is perhaps a meaningless semantic controversy. Nevertheless, it certainly seems to me that the phrase "close the achievement gap" strongly implies a desire to narrow or eliminate (i.e., "close") the differential achievement level (i.e., "gap") between high-SES and low-SES rather than a desire to bring low-SES students up to a minimum level of achievement. That latter goal doesn't seem to have any particular relationship to the concept of a "gap" that ought to be "closed." Which returns me to what I think I was saying initially, namely that our education system could serve low-SES students better, and we ought to endeavor to do so. Doing better on this score would have a variety of benefits, but a significant reduction in inequality, even inequality in educational achievement, isn't likely going to be among those benefits.

Andrew Rotherham wants to note that in policy terms No Child Left Behind is in fact geared toward Carey's goals. That is an important point, and I think NCLB was a pretty good bill. To me, though, the "close the achievement gap" rhetoric lies somewhere between pernicious and misleading. As Rotherham emphasizes, there's no measures a liberal society can take to prevent socioeconomic inequalities becoming educational inequalities through the mechanism of higher levels of high-SES parental investment in education. So insofar as you care about educational inequities, you really need to tackle broader socioeconomic inequities.