The same result is true in second language acquisition research. Those who are provided with more comprehensible input in second and foreign language classes do better on all tests related to real communication. On grammar tests, there is usually no difference between the groups, and occasionally the comprehensible input students actually do better (Krashen, 1994a).

The analogous hypothesis in terms of concept learning is the "problem-solving" hypothesis: Our knowledge of facts and concepts is the result of our attempts to solve interesting problems.

Research evidence for the problem-solving hypothesis is reviewed in several places, e.g. Smith (1975, 1998), Krashen (1990, 1995), but I cannot resist one obvious piece of evidence: Smart people don't study, they try to solve problems. The world's great chemists, Linus Pauling and Glenn Seaborg, did not review the periodic table each morning. Their encyclopedic knowledge of chemistry (and physics) came from decades of attempting to solve fundamental problems in chemistry.

Thus, Ayers' project in calculating the volume of a silo not only resulted in deeper knowledge of mathematics, but may also have resulted in considerable skill development. In addition, it was probably more enjoyable than worksheets.

Having fun

Despite incredible claims that children actually enjoy doing mindless exercises, fueled no doubt by infomercials, there is good evidence that students enjoy real reading and hearing stories (Krashen, 1994b; Von Sprecken, Kim, and Krashen, 1998; Ramos and Krashen, 1998) and that they like these activities better than drills (Greaney, 1970; McQuillan, 1994). There is also evidence that comprehensible input-based methods for second and foreign language teaching are more enjoyable than grammar -based methods (Krashen, 1994b), and it is reasonable to hypothesize that solving interesting problems is more enjoyable than memorizing facts. What is best for children appears to be what is most enjoyable.

But this is not to say that if an activity is pleasant it is always good for children; "if it feels good, it must be good for you" is inaccurate. Rather, what appears to be correct is the following: If an activity is helpful for language, literacy, or concept development, children will find it to be interesting and enjoyable.

Conclusions

There is considerable evidence supporting the validity of Ayers' criticisms of the testing/skill building movement. Interesting reading and problem-solving will lead to acceptable levels of skills, and there is no need for delayed gratification and tough love. Children and teachers have the right to enjoy school at all grade levels.

2