Diane Ravitch is a historian of the (US) school system. While this does not immediately qualify her as an advisor on how to best organise schools, it does ensure that she has a thorough knowledge of the system's inner workings. Also, she has served on several (mostly conservative) think-tanks on educational matters and advised the Bush (II) administration on education reforms. She knows a lot about the subject – and what is more: she has completely changed her beliefs, thus proving that she also thinks and re-thinks her positions.

The book's title sounds more optimistic than the book itself. Ravitch more or less describes and analyses the bad shape the American schools are in – she gives little forward-looking thoughts on how to improve it except by giving up on the disastrous reforms of the last two or three decades.

Ravitch identifies two basic factors responsible for the American schools' dilemma:

(1) Belief in output-orientation

The American system used to follow non-formal consent on what should be taught. Under the reforms of the 1990s, the authorities seized taking an interest in what is taught (input) but instead focussed on what is learned (output). This sounds either like just a change in wording (after all, what is learned must have been taught before) or a sensible readjustment (teaching is not an end initself – learning is the final goal of teaching and should therefore be the authorities' focus). But it is not. Because on the one hand, it produces garbled responsibilities: The scool boards and state governments can instruct and hold responsible only the teachers, not the learners. It is only fair to hold them responsible for how well they do their job, i.e. what and how they teach. It is blatantly unfair to hold teachers responsible for how well pupils learn because they are only one amongst many influencing factors like social background, parents' education, school resources, etc.

Also, output can not be prescribed, but must be measured. This has lead to a true orgy in American schools, where every pupil is tested every year in literacy and in mathematics. Apart from the stress for teachers, parents, and pupils, apart from the immense costs, this has led to all the problems of test-oriented teaching: Teaching-to-the-test focusses on short-term results, not on true education. It tends to teach how to take tests instead of teach the subject at hand (e.g. it is never a good idea to leave a question in a multiple-choice unanswered, you might just tick the right answer by chance). It tempts schools and teachers to rig results in one way or the other. And, most importantly, it degrades everything that is not literacy or maths, but still important for a good education. Think of the arts, sports, an open and democratic mind etc.

(2) Accountability

A main tenet of capitalist economists is that (economic) rewards for good manufacturers and punishments for bad ones leads to ever-improved performance and better goods and services for customers. So, why not apply this principle to education, reasoned the scholl reformers of the free-market era. They gave schools and teachers pretty much all possible liberties to reach targets on pupils' success, i.e. testing results in literacy and mathematics. However, if targets were not reached, teachers and schools had to pay the price: teachers were sacked, schools were closed – in the hope that new teachers and new schools would produce better results.

Of course, this is unfair – teachers are not the only, maybe not even the most important factor in school success, see above. But it is not only unfair, it is also counter-productive: sacked teachers and closed schools take with them experience and expertise that may be forever lost. There is little reason to believe that newly employed teachers or newly opened schools will even show comparable results. The market-happy idea to lift all requirements on teacher training certainly does not help. There may be naturals who are good teachers even withouit training and the not-so-good ones can be disposed of once their poor performance is statistically established. Nobody seems to have noticed that this is cynical towards teachers who may lose their jobs after only a few years of teaching. And of course, since bad teachers can only be identified through pupils' test results, they must be let loose on children who act as educational guinea pigs. It seems much more sensible to invest in teacher training and only employ good ones instead of employing everybody and than get rid of most of them.

An interesting twist is how private schools on public funding ensure optimal test results: They see to it that their intake comes from well-educated families. Pupils who turn out to be a burden to the school's test statistics may be »counselled out« or re-labelled as special needs students and thus taken out of the general statistics.

Since many of the ideas already failed in the US are still on the program of conservative or »liberal« reformers in Germany, this book is of eminent importance to German teachers and generally all those involved in school.

As I have already mentioned, Ravitch gives little hints on how to actually improve schools apart from returning to the system of the pre-reforms era. One important point she stresses again and again is that success at school closely correlates to financial conditions at home. children from poor families who have a lot of worries more important than school, who may not even be able to get enough proper food and rest, will always perform below their potential. The best way to improve school performance may therefore be to spend money not on schools, but on social programmes.

Another point is that good teachers produce good results. So it seems like a sensible idea to invest in teacher training programmes.