The reaction to Richard Dawkins's tweet about Islam and science has been unswervingly negative. There's a good reason for that. Lumping Muslims into an amorphous mob of scientific illiterates may not strictly be "racist", but it is deeply insensitive.
On his blog, Dawkins suggests that even Abdus Salam should not be considered a Muslim Nobel laureate. Salam shared the Prize in 1979 with Steven Weinberg for his work on the weak nuclear force. However, he was a member of the Ammadiyya sect in his native Pakistan, which the Sunni majority consider non-orthodox. To say that this made Salam any less of a Muslim is just doing the fundamentalists' work for them. Certainly, he was a most devout man who worked hard to improve scientific education in the developing world.
The case of Salam makes a mockery of Dawkins's efforts to group all Muslims together. And if they happen to have won the Nobel Prize for physics, he doesn't seem to count them as Muslims at all. But the second half of his infamous tweet is also based on ignorance: "They did great things in the Middle Ages, though" says Dawkins. This seems to be a reference to the popular narrative of a scientific flowering under early Islam that was later subsumed under a wave of obscurantism.
Dawkins's vision of a lost Islamic Golden Age sounds similar to the widely-believed trope that Christianity extinguished ancient Greek science by closing the schools in Athens and burning down the Great Library of Alexandria. In fact, when they closed in 529AD, the neo-Platonic mysticism taught in the Athenian schools in no way resembled science. As for the destruction of the Library of Alexandria, it never happened. Even the attempt by Edward Gibbon to pin the loss of the subsidiary Serapeum library on Christians was based on a misreading of the sources.
Likewise, Islamic science wasn't snuffed out by the fundamentalist teaching of Al-Ghazali (d. 1111). The astronomical work of Nasir al-Tusi (d. 1274) and Idn al-Shatir (d. 1375) alone refutes that theory. The mathematical models of both these scholars were used, unacknowledged, by Nicolas Copernicus (d. 1543) in The Revolutions of the Heavenly Spheres. Rather, history turns out to be a lot more complicated than some great battle between science and religion.
As George Saliba notes in his Islamic Science and the Making of the European Renaissance, the question of why modern science didn't arise in the Muslim world is the wrong one to ask. It didn't arise in all sorts of advanced civilisations including China or Japan; ancient Greece and Rome; or Sassanid Persia and its great antagonist Byzantium. Instead, we should be wondering why a recognisably modern science had arisen in the West by the end of the nineteenth century. That this didn't happen elsewhere isn't because of the deficiencies of other societies. It's just that there was a unique conjunction of historical contingencies in one place and time. Exactly what those contingencies were remains a matter of much debate.
As for the Abbasid Caliphate, whatever the nature of the science being practiced therein, it wasn't "modern". Islamic apologists don't help themselves by anachronistically searching for the roots of experimental science in ninth-century Baghdad. They won't find them. What a careful study of this period reveals is the sheer variety of theological and philosophical belief among Christians, Jews and even pagans, as well as Muslims.
All this diversity could be politically unwelcome. When, in 833AD, the Caliph al-Mamun launched the Mihna, a persecution of Muslims who believed the Koran to be eternal and uncreated, he wasn't acting in the name of reason. His campaign is sometimes called the "Rational Inquisition": a double-anachronism. Inquisition is a particular legal system developed by jurists in thirteenth- century Italy. Although it was initially used by the Church to combat heresy, it still forms the core of the criminal law in continental Europe today. And the Caliphs were no rationalists. They took their status as the guardians of the Prophet's legacy very seriously indeed. Al-Mamun, like the Roman Emperor Constantine at Nicea, simply wanted to impose some sort of uniformity on the state religion. Herding Christians into agreement was hard enough, but early Islam had no detailed body of dogma and certainly no magisterium to tell people what to believe. Al-Mamun's campaign was an inevitable failure and Muslims have resisted centralised doctrinal authority ever since.
The fascinating mathematics and natural philosophy of this period probably owes much to the heterodox nature of the society in which it arose. It might also have been the reason that no doctrine ever reached critical mass so that it could dominate the others. Either way, Richard Dawkins's vision of an undifferentiated mass of medieval Muslims achieving great things is fantasy. The contribution made by Muslim scholars to the body of knowledge that makes up science today is of a fairly typical volume for a major civilisation. In any case, the image of a golden age followed by a fall from grace is not helpful or accurate. The history of Islam deserves better.