Second, the inadequacy of traditional Islamic languages—Arabic, Persian, Urdu—is an important contributory reason. About 80% of the world's scientific literature appears first in English, and few traditional languages in the developing world have adequately adapted to new linguistic demands.In what sense can a language be inadequate for a purpose? What I take him to be referring to is the inadequacy of technical terminology. Specialists in any field have to learn a set of fairly complicated ideas to which they can refer concisely and unambiguously (phoneme, wh-movement, coronal, theta-role; integration, isomorphism, standard deviation...) Such terms often do not refer to anything normally noticed by people, and therefore have no equivalent in any language until one is created or borrowed. Various specialists or committees have undertaken to create such terms (in Arabic, at least, they generally eschew the idea of borrowing them.) But in many cases a chaos of alternative terms is spread. For "linguistics" alone, different Arabic dictionaries will suggest اللسانيات، الألسنيات، اللغويات، علم اللغة, and even other terms. I have three dictionaries of linguistic terminology in Arabic sitting on my shelf; randomly looking up "retroflex", for example, I find ارتدادي، التوائي، انقلابي all given as translations.
One might expect that the efforts of specialists to communicate with each other would end this problem, with the community of linguists (say) rapidly converging on a single term and abandoning the rest, just as such synonyms for "retroflex" as "cerebral" or "cacuminal" have largely disappeared in English. But there we have a vicious circle. At present, to be a good specialist in many fields, you need to have studied them in some Western language, and to be following a literature on them that's largely in a Western language, and to be communicating with colleagues who mostly speak that same language. In fact, given how little on average is spent on research in the Islamic world, in many such fields the odds are high that you won't even be able to find employment without going to or staying in the West, further reducing your opportunities to talk about, or teach, the subject in your own language - and if you do stay in your own country, you may find that specialist terminology dictionaries, especially those printed in other countries, are hard to find. So if ambiguities or misunderstandings come up, the easy thing to do is to switch to English or French or the like; the ideological incentive to use your own language is not supplemented by any significant material or practical incentive. And thus the language gets slowly pressured out of another domain. It's not inevitable, but to change it you'd have to create more incentives and more opportunities for people to stay and to teach in their own countries.
Of course, for Arabic in particular but to a lesser extent for Urdu and Persian, there is a second factor to be considered: diglossia, the wide gap between the language spoken in everyday conversation and the one considered suitable for writing or teaching in. This in itself has some negative implications for teaching science, although the obstacles it sets up to participation by the masses are far less than those that use of an unrelated foreign language like English or French does. But that is another topic for another time.