nications albeit in forms modified beyond anything we can imagine.11 It would have been less “efficient” than what actually emerged, but nobody would have known, just as we cannot know of possible techniques that were for some reason beyond our reach but just never occurred to anyone.
This example is meant to illustrate the basic point that, given a certain basis of knowledge, the right institutional set-up and a certain amount of luck, many inventions or their equivalents seem to have been quite probable ex ante and some perhaps inevitable. But this only pushes the question one step back: how likely was it for the background knowledge itself to emerge? As I have argued elsewhere, the scientific revolution of the seventeenth century created what might best be called an “industrial enlightenment” in which natural philosophers and experimentalists teamed up with industrialists and engineers, and jointly created the kind of environment that was uniquely favorable to rapid innovation.12 Suppose these intellectual developments had not happened in Western Europe. Would some other area in the world have followed a somewhat similar pattern? Would, for example, Africans have eventually invented a steam engine if left alone by Europeans? The critical question is not whether there is coal in Africa (there is) or whether Africans had the iron-working skills (they did), but whether the underlying knowledge of atmospheric pressure would ever have occurred in an African setting.
The assumptions underlying any such analysis should be made explicit. First, they assume that there are natural regularities and phenomena that transcend social context, and that while their representation may vary amongst different cultures, they ultimately represent self-evident properties that are naturally determined but discoverable and comprehensible by humans. Nature poses certain challenges and constraints that matter to the human material condition, and overcoming these constraints is what technology is all about. To overcome them, we need to know things that can serve
11The Chappe semaphore telegraph, operating through France as well as in other parts of Western Europe, was quite successful: it could transmit under optimal conditions a bit of information from Paris to Toulon in 12 minutes in contrast with the two full days it would take a messenger on horseback. A 100-signal telegram from Paris to Bordeaux in 1820 took ninety five minutes, in 1840 half that. Given that a “signal” was picked from a code book with tens of thousands options, this was a huge amount of information. The optical telegraph at its peak covered 5000 miles and included 530 relay stations. The Chappe system was a government monopoly and did not serve as a means of transmission of private information, yet in the absence of the electrical telegraph there is no reason why it could not have played a much larger role. Another widely used visual telegraph was developed in 1795 by George Murray in England. This system rapidly caught on in England and in the United States, where a number of sites bearing the name Telegraph Hill or Signal Hill can still be found, particularly in coastal regions. Cf. Alexander J. Field “French optical telegraphy, 1793-1855: hardware, software, administration.” Technology and Culture 35 (1994): 315-48; Daniel Headrick, When Information Came of Age: Technologies of Knowledge in the Age of Reason and Revolution, 1700-1850. New York: Oxford University Press, 2000.
12Particularly in Scotland the connection between scientists such as William Cullen and Joseph Black and an audience that demanded that knowledge be made socially useful was quite close. Such connections of course could converge within a single mind: HumphryDavy, Count Rumford, and later Charles Wheatstone and Lord Kelvin all represent examples of scientist-inventors, in which the epistemic base and its application to a specific technique took place in the confines of a single mind.