With this book, Jon Agar has done us all a great service. He has lifted the story of the invention and development of the electronic digital computer out of its specialist niche and placed it squarely in the context of the long history of the interest of the British government (among others) in making its functions more routine, methodical and "mechanical".
Terms such as "machinery of government", "wheels of justice", even "bureaucracy" are so ingrained in our daily vocabularies that it takes a special effort to disengage them, analyse their meanings and trace their origins. In his introductory chapters, the author takes on this task and not only succeeds but does so in a readable and fascinating manner.
Students of British history will find descriptions of a host of familiar people, but from a perspective that no one to my knowledge has examined before (save the exception that proves the rule, Charles Babbage).
Agar argues that it has been the goal of government to make its functions more routine and therefore "mechanical" from the beginnings of civilisation, but that it was in the 19th century that these efforts accelerated, in conjunction with advances in record-keeping. And it was not only a case of the development of mechanical information-processing devices but also the introduction of standardised forms, methods of organising office work and so on. Nor was it a question of "technological determinism", in which advances in calculating machinery that caused British government to evolve in certain directions. Rather, the British government became more machine-like while, in tandem, machines were invented and marketed to process statistical information to serve it. Seen in this context, the invention of the electronic digital computer is but another link in this unbroken chain of activity.
Standard histories of the computer emphasise the computational demands of the Second World War: in America, where machines were developed for computing firing tables, and in the UK, where computers were developed to decode intercepted German radio traffic. Agar does not reject either of these creation stories, and indeed spends considerable time on the latter.
But he places them in the context of a general need for the Allies to gather, process and make practical use of information of all kinds. That includes information on citizens, to ensure, among other things, that no able-bodied men were shirking their military obligations. (He does not dwell on the parallel use of punched cards by the Nazis to aid in identifying residents under Reich control for shipment to concentration camps, but that was but the flip side of the same coin, at least as far as using data-processing machines was concerned.)
Turning to the digital computer, however, Agar finds that the story becomes more complicated. The computer is a machine, to be sure, but unlike the comptometers and Hollerith tabulators of an earlier age, it has an ability to modify its actions based on results previously computed. As in previous ages, governments - especially, in this case, the US military establishment - embraced technology as a way to reduce uncertainty. But the plastic nature of the computer made that goal as elusive as ever. Agar cites the famous film Dr Strangelove , which brilliantly shows both the extent to which computers have become a part of the way modern wars are fought, and how futile it is to think that one can turn over basic decisions about war to such machines, no matter how artificially "intelligent" they are claimed to be. By extension, the same may be said of the other functions of government.
In the post-Cold War era, when personal computers and internet connections have brought all this processing power to the masses, the notion that a government can make its functions more and more mechanical seems even more absurd. I say that even though I am alarmed by plans by US intelligence agencies to gather and "mine" statistical data on its citizens in the name of national security. Thanks to this book, I am able to recognise something that otherwise would have escaped notice: the clever use of the metaphor "mining" in the above description. In any event, I recommend this book to those who are following the computerisation of our world but who find most narratives purporting to explain it lacking.
Paul Ceruzzi is curator of aerospace electronics and computing, National Air and Space Museum, Smithsonian Institution, Washington DC, US.
The Government Machine: A Revolutionary History of the Computer
Author - Jon Agar
Publisher - MIT Press
Pages - 554
Price - £32.95
ISBN - 0 262 01202 2