Gordon Corera, best known as the security correspondent for BBC News, somehow finds time to write authoritative, well-researched and readable books on intelligence. Here he explores the evolution of computers from what used to be called signals intelligence to their transforming role in today’s intelligence world. The result is an informative, balanced and revealing survey of the field in which, I suspect, most experts will find something new.
He starts with an event that took place 101 years ago next month, when the British dredger Alert set off from Dover in the early hours to cut the German undersea telegraph cables. This inconspicuous act meant that throughout the first world war Germany’s overseas communications had to be sent by radio, which in turn meant they could be intercepted and eventually decoded. One of these intercepts was later instrumental in bringing America into the war, which helped shape our modern world.
The story of Enigma, Bletchley Park, Turing et al is well known (though that of Room 40, the 1914–1918 equivalent, less so). Corera’s account gives due credit to Tommy Flowers, the Post Office engineer who made Turing’s concept of a universal computer a reality by building Colossus. He also makes welcome mention of mathematicians other than Turing — the young Bill Tutte, for example, who stared at a wall for months, twiddling his pencil, as he conceptualised Tunny (the harder-than-Enigma machine that carried German High Command signals) and the maths that must lie behind it, without ever having seen one.
Corera next traces the evolution of computing following the second world war, showing how intimately entwined it was with Cold War signals intelligence. Of particular interest to us now is how the impermeability of most Russian ciphers led to what we call bulk data analysis. You may not be able to read the signals, but if you record them all, establishing the where and when and correlating them not just with each other but with observable events, you identify patterns. When you’ve done that you can identify what is normal, and once you’ve done that you can recognise the abnormal, which is what you’re after. That early computerised data-trawling led directly to the kind of market research that bodies such as Google and Facebook routinely do on us now, and to what governments can do when hunting terrorists, spies and criminals, if the law allows.
Why then is the computer industry and the internet dominated by America, just as, a century ago, Britain dominated the world’s telegraph system? ‘The story of the British computer industry after the war’, says Corera, ‘was one of brilliant engineering but business failure’. The reason seems to be not just vastly greater US resources but the closer linking of its defence and intelligence organisations with the commercial world, with people and ideas moving more freely between them.
Part of the legacy of Bletchley was the hermetic sealing of our intelligence bodies from the outside world, which meant that GCHQ neither fed into — nor always benefited from — commercial developments. An example is Phil Zimmermann’s PGP (Pretty Good Privacy) encryption package, available to all on the internet. It and similar techniques offer much to banks and businesses — also to any terrorist, criminal or paedophile with much to hide — but in fact two individuals in GCHQ had worked out the technique years before Silicon Valley. They thought it better not to tell anyone.
In recent decades the means of communication and encryption that were once solely the province of defence and intelligence agencies have become everybody’s. This in turn means that a state seeking to frustrate its enemies — however defined — has to fish in the common internet sea rather than the secret streams of yesterday. Corera’s discussion of today’s privacy-versus-intrusion arguments, dramatised by the US defector Edward Snowden and the privacy-evangelist-cum-rape-suspect Julian Assange, sets out the issues and motives fairly and accurately. But it remains a puzzle why some who live under the most benign governments that have ever existed should do so much to help some of the least benign.
Another issue examined by Corera is the last Labour government’s permitting the Chinese company Huawei to become our telecoms infrastructure provider, in the course of which the Chief Information Officer, who was meant to look after our cyber security, jumped ship to do the same for Huawei. Whatever its professed intentions, no Chinese company can afford to say no to the Chinese government — and we know from the blizzard of attacks against British government and industry systems (about 70 a month) what their attitude is. Not to mention the Russians.
This must at the very least represent a theoretical long-term compromise of national security and it would be interesting to know which security dogs did not bark. Corera is too sensible in his wide-ranging and well-sourced survey to name names without hard evidence, but they are not difficult to work out.
The nonagenarian Brian Stewart’s musings on his long career in intelligence-related work — ably aided by the researcher Samantha Newbery — have the aroma of a good whisky, well-distilled. Rather than offer a blow-by-blow account, Stewart takes a historical and philosophical perspective, rightly emphasising that ‘good assessment is the key to good intelligence’ and regretfully observing that ‘it has been well said that customers sometimes use intelligence as a drunk uses a lamp post: for support, not illumination’. This was of course particularly the case in the run-up to the Iraq war, of which — as a former secretary to the Joint Intelligence Committee — he is particularly critical. His book reads like an extended ambassadorial valedictory, as they used to be — and is all the better for it.