When I was in the Royal Air Force in the 1950s, learning about electronics as part of my military service, my fellow airmen and I were ushered into a hut, told to sign the Official Secrets Act and shown how to repair an Enigma-like cipher machine that sent coded messages. We were told to take notes for an end-of-course exam. As we left the hut, the armed guard on the door collected our notes and destroyed them. I passed the exam, but always worried about going to jail if I told anyone what I had learned. Now, if I wanted to, I could. Patents filed on the electro-mechanical coding drums have finally been published. Whether anyone gets to read them is quite another matter…
ON CHRISTMAS DAY 1973, the US Patent Office published details of a tongue-twisting group of chemicals called O-alkyl S-dialkyl-phosphonothiolates. Why then? Here’s a clue. The patent had been filed by the British minister of supply a full 18 years earlier. To a suspicious journalist who knows a thing or two about patents, the odd timing and the long delay can only point to one thing: whatever this stuff was, someone somewhere didn’t want anyone else to know about it. After a quick scan, it was obvious why. Those chemicals are nerve gases.
Because patents contain a wealth of technical detail, declassified filings can offer fascinating insights into the history of military secrets. But first you have to find them. And that’s not easy for the security services play a peculiar game with inventions they want to keep quiet. If the technology is sensitive, the patent is branded secret the moment it is filed, then locked away and not even indexed. Apart from a spook or two and a patent examiner with high-level security clearance, no one knows it exists. Then, some unpredictable time later, an anonymous official in a shadowy office decides the invention is no longer a threat and can be published. The papers are then slipped onto the shelves of patent libraries at times when patent sleuths are otherwise engaged – eating turkey, perhaps.
Advertisement
Until recently, searching for declassified patents was harder than looking for the proverbial needle in a haystack. More than 10,000 new patents are published each week in the English language alone. But now that the world’s major patent offices in Europe and the US have online databases, anyone with an internet connection and a few practical tricks of the patent trade can play armchair spy.
For the past 30 years, I have searched each week’s batch of published patents looking for interesting new ideas to report in 91av. Because there is no index for “interesting” patents, the only way to find them was the hard way, speed-reading several thousand uninteresting ones. After a while, I discovered how to pick up the trail of an old secret.
First, beware exciting titles. Take Nelson Waterbury’s “buoyant bulletproof combat uniform”, which enables a marine who falls overboard to carry on fighting. The title screams secret, but one look at the inventor’s sketches – more Captain Marvel than Ministry of Defence – and it’s obvious why the US Patent Office didn’t bother locking it away. And if the title includes the word secret, you can probably dismiss it – although patent examiners should check carefully before doing the same.
In 1941, pianist George Antheil and Hedy Markey, better known as Hollywood star Hedy Lamarr, filed for a “secret communication system” that spread a signal across dozens of carrier frequencies to foil eavesdroppers and jammers. This spread-spectrum system was later used for military communications and became the basis of modern 3G cellphones. The invention would undoubtedly have been useful to the military, yet it was published within months and shortly after the Japanese attack on Pearl Harbor. It seems the US Patent Office didn’t take seriously something invented by a pianist and movie star. Yet 10 years earlier Texas Instruments had suggested a “dual band communication receiver” that split a message between just two frequencies. Washington kept that patent secret until 1990.
Tapping in an obvious keyword isn’t much help either. If you want to know about Alan Turing’s work at Bletchley Park, the UK’s wartime code-breaking centre, you might think “Turing” would turn up something useful. All you find are a couple of post-war patents on data storage. Bletchley’s cryptography patents, it seems, remain classified. But you can get an idea of what such patents might contain by looking at what the Americans were doing before the war. Cryptography research carried out for the US navy by Chester Wood, Bern Anderson and Donald Seiler between 1936 and 1938 was released from the US patent vaults in 1979. Details of William Friedman’s work for the US government on a five-drum cipher machine, begun in 1933, was released in 2000.
“Anyone with an internet connection can play armchair spy”
Not all nations have always been so secretive. In 1920s Germany, with its war-ravaged economy, protecting commercial interests seems to have taken priority. Far from keeping the Enigma cipher machine secret, manufacturer Chiffriermaschinen applied for UK patents, providing working drawings and full mechanical details, helpfully translated into English. By 1931 anyone could read about it in the UK Patent Office library.
It is unlikely that the Bletchley Park code-breakers and the spies who risked their lives stealing details of Enigma machines knew about these patents. The Enigma story has never included a heroic patent searcher wading through pages of small print. Yet according to a UK government technical paper in 1982, the patents could have solved many of Bletchley’s code-breaking problems “at a stroke”.
A visit to the Patent Office did pay off for Allied bomb-disposal squads faced with particularly dangerous German fuses. Rheinmetall, the German company that made the fuses, had protected its business interests by filing its patents in the UK. It did this in 1931 – two years before Hitler came to power. The company even offered to sell the design to the British Air Ministry. The ministry said “no thanks”, but 10 years later, someone with a long memory recalled the sales pitch and did a patent search under the company’s name. Sure enough, Rheinmetall’s patent revealed the secret of the double-capacitor circuit that had been killing sappers as they tried to disarm unexploded bombs. Doubtless this is why the US Patent Office slapped a 32-year secrecy order on a patent for a similar capacitor fuse filed by the University of Illinois in 1948.
If that seems paranoid, there are instances where more caution was required. Between 1951 and 1962 a clutch of British and American government agencies filed a series of patents for obscure organic compounds containing phosphorus and sulphur. None of the filings spelled out what the compounds really were – deadly nerve gases. By the early 1970s most of the patents had been published in the US and the UK, including the one sneaked out at Christmas. The “recipes” for making nerve gases were then freely available for anyone to read.
In 1975 the security services realised their mistake. All copies were whisked from library shelves, microfiche copies blacked out and a retroactive law passed to allow something completely new in the history of patenting: the offending patents were “unpublished”. In the UK, defence minister Fred Mulley later admitted, “there is no possibility now of closing all the doors”.
The internet has made his prediction come horribly true. Although the nerve gas patents never returned to the library shelves, if you know how to find them (and in this case, I’m not telling) they can now be freely downloaded from one of the official online websites.
If you think that’s scary, how about this: working details of the nuclear reactors built in the 1940s by Enrico Fermi and Leo Szilard were published some 15 years later. But vital reactor safety-control systems developed for the US Department of Energy in the 1950s were judged too sensitive to release until 2004. A formula for preventing corrosion of uranium by coating it with a copper-tin alloy was also kept secret until last year.
It would be nice to think that there is a well-considered reason for judging it safe to publish all this nuclear information. Unfortunately the secrecy surrounding the process means we have no way of knowing. And even after 30 years, I still can’t fathom what goes on in the censor’s mind. If you think you can detect any logic to the process, ponder the following.
In 1942, British inventor Barnes Wallis filed a patent for a way of making a cylindrical bomb spin, so that when it hits water it skims over the surface until it slams into its target – a dam, say. In 1954, when The Dam Busters appeared in cinemas everywhere, millions of people learned how to make a bouncing bomb. The patent was declassified in 1963.
