National Post

Friday, May 30, 2008

Presented by

The necessity of HUMINT

Colby Cosh,  National Post  Published: Friday, May 30, 2008

CBC News scored a fascinating coup on Monday when it disclosed the existence of a dedicated Canadian human intelligence (HUMINT) unit in Afghanistan-- one, according to Brigadier-General Denis Thompson, that has existed in the combat theatre "essentially" since the commencement of hostilities. According to documents obtained by the network, the Canadian Forces (CF) will spend $27-million on equipment for the unit over the next three years.

Unfortunately, the new information in the CBC's story was handled in the usual slightly tone-deaf manner with which civilian reporters treat military news; making matters worse was a canned response from NDP defence critic Dawn Black, who complained that human intelligence gathering in a war zone "raises all kinds of red flags" (say, what have the New Democrats got against red flags?) and stated, "There's never been a debate in Canada that I am aware of on running an intelligence company out of the Canadian Forces."

Guys, honestly:Why even bother having a defence critic?

That scary acronym HUMINT is, in truth, nothing more that mil-speak for any relevant knowledge gathered by an army directly from human sources. Other types include signals intelligence (SIGINT), obtained by intercepting and decrypting the enemy's (or someone else's) information transmissions, and imagery intelligence (IMINT), which comes from the study of photographs taken from the air or space. It is not news that the CF has been trying to strengthen its HUMINT-gathering capacity for years; indeed, it has been openly recruiting bright, curious soldiers for the purpose.

Most of us now know how important SIGINT was in determining Allied success in the Second World War; victory in Europe would have been delayed considerably if high-level German cryptographic traffic hadn't been cracked by the British, who essentially invented the digital computer for the purpose. Yet the importance of human intelligence can scarcely be overstated, though it is sometimes neglected by the historians. The invasion of Normandy could not have succeeded without an enormous layer of resistance-provided HUMINT, covering everything from the quality of glider-landing sites to the dispositions of Axis forces guarding bridgeheads. (By contrast, better Allied HUMINT might have saved the failed Market Garden offensive of 1944 by tipping commanders to the presence of two freshly arrived SS Panzer divisions in the neighbourhood of the main thrust.)

Nevertheless, as the shooting war turned into a balance of terror between nuclear-armed superstates, HUMINT began to take more and more of a back seat to technological methods of intelligence-gathering in the military and espionage directorates of Western democracies, and this was particularly true of Canada. But in the 1990s, theorists were clever enough to see that emerging wars of the immediate future-- strikes at "non-state actors," counterterror operations, humanitarian interventions -- would require superior HUMINT and more trainees. If your enemy is embedded amongst the people of an occupied country, you need to be able to talk to the people.

Talking, of course, probably isn't the reason the CF intends to burn through $27-million. Much of what Canada's HUMINT company is probably getting up to has very little to do with "spies" as such. The religious, all-male, communal nature of the Taliban makes it virtually impossible to infiltrate by means of cash (consider how little publicly known progress has been yielded by means of the bounties on Osama bin Laden's head) or other inducements.

Anyway, even under ideal conditions, much of the work of a HUMINT agency is sophisticated bookkeeping. Some of the expense will no doubt go toward providing a credible security cover for installations in the region: The Taliban cannot be allowed to blind our intelligence apparatus at the cost of one suicide bomber. But a lot of it will be going toward computers and software.

Tips and reports from Afghan civilians opposed to the Taliban must be stored in a way that makes them available for retrieval, rated for reliability and plausibility and turned into memoranda for the timely use of soldiers and staff. The toughest task of all is linking multiple fragments of intelligence together so that they combine into a trustworthy picture of, say, the location of a bomb factory or the date of an attack. Military software developers have been putting a great deal of effort into applying artificial intelligence to HUMINT gathering: arming computers with natural-language recognition abilities would help them navigate databases and put up flags when pieces of evidence point in a common direction.

It is absurd to demand a "debate" on whether a fighting force abroad should have a HUMINT apparatus; it would be exactly like debating whether it should carry ammunition. And the existence of such an apparatus can only raise "red flags" in the eyes of a person who has never devoted a moment's thought or study to how armies fight.