Intelligence Studies Review

Professor Philip H J Davies 

Spies and their Shibboleths  

Every profession has its terms of art that have what appears – at least to outsiders to outsiders – a disproportionate significance to its practitioners.  Intelligence has many of these.  They serve simultaneously as shibboleths and what might be called professional virtue signalling as well as crucial practical distinctions.  As much as anything else, they convey to the listener that the speaker knows what they are talking about and should be taken seriously.  Human intelligence (HUMINT) operators in Anglophone national agencies like the Secret Intelligence Service (SIS, aka MI6) and the Central Intelligence Agency (CIA) draw a strict distinction between ‘agents’ and ‘intelligence officers’.  Intelligence officers are career employees of intelligence agencies.  Agents are the human sources of information who are recruited and run by intelligence officers.  The fact that US federal law enforcement agencies like the FBI insist on calling their intelligence officers ‘special agents’ and their agents ‘agents’ (and sometimes ‘informants’) probably does not improve notoriously fraught relations between them and the intelligence community.  For warning intelligence practitioners, the flash point is the difference between an indicator, i.e. what you are looking for, and an indication which is what you have found.  To confuse or conflate the two in front of a warning analyst will prompt a fixed stare, clenched teeth and moment of stony silence (helpful hint – you might just recover the situation by referring to that ‘indicator’ being ‘activated’). 

Counterintelligence practitioners have their own red flag, and that is the distinction between a penetration agent and a double agent.  This rests on HUMINTERs’ distinction between intelligence officer and agent.  A double agent is an agent – a source – working for one organisation who has been discovered by another and has been ‘turned’.  They subsequently work for the second agency while to all appearances continuing to serve their original controllers.  A ‘double agent’, therefore, is a source who has been doubled.ⁱ  The famous stable of ‘double cross’ agents during the Second War like the charismatic career criminal Eddie Chapman, codenamed ZIGZAG, and one-man strategic deception service Juan Pujol, aka GARBO, were double agents.  They had been recruited by the German Abwehr as sources and, once detected and identified, were played back on the Germans. 

Oleg Gordievsky, on the other hand, was a KGB intelligence officer who was then recruited as a source by SIS.  He was, therefore, an agent of the SIS of a very specific sort referred to as a penetration agent. Gordievsky provided a route for SIS to penetrate the KGB to acquire information and, later in Gordievsky’s run, and act as a conduit of influence.  By the time he defected he had already fallen under suspicion and undergone a preliminary interrogation.  Had he not managed to resist that interrogation he might well have been ‘flipped’ and played back on the SIS – i.e. doubled – in which case he could then have been described as a ‘double agent’.  But this did not happen and Gordievsky remained a sound, reliable penetration agent. 

A Difference That Makes a Difference 

Double agents and penetration agents offer very different kinds and depths of information and run very different risks.  The double agent does not have access to the inner workings of the organisation that originally recruited them.  Indeed, their access to and knowledge of that organisation is very strictly controlled by that organisation.  Indeed, what ‘knowledge’ they have may even be false if they have been recruited under a ‘false flag’.  That is to say, they may have been targeted by officers pretending to work for an entirely different agency or country from that which they actually serve – by either the first or second agency recruiting them (or, indeed, both).  The double agent will, however, be able to identify their handler(s) or case officer(s)ⁱⁱ and be able to provide details of techniques, tactics and procedures (TTPs) employed by those handlers such as recruitment technique, communications methods, operational skills or ‘tradecraft’.  And the requirements against which they are tasked can be exploited for clues about an adversary’s interests, intentions and sometimes blind spots that might be exploited.  And they can, of course, be employed to feed their handlers false or misleading formation. But in the last analysis, the double agent can offer only a ‘worm’s eye view’ of the agency in question. 

By contrast, penetration agents provide deeper, richer veins of higher-level information.  Depending on where they are in their agency they can provide detailed and potentially vital information.  Penetration agents inside a hostile foreign intelligence service can reveal a wide range of operational activities by that service, including identifying or helping to identify that service’s penetration and double agents directed against oneself.  They can provide strategic information about their agency’s objectives and capabilities, its organisation or ‘order of battle’ (ORBAT), insights into key personalities and their relationships with each other and external figures in the wider regime.  They can help identify rivalries, personal antipathies and other fracture lines that might be exploited by covert political and psychological operations.  They will also be able to provide detail about TTPs and technologies or ‘kit’.  And a long-standing piece of SIS practice is that an agent well positioned to inform will also be well positioned to influence.  And Gordievsky’s position as an agent of influence, providing the intelligence briefing to Soviet leader Mikhail Gorbachev during a visit to London, is well known. 

The risks they take are also proportional to their level of access.  For most of their time double agents operate in a relatively low-risk environment.  On the whole, the wartime Double Cross agents lived in the comparatively benign environment of UK or allied-occupied territory.  They might occasionally face the hazard of meeting their Abwehr controllers abroad, but typically in neutral states rather than Axis occupied Europe.  Indeed, Chapman often appears to have been entirely too comfortable, even when meeting his controllers face to face on the Continent. By contrast, the penetration agent spends almost their entire operational lifetime in the heart of enemy territory, within the reach and potentially under the gaze of their agency’s internal security, the regime’s counterintelligence and policing agencies.  Even when posted abroad, the risk of being abruptly recalled and repatriated conscious or unconscious is ever-present.  Communication with them if they are posted to their agency’s headquarters is more difficult and less frequent.  And this reduces the likelihood of successfully exfiltrating them if they come under suspicious – which makes Gordievsky’s escape all the more remarkable. 

Penetration agents, then, offer both far higher value while taking far greater risks than double agents.  With that in mind, to describe Oleg Gordievsky as a ‘double agent’ is not merely inaccurate, it also does him something of a disservice.