Peter Gill
Frank Cass, London, 1993
Academia’s a swine. Writing an essay on International Relations (the ideological version of Foreign Office ‘realism’) for my Politics MA, I managed to smuggle in a few references to actual politics — European Nuclear Nuclear Disarmament, the SNP, and ‘independence within Europe’, that kind of thing. Flushed with success, I told a friend all about it. ‘How fucking tedious,’ he said. The bastard was right, too.
Likewise with parapolitics: it’s not impossible to bend the subject into academically respectable shape, but you run the risk of leaving out what makes it interesting to the rest of us. You can have a discussion of bureaucratic and organisational models of the modern state, concluding that inter-agency rivalry is to be expected — or you can have an overview of the MI5/MI6 turf wars. You can’t, yet, have both.
Which is not to say this book is ‘Parapolitics for Beginners (with sociology degrees)’. Some of Gill’s academic digressions are distinctly useful: the work of Graham Allison is well worth following up, for example. On the other hand, the index records five references to Foucault, which was four too many for me. It’s good to define one’s terms, too, but did we really need a definition of the concept of ‘concept’?
But these scars of academia are the only real flaw in an ambitious and potentially very useful book. Gill aims to provide the groundwork for an inquiry into the British security intelligence agencies, the Security Service and police Special Branches. To that end he reviews what is known about the operations and the remit of British security intelligence agencies and their counterparts in the US, Canada and Australia. Although this focus precludes any sustained attention to the increasingly important European context for British intelligence, as a framework for country-by-country comparisons it works well.
Gill uses a four-layer model of the state and society: first there is the secret state, then the executive (the government of the day plus its civil service). Next comes ‘other government agencies’ — a category including the courts as well as Parliament — and society at large. The secret state’s relations with the other levels is examined using the two governing concepts of penetration and autonomy. Penetration signifies the capacity of an institution at one level to penetrate ‘outwards’, either to gather information or to take action with society; autonomy refers to the capacity to resist attempts at invigilation or direction from ‘outside’.
As an overall picture Gill proposes the ‘Gore-Tex state’: at each level (secret state, executive, government) it’s easier for the influence of the inner level to get out than for the influence of the outer level to get in. Different layers within the state may have different degrees of autonomy and penetrative ability; these will tend to vary together — that is, a low-penetration secret state will be unlikely to be have a high degree of autonomy — although this can’t be assumed. Within this framework Gill proposes three main modes of the secret state, ranked by penetration and autonomy: ‘independent security state’ (high), ‘political police’ (medium) and ‘domestic intelligence bureau’ (low).
That completes the background, with the exception of the small question of what security intelligence agencies are actually for. Gill defines ‘security intelligence’ as ‘the state’s gathering of information about and attempts to counter perceived threats to its security deriving from espionage, sabotage, foreign-influenced activities, political violence and subversion’. Based on real-world definitions, this provokes a host of questions: should the same agency have charge of information-gathering and ‘attempts to counter’? (Gill appears to regard this as preferable to a proliferation of agencies.) Who perceives these ‘perceived threats’? What, in peacetime, is ‘sabotgage’? Who identifies activities as ‘foreign-influenced’ and what standards of proof apply? What counts as ‘violence’ and when is it ‘political’?
Given the existence of some sort of security intelligence agency, questions like these are bound to arise and have no final answer. The best to be hoped for is that they can be addressed in practice, through the kind of external vigilation which was built into the Canadian system in 1984 (and which in our case we have not got).
Larger questions are posed by the final item on the security intelligence shopping list, ‘subversion’. The closest thing to a definition in the British context appears in the 1989 Security Service Act: ‘actions intended to overthrow or undermine parliamentary democracy by political, industrial or violent means’. With no necessary implication of violence or illegality, this is a definition broad enough to enfold every last Trot paper-buyer and Charter 88 signatory. (Guilty on both counts, your honour.)
The Canadian Security Intelligence Act of 1984 includes a slightly more intelligent attempt at a definition: ‘activities directed toward undermining by covert unlawful acts, or directed towards or intended ultimately to lead to the overthrow by violence of the constitutionally established system of government in Canada’. Like Gill’s definition of ‘security intelligence’, the strength of this definition is in the problems it foregrounds. The first clause is particularly problematic. ‘Lawful advocacy, protest and dissent’ is specifically excluded from the scope of the Act; terrorism (‘serious violence against persons or property for the purpose of achieving a political objective’) is covered under another heading. The clause thus seems designed to target non-violent covert unlawful acts which threaten to undermine the state. Covert civil disobedience, anyone?
This just leaves the catch-all category of activities ‘intended ultimately to lead to’, etc (my emphasis). This plainly brings any organisation which claims to have revolutionary aims into the target category, even if those organisations do nothing more covert or unlawful than launching the odd anti-fascist front organisation. And if an organisation is committed, ultimately, to overthrowing the government, this must make any and all of its members and sympathisers legitimate targets for surveillance. They may not appear to be up to very much, but keep watching: after all, ‘we do not know today what we will need tomorrow’, in the words of Merseyside Police standing orders (1983).
In this way the concept of ‘subversion’ leads logically to the indiscriminate collection of information about whole categories of people: in Gill’s words, ‘the blanket surveillance of revolutionary rhetoric as a substitute for making realistic threat assessments’. The conclusion must be that ‘subversion’ owes more to state paranoia than to a realistic assessment of anything. To its credit, the Canadian Security Intelligence Review Committee (SIRC) recommended in 1987 that the counter-subversion branch of the Canadian Security Intelligence Service be closed and its work split between the counter-terrorism and counter-intelligence branches. Two years later SIRC followed up by recommending the repeal of the ‘subversion’ paragraph. The branch was closed, but the paragraph remains: two steps forward.
In the remainder of the book Gill takes the concepts of penetration and autonomy in turn and considers them as they apply to the actual activities of agencies in the countries under consideration. Penetration covers intelligence-gathering and the dissemination of information and disinformation, as well as ‘countering’ activities; autonomy covers the internal organisation of agencies and questions of oversight and control. The results of this comparison are detailed and informative. In some areas there is very little British information under consideration: this, of course, tells its own story.
‘The general mystification surrounding intelligence matters in the UK’ is one problem: Gill instances, and cheerfully demolishes, the puerile argument that invigilation is incompatible with official secrey (those who know can’t speak). Sheer ignorance is another: Gill makes it abundantly clear how little we know in Britain about some fairly large areas — ‘the rules (if any) and practices of security intelligence officials’, for example. In some areas there may be nothing to know: the chapters on oversight and control would have been particularly thin if limited to British examples. Then there is official disinformation, another area in which Britain excels. As Home Secretary, Leon Brittan spent two weeks investigating MI5, concluding that ‘the Security Service has carried out no operation, investigation, surveillance or action against any individual otherwise than for the purposes laid down in its directive.’ The words ‘lying’ and ‘bastard’ come to mind, as they say. Gill takes a deep breath and contents himself with observing that ‘such an inquiry, to be carried out properly, would have taken many researchers many months.’
Readers of this book will have been aware that we’ve got a problem with our secret state — ‘summary killings and disinformation campaigns against elected ministers raise major questions about the organisation, control and oversight of security intelligence agencies’, as Gill observes. The question is what kind of a problem? I sense that Gill may have set out to write an argument for institutional reform — structure, function, history of abuses, underlying causes, proposals for change: instead, he has succeeded in documenting the barriers to be surmounted before such a reform can even be considered.
This has the merit of giving us something to focus on. Even relatively minor reforms — abolition of the Security Service or the Special Branch, say — may be too much to hope for from any prospective government; but a Bill of Rights, reform of the state’s prerogative powers, a Freedom of Information Act and a few other cheap and simple measures — Gill’s proposed tribunal of state for one — could be enacted by any government with pretensions to reform. The reform of the secret state would be as far off as ever, but at least we’d be able to establish what needs doing.