Ken Robertson 1947–2021: foundations, innovations and legacies

Oversight is no substitute for adequate internal mechanisms for ensuring effective and efficient intelligence.¹

Ken Robertson, 1988

On the morning of 14 December 2021 the UK’s intelligence studies community lost one of its most important founding architects, Dr Kenneth Gordon Robertson. Ken was one of the first British academics to make a career of the scholarly study of the intelligence services and their role in government. He was also one of the founding members of the Study Group on Intelligence (SGI).² From the beginning, however, he stood apart from his contemporaries as the only social scientist working on a subject dominated in Britain by historians. Moreover, Ken was not a political scientist or international relations scholar but a sociologist, something that would prove both his greatest intellectual strength but his most difficult professional vulnerability. For much of the first decade of his career, he would remain the only social scientist in the British intelligence studies community. As a small number of others became involved at the cusp of the 1990s, he still found himself the sole sociologist in the field, at least until I registered as his PhD student in 1992 and became the only other sociologist. Ken was, therefore, a trail-blazer, mentor and occasional dissenter in the founding years of British intelligence studies.

Ken was born in Aberdeen on 14 March 1947 taking his first degree at the University of Aberdeen where he won the Milne Prize in Economic History. From there, he moved to the University of Reading to take his MA and then PhD under the supervision of Professor Stanislav ‘Stash’ Andreski. Andreski was a formidable figure, a Polish war hero and ardent anti-Communist, a combative scholar and founder of Reading’s Department of Sociology. Ken was, to all accounts, his premier disciple and while still engaged on his doctoral work, Ken was awarded a lectureship at Reading’s Department of Sociology in 1979. Ken was no less a keen and incisive a critical intellect than his supervisor, but was by temperament more understated and urbane. These would prove valuable assets of personal style to have at a time when scholarly study of the intelligence services was in its infancy in Britain. The subject was still regarded with suspicion backed by draconian official secrets legislation in official quarters, and in academia as both methodologically unfeasible because of that secrecy and too grubby and disreputable in its substance to concern respectable scholars. To Ken, however, these reasons were precisely why it was so important make intelligence studies happen in this country.

Phase 1: the sociology of secrecy

Ken Robertson came to intelligence studies somewhat obliquely through his doctoral work on official secrecy. He once admitted that his early, youthful ambition had been to craft an omnibus ‘sociology of secrecy’. This quickly proved too intractably complex and diverse a question to fit within the confines of a doctoral project (or, indeed, within the grasp of any individual scholar).³ Amidst the political scandals and public controversies of the day, however, he realised that the debate over official secrecy in government was both amenable to sociological scrutiny and in need of a dispassionate voice parsing the issue. Throughout his career, Ken’s work would be shaped by a fundamental distrust of conventional wisdoms and the partisan orthodoxies of disputing camps and schools of thought. And this formed the temperamental, as it were, point of departure for his approach to the official secrecy debate. By and large, he noted, that the debate had polarized into fairly stable pro and con camps arguing over whether official secrecy was, bluntly put a necessary thing with specific limitations or a bad thing with selected exceptions. Ken was struck by the fact that the exceptions on either side tended to be matters with which the authors had an ideological sympathy rather than following logically from either the pro or con case.⁴ In caricature, those in favour of secrecy (of which there were few in scholarly circles) found justifications largely in reasons of state and those against because of a presupposition that secrecy was required chiefly because of either bad conscience or the desire to monopolise knowledge as power. Ken’s argument was that one first needed to ask the question of why official secrecy existed at all, and then why it manifested itself to varying degrees in different democratic countries. Once one empirically identified the drivers of secrecy one would then be better positioned to consider one’s options for greater openness and transparency.

There are two especially important aspects of Ken’s approach to his work that appear in this early phase and which need to be appreciated at this point because they would prove vital to his later work and arguably one of the most important legacies of his thinking in the evolution of British intelligence studies. The first of these would be a consistent theme throughout his career, and that was the importance of comparative method. This was to avoid the potential idiosyncrasies of isolated cases studies, and to better flesh out common functions and forces underlying different versions of the same institutions and their processes. The second is that Ken’s sociology was always that of what Robert Merton called the ‘middle range’, that is, empirical research ‘intermediate to general theories of social systems which are too remote from particular classes of social behaviour, organization and change to account for what is observed, and to those detailed, orderly descriptions of particulars that are not generalized at all’.⁵ Ken was little inclined to become bogged down in abstract theoretical arcana dissociated from the ‘work-a-day world’ of day-to-day practicality at the coal face of institutions and processes. Ken’s sociology was fundamentally empirical, evidential and observational. To be sure, his later work (as we shall see) would adamantly make the case for conceptual thought on intelligence modelled on American scholarship and the US intelligence community’s ‘intelligence theory’ but, as we shall see in greater detail below, this was theory as a middle-range conceptual instrument applied to very empirical questions. To be sure, Ken may not have explicitly invoked the Mertonian ‘middle range’ but he was profoundly influenced by Chicago’s Edward Shils (himself an erstwhile intelligence ‘practitioner’ from the wartime US Office of Strategic Services). Shils was a product of the same US sociological paradigm and professional community as Merton, and a significant proportion of whose work occupied Merton’s intermediate realm.⁶ Much of Ken’s early work was conducted during Shils’ interval at Cambridge, during which time Ken fell into his intellectual orbit and came to regard Shils as something of an inspiration and mentor as well as intellectual influence.⁷ In the latter years of Ken’s career, he did indeed eventually engage with theorists like Michel Foucault and Anthony Giddens, but always to drag their rarified formulations back down to the gritty nuts and bolts world of what might be termed, in an appropriately sociological idiom, ‘doing’ intelligence.

Neither bad conscience nor domination

Ken’s 1982 Public Secrets: A Study in the Development of Government Secrecy was an exemplary application of what John Stuart Mill termed the ‘argument from difference’, or what is today rather oddly termed ‘most similar systems’ (MSS).⁸ Setting Britain as case for a high level of official secrecy, the USA at opposite end of the spectrum, and Sweden as an intermediate case, Ken’s design was elegant and his results more plausible and robust that some more high profile comparative studies of the same period.⁹ They also met the Durkheimian sociological goal of detecting and demonstrating social forces at work of which members of the affected communities might only be dimly aware, if at all.

What Ken looked for were drivers of variation. What the evidence indicated, he argued, was that the level of government secrecy was not chiefly a consequence of ‘ideology, inefficiency corruption, national security, good or bad politicians, right- or left-wing government, pressure groups or paternalism’. Rather, ‘the level of secrecy is affected by two factors: the degree to which the civil service or administration is clearly subordinated to elected representatives, and the extent to which that responsibility is undivided and hierarchical’.¹⁰ This is not to say that those in power were never guilty of malfeasance that they might conceal with official secrecy, nor that there was no compelling case for secrecy in national security and national strategy. Rather, secrecy reflected the consolidation of power and the need for national government to speak with one voice or not. Consequently, the evolution of Britain’s Cabinet system of government required that the Home Civil Service be silenced to avoid officialdom becoming a competing public voice that might challenge the judgment and authority of Ministers from a putative position of greater substantive expertise and experience. On the other hand, the checks and balances of the US Constitutional model created an incentive for the different branches of government (especially Congress) to flush official information out into the open and use it to test, challenge and limit the authority of the other branches and their actions in the name of the nation as a whole (especially the President and Executive Branch). In Britain, therefore, secrecy was an unintended consequence of Parliament wresting power away from the Crown, while in America openness was an unintended consequence of setting ambition against ambition to forestall tyranny.

In a sense, Ken took secrecy rather than openness as the default condition and examined its fluctuation. This was quite an ingenious and counterintuitive manoeuvre, and actually quite a reasonable ploy given how individuals and households naturally resist disclosure of their own internal affairs at least as jealously, albeit with less enforceable power, as firms and governments. After all, from the perspective of 2022, what are privacy activists in the post-Snowden era doing if not defending the individual’s entitlement to secrecy?

Phase 2: from secrecy to secret services

In retrospect, it appears an almost inevitable evolution that a concern with official secrecy in the state would lead naturally to an interest in the secret offices of state. In the event, an opportunity to spend the 1982–1983 academic year attached to the University of North Carolina as a Visiting Professor while Public Secrets was in press cast the die. Ken took to America and the American academic world with enthusiasm. While there, he developed a close intellectual relationship with Georgetown University’s Roy Godson and his well-resourced and well-connected Consortium for the Study of Intelligence. One of Ken’s early initiatives would be a 1984 transatlantic conference in collaboration between SGI and Godson’s Consortium, an event that would set something of a precedent in having a retired Deputy Chief of the Secret Intelligence Service give a public lecture on (his personal view of) intelligence¹¹ and provide the materials for Ken’s well-received edited 1987 volume British and American approaches to intelligence.

Ken’s essay in the 1987 volume was the bridging work between his studies of secrecy and of intelligence.¹² As in his doctoral work, he avoided the convenient route of conventionalised ‘pro’ and ‘con’ arguments, or as he would later put it, ‘democracy “versus” intelligence’. Instead, he built explicitly and closely on his analysis of the drivers of secrecy and disclosure developed in Public Secrets, fusing that with the conceptual armature of intelligence theory that he had encountered while at North Carolina. Looking across the two national intelligence systems he identified the common tasks both communities needed to perform, at the heart of which (and to which we shall return) was requirements. ‘The debate over secrecy’ he warned ‘must be informed by a discussion as to the nature of intelligence requirements and the tasks which intelligence services are to perform’.¹³ All the other activities of intelligence – collection, analysis, getting the right message to the right people, management of funds and so forth – were ultimately driven and shaped by those requirements.

There were, he argued, intrinsic features of intelligence that made secrecy necessary. Arguably the two most important were the security necessary to protect operations from adversaries’ countermeasures, and the need for advice to policymakers to be confidential, firstly to avoid internal policy discussions from being politicized and secondly because the eventual decisions taken are the policy-maker’s responsibility and not those of their advisers. The commonly expressed concern from governmental quarters about a ‘slippery slope’ of uncontrollable progressive exposure was largely about ensuring the integrity of the other two. But there were differences in the degree to which these motivations for secrecy pertained in Britain and the United States that were special cases of the dynamics he had examined in 1982. Official resistance to increased transparency concerning British intelligence, he argued, was largely a special case of the existing political and Constitutional drivers of secrecy amplified by the specific secrecy needs of intelligence. On the other hand, in the USA, interagency tensions and rivalries between intelligence organizations acted as a force multiplier on the inherent centrifugal tendencies towards disclosure already generated by America’s Constitutional architecture.¹⁴

Robertsonian realism

Ken had a distinctive intellectual idiom and style in his work, but because he retained such a trenchant distance from intellectual camps and orthodoxies he struggled to give a name to that approach. As we have seen, what he particularly opposed were arguments about real-world policies and actions derived from first principles in the abstract in preference to thorough empirical investigation and characterisation of the problem at hand. Consequently, he made some limited forays into articulating his approach as what he called ‘realism’. ‘Realism’, of course, is a fraught term to use at the best of times, carrying with it an assortment of highly charged associations from debates in other fields, especially international relations (IR) theory. Ken’s idea was not, however, intended as a simple appropriation of the IR use of the term, and he clearly hoped to flesh his version of ‘realism’ out in greater detail in due course, refining it significantly in his final monograph in 1999.

To be sure, in his initial effort he emulated Hans Morgenthau’s original juxtaposition of ‘realism’¹⁵ as a theory of how the world actually works as opposed to normative ‘idealist’ theories of how it ought to work. In the editorial essay of a 1987 special issue of Defence Analysis entitled ‘An Agenda for Intelligence Research’, he observed that American intelligence studies had become two clearly demarcated camps with their own orthodoxies in which

The Idealists have concentrated on ethics and democracy theory, focusing on such issues as accountability, constitutional law, and the morality of intervention with little concern for questions of efficiency and effectiveness. The Realists have concentrated on performance and producing intelligence theories to the neglect of issues which concern Idealists.¹⁶

This is clearly not ‘realism’ in the typical IR sense, and to some degree the terms were being used at this point as flags of convenience or as near approximates for what Ken was trying to get at.

In 1999, he revisited and sought to generalise the notion, and bring it into closer alignment with the kind of critical analysis that had characterised Public Secrets. In Secrecy and Open Government ‘Idealism’ was replaced with ‘normative’ and ‘surveillance theory’ approaches. In his view, normative theory descended quickly into arm-chair moralising,¹⁷ while surveillance theory rested on an a priori unexamined and untested presupposition of the notion of ‘hegemony’ and its variations thereby ‘creating a picture of sinister forces manipulating citizens for some vague, and unspecified, purpose’. Ken saw theories of surveillance and hegemony as substituting something akin to conspiracy theory for any detailed empirical analysis of structures and processes. Power, he acknowledged, was a correct and necessary point of departure but not because of a caricature notion of a collusive, ‘greedy state’ (as he put it). Rather, it was because democratic governments ‘live and work in an environment of competing interests’ in which ‘it is the fragmented and diverse nature of interests in the democratic states that give rise to anxieties about power’.¹⁸

The ‘realist’ approach Ken advocated was concerned with developing a clear and detailed empirical understanding of matters like secrecy and intelligence in preference to vague accounts dependant on sweeping theoretical generalisations. The principle was to investigate the why and how at a very specific level first, and derive the pro or con position as a result. Consequently, he asserted that his realism approached ‘secrecy as a reflection of power, not values, and believes that there are almost as many reasons for secrecy as for disclosure’. ‘The realist’ he asserted ‘takes the view that secrecy is not a matter of conviction but of the balance of forces within the political system’ and that ‘changes in the degree of secrecy will only follow changes in the balance of power’.¹⁹ Ken’s dissatisfaction with the prevailing accounts of secrecy was the same as that driving his rejection of a ‘traditional discussion’ of the intelligence services that ‘has too often been focused on democracy “versus” intelligence, and accountability rather than management’.²⁰

It is, therefore, perhaps most useful to see Ken’s notion of realism as rooted not in some version of IR realism but in the ‘sociological realism’ of Emile Durkheim or August Comte. In sociological realism, social phenomena like institutions and cultures have an ontological status of their own as what Durkheim called ‘social facts’, with their own functional principles and prerequisites not reducible to the individuals navigating and making choices within those institutions and cultures. There is an irony here because Ken firmly placed himself and his focus on public bureaucracies and organisations within the competing ‘classical’ sociological tradition of Max Weber.²¹ In the last analysis, such arcane theoretical distinctions mattered less to Ken than generating granular and accurate accounts of official secrecy and the intelligence services.

Ken Robertson versus the ‘British School’

Even at the turn of the 1980s British intelligence studies was already clearly crystalizing around a historical orientation, interests and methodological conventions. As the solitary social scientist, Ken picked up on this emerging orthodoxy, both identifying and challenging it in print the year before Donald Cameron ‘D.W’ Watt’s influential 1988 review article announcing the emergence a ‘British school of intelligence studies’.²² In a striking instance of the sort of irony that would nag and frustrate Ken throughout his career, Watt pointed out with approval Ken’s British and American Approaches to Intelligence as ‘a collection in which British historical and American contemporary policy approaches are shown side by side’ – but then failed to note that Ken’s own contribution completely failed to fit the British mould!

The Approaches volume was not the only place in which Ken took up the issue of the difference between intelligence studies on the two sides of the Atlantic. In is 1987 ‘Agenda’, he pointed out that the differing approaches were not intellectual accidents of history. Where American discussions were shaped by ‘allegations of abuses’, British writing either focused on ‘heroic’ revelations about wartime special operations, ‘Double Cross’ and ULTRA, or on bitter disclosures of repeated and often long-term Soviet penetration of the ‘establishment’.²³ That the approaches in Britain and America were so different was, therefore, almost inevitable. But, in Ken’s view, there were costs and trade-offs attached to taking either route and each tradition needed to learn from the other.

Ken made it his mission to import American intelligence theory into British intelligence scholarship. In his ‘Agenda’ piece he criticized America’s intelligence studies Idealists and Realists for largely talking past one another and for an interest in historical research ‘only in so far as it can be plundered for “lessons”’, judged by ‘post-McCarthy and post-Vietnam’ values and a ‘handmaiden to controversy’.²⁴ On the other hand

the study of intelligence in the USA also has many strengths, one of which is the theoretical sophistication which exists in such concepts as requirements, surprise, analysis and surveillance … The development of conceptual frameworks for the analysis of intelligence has meant that many of the problems associated with it have been accurately identified and discussed at a theoretical level. No other society has placed so much effort in this direction, or achieved so much good sense.²⁵

By contrast, Britain’s historical work had produced a commendable emphasis on factual evidence and accuracy, and largely discussed intelligence in its wider context and relationship to events. On the other hand, British scholarship

has not produced a clear and accurate description of the organization, politics and activities of British intelligence, despite some valiant efforts … Furthermore, there is no agreed framework, or even choice of framework, for the study of intelligence in the UK … this has left historians without a common vocabulary and without a conception of ‘good’ intelligence practice with which to assess the past.²⁶

The agenda Ken set out was as much a statement of personal intent as it was a gentle chiding of friends and colleagues on both sides of the Atlantic.

There can also be little doubt that the sociologist in Ken had seen something in the American conceptual armature that even the architects of intelligence theory may not necessarily have intended. It offered a schema for identifying and understanding the raison d’être of intelligence and its functional exigences that could be used to interrogate and evaluate how and why specific agencies and specific intelligence communities existed, what their operation needed to entail, and to evaluate their fitness for purpose. What American intelligence theory offered was just the sort of empirically oriented analytical framework that would be perfectly suited to a Mertonian sociology of the middle range.

Study Group(s) on Intelligence

Ken was also a keen institution-builder, not as a self-regarding builder of empires but as someone committed to establishing an intellectual commons for the study of intelligence. Around late 1981/early 1982, he had already set about contacting fellow scholars and retired intelligence practitioners sympathetic with the idea of well-informed academic and public debate on intelligence organisations and their roles in government and politics.²⁷ It was during this interval he and Christopher Andrew first crossed paths and the development of SGI became a collaboration that would continue throughout Ken’s career. At the time, Andrew had the edited volume The Missing Dimension²⁸ in press whilst Ken was awaiting publication of Public Secrets and was about to set off to his formative year at North Carolina. Ken would consequently find himself abroad during much of the early discussions around the establishment of the Study Group. By 1983, Ken and his allies found that they already had a starting mailing list of 40 potential participants. The process was not an easy one, with on-going resistance and suspicion from official quarters, and even some concern that 40 participants might be too large to be manageable given the expected sensitivity of the discussions. But both momentum and confidence quickly gathered, and SGI had its first meeting on 12 December 1983.

SGI subsequently established itself as not only the first intelligence studies body in the UK but the principal setting for British academics to engage with current as well as former UK intelligence officials. In its initial form, SGI had a minimum of formal organization and an equally minimalist executive that consisted of D.C. Watt as the Chairman, Christopher Andrew as Deputy Chairman, and Ken Robertson as Secretary. In 1994, the executive underwent something of reorganization under which Andrew became Chairman and Ken Vice-Chairman. If SGI had a significant limitation, however, it was its relative closure. SGI was, and would remain, a primarily invitational entity. This was a necessary quid pro quo to assuage the sensitivities of both government points of contact and the former intelligence officers who were involved. Indeed, SGI members were prone, between themselves, to liken it a gentleman’s club in idiom and almost in exclusivity. This would become a crucial consideration in Ken’s next institution-building venture.

In 1993, Ken and I wrote²⁹ to the UK Political Studies Association (PSA) to propose a panel at the upcoming PSA Conference in Swansea. In response, the PSA invited Ken to establish a PSA Specialist Group on Intelligence and Security Studies. The opportunity to create an institutional base for intelligence studies as an undertaking within political studies was not to be missed. A PSA Specialist Group³⁰ would create an academic forum that could be more open than SGI and therefore have greater potential for shaping a wider but still well-informed and social science-oriented discussion about intelligence. Ken set about giving shape to the PSA Security and Intelligence Studies Group (SISG) as its first chair, retaining that role until his retirement from academia in 1999. Under Ken’s leadership SISG grew rapidly and was quite a substantial enterprise by the time it was passed on to a new leadership team in the form of Peter Gill and Mark Phythian who continued to grow SISG into a sizeable, sustained and highly productive operation.

Methodology and the comparative agenda

Ken’s 1987 ‘Agenda’ entailed more than just encouraging a fusion of British and American approaches to intelligence studies. It also made the case for a specific sociological contribution through methodological sophistication and research design strategies. Academics on both sides of the Atlantic frequently complained about the lack of documentary sources, and Ken saw this as too often an excuse ‘for substituting speculation’. What was needed, he argued, was better use of the available documents, ‘structured interviewing of former intelligence officers’, and the adoption of comparative methods.³¹ This is, of course, exactly what he did in his work on the UK and the USA, but this was really only a starting point.

In 1988, Ken secured a Canadian Studies Research Award from the Canadian High Commission in London. This led to a thorough and almost mischievous article that, in Ken’s inimitable style, once again challenged the established intellectual comfort zones of the prevailing camps of thought and debate regarding Canada’s intelligence agencies. In particular, his goal was to deflate what he perceived as a certain ‘normative’ and idealist conformity of thought in Canadian intelligence literature. At the time – and arguably throughout subsequent decades³² – Canadian intelligence writing was been dominated by the moralizing conventional wisdoms of a discourse framed almost entirely in terms of ‘intelligence ”versus” democracy'. As in the United States, public discussion and awareness of intelligence in Canada had taken shape during the 1970s amidst a sustained furore over intelligence abuses. In Canada’s case, this was driven by allegations of misconduct by the Royal Canadian Mounted Police (RCMP) Security Service. Focused on a Royal Commission led by Judge David McDonald, the furore had culminated in the abolition of the RCMP Security Service, its replacement with a new Canadian Security Intelligence Service (CSIS) and the establishment of an independent oversight body called the Security Intelligence Review Committee (SIRC).

Applying much the same middle range analysis that he had employed on the UK and USA, Ken warned that ‘concern with “intelligence abuse” … diverted attention away from the role of intelligence in policymaking and the requirements of politicians for good quality and relevant intelligence if they are to make “sensible” decisions on a whole range of policy issues’.³³ ‘The creation of CSIS and … SIRC’ he admonished ‘can easily mislead an observer into believing that change outweighs continuity but an examination of the total Canadian intelligence effort shows instead striking continuity’.³⁴ CSIS, he argued, was ‘in many ways peripheral’ and ‘an ad hoc addition’ to Canada’s intelligence enterprise.³⁵ Rather than starting with the cause celebre of abuses and putative accountability deficits, Ken’s point of departure was to ask about Canada’s national security posture, how this drove national intelligence requirements, how the community was organised and managed, and only then did he examine accountability, oversight and the role of CSIS.

From a policy support point of view, he argued, CSIS was neither the largest nor most important component of Canadian intelligence community. Far more significant, and entirely untouched by McDonald or new legislation, were Canada’s signals intelligence agency, the Communications Security Establishment (CSE) which was ‘arguably by far the largest and most important member of the Canadian intelligence community’, the Department of National Defence Intelligence Division (DID) and the analytic capabilities in External Affairs and the Privy Council Office.³⁶ Domestic security intelligence may have had a shaking up, but the lion’s share of intelligence activity, infrastructure and expenditure in Canada was almost entirely scandal free and carried on much as it had before the scandals of the 1970s.

Axioms and core concepts

While he used them to good effect at home and abroad, Ken did not just uncritically import US concepts and ‘intelligence theory’. He was aware that there were keenly argued, long-running disputes about basic concepts, definitions and the fundamental role and function of intelligence in government. For example, while almost all of his work drew on the essential constituents of the so-called ‘intelligence cycle’ it is striking that he never once used the term. He was also well aware of the debate over the proximity of intelligence to policymaking between what Arthur Hulnick has called the ‘traditionalists’ and ‘activists’.³⁷

It was while parsing such crucial definitions and terms of reference that Ken crafted his best remembered pithy proposition – that intelligence should be defined as the ‘secret collection of someone else’s secrets’.³⁸ So catchy was this axiomatic meme that today one can find it not only invoked with gratifying frequency, but less gratifyingly attributed to all sorts of grandees of intelligence studies other than its originator. Less well remembered is a slightly revised version he also offered in 1987, that is, ‘the collection of other people’s secrets and the protection of one’s own’,³⁹ a reminder of the vital importance of counterintelligence and its indispensability in the intelligence enterprise.

Ken’s emphasis on covert collection did not arise from a simplistic equation of intelligence with espionage or an interest in generating revelations about what took place behind the veil of secrecy. Nor either did he dismiss the significance of open source intelligence (OSINT). However, he pointed out, ‘although the gathering of information from open sources is a vital and significant part of all intelligence activity, it does not form the distinguishing character of such activity’.⁴⁰ The notion of ‘distinguishing character’ is crucial, as Ken’s concern was to identify how intelligence entailed different issues from common and garden-variety information collection activities like ‘the businessman or the housewife planning their budgets or their next vacation’.⁴¹ Intelligence became a matter of public concern chiefly because it entailed the covert use of what we would today term intrusive investigatory powers. ‘Most people’ he argued ‘would be willing to concede that if one could show that a particular response by the state was harmless to civil liberties, the state should be allowed wide discretion in its use’.⁴² OSINT, at least at that time,⁴³ had at most minimal civil liberties and consequently disclosure or regulatory implications.

But even if intrusive potential were the distinguishing feature in intelligence, Ken further argued this was a consequence of an even more fundamental issue. That more fundamental issue was the information needs that drove the establishment and operation of intelligence services. Intelligence services were, he was fond of pointing out, expensive, difficult and often risky organizations to establish and operate. Such an enterprise required a compelling motive to brave the risks and pay the costs. The basic driver in Ken’s analysis was, of course, requirements and, underlying those, the presence of threats. It was axiomatic to Ken that ‘a threat is not simply an unknown factor which may affect one’s interests but something capable of causing harm or injury’ and where ‘the seriousness of the threat depends on the degree of harm … and the likelihood of the threat being carried out’.⁴⁴ Consequently, if intelligence were defined as secretly acquiring the secrets of others then ‘an intelligence service is an organization devoted to collecting, by clandestine means, the secrets of those who have the capacity and intention to inflict harm on the interests, goals and values of a nation state’.⁴⁵ In fact, the heart of the matter was not the covert collection itself, but the intelligence requirement it was undertaken to meet.

Requirements, transparency and fitness for purpose

The question of intelligence requirements was, therefore, Ken’s skeleton key for understanding almost everything else about the establishment, operation and governance of intelligence organizations. Arguably this is where Ken differed most fundamentally with surveillance theorists by rejecting the notion that governments collected intelligence sui generis because knowledge is power and secretly collecting other people’s secrets therefore a natural and inevitable handmaiden to hegemony. Intelligence agencies existed because of intelligence requirements and priorities which, in turn, largely dealt with threats. Tasking processes and mechanisms were vital to managing intelligence because ‘[c]ollecting unnecessary information is wasteful, inefficient, and creates unnecessary risks’ and trying to capture ‘all’ potentially useful information ‘is to create a bureaucratic monster, producing much employment but little else’.⁴⁶ Intelligence requirements were both raison d’etre and raison de faire to the intelligence community.

Furthermore, he proposed, many of the debates about the activities of intelligence services were less matters of dissatisfaction with the fact that the agencies existed and operated than the ends towards which they were directed. Consensus around the nature and existence of threats was fundamental to any sense of public consensus around the activities of the agencies. ‘Moral disagreements’ about intelligence activities ‘are often also policy disagreements’.⁴⁷ Such consensus was, for instance, more likely to be forthcoming about a wartime adversary than terrorists in peacetime whom the public tend to see as a law enforcement problem rather than an existential menace.⁴⁸

Requirements also lay at the centre of his commentary on the relative stability of Canadian intelligence. Canadians during the Cold War may not have seen themselves principal or front-line participants in that struggle but the government had ‘no wish to be seen as the “soft underbelly” of North America and NATO’, and ‘Canada’s ready willingness to utilize its armed forces for the purposes of peacekeeping … means that it has intelligence requirements concerning many parts of the globe’. ‘Shifting patterns of demand and in the nature and type of refugees’ also had potential ‘strategic consequences’ that drove intelligence requirements. Domestic terrorism might have been ‘in remission’ and foreign espionage and subversion remote and abstract matters to the citizen in the street, but even Canada’s non-confrontation international commitments entailed the need for a substantial, sovereign intelligence capability and effort.⁴⁹

That being said, Ken was fully aware that the requirements process was fraught with perverse incentives. Tasking was often ‘the least formalized and coherent part of the intelligence process because of the incentive each party, collector or user, has to avoid any blame for the deficiencies that may exist’. Users had a motive to ‘deliberately create uncertainty’ so that they might ‘blame the collectors for collecting too little information or the wrong kind of information, thus avoiding their own responsibilities’. By the same token, the agencies had an interest in receiving requirements vague enough to ‘give them more flexibility or room to manoeuvre’.⁵⁰ The danger was that

Many of the abuses which have occurred and the fears which they have generated have arisen as a result of a lack of clarity over: the nature of the threat which society faces, the purpose for which information is being collected and the uses to which it is being put. The responsibility for this situation lies with legislators and the executive who have failed to give proper guidance in these key areas. An effective intelligence service is less dangerous to civil liberties since it is less likely to collect information which is not related to a specific goal and for which there is no clear purpose.⁵¹

The problem with security intelligence

An often inadequately acknowledged aspect of Ken’s work was a persistent concern about the governance of domestic, security intelligence organizations and activities. His concern on this front was less about the existence or extent of intrusive powers in domestic intelligence as such, but their management and the degree to which they failed, or appeared to fail, to meet his criteria for effective intelligence management and control. Of the British system, he complained that for foreign intelligence ‘There is a structure, the JIC [Joint Intelligence Committee], which deals with tasking, analysis, and dissemination of the product, but no equivalent mechanism has ever been claimed for domestic intelligence’.⁵² He also raised a similar concern about the USA albeit in less detail.⁵³ And, while his discussion of Canada was intentionally dismissive of the widespread fixation on oversight and CSIS, he was very concerned about the governance of security intelligence in the Privy Council Office. While there was top-level provision for foreign intelligence coordination, tasking and analysis through the Intelligence Advisory Committee, the domestic Security Advisory Committee ‘has no role in producing assessments nor any responsibility for tasking or coordinating security operations’.⁵⁴

He further observed that the suite of collection disciplines required for domestic intelligence differed from foreign intelligence. Both might share ‘interception of communications, the placing of “bugging devices” and the use of human agents’ but ‘the use of spy satellites and … intelligence from radar signals is generally limited for foreign intelligence services’. Moreover, security intelligence was more likely to be concerned with support to law enforcement than ‘evaluation of trends and long term planning of resources and responses’ and ‘less likely to involve the military, the diplomatic corps, the specialist coding and cryptography departments’ which therefore ‘reduces the need for interagency bodies’ with ‘less of an incentive to form specialized analysis bodies’. Nonetheless, he argued ‘these are not good reasons for the neglect of analysis within domestic intelligence’ and concomitant risks to both the quality of intelligence products and to civil liberties.⁵⁵ Given that domestic security services operated within a nation state’s own civil space, effective management and supervision of their activities had more significant implications for civil society than foreign intelligence. Despite this, there was a persistent and widespread lack of convincing and demonstrable effective internal, governmental supervision and direction comparable to that for foreign intelligence.

Intelligence and open government

It is important to keep in mind that while Ken may have been no friend of ‘moralizers’ and theorists of hegemony, much of his career had, in fact, been conducted in pursuit of greater openness about intelligence. SGI had, of course, been established to foster a better-informed academic discussion of intelligence, and SISG to do much the same thing in a significantly more open and accessible fashion. Indeed, in 1989 he made a highly effective case for greater government openness, noting of the then-prevailing doctrine of the intelligence agencies being accountable to Parliament via Ministers:

Ministerial control can only be considered adequate if the mechanisms exist to make such control a reality. A statement that Ministers exercise control cannot be persuasive if the public is unaware of how this control operates in practice. This constitutional device is only credible … if the principles and organizational structures are available to Parliament and therefore to the public.⁵⁶

It is important to keep in mind that what Ken advocated in the first instance was not specific forms of external scrutiny and intervention, but better communication and understanding of the systems that were in place, and more credible evidence that they worked and how. ‘The first priority’ he admonished ‘must be effective internal control with the desirability of external control, whether through new laws or committees, coming second’. Emphasizing safeguards over effective management and direction risked building an elaborate oversight architecture on administrative quicksand, potentially ‘constructing safeguards for an instrument which performs no useful tasks whatsoever’.⁵⁷

That being said, he was pessimistic throughout the 1980s about the structural, political forces limiting or preventing any dramatic changes in official secrecy or intelligence oversight in the UK. British public awareness and interest in intelligence was, as we have noted, shaped principally by historical and, in the case of the Falklands conflict, more recent revelations of failure and success rather than significant, plausible allegations of intelligence abuses. The various options available in the 1980s, such as an internal, senior Civil Service committee of ‘wise men’ would be too closed to provide credible ‘public’ assurance, while a Parliamentary committee ran the risk of being dragged into the partisan political hurly-burly of the Commons.⁵⁸ In the meantime, the British Constitutional predisposition to secrecy offered little likelihood of change to the structural drivers of official secrecy as they affected the intelligence community.

Ken was not, therefore, averse to some form of external oversight but he was concerned that if it developed it should not pre-empt effective internal direction and assurance of efficiency and effectiveness. Here again, he sidestepped conventionalised arguments for and against intelligence openness. Rather, he acknowledged on the American precedent that legislative intelligence oversight was entirely feasible but if it was going to develop it would come with complement costs and unintended consequences that transparency advocates had to acknowledge, and people had to be willing to live with those trade-offs. ‘Bringing intelligence into current political debates’ he warned in 1987 ‘has the result of “politicizing” the intelligence services in a way that is undesirable’.⁵⁹ In the event, transformations in international governance standards and pressures at the European level, and a post-Cold War suite of intelligence requirements with more direct relevance to the average citizen such as terrorism, organized crime and the intelligence needs of peace support operations, did indeed drive changes in the distribution of power interests that made external oversight feasible. At the same time, the establishment of Parliament’s Intelligence and Security Committee (ISC) created something of a converse to his earlier concern about openness as politicization. Rather than simply a vehicle for external scrutiny of the agencies, he warned, the Committee would tend to act as interlocutors between the intelligence community and civil society and a new forum for Whitehall organizational politics to play out. Consequently ‘the existence of the ISC has created a new avenue for influencing opinion – if you can convince the Committee that your proposals are the right ones they will do the advocacy for you!’⁶⁰ Not without reason did Ken in private often refer to his form of realism, only half in jest, as ‘cynical realism’.

Phase 3: full circle and the closing act

During the next half decade, Ken would move laterally in his discussion of intelligence, away from national intelligence communities and into the wider study of intelligence and related institutions. Here again, Ken would prove himself ahead of events and an effective critic of conventionalised lines of debate. He became involved with a suite of policing and law enforcement issues and produced a series of pieces on the emergence of law enforcement intelligence just as British policing was giving a shape and name to its National Intelligence Model (NIM). Also using policing and law enforcement intelligence as his route of entry, he conducted an incisive analysis of the security implications of the impending 1992 Maastricht Treaty and its commitments to open borders and the free movement of people and capital. Here, he accurately forecast the establishment in the UK of what would become the National Crime Agency.⁶¹ By the mid-1990s, dramatic changes in British governmental transparency and oversight of the intelligence community prompted Ken to return in his final academic years to the topic of secrecy from which he had begun. In the short but crisply reasoned Secrecy and Open Government: Why Governments Want You to Know (1999), he argued that the changes in official secrecy, especially in the UK, arose from significant changes in the balance between various power interests, the implications of an ever more information-intensive society and a post-Cold War suite of intelligence requirements where the Security and Intelligence Agencies actively needed to communicate their missions to the public more systematically. As his academic career drew to a close, he permitted himself a final, parting shot at the case for open government inspired by a fear of governmental abuses of power:

One has only to consider that in the USA Watergate occurred after the [Freedom of Information] Act of 1966, and that, although the Act was strengthened in 1974, this did not prevent the Contra Affair, the arms to Iran scandal, or allegations of sleaze in the White House … one could go through the list of all of the main scandals that have affected an American president since the Second World War and show that FOI was irrelevant to every single one.⁶²

Sea-changes in sociology

Ken Robertson struggled throughout his career against a succession of challenges. At the beginning, he wrestled with official resistance to scholarly engagement with intelligence, and academic prejudices against such an endeavour. As his work developed and the publications flowed, he struggled to locate himself in disciplinary terms. He was, as we have seen, at his core a sociologist, but he rarely referred to ‘sociology’ in his work on intelligence. When challenging the ‘British school’ and championing the conceptual richness of the ‘American’ he consistently referred to ‘political science’ or more vaguely ‘social science’. By the same token, in recent decades sociology has not been a hospitable disciplinary climate for any scholar to study intelligence unless their sole aim was to decry pervasive and pernicious surveillance and write the intelligence community off as the agents of power interests and hegemony. To the generations of sociologists trained after Alvin W. Gouldner’s 1970 Coming Crisis of Western Sociology, Merton and any other author redolent of ‘structural-functional’ thought were well and truly out of fashion and verged on anathema. Indeed, on one occasion in the mid-1990s, Ken had the galling experience of being derided during a British Sociological Association conference panel as a ‘Sunday Times sociologist’.

And, of course, Ken and his work were at their essence interdisciplinary and not the stuff of disciplinarily circumscribed orthodox sociology. He may have been a sociologist by training, but much of his work was political science, much of that belonged more to public administration and public policy than anywhere else, and a great deal of his analysis drew centrally on historical evidence. Despite persistent approving calls over the decades for interdisciplinary scholarship in almost every sphere, the intellectual and managerial stovepiping of academic work has always run against such efforts. And so, despite being a lecturer and, from 1986, senior lecturer in sociology at the University of Reading, he transferred his flag as far possible and as long as possible to Reading’s explicitly interdisciplinary Graduate School of European and International Studies. In that capacity he served as its director for a decade between 1986 and 1996 and almost all of his publications identified him this role rather than his position in the Department of Sociology.

Ken’s career also spanned a period of intense and profound change in the British academic environment, much of which proved increasingly inclement to his work and working style. Ken’s real forte was as a talented academic essayist. He was most adept at pithy, thought-provoking observations and singling out fragile pressure points in comfortable conventional wisdoms and orthodoxies, then adroitly poking them with often enviably quotable intellectual acupuncture needles. But he did not write stereotypical ‘sociology’ and did not publish in properly kitemarked sociology periodicals. Rather, he was in the sociological export business, applying what C. Wright Mills called the ‘sociological imagination’⁶³ to other fields and disciplines where it would provide alternative and innovative insights and sober second thought. But the British academic world was increasingly in the thrall of government-mandated audits of research in the higher education sector, at the time known as the Research Assessment Exercise or RAE. Intended to generate some sort of value for money appraisal of UK research, the RAE also served to intensify and reinforce disciplinary boundaries. There was little space in such a regime for a sociologist who published almost anywhere else in the social sciences other than sociology. Success in the RAE was essential, however, for a range of reputational and financial reasons, and pressure on academics to conform became relentless. Intelligence studies might need and benefit from sociology, but sociology did not need or want intelligence studies.

By late 1998 Ken had elected to leave academia, and with it the study of intelligence, and moved into the booming and far more lucrative world of private sector research and analysis. As he did so he remained determined to foster the future development of social science-oriented intelligence studies in the UK. In 2002, he donated the lion’s share of his personal library of intelligence materials to the University of Aberystwyth’s Department of International Politics where they became the Ken Robertson Collection in the Aberystwyth library. In appreciation, Aberystwyth established a prize in his name for the best undergraduate performance in the field of intelligence.⁶⁴ Beyond this, he drew a line under his academic career, and his passion for intelligence studies, and made a clean break to set out on a very different future.

Legacies and reflections

Ken’s work was, of course, shaped by the debates of the day, and limited by the granularity of the information available, especially about the structure and operation of the UK and Canadian intelligence communities. Some of Ken’s persistent concerns about domestic and security intelligence analysis and assessment reflected the lack of information in the public domain about MI5ʹs internal organisation and functions. Defence Intelligence, in those days officially designated the Defence Intelligence Staff (DIS), makes only two appearances in his work and then only in passing.⁶⁵ Less clear in the 1980s was the degree to which the JIC as a ‘national’ assessment entity concentrated on producing what Michael Herman – whom we also lost in 2021 – once called ‘high powered reports for high powered people’⁶⁶ while DIS actually did most of the analytical heavy lifting on national intelligence requirements for less august consumers. To a degree his concern about security intelligence analysis would today largely be allayed by our better understanding of the integral relationship between investigations and analysis in that sphere.⁶⁷ On the other hand, the UK has had a chronically difficult time coming up with a credible security intelligence tasking formula that can navigate the Scylla and Charybdis of effective and explicit tasking while ensuring independence from political interference.⁶⁸

It is impossible not to point out that the political use and abuse of intelligence prior to the 2003 invasion of Iraq was a catastrophic manifestation of the kind transparency-driven politicisation of which Ken warned in 1987 and 1988.⁶⁹ On the other hand, it is possible to take a certain smug satisfaction in the degree to which the ISC’s statutory marching orders were principally concerned with the kind of management and value for money concerns Ken advocated rather than the restriction and regulation of the intelligence community’s activities and powers. And finally, one also need only look at the ISC’s interventions on matters like diversity in the intelligence community workforce and the Investigatory Powers Act 2016 to see how the Committee’s role as scrutineer has so easily slipped from interlocutor to advocacy and influence on behalf of the agencies.⁷⁰

In his private politics, Ken was cipher. He was no natural ally of the ideologically committed left, but had as little time for the hidebound right with its time-worn petty prejudices and resistance to change. He loved fine dining and high culture, and was to all accounts a formidable and ruthless adversary on lawn bowling’s field of honour. And he was an academic by vocation, in its proper use as calling, as well as career. Of his sincere and selfless passion for the academic study and public understanding of intelligence there could be no doubt. One did not have to be one of his registered acolytes for him to be generous with his advice, support, assistance and encouragement as many then aspiring but now well-established scholars in the field will recall.

In late 2021, Ken fell prey to a cancer that he had previously fought into remission, and after a brief but courageous struggle succumbed at last. And while it has been more than two decades since he was an active voice in intelligence studies, Ken’s impact on the field was both profound and permanent. Today, few if any ‘British school’ historians of intelligence would ply their trade without first securing a command of the core concepts of the intelligence profession and an understanding of (still mainly American) ‘intelligence theory’. And some have made their own contributions to theoretical analysis and understanding of the intelligence function. That being said, social science scholars who work on intelligence in Britain remain woefully few and far between. Nonetheless, in their impact on both the literature and the governance and professional practice of the intelligence community they have consistently punched well above their numerical weight. In the last analysis, Ken’s contribution to intelligence studies in Britain has been so fundamental that his methods and ideas have become part of the tacit knowledge and conscience collective of the field. We draw constantly on his thought and insights today without really being consciously aware of doing so.

K.G. Robertson: selected and key publications

• Secrecy and Open Government: Why Governments Want You to Know (London: Macmillan, 1999)

• War, Resistance and Intelligence: Essays in Honour of MRD Foot (Barnsley, UK: Leo Cooper/Pen & Sword, 1999)

• ‘Recent Reform of Intelligence in the UK: Democratization or Risk Management?’. Intelligence and National Security 13, no. 2 (1998): 144–158.

• ‘Police intelligence co-operation in Europe – rhetoric or reality?’ in Blaise Cronin (ed.), Information, Development and Social Intelligence (London, Taylor Graham, 1996), 68–80

• ‘Police Co-operation in Europe – the Intelligence Dimension’ in London, Pinter Publishers, 1994.

• 1992: The Security Implications (London: Institute for European Defence and Strategic Studies, 1989)

• ‘Police Intelligence Co-operation in Europe – problems and prospects’. Journal of Economic and Social Intelligence 3, no. 1 (1993): 21–33.

• ‘Sécurité et secret au Royaume-Uni’, Les Cahiers de la Sécurité Intérieure, 10 (Août-Octobre 1992): 247–266.

• ‘Crime, Frontier Controls, and 1992’ in Susan Flood (ed), Illicit Drugs and Organised Crime – Issues for a Unified Europe, (Chicago: The University of Illinois at Chicago, 1991): 67–82.

• ‘Terrorism: Europe Without Borders’, Terrorism – an international Journal 14 no. 2 (1991): 105–110.

• ‘1992 – The Security Implications’ in Gerald Frost (ed), Europe in Turmoil – the struggle for pluralism, (London, Adamantine Press, 1991) 171–200.

• ‘Canadian Intelligence Policy: The Role and Future of CSIS’, International Journal of Intelligence and Counterintelligence 3, no. 2 (1989): 225–248.

• ‘The Study of Intelligence in the United States’ in Roy Godson (ed), Comparing Foreign Intelligence – the US, the USSR, the UK and the Third World, (Washington, D.C. and London, Pergamon-Brasseys, 1988) 7–42.

• ‘Accountable Intelligence – the British Experience’, Conflict Quarterly 8, no. 1 (Winter 1988): 13–28.

• ‘Northern Ireland – Change, Continuity and Trends’ in Anat Kurz (ed), Contemporary Trends in World Terrorism, (New York, Praeger, 1987) 32–42.

• Defence Analysis Special Issue: Intelligence 3 no. 4 (June 1987) including his essay “An Agenda for Intelligence Research’ 95–101

• ‘Intelligence, Terrorism and Civil Liberties’, Conflict Quarterly 7, no. 2 (Spring 1987): 43–62

• British and American Approaches to Intelligence (London: Macmillan, 1987) including his chapter ‘The Politics of Secret Intelligence – British and American Attitudes’ 244–272

• ‘Review Article: Intelligence Requirements for the 1980s’ Intelligence and National Security 2, no. 4 (October 1987): 157–167

• ‘The Sociology of Spying’ Reviewing Sociology 3, no. 2 (Summer 1982): 4–12

• Public Secrets: A Study in the Development of Government Secrecy (London: Macmillan, 1982)

Acknowledgements

I am deeply grateful to Ken’s widow, Deborah Robertson, for all of her support, help and advice in preparing this obituary.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Philip H.J. Davies

Professor Philip H.J. Davies is Director of the Brunel University Centre for Intelligence and Security Studies. Between 1992 and 1997 he undertook his PhD with Ken Robertson at the University of Reading. He has subsequently lectured in Singapore for the University of London, at Reading University, the University of Malaya and since 2003 at Brunel University London.

Notes

1. ‘Accountable Intelligence,’ 25.

2. Sometimes given as the British Study Group on Intelligence, although the organisation has never actually referred to itself in this fashion.

3. Review of A Pack of Lies

4. Ken singled Sissella Bok out on this front most often, but the point was a more general one, see e.g., ‘The Politics of Secret Intelligence,’ 248.

5. Merton, Social Theory and Social Structure, 39.

6. One could readily locate Shils’ early work on pattern variables and roles in Merton’s middle range, as well as his later work on secrecy in the United States. See, e.g., Parsons, Bales and Shils, ‘Phase Movement in Relation to Motivation, Symbol Formation and Role Structure’ and Shils’ monograph The Torment of Secrecy.

7. Public Secrets ix.

8. I say odd because in most cases the systems compared are merely highly similar or similar enough for a certain purpose, and not the most similar systems possible or in existence.

9. One can draw a striking contrast between the robustness of Ken’s findings as against the awkward and structured argument about the causes of ‘social’ revolution, as opposed to the merely political, developed by Theda Skocpol in her influential but controversial States and Social Revolution, or the anodyne conclusions reached by Aaron Wildavsky in Budgeting: a Comparative Theory of the Budgeting Process.

10. Public Secrets, 180.

11. Bruce Lockhart, ‘Intelligence: A British View.’

12. ‘The Politics of Secret Intelligence: British and American Attitudes.’ In fact, his first publication on intelligence was a review article of several major publications on intelligence entitled 'The Sociology of Spying'. But British and American Approaches was where he first began to craft a systematic approach to the subject.

13. ‘Politics of Secret Intelligence,’ 246.

14. ”Politics of Secret Intelligence” passim.

15. Morgenthau, Politics Among Nations.

16. Robertson, ‘An Agenda for Intelligence Research,’ 98. For a later and more detailed version of this distinction, see also his 'The Study of Intelligence in the United States' passim but especially 12-21 and 23-34. This discussion, however, adds little to the conceptual detail of his Idealist-Realist distinction.

17. Secrecy and Open Government, 9–23.

18. Ibid., 39.

19. Ibid., 9–10.

20. ‘Accountable Intelligence,’ 25.

21. Weber’s thought is especially and explicitly central to his early writing on secrecy, see Public Secrets, 19–21.

22. Watt, ‘Intelligence Studies: Emergence of a British School’. For an instructive contemporary response and counterpoint to Watt and meditation on a ‘Canadian school’, see Stuart Farson, ‘Schools of Thought’.

23. See, variously, ‘An Agenda for Intelligence Research,’ 96; ‘The Politics of Secret Intelligence,’ 262; ‘Accountable Intelligence’ passim. Interestingly, he makes no reference in this article to the recent failure of strategic intelligence warning on the Falklands and Lord Franks’ Falkland Islands Review, although it featured elsewhere in his work during the same period. It is worth pointing out that Lawrence Freedman has since challenged the conventional wisdom about the Falklands as a classic warning intelligence failure in Volume 1 of his Official History of the Falklands Campaign.

24. ‘An Agenda for Intelligence Research,’ 97.

25. Ibid., 99.

26. Ibid., 100.

27. It is difficult to date the exact beginning of this process or how it commenced, but from Ken’s papers at is clear that the effort was already well under way by the time he and Christopher Andrew began corresponding regularly from February 1982.

28. Andrew and Dilks, The Missing Dimension.

29. My role being, as a recently installed PhD student, chiefly akin to a Civil Servant drafting on behalf of Ken’s Minister.

30. A British International Studies Association specialist group would evolve shortly thereafter, eventually linked up with and operating jointly with SISG.

31. ‘An Agenda for Intelligence Research,’ 98, 101, emphasis added.

32. Recent efforts such as the establishment of the Canadian Foreign Intelligence History Project has emerged as something of an antidote to this trend, see https://carleton.ca/csids/canadian-foreign-intelligence-history-project/ (downloaded 31 January 2022). To be fair, there have also been a vociferous minority of ‘realist’ thinkers, but most of these have been former practitioners rather than academics.

33. ‘Canadian Intelligence Policy,’ 225.

34. ibid., 226.

35. ibid., 227, 235.

36. ‘Canadian Intelligence Policy,’ 225, 233–4. The Canadian PCO is also often referred to as their Cabinet Office, and this is how Ken refers to it. To avoid any confusion with the UK system in a short piece like this I shall use PCO to refer to the Canadian body and confine Cabinet Office to the UK.

37. Ken was, of course, writing at much the same time that Hulnick was developing his typology and long before Stephen Marrin helpfully recast the discussion in terms of ‘proximity’. Rather, he focused on personalities such as Sherman Kent, Roger Hilsman and Wilmoore Kendal and their writing, see e.g. 'The Study of Intelligence in the United States' 9-12. See variously Arthur Hulnick ‘The Intelligence Producer-Consumer Linkage, Stephen Marrin ‘At Arms Length or At the Elbow?’ and Robertson, ‘Intelligence, Terrorism and Civil Liberties,’ 45 and ‘The Study of Intelligence in the United States’ passim.

38. ‘Intelligence, Terrorism and Civil Liberties,’ 46–47, emphasis added.

39. ‘The Politics of Secret Intelligence,’ 263, emphasis added.

40. ‘Intelligence, Terrorism and Civil Liberties,’ 46–47.

41. ‘Canadian Intelligence Policy,’ 227.

42. ‘Intelligence Terrorism and Civil Liberties,’ 52.

43. This has been transformed by contemporary information and communications and, at least in the UK, government OSINT activities are increasingly treated as subject to the surveillance provisions of the Regulation of Investigatory Powers Act 2000.

44. ‘Intelligence, Terrorism and Civil Liberties,’ 46.

45. ibid., 47.

46. ibid., 47.

47. ‘Politics of Secret Intelligence,’ 246.

48. ‘Intelligence, Terrorism and Civil Liberties,’ 50–51.

49. ‘Canadian Intelligence Policy,’ 230–1.

50. ‘Intelligence, Terrorism and Civil Liberties,’ 47.

51. ibid., 58–59, emphasis added.

52. ‘Accountable Intelligence,’ 19; he originally raised this concern in ‘The Politics of Secret Intelligence,’ 264 and ‘Intelligence, Terrorism and Civil Liberties,’ 48.

53. ‘Politics of Secret Intelligence,’ 250; ‘Intelligence, Terrorism and Civil Liberties,’ 48.

54. ‘Canadian Intelligence Policy,’ 234.

55. ‘Intelligence, Terrorism and Civil Liberties,’ 48.

56. ‘Accountable Intelligence,’ 17.

57. Ibid., 25.

58. ibid., 14–16.

59. ‘The Politics of Secret Intelligence,’ 267.

60. ‘Recent Reform of Intelligence in the UK,’ 154.

61. 1992: The Security Implications.

62. Secrecy and Open Government, 157. And yes, for all of you knew him, you did read that in his voice.

63. Mills, The Sociological Imagination.

64. I am deeply grateful for Professor Peter Jackson and Professor Len Scott for their recollection of these events and their details. See also Centre for Intelligence and Security Studies ‘The Ken Robertson Intelligence Studies Prize’.

65. ‘Accountable Intelligence,’ 21; Secrecy and Open Government, 100.

66. Quoted in Davies, ‘Organizational Politics and the Development of Britain’s Intelligence Producer/Consumer Interface,’ 113.

67. See, for example, the security intelligence case studies deployed in e.g., Central Intelligence Agency A Tradecraft Primer: Structured Analytic Techniques for Improving Intelligence Analysis or Heuer and Person, Structured Analytic Techniques for Intelligence Analysts.

68. Davies, Intelligence and Government in Britain and the United States, 251, 262.

69. As Anthony Glees and I argued at some length in our Spinning the Spies.

70. See, e.g., ISC, Diversity and Inclusion in the UK Intelligence Community (London: Her Majesty’s Stationery Office, 2018) and Privacy and Security: A Modern and Transparent Legal Framework (London: Her Majesty’s Stationery Office, 2018).