Secret Intelligence Service and National Security

The revolution in military affairs has given rise to powerful strategic tools such as effects based operations (EBO), mirrored by the concept of intelligence-led policing in law enforcement. Some advocates of intelligence change argue that the role of intelligence be expanded to provide the analytical power-house for ‘whole of government’ decision-making in relation not just to traditional threats, but also to this new range of threats—a kind of EBO for the whole of government.

This article argues for a more limited view of intelligence and its role—one that recognises the inherently human, and hence secretive, quality of intelligence as a means for dealing with human-generated competition. A nation’s intelligence apparatus is only one small part of the wider machinery for delivering policy and executive action. Traditionally, the role of intelligence within this wider structure was to counter threat from some kind of human collective opposition—whether a country, a crime group or a terrorist organisation.

Intelligence was regarded as a highly specific undertaking to give advantage over that threat in the form of knowledge, insight and predictive capacity. According to this model, advantage was sought over a human threat capable of learning and adapting. Intelligence therefore needed to be secret to deliver an advantage. To protect the ‘intelligence advantage’, countries also developed counter-intelligence organizations such as the Australian Security Intelligence Organisation (ASIO) and MI5 and encoding and decoding organizations such as the US National Security Agency and the Australian Defence Signals Directorate.

Today there is broad consensus that the threats we confront have expanded beyond the typical military or counter-intelligence threats of the past, especially those of the Cold War. This expanded range of threat falls into a major category and two sub-categories. The major category can be termed ‘non-conventional’ threats, ones that do not fall into the state-on-state category. They include environmental threats, threats of pandemic disease, terrorism and transnational crime. Volume 3 Number 3 (August 2007) – 55 – Security Challenges

This broad category of non-conventional threat can be further divided between those threats of a human agency (terrorism, crime, people smuggling and trafficking) and those of a non-human agency (climate change and other types of environmental threat, natural disasters, pandemic disease). These two sub-categories are, however, closely linked, as demonstrated by Thomas Homer-Dixon and others. 1 They are linked in two ways. First, they are linked in the sense that so-called non-human agency threats such as climate change can give rise to instability.

Instability can in turn give rise to many of the human generated conventional and non-conventional threats mentioned above. Second, threats like climate change are also linked with human agency in that they are often caused by human intervention. Changes in human behaviour are therefore necessary to remedy such threats. Even though these two sub-categories of threat are linked, they give rise to very different implications for the role of intelligence. On the one hand, the role of intelligence in countering human-related, nonconventional threat is relatively clear-cut and traditional.

It includes counterterrorism, police intelligence, customs intelligence, coast watch intelligence and so on. All of these intelligence activities are characterised by the fact that they involve secret information that would be compromised if it were to leak out, and through its compromise would give the opposition (or threat) an intelligence advantage, or sacrifice the advantage enjoyed by the state. On the other hand, the non-conventional threats such as climate change, natural disasters or pandemic disease, entail no such secret, tightly held intelligence response.

On the contrary, to counter such threats, especially in a liberal democracy like Australia, governments need to engage in a public dialogue with experts on the threat, whether those experts are scientists, journalists, medical or other experts operating in the public domain. This dialogue has to be public because the public needs to be taken along with the broad strategic changes required to deal with this type of threat. It also needs to be public and transparent because the scientific method is evidence based and depends on the capacity for peer review.

Moreover, it is often difficult to achieve major changes in attitude to such threats in liberal democracies unless there is some kind of ‘tipping point’, either in terms of the concrete effects of the threat (catastrophic drought, for example, in the case of climate change) or public consensus on the science, or both. The concept of threat needs to be actualised right through the 1 Thomas Homer-Dixon and Jessica Blitt (eds), Ecoviolence: Links Among Environment, Population and Security, Lanham, MD, Bowman and Littlefield, 1998, ‘Introduction: A Theoretical Overview’. – 56 – Volume 3 Number 3 (August 2007)

Security Challenges community before radical behavioural change can occur. This is because the political system of liberal democracies is usually geared to the short-term advantage of governments according to the political cycle rather than the long-term advantage of the nation. In order to move beyond populist politics, the whole community needs to be aware of the circumstances and prepared to make sacrifices to deal with the threat. Indeed, it is well known by intelligence specialists that long-term assessments, while they may eventually prove accurate, are almost never acted upon by policy.

For example, nearly twenty-five years ago a leading Australian intelligence assessment agency employed a scientist—the only person in the organization working on scientific issues at that time. The scientist predicted that overpopulation, scarcity of water and climate change would result in vast changes for the worse in the Middle East, South East Asia and South Asia, forcing major, economically induced out-migration. Such events, he assessed, would eventually challenge Australia’s security. Today such an assessment would be commonplace. But at the time, no action was taken.

Although the mandate of the organization was to predict long-term change, the political system was not equipped to deal with uncertain judgements about what was then considered the ‘deep future’. Nor could a secret intelligence report enable governments to deal with such predictions through debate in the public domain. Further, since threats like climate change constitute threats to the ‘global commons’, by definition they can only be addressed by global cooperation rather than competition. A ‘beggar your neighbour’ approach will only lead the globe deeper into trouble.

The implications for intelligence are significant. In terms of threats like climate change, pandemic disease and catastrophic economic change, intelligence ceases to concern itself with achieving an advantage over an enemy or competitor. So the question therefore arises: are secret intelligence agencies appropriate organizations to advise on such threats? Despite the intrinsic problems associated with the use of intelligence to analyse threats of this nature, increasingly, intelligence agencies are being coopted to advise on them.

For example, we learn from the Sydney Morning Herald of 10 April 2007 that the Office of National Assessments (ONA) has been commissioned by the government to determine the security implications of climate change. We further learn from the ABC news on 23 May 2006, which describes the ONA Director General being quizzed by the Senate Estimates Committee, that ONA has received multiple taskings of this nature. But the public are prevented from accessing the outcome on the grounds that the ONA is an intelligence organization operating in the secret realm. Volume 3 Number 3 (August 2007) – 57 –

Security Challenges True, the ONA is being asked to look at the security implications. But to do so, it would need to make a sound assessment of the nature and extent of climate change. No doubt the ONA now has a few scientifically trained people working on this and similar issues. But no doubt also, it will be locked in earnest consultation with the Commonwealth Science and Research Organisation (CSIRO), the Australian Bureau of Agricultural and Resource Economics (ABARE) and similar agencies and institutions. And it will also be carefully perusing the reports of the International Panel on Climate Change.

In other words, ONA is not, in itself, able to report on this issue. Its role is, rather, to organize, validate and valorize all the noise out in the public domain concerning the issue. The same evidently applies in the US. According to Anne Harrington, Director of the Committee of International Security in the National Academy of Science in Washington, If you get the intelligence community to apply some of its analytical capabilities to this issue [climate change], it could be compelling to whoever 2 is sitting in the White House.

But why should the Central Intelligence Agency suddenly have authority on this issue when the world’s leading scientific specialists, who have spent lifetimes working on the issue, have been studiously ignored—and some even silenced—by the White House for the past seven years? All this leads us to ask what, exactly, should be the role of intelligence in the so-called ‘new security environment’? And how should intelligence fit in with other government structures to provide an analytical capacity in these areas? Intelligence and Its Purposes

The narrower view of intelligence agencies suggested above—that is, organizations that deal fundamentally with human competition and therefore by nature exist in the secret domain—has not so far been widely accepted. The advent of non-conventional threat has generated considerable discussion in the ranks of those advocating intelligence reform. Various commentators have called for a broadening of the informational base of the traditionally tightly held intelligence agencies and a more ‘whole-ofgovernment’ approach.

However, very little of this discussion has drawn the distinction between human-induced and non-human induced threat in relation to the role of intelligence. Nor has it distinguished between longterm threat to society and the ‘global commons’ caused by environmental issues and short-term threat generated by problems such as transnational crime and terrorism. 2 Tom Allard, Mark Forbes and agencies, ‘US braces for global warring’, Sydney Morning Herald, 10 April 2007. – 58 –

Volume 3 Number 3 (August 2007) Security Challenges That is not to say that reform is not necessary, but rather that we need to be very clear what role intelligence should play in dealing with such threats and also where it should fit in to the wider ‘machinery of government’. Nor is it to claim that intelligence has no role at all in these matters: one very important role is to assess the security implications of issues like climate change, in order to prepare the state to meet those potential threats.

Such a position does not imply, moreover, that some broader reform of the machinery of government would not be advantageous. But in deciding on the nature of that reform, we need to ensure that the tail of intelligence reform does not wag the dog of machinery of government reform. Nor is it to say that a discussion of intelligence reform should be considered in isolation from a discussion of wider issues to do with the machinery of government. Intelligence obviously has to be fitted in with the machinery of government, and how it fits in is important.

Rogers correctly argues that the “practice of strategic intelligence is at its best when it is in counterpoise with strategic thinking [on the part of decision-makers and policy makers]”. 3 It follows that correct structure in the machinery of government should facilitate the connectivity between intelligence and policy on the one hand, and the consequent practice of strategic thinking in policy development on the other. But the problem here is that those involved with structuring intelligence do not necessarily have a say in the structuring of the machinery of government.

And in any case, in a liberal democratic, federal structure such as Australia’s, a considerable proportion of government process is dictated by relatively immutable conventions and constitutions. We should also note that this issue of where intelligence sits in organisational structures is relevant both within a particular organization that uses intelligence, and also within the wider structures of state as supported by intelligence. In one case intelligence is embedded within the agency, in the other, it is embedded within the machinery of government.

These two types and uses of intelligence may require very different structural approaches. In the case of the latter (intelligence agencies embedded in the machinery of government) it is the role of intelligence agencies to draw intelligence up and enable it effectively to be used in national policy-making. The structures used to ‘draw intelligence up’ are entirely proper considerations of a paper such as this on intelligence. This issue of the drawing up of intelligence covers the question of how a peak agency such as the ONA can best garner the wide range of intelligence that is required in today’s expanded threat environment.

This expanded 3 Kevin Rogers ‘Developments in Australian Strategic Criminal Intelligence’ in Ratcliffe (ed) Strategic Thinking in Criminal Intelligence (Sydney: The Federation Press, 2004), p23. Volume 3 Number 3 (August 2007) – 59 – Security Challenges environment requires that the entire range of agencies producing intelligence be included—agencies such as the Australian Federal Police (AFP), Customs, Immigration, Coastwatch, Quarantine and the Australian Crime Commission (ACC). Structures to support this expanded role were discussed in an earlier paper by this author. But (alluding to the issues raised above), the expanded role for intelligence does not—or rather should not—include agencies such as the CSIRO, ABARE, the Productivity Commission, Geosciences Australia, the Department of Health, and so on. Unfortunately, this distinction is not always recognised or agreed in discussions about intelligence reform. For example, some have begun to question whether concepts like effects based operations (EBO), which in turn have been spawned by the new intelligence environment and new technologies, cannot be applied in a ‘whole-of-government’ way. According to this view, the three-way relationship between intelligence, policy and operations could be seen to apply across the spectrum of government decision-making, thus incorporating all departments of state and agencies in an endeavour to achieve a strategic outcome. Although such a project would be ambitious, ‘whole-of-government’ possibly can and should be made to function in a strategic sense. But it should be recognised that intelligence is not central to the process in the same way as it is central to EBO in a military setting or to intelligence-led policing in law enforcement.

Indeed, in the author’s view, intelligence is a highly specific function to do with human competition and human enemies. It is not only inappropriate for wider use, but such use could be positively harmful in terms of the needed outcomes in government decision making in a democratic setting. Certainly, good strategic intelligence should be suggestive of courses of action, but only suggestive in the sense that the knowledge brought forward is suggestive. Intelligence can also comment on implications of actions when specifically asked to do so, but should not go the extra step of recommending options.

It is not the role of intelligence to present analytical options to decision-makers in the same way as that is the role of a department of state or ministerial staff. The temptation to use intelligence agencies to support a strategic, ‘whole-ofgovernment’ approach is quite strong, however. Traditionally, intelligence agencies have been very close to the seat of power. Indeed, they were born Sandy Gordon, ‘Re-Shaping Australian Intelligence’, Security Challenges, vol. 1, no. 1 (November 2005), pp. 27-58. Brice Pacey, ‘National Effects-Based Approach: A Policy Discussion Paper’, Working Paper, no. 381, Canberra, Strategic and Defence Studies Centre, Australian National University, 2003, passim. Pacey is not, however, arguing for a central role for intelligence in this enterprise. . 4 – 60 – Volume 3 Number 3 (August 2007) Security Challenges of a one-to-one relationship, in which the intelligence chief sat at the leader’s right hand. This was both to provide immediacy and preserve secrecy. The idea of a separation between intelligence and policy is a relatively recent one.

The only separation that was once required was that between master and servant. The temptation to use intelligence agencies as analytical determinants of policy is even stronger in situations where there is no real alternative to the analytical powerhouses that some intelligence agencies can bring to bear. Moreover, governments that use intelligence agencies to consider politically sensitive issues like climate change have the added advantage of keeping such consideration outside the public domain and the scrutiny of oppositions.

This is because once a matter is within the purview of intelligence, governments can claim they can ‘neither confirm or deny’ questions in respect of them. But as argued above, this is essentially a misuse of intelligence. Intelligence in National Strategic Decision-Making At the moment in Australia, national intelligence exists within a relatively tight framework of the Australian Intelligence Community (AIC), oversighted by a small and powerful group of departments and ministers, particularly Prime Minister and Cabinet (PM&C), Defence, Foreign Affairs and AttorneyGeneral’s Department.

This tight structure is reinforced by the restricted membership of the National Security Committee of Cabinet (NSCC) and the Secretaries’ Committee on National Security (SCONS), which proffers advice to the NSCC. It is a structure that in its essence was bequeathed by the Cold War, with minor modifications as a result of the Flood Report and other developments. As such, it was designed to deal with state-on-state threat and the threat of spying and political violence rather than the broader range of threats we now confront.

Such a tight structure has both advantages and disadvantages. The advantages are that the intelligence agencies of the AIC are well plugged in to the Canberra policy environment and have a nuanced appreciation of what the government wants to know. Equally, this tight structure allows for rapid, consensus decision-making when needed. Further, the agencies of the AIC, particularly the ONA, represent a collection of individuals capable of high-level strategic thinking. The disadvantages of such a tight system are well known.

The 9/11 Commission and Butler report chronicled the distortion of the intelligence process to serve particular policy needs, or at least perceived needs. Given the tight inter-relationship between the government, key departments and intelligence agencies in Australia, such distortions are also possible here. Volume 3 Number 3 (August 2007) – 61 – Security Challenges The Butler committee report also emphasised the dangers of a filtration system such as the UK Joint Intelligence Committee (JIC) standing between the judgements of experts and policy-making bodies.

Some of the worst errors evident in the so-called ‘Dirty Dossier’ arose because the expertise in the Defence Intelligence Staff was filtered out or distorted, either as a result of a classic ‘broken telephone’ situation or through pre-judging the policy bias on the part of the JIC. In Australia, the system would be equally vulnerable should the findings of organizations like ABARE and the CSIRO be filtered through a small, nonexpert (in the disciplines involved) organization like the ONA. How suitable is the present Australian structure in light of the changes to the regional, global and technological environments?

Before considering this important issue, let us consider the needs of a system designed effectively to operate in the new environment. x As discussed above, such a system would need to delineate a specific role for intelligence, one that relates to human-on-human competition, and that in turn necessitates a secret approach to intelligence. Such a system would also need to be capable of melding the intelligence product with product from other agencies working on issues that do not require secret intelligence and with other open sources. It would need to be flexible.

That is, it would need the capacity to draw to a greater or lesser degree on a ‘whole-of-government’ approach for supporting information and judgment, depending on the urgency and nature of the threat and degree of secrecy needed. In other words, it would need to have the administrative means to ‘slide up and down the scale’ between a narrow, decision-making capacity at the top and a broad consensus model below, one that included information and analysis from a range of agencies, not only intelligence agencies but also economic and scientific agencies.

In some instances, it would need to shape decisions for the longerterm. Yet it would also need to be capable of making adjustments in light of the evolving evidence. Such decisions would need to be maintained well outside the life span of a typical Australian Government. At times it would need to draw in two, or perhaps even three, levels of government, as already evident in the case of terrorism and pandemic threats such as SARS and ‘Bird Flu’. It would need to be well connected internationally in order to draw on available information and assessment. x x x x – 62 – Volume 3 Number 3 (August 2007)

Security Challenges x Above all, it would need a powerful apex analytical and organisational capacity containing a range of expertise in order to bring together diverse lines of information, identify real problems, set priorities between them and devise viable strategies to deal with those priorities. This role should not be undertaken by existing intelligence agencies, because their role should be confined to the analysis and dissemination of intelligence. To use them in the dual role of policy advice and intelligence analysis and dissemination would be unduly to compromise the intelligence role.

How well does the present system meet the needs outlined above? In some respects, quite well. It basically consists of a tight core capable of being expanded to meet a broadening of threat, with the NSCC providing a sort of ‘inner cabinet’, surrounded by a tight supportive framework consisting of key departments, SCONS and the AIC. It is a highly functional arrangement for an environment requiring a high degree of secrecy and relatively rapid decision-making. Where the nature of threat broadens, for example in the case of climate change, the current system is capable of rapid expansion.

Members of Cabinet, such as the environment minister, can be brought into the NSCC where necessary. The AFP Commissioner, CEO of Customs and others can also be inducted into the SCONS when necessary. Within PM&C, the National Security Division (established in 2003) provides a potential analytical unit that is not confined to intelligence, but that can range over the available government and non-government agencies, given its location on the ‘commanding heights’ of PM&C. The present system falls down in a number of respects, however. It is deficient in that certain information deemed intelligence in the narrower sense outlined above is still not fully drawn into the AIC information network and database (AICNET). Nor are the organizations generating this intelligence (such as the AFP, Customs and the ACC) included in the tight deliberative network at the apex of government decision-making, at least not on a day-today basis. These exclusions cause a deficit in knowledge and analysis of non-conventional, but human-induced, threat. This deficit was discussed in greater detail in an earlier paper. 6 6 Gordon, op cit.

At the time of final preparation of this paper, the government has announced a new system of combining the databases of Customs, Immigration and ASIO. One might well ask why this is only being done now, six years after the events of 9/11? Volume 3 Number 3 (August 2007) – 63 – Security Challenges x It is not capable of adopting strategic thinking across all levels of government in the federal structure—the so-called ‘EBO of governance’. Because of the requirements of the liberal democratic federal structure, however, this problem may not be amenable to an entirely satisfactory solution.

Even given the restraints imposed by our political system, there is inadequate capacity at the top to analyse, identify and give priority order to threat. Although the National Security Division of PM&C seems to be set up to undertake this role, according to Pacey, the division is still limited by coordination roles and the need to deal with immediate crises rather than provide long-term analysis. 7 In counter-terrorism, for example, the role of PM&C is to provide a multi-government and multi-disciplinary platform.

One suspects that this demanding role diminishes its capacity to deliver long-term policy advice. In the absence of an appropriate analytical unit outside the confines of intelligence, there currently seems to be a growing de facto move to recruit the ONA for this top-level analytical role, as discussed above. But, as also argued above, the ‘heavy lifting’ on matters like climate change should not be conducted by a secret intelligence organization at all, but through transparent, evidence-based techniques that are well tried and understood in scientific organizations.

Moreover, to place an organization like the ONA at the apex of the policy advice structure is, at least in a de facto sense, to bring it directly into the policy-making apparatus—hitherto considered anathema for an intelligence organization. x x Therefore, if we accept the more limited role for intelligence advocated in this paper, we are left with a potential deficit in terms of an apex analytical unit— the intellectual powerhouse of ‘EBO of governance’. The main candidate for fulfilling such a role seems to be the National Security Division of PM&C.

And in fact, more and more of the capacity relating to security in areas requiring a multi-disciplinary approach is now located in PM&C. This includes terrorism, energy security, pandemic disease, nuclear energy and intelligence. Conclusion and Issues for the Next Government In light of the profound changes in the security environment we have witnessed in recent years, those responsible for shaping the way governments will deal with long-term structural change confront a choice. 7 Pacey, op. cit. , p. 5. – 64 – Volume 3 Number 3 (August 2007)

Security Challenges On the one hand, they can advocate an expansion of the role of intelligence to cover the broader nature of the threat we now confront. In a de facto sense, the current arrangement seems to be drifting toward this kind of arrangement. On the other, they can continue to regard intelligence as an essentially secret function designed to give advantage and deny advantage in terms of human competition, whether of the state-on-state variety or threats from criminal and terrorist groups. In this paper we have argued the latter position.

We have done so because of concerns about the nature of intelligence and how it differs from policy analysis, the nature of scientific inquiry and the democratic need for debate and consensus. We have further argued that, while it may be possible to achieve something close to a strategic process on a ‘whole-of-government’ basis, such a process cannot be driven by intelligence; and nor is it correctly placed within a discussion of intelligence and its role. We are of the view that additional analytical capacity required to support ‘EBO of governance’ should exist in the form of an expanded staff specifically advising the NSCC.

While the National Security Division of PM&C would seem to be the logical candidate for such a role, several changes would be needed to provide the kind of analytical capacity described above. Ideally, a unit of this nature should be administratively removed from the day-to-day, short-term contingencies and coordination functions normally undertaken by a division of PM&C. That is not to say that the unit should be entirely administratively removed from PM&C. But it might be a statutory body linked in a similar way to the ONA.

Or it might be more directly associated with the Cabinet Division. Further, the unit would require an expanded ability to provide advice on a ‘holistic’ basis, with a range of expertise covering scientific, health, intelligence, economic and defence issues. Already the germ of such a capability is contained within National Security Division. The existence of such a support unit would act as a buffer between the Cabinet and intelligence agencies and ensure that the latter continue to function as providers of intelligence rather than strategic advice.

It would provide the capacity to meld factual and analytical work from both the intelligence agencies and those agencies outside the AIC, such as the various government scientific and economic agencies and non-government agencies. Its interface with such agencies would be far easier than between intelligence and outside agencies, given the role of intelligence in protecting information from human competition. Volume 3 Number 3 (August 2007) – 65 – Security Challenges

In light of this position, the following issues emerge for the next government: x Australia’s national intelligence database should be expanded more fully to incorporate information and intelligence from agencies outside the AIC, such as the AFP, Customs, Coastwatch and the ACC. Details of how this might be accomplished are set out in an earlier paper by this author. 8 Leaving aside the security aspects of issues like climate change, pandemic disease and radical economic change, government should recognise that such issues in themselves are not conducive to analysis and advice from intelligence agencies.

Rather, a ‘whole of government’ analytical and strategic capacity should be incorporated into the advice mechanisms serving the NSCC. This unit should incorporate the work of intelligence agencies, but also range far more broadly across government and non-government agencies. It should posses a ‘holistic’ capability— that is, it should include scientific, health and economic professionals as well as national security experts. It should not be constrained by the day-to-day needs of servicing a busy department like PM&C. x Sandy Gordon joined the Australian Public Service in 1977, subsequently working in the Office of National Assessments, AusAID and as Executive Director of the Asian Studies Council and Australian Literacy Council. In 1990 he became a Fellow at the Strategic and Defence Studies Centre, Australian National University, where he worked on South Asia and the Indian Ocean. In 1997 he was appointed head of intelligence in the AFP, a position he held until 2000.

He then became Co-Chair of the Council for Security Cooperation in the Asia-Pacific Transnational Crime Working Group and a member of the National Expert Advisory Committee on Illicit Drugs. Between 2003 and 2005, he lectured on terrorism and transnational crime at the Australian Defence Force Academy, University of New South Wales. He is currently Associate Professor, Centre for Transnational Crime Prevention, University of Wollongong. thegordons@homemail. com. au. 8 Gordon, op. cit. – 66 – Volume 3 Number 3 (August 2007)