Commentary No. 88: Between Probity and Proficiency: Challenge and Change within the Australian Intelligence Community

Commentary No. 88

Michael Wesley

April 2006
Unclassified

Editors Note:

Professor Michael WESLEY is the Director of the Griffith Asia Institute at Griffith University, Australia. Prior to taking up this position in July 2004, he was the Assistant Director General for Transnational Issues at the Office of National Assessments. Before he joined ONA, he was a Senior Lecturer in the School of Politics and International Relations at the University of New South Wales and a Research Fellow at the Asia-Australia Institute, also at UNSW.

Disclaimer: Publication of an article in the Commentary series does not imply CSIS authentication of the information nor CSIS endorsement of the author's views.

1. Introduction

The Australian Intelligence Community (AIC) appears to undergo significant reform at quarter-century intervals. Australia’s core collection agencies, the Australian Secret Intelligence Service (ASIS), the Defence Signals Directorate (DSD) and the Australian Security Intelligence Organisation (ASIO) were developed at the end of the Second World War. Then in the late 1970s, the Commonwealth created a new intelligence analysis and co-ordination agency in the Office of National Assessments (ONA), separate from the Defence Intelligence Organisation (DIO) and formalized the principles of functional differentiation and oversight. These arrangements operated largely undisturbed by organizational reforms to the rest of the bureaucracy until 2004, when as a result of new pressures, controversies, inquiries and reports, a new set of reforms were enacted. Both the everyday operation of the AIC and each period of major reforms to Australia’s intelligence architecture have been guided by three powerful organizing principles: functional specialization in intelligence collection; the separation of intelligence collection and assessment; and the separation of domestic and foreign intelligence collection and assessment. Recently, at a time when some of Australia’s closest allies have considered major structural changes to their intelligence communities, Canberra has engaged in more modest reform, which has underlined and strengthened, rather than challenged, these basic organizing principles.

The most recent reforms to the AIC are a response to two linked sets of stimuli. One is the quickening pace of globalization, which is changing the nature, volume and urgency of the demands on all states’ intelligence agencies. If the 1990s were an era that had celebrated the benefits of globalization, the key preoccupation of the first decade of the twenty-first century would be to come to terms with globalization’s dark sides. The other stimulus was a series of high-profile controversies relating to the AIC. In 1999 the defence intelligence agencies became embroiled in allegations that intelligence on the activities of Indonesian-backed militias in East Timor had been ignored and withheld from field units of the Australian Defence Force and from Australia’s closest allies. The scandal deepened with the suicide of an Australian Defence Intelligence Liaison Officer in Washington, and the charges were reiterated in 2004 in the midst of later intelligence controversies. In 2001, ONA was criticized for fabricating intelligence to support the government’s case over the actions of asylum seekers, and DSD was accused of illegally eavesdropping on Australian citizens acting on behalf of the asylum-seekers. The October 2002 Bali bombing, which caused the deaths of 88 Australians, saw questions raised about whether Australian intelligence had taken sufficiently seriously the threat of violent Islamic groups in Southeast Asia. Then in 2003, the AIC came under sustained scrutiny for its role in providing justifications for Australia’s participation in the March 2003 invasion of Iraq.

The result was a series of commissioned and parliamentary inquiries into the operations and adequacy of the AIC. Particularly in the context of a bitter partisan debate over the invasion of Iraq in the lead-up to the 2004 federal election, these inquiries attracted unprecedented attention to the intelligence agencies. The culmination of the review process came with the inquiry conducted in 2004 by former ONA Director-General Philip FLOOD, which resulted in a package of 23 reform proposals being presented to the government, all but one of which were implemented.¹ My aim in this article is to examine the Australian government’s recent reforms to the AIC in light of new pressures facing intelligence agencies and allegations of “intelligence failures” against the AIC.

1. Pressures on Intelligence

Australia’s intelligence agencies, like those of its closest allies, face a brace of challenges arising from the processes of globalization. Over more than a decade, globalization has transformed the nature of threats faced by modern societies, the capacities of both intelligence agencies and their targets, and the political and diplomatic context within which intelligence agencies operate. The challenges posed to intelligence agencies by transnational threats are complex, and most intelligence communities are still evolving in response. While terrorism is a top national security priority in most Western countries, the proportion of intelligence resources devoted to other transnational challenges is very small. Currently in Australia as in many other countries, the challenge is to ensure that the intelligence effort against terrorism does not too drastically pull resources away from traditional targets and tasks. This may change if other transnational issues have as dramatic an effect on the public psyche as terrorism has recently.

Transnational challenges

The demise of Cold War confrontation and the forces of globalization have raised transnational threats to the forefront of national security, and thus brought them to the centre of the work of intelligence agencies. Transnational issues can be defined as those that arise from within one or more societies (rather than from within the decision structure and resources of the state) and are transmitted or replicated across national borders to threaten the security or values of society—and ultimately the security or viability of the state. They include terrorism, sub-state proliferation of WMD and missile technology, organized crime, drug production and distribution, serious disease epidemics, irregular people movements, illegal resource exploitation, and disruptive doctrinal movements.² Most transnational issues are the work of decentralized, transactional networks of non-state actors distributed across several societies, interacting amidst the vast quantity of transactions occurring daily in a globalized world. Linkages among transnational actors are increasingly made on the basis of either shared general goals or common interests in a particular transaction, rather than being based on ethnic- or family-based loyalties or formal command structures. As a result, the most dangerous transnational threats we face are beginning to exhibit what biologists term ‘emergent behaviour’, where the dynamic interactions of agents reacting to local situations give rise to collective behaviours that are more flexible and complex than a centralized, hierarchic organization could manage. Globalized communications enable far-flung groups to observe and adapt the most effective techniques they see used elsewhere. Often the most dangerous transnational flows are not people or goods but information—be it the doctrine of world jihad, gas centrifuge designs, synthetic amphetamine recipes, or how to make shape-charges.

Transnational threats challenge long-established understandings of security and international relations. States such as Australia have come to realize that the major contemporary threats they face are not those posed by other states to their military-industrial power or their institutions of state. Rather, transnational threats, from terrorism to crime and drugs and disease pandemics, endanger the understandings and practices of civic association that are so basic to the functioning of modern societies. Not only does globalization transmit transnational threats across borders more rapidly and extensively than ever before, it has magnified what Emile DURKHEIM described as the greater “organicism” of modern societies—meaning the advancing interdependence and greater functional specialization of actors and institutions.³ The growing specialization and patterns of interaction in modern societies, their growing dependence on information technologies and “just-in-time” inventory and production systems, and their increasing impatience with redundancy have led to huge increases in their vulnerability to catastrophic disruption.⁴ As a result, transnational threats are blurring the distinction between domestic and international intelligence collection, as well as that between the public/state sphere and the private/societal sphere of security.⁵ There has been a major transformation in the role of foreign intelligence organizations. Rather than concentrating only on discovering the capabilities and intentions of other states, foreign collection and analysis agencies have been drawn increasingly into the realm of their domestic security counterparts, monitoring the actions, intentions and influences of individuals and groups that could eventually threaten the lives and livelihoods of Australians. And as transnational issues widen the range of security threats to states, more government departments are being drawn into the business of national security and are becoming the consumers and collectors of intelligence. The vulnerability of societies and their infrastructure to terrorist attack has brought immigration, health and transport departments into security structures and made the information available through their bureaucratic networks vital to intelligence early warning systems.

Transnational threats, by their nature and volume, pose major difficulties for intelligence organizations accustomed to focusing on the activities of states and their agents. The standard methods of Western intelligence agencies were originally developed to collect and interpret intelligence on a few secretive but rigidly hierarchic and centralized states—usually on policies, dispositions and capabilities that were developing in a single, coherent direction over decades. The major challenge facing intelligence collection and assessment on decentralized, transactional networks arises from their absence of a clear hierarchy of command and control. Consequently, the real challenge for collection and analysis is assessing the value of the intelligence gained. Reporting on the intentions of one or several actors in a transnational network gives intelligence agencies a highly uncertain picture of the intentions and capabilities of the network as a whole. And as decentralized actors constantly adapt their behaviour to deal with local requirements, it is difficult to build a dependable picture of the network incrementally. There are challenges here for both collection and analysis agencies.

Collectors of intelligence on transnational threats increasingly have to do their own on-the-spot analysis in order to validate information and prioritize collection targets against a rapidly evolving organization. Advances in information technology, and greater publicity about the methods of collection agencies have complicated the process of intelligence collection. Individuals and organizations are now able to acquire cheap and highly effective encryption technologies and use a range of Internet innovations to disguise their communications. Even as SIGINT organizations race to keep one step ahead of these technologies and tricks, transnational actors have resorted to a range of “low-tech” strategies, placing greater pressure, in turn, on HUMINT collectors to find clues to their activities. And as states increasingly deploy to anticipate and block transnational threats, collection agencies are devoting more and more of their resources to supporting overseas operations of defence and police forces, as well as their own disruption operations, as opposed to collecting for analysis. At the same time, the state’s deployed defence and police forces are generating large volumes of valuable intelligence that need to be gathered and evaluated by the collection and analysis organizations.

Against transnational threats, intelligence analysis agencies face an asymmetric contest. For terrorists and criminals in decentralized networks, every problem is local and dealt with locally, whereas the intelligence effort is required to try to grasp the entire problem. The art becomes that of abstracting and anticipating behavioural patterns and possibilities, rather than seeking definite answers. In the words of one commentator, against transnational threats, “many questions cannot be resolved as a matter of logic because they depend on events yet to happen, such as human intentions that will unfold or factors that will converge by chance.”⁶ Transnational intelligence analysis is required not only to assess warnings of emerging threats and reports on strategic trends in transnational threats, it is also being drawn into tactical analysis geared towards enforcement operations, greatly expanding the demands on the limited resources of analytical agencies.

Political and diplomatic contexts

In creating a public climate of apprehension and a new set of imperatives for government action, transnational threats have also changed the political context in which intelligence agencies operate. Particularly since the September 11 attacks, intelligence agencies have evolved in the public mind from arcane “spy-versus-spy” operations toward a more accepted role in protecting society, more akin to the defence forces and law enforcement agencies. Intense media reporting of intelligence inquiries both in Australia and overseas and a profusion of polemics and exposés, often by disgruntled former intelligence staff, have raised public awareness and knowledge of the AIC and its practices. One frustration of many intelligence professionals is that intelligence agencies are unable to answer many of their critics, which leaves the field open to distortions and misunderstandings. As the intelligence agencies have received greater funding and powers in the aftermath of September 11, it is inevitable that public expectations of their omniscience will have been reinforced. Repeated references to “intelligence failures” and a public culture of looking for someone to blame for terrorist attacks have worked against government attempts to lower public expectations. There is still a very low level of public understanding of the capacities and limitations of intelligence agencies, an issue that calls for better public information. While the majority of the Australian public appears supportive of the role of intelligence in combating terrorism, some critics have worried that their role represents the vanguard of a national security state. The intense debate over ASIO’s enhanced powers of detention and questioning has heightened public awareness of individuals’ legal rights, and prominent role of the AIC leading up to and after the war in Iraq has decreased the willingness of some groups and individuals to co-operate with the intelligence community.

The impact of terrorism on democratic politics has greatly complicated the political context within which intelligence agencies work. In Australia, the psychological impact of the September 11 attacks coincided with a sudden rise in the number of sea-borne asylum seekers heading for the island continent to create a pervasive atmosphere of threat, in which stewardship of national security joined the capacity to manage the national economy as the two key tasks of government in the minds of the vast majority of the electorate. Governments wanting to burnish their “national security credentials” in the face of oppositions equally keen to tarnish those credentials have elevated public safety to the highest possible policy priority. In a climate of public fear of terrorism, nothing can be left undone in the interests of public safety. Intelligence and law enforcement agencies, identified as the best way of protecting society from transnational threats, have been granted greater resources and powers. Amid the partisan competition over national security credentials, media and parliamentary scrutiny has grown. Opposition parties are increasingly keen to use parliamentary committees—and especially the annual Senate Estimates Committee hearings—to probe intelligence chiefs over issues thought to erode the electorate’s trust in the government’s management of national security. Intelligence agencies and their political masters realize that any event deemed an “intelligence failure” could and has resulted in their practices, thoroughness and judgments being subjected to the detailed scrutiny of the media and parliament, both acting with the wisdom of hindsight.

In this atmosphere, overlooked, misjudged or unactioned intelligence information can be political dynamite. Intelligence agencies must ensure that all credible threat intelligence is passed on to government, and governments are expected to pass threat warnings on to the public in the form of travel advisories. Placed squarely in the front line of protecting society from transnational threats, intelligence agencies have become worried they will miss the significance of a piece of intelligence warning of an impending attack. Facing the choice of not reporting a piece of intelligence and risking it might be genuine, however spurious it may look, the natural temptation is to report it. If not resisted, the onus of evaluating and reacting to threat intelligence would be pushed to higher and higher levels of government, challenging the very reason for being of intelligence analysis agencies. The political imperatives of transnational threats also magnify the time pressures on agencies confronted by an avalanche of intelligence information of highly variable quality. Intelligence organizations are required to present their customers with high-quality intelligence and hard baseline judgments and predictions in a timely manner, while the intelligence is still ‘fresh’. This leads inevitably to evaluation and analysis having to make judgments on the basis of the information at hand, without being able to await confirmation or a fuller picture. It also means that collectors and analysts are rarely in a position in which they can spend time trying to verify the information they have received. The need for real-time analysis has only been increased by the flow of raw intelligence to ministerial offices and policy departments—a development that is likely to be permanent given ministerial appetites for intelligence and the collection agencies’ interest in expanding their customer base—as ministerial offices and policy departments increasingly demand immediate analysis of the credibility of incoming intelligence. To add to these pressures, Australian government ministers and their senior advisors, many of whom have occupied national security portfolios for a decade and regularly meet to discuss intelligence-informed issues in the powerful National Security Committee of Cabinet (NSCC), have become much more experienced and discerning consumers of intelligence.

The rise of transnational threats has also seen Australia’s intelligence agencies drawn into the world of diplomacy. While the underlying adversarial logic of intelligence still persists among states, the very nature of transnational threats mandates greater co-operation among countries. Developed states’ concerns about “failing states”—weak parts of the fabric of international order in which terrorists, criminals and proliferators lodge and transit—have coincided with developing states’ concerns to reassure the outside world, and in particular investors and tourists, of their commitment to good governance. Transnational threats have increased the intimacy of Australia’s UKUSA intelligence-sharing relationship with America, Britain and Canada, even while eroding its geographic specializations. In the words of one manager in a collection agency, “simply watching your ‘patch’ won’t cut it any more with customers or partners.” Australia’s intelligence and law enforcement agencies have been at the forefront of regional co-operation against transnational threats: collaborating with counterpart agencies in field investigations, training their counterparts in regional countries, and participating in an ever-expanding network of exchange of intelligence material and judgments on transnational threats outside of traditional alliance structures. In addition to the time demands on already stretched organizations, “intelligence diplomacy” and the resulting sharing of intelligence and tradecraft give rise to fears about the protection of sources and methods and the effect that co-operation will have on these states’ future counter-intelligence capabilities. Expanded intelligence liaison has also delivered a greater amount of intelligence material of uncertain reliability for intelligence analysis organisations to cope with.

2. Intelligence Failures

Despite the controversy surrounding Australia’s intelligence agencies since 1999, two issues had become the focus of inquiry and debate by 2004: whether the AIC had underestimated and overlooked the threat of radical Islam in Southeast Asia prior to the October 2002 Bali bombings; and whether the AIC had oversold the possibility that Iraq possessed weapons of mass destruction (WMD) before the March 2003 invasion. Australia’s controversies and inquiries closely mirrored those in the United States and United Kingdom over terrorist attacks (September 11 and Iraq for the US, the Bali bombing and Iraq for the UK), and many of the accusations, arguments and judgements made in Australia were influenced by those in America and Britain. In Australia, as in the United States and the United Kingdom, three basic categories of intelligence failure were highlighted. First was a range of issues relating to the collection and distribution of intelligence: overloading and failing to properly evaluate sources; allowing assumptions to guide collection targeting; failing to pass information between agencies. Second were failures of imagination: either allowing preconceptions to blinker creative thinking or allowing “satisficing”⁷ deductions to crowd out the consideration of alternative explanations. Third were issues relating to “politicization”: whether intelligence agencies had allowed their judgements to be influenced either overtly or subliminally by the preferences of governments seeking to make compelling cases to go to war. The first two categories are familiar shortcomings of intelligence: problems of source evaluation, information-sharing and cognitive rigidity have been recognized for a long time, and have spawned a considerable literature. The third category, raising issues of the intelligence-policy interface and the public use of intelligence, is much newer. The recurring nature of “old” intelligence failures poses particular challenges to intelligence reform; while “new” challenges raise a range of different considerations.

Familiar failures

The failure of the Iraq Survey Group (ISG) to find evidence of the WMD that the American, British and Australian governments had alleged Iraq possessed prior to their invasion was seized on by opponents of the war to support charges that the coalition governments had misled their electorates over the reasons for going to war. As the governments’ prewar statements on Iraq’s WMD stocks and activities were picked over, it was inevitable that the intelligence agencies, on whose material and judgements these claims had been based, would become drawn into the controversy. As the debate continued, three particular pieces of intelligence that had contributed to judgements that Iraq was developing a “breakout” WMD capability came under the spotlight. They were reports that Iraq had attempted to negotiate the sale of uranium from Niger; reports that Iraq had imported large quantities of high-grade aluminium tubes and magnets that could be used to construct gas centrifuges for enriching uranium; and that the Saddam HUSSEIN regime had developed several mobile laboratories for manufacturing biological weapons (BW). Each of these pieces of intelligence were cited by the Australian, British and American governments in speeches arguing the case for war, and each of them had been publicly discredited shortly after the war. The Italian newspaper La Repubblica published photos of obviously forged documents on which at least some of the reports of Iraq’s attempts to buy uranium from Niger had been based, and former American diplomat Joseph WILSON made public the negative findings of his official mission to Niger to check the claims.⁸ Scientific experts testified to the International Atomic Energy Agency (IAEA) prior to the war that the aluminium tubes, unless significantly modified, were unsuitable for use in a uranium enrichment centrifuge. And the vehicles thought to have been mobile BW laboratories were found by the ISG to have been intended for other purposes; in late 2005, officials in Germany’s Federal Intelligence Service revealed that the CIA had been overly credulous in accepting claims of an unreliable Iraqi defector codenamed “Curveball” about the mobile BW laboratories.⁹

The post-war intelligence inquiries in the United States, the United Kingdom and Australia all criticized their respective collection and analysis agencies for failing to properly scrutinize the claims of HUMINT sources. The Australian inquiry, held by the Parliamentary Joint Committee on ASIO, ASIS and DSD (PJCAAD), cited the arguments of a witness to the UK’s Hutton inquiry, that human intelligence collection agencies have an incentive to overemphasise the reliability of their sources in the interests of demonstrating their relevance and usefulness to intelligence consumers. The Butler inquiry in Britain found that intelligence agencies’ increasing sense of urgency to clarify Iraq’s WMD holdings and capacities had led HUMINT collectors to put more pressure on sources: either relying more heavily on untried sources, accepting reporting chains whose length would have been unacceptable in other circumstances, or pressing sources to report on issues outside of their areas of expertise.¹⁰ In Australia, ASIS escaped much of the criticism endured by its American and British counterparts due to the fact that perhaps 97 per cent of the raw intelligence informing Australia’s intelligence judgements came from allied sources. This, however, raised specific issues for Australian intelligence. Shared human intelligence must ultimately be taken on trust, with the assumption that allied HUMINT collectors would only pass on intelligence they had deemed to be credible. Source descriptors on pieces of HUMINT must necessarily be general in the interests of source protection; and given the massive increase in the amount of reporting collection agencies have been required to do, these descriptors had tended to become formulaic, choices from a “pull-down” menu on a computer program. Despite these caveats, however, Australia’s analysis agencies were accused of being insufficiently critical in assessing their claims. The Office of National Assessments in particular was accused, like its counterparts in the United States and United Kingdom, of drawing too strong a set of judgements from the available intelligence. As the PJCAAD sternly admonished in its report, “Decisions to go to war, with the potential to cost many lives, must only be taken on the basis of the soundest information, information that Australian agencies can reasonably rely upon.”¹¹

As the shock of the Bali bombings percolated through Australian society, questions began to be asked whether this terrorist attack had been preceded by the same type of intelligence failures that were emerging concerning the September 11 attacks. Attention soon focused on the Web site of the Department of Foreign Affairs and Trade (DFAT), which warned at the time of the attacks that the threat of violence in Indonesia was “high”, but also stated that “tourist services in other parts of the country, including Bali, [were] operating normally.” It also emerged that the United States had issued a travel warning concerning Bali in late September 2002, and that the United States Ambassador to Australia had delivered a warning concerning impending attacks in Bali to the Australian government the day before the attacks occurred.¹² In the United States, as the September 11 post-mortems continued, evidence was emerging that the American intelligence community had warnings of impending attacks, but that different agencies had either failed to pass information to each other, or had failed to act on information that was passed on; the 9/11 Commission later spoke of a “void” between coverage of foreign and domestic threats.¹³ Was Bali similarly a “preventable” tragedy, allowed to occur because Australian agencies had failed to communicate?

A parliamentary inquiry found that no specific intelligence could have warned of an impending attack in Bali in October 2002. Instead, controversy focused on another site: that intelligence assessments about the danger posed to foreigners in Indonesia by radical Islamic groups was not “adequately reflected” in the DFAT Travel Advice.¹⁴ In their testimony to the Inquiry, DFAT and ASIO had admitted there was no joint consultation on the preparation or wording of Travel Advice, and that their roles had been “too compartmentalised”. But beyond this isolated problem, there was a degree of confidence in the AIC that the same scale of communication and responsibility failure would not occur in Australia, where ASIO had long been understood to be the lead agency on terrorist threats. The ASIO Act makes clear that its responsibility is security intelligence—whether domestic or foreign – not simply domestic intelligence. It had always been the case that ASIO, in addition to its own domestic intelligence collection, gains full access to all relevant foreign intelligence collected by the AIC.

But this does not mean there is complacency in the AIC about potential communication problems. The proliferation of parts of government involved in intelligence in the age of transnational threats necessitates good communication with agencies unfamiliar with the practices and understandings of the AIC. Here the wider intelligence community faces issues of maintaining security, information technology connectivity, and most importantly, differences in informational culture. Organizations with different informational cultures have different understandings of what is important intelligence and what is background, what can and should be shared and what shouldn’t. The biggest potential challenge here lies in establishing effective and seamless communications with the Australian Federal Police (AFP), whose external operations are the source of a wealth of valuable intelligence on transnational threats. Intelligence work is essentially forward-looking, seeking to predict and forestall emerging threats to state and society; police work usually involves collecting evidence on a case-by-case basis after a crime has been committed. As one witness told the 9/11 Commission, foreign intelligence organizations tend to play “zone defence”, watching for patterns and developments in geographic areas of responsibility, whereas law enforcement agencies play “man on man”, seeking and tracking identified perpetrators of crime.¹⁵ Information-sharing difficulties abound: intelligence agencies are unwilling to let covert intelligence material be used in making criminal cases against the perpetrators of transnational crimes, while police won’t allow evidence crucial to an ongoing investigation be passed outside the bounds of the investigation. Often police and intelligence agencies fail to realize they have information that is valuable to each other.

Another charge levelled at Australian intelligence agencies, and also their counterparts in the United States and the United Kingdom, was that their judgements about Iraq’s WMD capabilities represented a failure of critical thinking. Intelligence assessment agencies, faced with uncertain and incomplete intelligence on Iraq’s WMD activities, inferred from Baghdad’s patterns of past weapons acquisitions and extensive attempts to avoid arms inspections that Iraq had WMD which it was trying to preserve and conceal. This was a judgement that came under extensive scrutiny after ISG inspections failed to find any WMD in Iraq after the war. The Flood Report argued that in the Australian agencies there was “a failure rigorously to challenge preconceptions or assumptions about the Iraqi regime’s intentions” and more broadly “little evidence of a consistent and rigorous culture of challenge to and engagement on intelligence reports from collectors, and limited evidence of dialogue on assessed material.”¹⁶ Given the prominence of “groupthink” or “cognitive dissonance”-type explanations in many scholarly discussions of intelligence failures,¹⁷ public critics also emphasized closed minds and lack of critical thinking in the AIC.¹⁸ With the luxury of hindsight, various inquiries offered alternative explanations for Saddam’s concealment activities: from bluff and national pride,¹⁹ to brinkmanship and miscalculation,²⁰ to the paranoid and chaotic politics of dictatorship.²¹ Intelligence analysis, like any interpretive activity, is prone to what Robert JERVIS calls ‘the parochialism of the expert’, where long experience and detailed knowledge give added certainty and a degree of uncontestability to judgments.²² In the case of Iraq’s concealment activities, the initial assessment led to a circular process of collection and analysis, as more and more intelligence on Iraq’s concealment activities was collected after late 2002, in turn serving to firm up judgments that these activities signified the existence of WMD.²³ This also points to another danger: as analysts have to devote increasing amounts of their time to digesting growing inflows of intelligence, there can be a tendency to start to think that the patterns emerging in the covert intelligence constitute the whole picture, when in reality covert intelligence can only provide a partial picture provided by the sources collection agencies can access and recruit. There is also an attendant danger in “fetishising” covert intelligence as inherently more valuable and indicative than other types of information.²⁴

The final familiar charge leveled at the AIC is that due to its close intelligence relationship with the United States and United Kingdom, it acted as an uncritical cipher in dragging Australia (again) into “other peoples’ wars”.²⁵ It has long been charged that the UKUSA intelligence community architecture, which connects the United States, United Kingdom, Canada, Australia and (formerly) New Zealand in an intimate intelligence-sharing relationship, is a major conduit for the “colonization” of Australia’s strategic thinking by its great power allies.²⁶ One commentator observed that there were “similar exaggerations and errors” in the Australian, American and British cases for war, leading to the strong implication that the AIC had simply regurgitated “previously digested US and British [intelligence] material” for the government.²⁷ Taking these charges seriously, the Parliamentary Committee Inquiry into Iraq’s WMD scrutinized what it called the AIC’s “external independence—that assessments are [driven by] Australia’s national interest, independent of the interests of other nations.”²⁸ Because of the geographic specializations of the UKUSA partners and the smaller size of the AIC, perhaps 97 per cent of the raw intelligence on Iraq’s WMD was sourced from American and British collectors. As the Flood Report observed, it is “difficult for analysts to query and challenge foreign-sourced material … in general, source descriptions were less than helpful for analysts, tending to be selected from a small group of standard phrases.”²⁹ Australian agencies also kept a close watch on American and British judgments about Iraq’s WMD and links to AL QAEDA. Yet these elements do not necessarily lead to the conclusion that Australian assessments simply parroted those of their American and British counterparts. Both the Parliamentary Committee and the Flood Inquiry found that Australian agencies had been “much more measured” than those in Washington and London about Iraq’s WMD capabilities. Australian agencies, for instance, rejected American claims about Saddam’s links to AL QAEDA and British claims about Iraq’s ability to deploy WMD within 45 minutes. Such small but significant differences among the allies suggest that the AIC took seriously its role of critically evaluating raw intelligence received and coming to independent judgments based on that material. That Australian agencies were not simply postboxes for passing on American and British assessments was demonstrated by a furore that developed in mid-2003, when they were roundly criticized for failing to pass on to the government US Bureau of Intelligence and Research (INR) doubts about intelligence reporting Iraqi attempts to source uranium from Africa.³⁰

New Problems

The Parliamentary Joint Committee Inquiry into Intelligence on Iraq’s WMD found that ONA’s judgments on Iraq’s WMD changed in tone almost overnight in mid-September 2002, when it began making “statements [on Iraq’s WMD capacities] in the indicative rather than the subjunctive mood, thereby denoting greater culpability on Iraq’s part and certainty on the part of the analysts.”³¹ The event causing ONA’s enunciation of more definite judgements on Iraq’s WMD was, it was implied, a request made by the office of Foreign Minister Alexander DOWNER for a document compiling unclassified information on Iraq’s WMD activities. This led to a storm of accusations of “politicisation” of the AIC. The Flood Report defined politicization as relating both to the content of what is reported, and to what is not reported. In the former case, there is potential for either overt pressure from government or policy departments to reach a particular judgement, or an unconscious identification by the analyst with a particular policy outcome. In the latter case, analysts and managers may choose, either consciously or unconsciously, not to report on a particular issue of potential sensitivity for government.³²

Critics identified politicization occurring in the AIC as a result of the involvement of heads of intelligence agencies on Prime Ministerial travels,³³ and in meetings of the National Security Committee of Cabinet,³⁴ or as the outcome of populating AIC management ranks with “politically astute” careerists from policy departments³⁵ who believe that their “career prospects [are] not enhanced by giving Ministers unpalatable advice or giving them information they did not want to hear.”³⁶ Others claimed that ONA had allowed its concern with policy relevance to slip into policy advocacy, meaning selectively reporting or shaping intelligence to provide the appearance that a certain policy is working or appropriate despite evidence to the contrary.³⁷ Some critics accused the AIC of not enough involvement in the policy process, implying it is the role of intelligence assessments to “challenge, confront, question [and] dissent from conventional Departmental wisdom,”³⁸ or arguing that the AIC’s fault was in “failing … to induce policy makers to learn,”³⁹ or that “If the decision maker, in this case the head of government, refuses to accept and act on vital intelligence, then intelligence officials have failed.”⁴⁰

The issue of the blurring of intelligence and policy roles goes to the heart of a foundational organizing principle of the AIC. The Hope Report, which in 1977 recommended the establishment of ONA and the formalization of co-ordination and oversight mechanisms for the AIC, quoted Walter Lippmann’s advice: “The only institutional safeguard [guaranteeing the objectivity and impartiality of analysis] is to separate as absolutely as it is possible to do so the staff which executes from the staff which investigates. The two should be parallel but quite distinct bodies intrinsically uninterested in each other’s personal success.” To the outside, the process of intelligence analysis slipping from policy-relevant but objective reporting to producing reports that are anticipated to please Ministers and support their policies looks intuitively simple, even inevitable. If irrelevance leads to marginalization, then surely “the pressures to tell policy-makers what they want to hear can become irresistible.”⁴¹ Yet care should be taken in leaping to such judgements. Democratic government is by nature risk-averse: given the political costs that can attend a policy mistake or a proven failure to take indications of an emerging danger seriously, the process of government is prepared to endure significant short-term costs to allay the possibility of a long-term risk. By their nature, intelligence agencies are intended to warn governments of emerging risks. In an age of growing media and parliamentary scrutiny into what governments did and knew, there is a greater danger for intelligence agencies and personnel in failing to report significant risks or evidence of policy shortcomings than there is in telling ministers what they don’t want to hear. No intelligence manager would last long if it came to light that he or she had glossed, ignored or refuted without good reason a development that could seriously affect the success of government policy. If the name of the game for intelligence is to capture ministerial attention, the presentation of discomforting or warning information, if adequately justified, will be much more compelling than merely presenting evidence for a policy’s wisdom and effectiveness. And a government which by design or acquiescence allows its intelligence agencies to slide towards policy advocacy risks depriving itself of a purpose-built mechanism for warning about impending dangers or policy failures. Intelligence agencies must remain policy-relevant—understanding which issues are important to the government, where more information would assist in reaching decisions, and how intelligence product is used to shape decision-making and policy outcomes. Their role should then be to report to government what the intelligence available suggests about the wisdom, sustainability and effectiveness of the policy. Against this background, charges of politicization of the AIC are serious indeed, and would require much more evidence than the apparent failure of judgment over Iraq’s WMD. Though dismissed by critics as an inevitable finding by a government “insider”, the Flood Report’s finding that there was no evidence of politicization of the AIC is justified.

A related issue of controversy was the public use of intelligence by government. This development had occurred earlier in Australia than in the United States or United Kingdom, when Prime Minister John Howard explicitly cited an ONA Report to support the government’s case against asylum-seekers just prior to the November 2001 federal election. Then, in the lead-up to the Iraq War, although the Australian government didn’t use intelligence material in public presentations as its allies in Washington and London did, it referred repeatedly to intelligence in publicly making the case for war. In mid-September 2002, the office of Foreign Minister Alexander Downer requested an unclassified document listing intelligence indicating Iraq’s possession of WMD capacity. As the Butler Inquiry commented in the British context, these developments were unprecedented in the direct use of intelligence to justify policy and in the commissioning of intelligence agencies to produce documents either for public release or as the basis for ministerial statements.⁴² Two major criticisms were made of the Australian government’s public use of intelligence. First, it was argued that the authority and objectivity of the AIC was being misused to bolster the government’s position. Of particular concern were government assertions that the intelligence agencies had “cleared” ministerial statements on the case for the war in Iraq. When it emerged in the PJCAAD Inquiry that ONA had focused only on the accuracy of specific references to intelligence, the Committee reacted angrily to the government’s casuistry:

“Accuracy must also encompass whether the picture being presented is complete. Ignoring significant elements of fact or opinion when citing intelligence assessments can have a distorting effect. A true and accurate interpretation must consider the total balance of the point of view being adduced in support of a policy.⁴³”

By misusing intelligence in this way, the government risks undermining “public confidence in the value of intelligence and its credibility.”⁴⁴ Second, there was concern that, by being forced from an intelligence reporting context into a public information one, the AIC had abandoned accuracy in favour of coherence and digestability. When writing for public consumption, it was suggested, the imperative is to highlight judgements rather than the detail and uncertainties of the evidence. This was the implication of the PJCAAD Report, which dated—without citing any evidence of causality—ONA’s downplaying of the uncertainties and gaps in the intelligence on Iraq’s WMD capacities from the preparation of the unclassified intelligence file for the Office of the Foreign Minister. Thus the public use of intelligence bolstered the charges of politicization of the AIC.

The temptations for governments to use intelligence publicly are unlikely to abate. In an era of transnational threats, which by their nature emerge unlooked for, the new security imperative is to act to prevent such threats materializing. As British Prime Minister Tony BLAIR told American President George W. BUSH at a meeting in Crawford, Texas in April 2002, to allow threats to develop through inaction or inattention “would be to ignore the lessons of 9/11.”⁴⁵ Yet in democracies, the public must still be provided with a rationale for policy, even if the motivations for the policy are not apparent. In this situation, intelligence is bound to be used to bolster the case for action. BLAIR told the Butler Inquiry in the United Kingdom that “the purpose of the [publicly released British intelligence] dossier was simply to say ‘this is why we think this is important … here is the intelligence that means this is not a fanciful view on our part, there is a real issue here.”⁴⁶ Here the line is thin between providing the electorate and opposition with a common, objective corpus of intelligence on which they can judge the wisdom of alternative policies, and using intelligence to bolster public support for government policy. Shadow Australian Foreign Minister Kevin RUDD accused the government of sliding towards the former in “cherrypicking” from intelligence assessments only those judgements that supported its case for war.⁴⁷ Here governments face a risk that if they are perceived to be misusing intelligence, or skewing it to partisan purposes, or if the intelligence they use proves inaccurate, it will devalue the authority of intelligence over time, and impede their ability to use it to explain policy actions against transnational threats.

3. The Reforms

While the familiar intelligence failures generated pressures to increase the proficiency of the AIC, the newer problems produced concerns over its probity. Taking these issues into account, the recent reforms have had three basic effects on the AIC: they have reinforced the organizing principles of the AIC while overhauling its co-ordination mechanisms; upgraded external oversight of the AIC; and increased the resourcing and size of most of the intelligence agencies.

Functional Differentiation and Co-ordination

In the face of evidence of “horizontal” intelligence failures—inappropriate scrutiny of sources, “compartmentalisation” of intelligence and warning roles, insufficient contestation of judgements—some commentators suggested following the American lead and merging the AIC and law enforcement agencies into a new national security department. This became the policy of the Opposition Australian Labour Party. But the Flood Report and the government rejected such radical restructuring, opting instead to reassert functional differentiation in areas where it had become blurred, and instead strengthen the AIC’s co-ordination mechanisms.⁴⁸ In his Report, FLOOD argued that “The clarity of the role played by each agency in the Australian Intelligence Community, together with minimal duplication of capability, is one of the strengths of the Australian system.”⁴⁹ Here was also a statement that the AIC was not affected by the type of information-sharing problems experienced in the United States,⁵⁰ or the United Kingdom,⁵¹ and that part of the reason for this was the AIC agencies’ clear sense of functional specialization and interdependence. There is a strong feeling inside the AIC that different intelligence functions require different organizational cultures. The main reassertion of functional differentiation occurred in deconflicting the roles of the two assessment agencies, ONA and DIO. The latter’s remit was carefully pruned to intelligence analysis in support of the Department of Defence and the Australian Defence Force’s requirements, while ONA’s role as the “peak national foreign intelligence assessment agency” was “reasserted”. FLOOD and the government rejected criticisms that such deconflicting removed a much needed source of contestation of intelligence judgements from the AIC, instead arguing that fostering a culture of debate and challenge internally to ONA and DIO is a more efficient use of resources.

The counterpart to reinforcing functional differentiation was to improve intelligence co-ordination. Co-ordination mechanisms have been forged at all levels. At the lowest level, Australia followed the United States and the United Kingdom in establishing a joint, 24-hour, all-source terrorist threat centre, termed the National Threat Analysis Centre (NTAC). Housed in ASIO, NTAC is staffed by secondees from all of the AIC agencies, plus the AFP, DFAT and the Department of Transport and Rural Services. It is acknowledged as the authoritative source of threat advice, and plays a regular role in DFAT’s preparation of Travel Advisories. At a higher level, the FLOOD reforms resulted in the creation of the Foreign Intelligence Co-ordination Committee (FICC), and the strengthening and better resourcing of the roles of ONA and its Director-General as co-ordinators of the AIC. The FICC formalizes a committee that had developed informally, known as the Heads of Intelligence Agencies Meeting (HIAM), in gathering together for a monthly co-ordination meeting the heads of all AIC agencies and the AFP, plus senior representatives from DFAT and the Departments of Prime Minister and Cabinet (PM&C). Chaired by the Director-General of ONA and resourced by a Secretariat within ONA, the FICC deals with co-ordination, strategic planning and evaluation within the AIC. The FICC has been provided with a formal interface with the policy process through its relationship to the newly created National Security Division within PM&C, and through the Secretaries’ Committee on National Security (SCONS) to the National Security Committee of Cabinet (NSCC). All of these mechanisms are intended to serve the functions of making sure the whole of government is more responsive to intelligence issues, and that intelligence judgements are more closely integrated into government policy-making.

The Australian government has been able to adopt such a relatively minimalist approach to horizontal co-ordination issues due to the AIC’s small size and established culture of collegiality. The new, formal co-ordination mechanisms are assisted by long-established understandings of roles and responsibilities. The regular drawing of AIC managers from DFAT and the Department of Defence, so often criticized as resulting in the “dumbing down” and de-professionalization of the AIC, has brought a culture of practical co-operation among the senior managers of the intelligence agencies. This commitment at senior levels drives co-operation at lower levels, ensuring a much more rigorous, judicious and collaborative approach to intelligence collection tasking and feedback; it also helps allay collectors’ concerns about the instrusiveness of too rigorous a process of co-ordination and oversight. So important is the “hidden lubricant” of co-operation within the AIC that many worry that the growth and professionalization that will follow the Flood reforms will threaten the community’s culture of easy collegiality.

External Oversight

A series of scandals involving Australia’s intelligence collection agencies over the years has resulted in the steady strengthening of external oversight mechanisms. The main innovations have been the creation of the Parliamentary Joint Committee on ASIO, ASIS and DSD and the Inspector-General of Intelligence and Security (IGIS), a statutory office with the responsibility to investigate and report regularly on the collection agencies’ operations and probity, with the ability to initiate investigations into collection agencies, and with the powers of a royal commission. The PJCAAD and IGIS have both played a prominent role in recent years in assessing the propriety and proficiency of the agencies of the AIC.

The Flood Report took the charges of politicization of the AIC seriously. Although it cleared Australian intelligence agencies of charges of politicization, it acknowledged their need to preserve their reputation for rigour and objectivity: “The functioning of Australia’s intelligence agencies is a matter of great public interest and scrutiny.” While conceding that foreign intelligence assessment agencies rarely threaten the rights or privacy of Australian citizens, FLOOD argued that extending parliamentary oversight to ONA and DIO “would enhance confidence in the Parliament and the public that the full range of intelligence agencies is accountable to a senior group of parliamentarians,” and recommended a concomitant name change to the Parliamentary Joint Committee on Intelligence and Security (PJCIS). With a similar logic, the Flood Report recommended extending the remit of IGIS to include ONA and DIO, including giving IGIS the capacity to initiate inquiries at his or her own discretion into the analysis agencies’ activities, and in particular periodically reviewing ONA’s statutory independence.

While the intentions of these reforms are laudable, it is difficult to see how they will work in practice without compromising the capacities of the analytical agencies to make hard, objective judgments. FLOOD seemed to recognize the dangers of exposing analytical agencies’ judgements to regular parliamentary review when he suggested that “Just as the advice that officials provide to Ministers is not disclosed in Senate Legislation Committee hearings, the judgments of assessment agencies should not be subject to parliamentary scrutiny.” Rather, in addition to parliamentary review of ONA’s and DIO’s “budgetary and administrative matters”, “the process by which ONA and DIO produce their assessments is an area which could be open to parliamentary scrutiny.”⁵² It is hard to see how a parliamentary committee will be able to confine itself to these matters. As the practice of Senate Estimates Committee hearings shows, remits to scrutinize “budgetary and administrative matters” are routinely overstepped. And it will be impossible properly to review the process by which assessments are produced without having some idea of what those assessments were. Added to these concerns are growing doubts that parliamentary committees can act as impartial overseers, with many commentators noting the increasingly bitter partisan divisions that are characterizing debates over issues of security and intelligence.⁵³ The inherent subjectivity of reviewing the process and outcomes of assessment processes is itself revealed in discrepancies in the two reviews on Iraq WMD: while the PJCAAD dated a hardening of ONA judgments from mid-September 2002,⁵⁴ the Flood Report claimed that these changes did not occur in ONA Reports until early 2003.⁵⁵ Faced with the possibility of partisan politics and subjectivity tainting the parliamentary review process, it is hard to see how intelligence assessment agencies will not start to factor in possible future parliamentary scrutiny when making their judgments. The danger is that the result will be a tendency to be more cautious and qualified in assessments, the very charges that many critics have argued result from the politicization that these oversight reforms are intended to address.

Resourcing and Expansion

Over many years, there has developed in Australia’s security and intelligence institutions a preference for maintaining small but very high-quality capabilities. The Australian Defence Force is by regional standards a small military organization, yet it takes pride in its very high levels of training and advanced technology. The AIC has similarly remained much smaller than the intelligence communities of its closest allies—the United States spends 100 times the amount Australia spends on intelligence; Britain spends four times Australia’s amount. Partly, the AIC has been able to afford the luxury of being small because it is conscious that it is a part of the most powerful intelligence-gathering collaboration in the world, the UKUSA agreement. The focus of Australia’s intelligence collection efforts is clearly spelled out in the Flood Report: “On Southeast Asia and the South Pacific, Australia needs to be a global leader in intelligence … Australia’s intelligence on North Asia should be exceptionally good … Intelligence on South Asia should be very good,”⁵⁶ implying that for intelligence on other regions, Australia is able to rely on its UKUSA partners. In terms of analysis, FLOOD argued that “Australia needs to maintain an independent capacity across the whole spectrum of issues on which the government needs advice. Australia must be able to critically assess intelligence sourced from others.”⁵⁷

Working under the political imperative to leave no stone unturned in protecting Australia from transnational threats, the government had begun to expand the budgets and powers of the frontline collection agencies in 2001. Australia’s spending on foreign and security intelligence increased from $332 million in 2000-2001 to $659 million in 2004-2005, before the adoption of the Flood Report reforms, with the vast majority of that increase going to ASIO, ASIS and DSD. Staff levels in ASIO and ASIS have or are projected to double in the coming years.⁵⁸

The analytical agencies’ preference for “boutique” capabilities came under sustained pressure during the controversies and inquiries. Parliamentary committees appeared genuinely astonished at how few analysts ONA had to deal with the growing amounts of intelligence received on issues such as Iraq and terrorism. The PJCAAD learned that ONA had two analysts with Middle East expertise and one with strategic analysis expertise working on Iraq’s WMD capacities, and that “their focus was not exclusively on Iraq during this time.”⁵⁹ FLOOD recommended a doubling of ONA’s staff from 74 to 145 and a budget increase from $13.1 million to $25 million per annum. He also recommended transferring the Open Source Unit from DFAT to ONA, a move which acknowledges open-source information as important intelligence, which responds to the growing emphasis on open-source information among Australia’s intelligence allies, and which will systematize the mining and use of open-source information and harmonize it with the tasking of covert intelligence collection.

Faced with such a sudden expansion in budgets and personnel, questions have been raised about the AIC’s ability to find suitable recruits while maintaining standards, and to “digest” such large numbers of new staff.⁶⁰ Generally, managers in intelligence agencies report that, contrary to media reports, applications have surged since September 11, benefiting from greater public interest in the intelligence community. Apart from certain language and technical skills, there has been little or no problem filling positions at more junior levels. The AIC has instituted a common process of security vetting and a common induction course for new recruits to all agencies. At more senior levels, or for certain skills sets, some agencies have found that their location in Canberra, away from the country’s major population centres, has hampered their ability to attract appropriate staff. Some agencies have run into problems arising from a shortage of mid-level managers with broad intelligence experience who are able to manage and mentor new staff and provide quality control. In areas of skills shortages, there has been some competition among agencies, but this has been managed collegially. The heads of the AIC agencies recognize that some staff mobility among agencies builds communication links and increases the stock of mutual understanding in the community.

4. Conclusion

The recent changes to the AIC have reinforced, rather than challenged, the basic organizing principles according to which it has always operated. The judgments of various inquiries and the government seem to have been that, if properly resourced and co-ordinated, the AIC should be able to cope adequately with emerging challenges and rectify past failings. It is too early to judge whether these judgments, and the reforms they have entailed, are indeed adequate to cope with the challenges posed by transnational threats, or the seemingly recurring “familiar” problems with intelligence processes, or especially the political pressures on intelligence agencies. If these challenges are too great to be solved by minor structural changes and resourcing upgrades, the next round of major reforms to the AIC may be closer than a quarter century away.

1. Commonwealth of Australia, Report of the Inquiry into Australian Intelligence Agencies, July 2004, Canberra: Goanna Print, 2004
   
2. Michael WESLEY, “Analysing Transnational Intelligence”, in Carole DUMAINE and L. Sergio GERMANI (eds) New Frontiers of Intelligence Analysis, Collected Conference Papers, Sherman Kent School for Intelligence Analysis, 2004
   
3. Emile DURKHEIM, The Divisions of Labour in Society, (New York: Free Press, 1984).
   
4. MICHAEL WESLEY, “Towards a Realist Ethics of Intervention”, Ethics and International Affairs, This problem has been somewhat more polemically discussed by Ulrich BECK in his Risk Society, Towards a New Modernity. trans. Mark RITTER, (London: Sage Publications, 1992).
   
5. Gregory F. TREVERTON, ‘Reshaping Intelligence to Share with “Ourselves”’, CSIS Commentary, No. 82, Summer 2003, p. 5.
   
6. Lee S. STRICKLAND, “The Information Shortcomings of 9/11”, Information Management Journal, 38:6, November/December 2004, p. 43
   
7. A course of action that is “good enough” - that satisfies. ALLISON, Graham T, “Essence of Decision-making, Exploring the Cuban Missile Crisis”, 1977.
   
8. The UKSIS has continued to assert the validity of the Iraq-Niger uranium claims, based on a HUMINT source it has refused to discuss further.
   
9. See Bob DROGIN and John GOETZ, “U.S. Whiffs on ‘Curveball’; Close Up”, The Los Angeles Times, 27 November 2005
   
10. The Rt. Hon. The Lord BUTLER of Brockwell KG GCB CVO, Review of Intelligence on Weapons of Mass Destruction, Report of a Committee of Privy Counsellors, House of Commons, London: The Stationery Office, 2004, p. 108
    
11. Commonwealth of Australia, Intelligence on Iraq’s Weapons of Mass Destruction, Report of the Parliamentary Joint Committee on ASIO, ASIS and DSD, Canberra:CanPrint Communications, December 2003, p. 74 (hereafter referred to as “PJCAAD Report”)
    
12. New York Times, 20 October 2002
    
13. Philip ZELIKOW et al., The 9/11 Commission Report: Final Report of the National Commission on Terrorist Attacks Upon the United States, Authorized Edition, New York: WW Norton and Company, 2004, p. 263
    
14. Commonwealth of Australia, Bali 2002: Security Threats to Australians in Southeast Asia, Report of the Senate Joint Foreign Affairs, Defence and Trade References Committee, Canberra: Senate Printing Unit, August 2004 (hereafter “JFADT Report”), p. xvii
    
15. 9/11 Commission Report, p. 268
    
16. Flood Report, pp. 25-6.
    
17. For example, Irving L. JANIS, Victims of Groupthink : A Psychological Study of Foreign-Policy Decisions and Fiascoes, Boston: Houghton Mifflin, 1972; Roberta WOHLSTETTER, Pearl Harbor : Warning and Decision, Stanford: Stanford University Press, 1962
    
18. See Paul MONK, “Why Didn’t We Think of That?”, Australian Financial Review, 2 November 2001; Lee CORDNER, “Reaping the Rewards: Executive Expectations of the Intelligence Investment”, Future Directions International, May 2003, p. 3.
    
19. PJCAAD Report, p. 80
    
20. Flood Report, p. 23
    
21. Butler Report, p. 15
    
22. Robert JERVIS, ‘What’s Wrong with the Intelligence Process?’, International Journal of Intelligence and Counter-Intelligence, Vol. 1, 1986, pp. 31-2.
    
23. Butler Report, p. 89.
    
24. WESLEY, “Analysing Transnational Intelligence”, p. 102
    
25. See Peter MANNING, “A Colonial State of Mind”, Griffith Review, 5, Spring 2004
    
26. See for example Joseph A. CAMILLERI, ANZUS : Australia's Predicament in the Nuclear Age, Melbourne: Macmillan, 1987
    
27. Geoffrey BARKER, Sexing It Up: Iraq, Intelligence and Australia, Sydney: UNSW Press, 2003, p. 79.
    
28. PJCAAD Report, p. 44.
    
29. FLOOD Report, p. 25.
    
30. “Australian Intelligence Agency Admits It Did Not Tell Government of US Doubts Over Iraqi Evidence”, Associated Press, 10 July 2003; “The PM and the Public Service”, Sydney Morning Herald, 15 July 2003; Allan BEHM, “How We Got Iraq’s WMDs Wrong”, The Australian, 14 July 2003
    
31. PJCAAD Report, p. 33
    
32. Flood Report, p. 105
    
33. Tom ALLARD, “The War of Secrets”, Sydney Morning Herald, 6 March 2004
    
34. Allan GYNGELL, “Closer Ties That Weaken the System”, Sydney Morning Herald, 24 February 2004
    
35. Andrew WILKIE, “A Lack of Intelligence”, Sydney Morning Herald, 31 May 2003
    
36. BARKER, Sexing It Up, p. 73; see also Philip KNIGHTLY, “Deadly Secrets”, The Bulletin, 122, 27 April 2004; Tom ALLARD, “Canberra Was Warned on Spy Reports, Says Ex-Analyst”, Sydney Morning Herald, 27 May 2003.
    
37. See “ONA’s Independence is Part of a Delicate Balance”, Sydney Morning Herald, 15 April 2004
    
38. Gary KLINTWORTH, “Flood Report: Mild Applause”, Asia Pacific Defence Reporter, September 2004, p. 12
    
39. Paul MONK, “Learning to See”, Griffith Review, 5, Spring 2004
    
40. CORDNER, “Reaping the Rewards”, p. 3
    
41. Desmond BALL, “Silent Witness: Australian Intelligence and East Timor”, The Pacific Review, 14:1, 2001, p. 36
    
42. Butler Report, p. 76
    
43. PJCAAD Report, pp. 93-94.
    
44. Ibid., 74
    
45. Quoted in the Butler Report, p. 71.
    
46. Quoted in ibid., p. 77
    
47. Quoted in Cynthia BANHAM, “Report is Proof of Deception”, Sydney Morning Herald, 2 March 2004
    
48. Verona BURGESS, “Ruddock Rejects Mergers”, Australian Financial Review, 11 May 2005
    
49. Flood Report, p. 69
    
50. 9/11 Commission Report, p. 353
    
51. Butler Report, p. 111
    
52. Flood Report, p. 58
    
53. See for example Carl UNGERER, “Issues in Australian Foreign Policy: January to June 2004”, Australian Journal of Politics and History, 50:4, 2004
    
54. See PJCAAD Report, p. 33
    
55. See Flood Report, p. 29
    
56. Ibid., pp. 12-13
    
57. Ibid., p. 15
    
58. “Australia Doubles Spy Agency”, SBS World News Headline Stories, 17 October 2005
    
59. PJCAAD Report, p. 47.
    
60. See for example Julie MACKEN, “I Spy a Problem on Intelligence”, Australian Financial Review, 24 November 2005; Deborah Snow, “A Shortage of Spooks”, Sydney Morning Herald, 15 November 2004
    

---

Commentary is a regular publication of the Analysis and Production Branch of CSIS. Inquires regarding submissions may be made to the Chairman of the Editorial Board at the following address:

The views expressed herein are those of the author, who may be contacted by writing to:

CSIS
P.O.Box 9732
Postal Station T
Ottawa, Ontario K1G 4G4
Fax: 613-842-1312

ISSN 1192-277X
Catalogue JS73-1/88