Direction, Collection, Analysis and Dissemination: Have modern methods of intelligence gathering significantly reduced the probability of ‘intelligence failures’

Direction, Collection, Analysis and Dissemination Hasan Q. Zulfiqar Have modern methods of intelligence gathering significantly reduced the probability of ‘intelligence failures’? 1000331 Autumn 2012 Word Count: 2,433 1 This study attempts to assess whether modern methods of intelligence gathering have significantly reduced the possibilities of intelligence failures. In light of the ‘technological revolution’ in intelligence gathering and analysis, there is a growing need to find better and effective ways to use this technology. The maelstrom introduction and usage of newer methods of signals intelligence (SIGINT) and imagery intelligence (IMINT) has meant that intelligence analysts are in danger of fallacious complacency, without first understanding the limitations of these methods. I begin by stating what these modern methods of intelligence gathering are; comparing the characteristics, benefits and shortcomings of SIGINT and IMINT with traditional human intelligence (HUMINT), in light of the intelligence cycle. I then go onto explain what exactly an intelligence failure is composed of and what the main causes of intelligence failures are. Here, I also distinguish an intelligence failure from a strategic surprise which is formed of and the result of the former. The analysis is then expanded into specific case study comparisons between those intelligence failures which involved primarily HUMINT gathering and those which overwhelmingly employed newer, more modern methods. Though the basis of my approach is theoretical in nature, I build upon empirical research and comparative analysis to find that if employed properly, modern methods of intelligence gathering theoretically have a slightly higher probability of reducing the chances of intelligence failures, albeit it inevitably depends on the analysis of said intelligence and how it is put to use. Therefore, the main findings lead to the conclusion that intelligence failures are not wholly dependent on the intelligence gathering methods employed but are mainly a result of inadequacies in the intelligence cycle. 2 Why is intelligence key? Intelligence, good and as complete as possible is vital for a nation’s strategic policy and direction. It is also necessary for any pre-emptive and preventative actions. Looking at current events with relation to Iran, Western powers need to know the exact location and defensive capabilities of all nuclear sites within the country, as missing even one or part of one may mean the risk of retaliatory nuclear strikes, or a dirty bomb. Therefore intelligence agencies and analysts are tasked to provide as complete a picture as possible for policy makers to act upon. Any gap in the threat assessment can have devastating consequences. This applies not only to offensive intelligence gathering but also counter-intelligence, which is primarily defensive in nature. Let us further sub divide the former into (i) Objective specific intelligence (OSI), which is to seek out specific strategies of the enemy and relay targeted information on a specific issue or operation, and (ii) Opportunities and vulnerabilities intelligence (OVI), which is to constantly monitor target activities in the chance of gleaning valuable information without knowing exactly what to look for - the bulk of the job which intelligence agencies do, day-to-day.1 The latter, counter-intelligence, can also be divided into sub categories; (i) Defensive Intelligence, which is to constantly maintain a high degree of operational security to guard against the enemy gaining a full picture of one’s own operations and knowledge – simply, hiding your own secrets. And, (ii) Deceptive Intelligence, which is to place and relay to the enemy, information about oneself which is misleading and wrong, with the aim of giving the opponent a distorted picture of oneself.2 Our concern, however, is primarily with offensive intelligence and the methods employed to gather this. 1 2 Odom, W. E. ‘Intelligence Analysis.’ Intelligence and National Security (Vol 23, No.3, 2008) Ibid. 3 An introduction of modern methods Megalithic change occurred with earnest in the ‘second oldest profession in the world’ with the widespread introduction of technological innovation by US Secretary of Defense Robert McNamara, in the early 1960’s.3 The creation of the National Security Agency (NSA), National Reconnaissance Office (NRO), and later the National Geospatial-Intelligence Agency (NGIA), with larger budgets, manpower and eventually, clout, than the Central Intelligence Agency (CIA) meant that there was a paradigm shift in the methods of collecting intelligence in the US and the world. Whole agencies are now dedicated to SIGINT and IMINT collection, whilst others devoted to the analysis of this data, dwarfing the resources allocated to the CIA and its exclusively HUMINT subsidiary, the National Clandestine Service (NCS). The attacks in New York and Washington in September 2001 accelerated the demand for more SIGINT and IMINT data from around the world. Budgets were increased dramatically and organizational transformation initiated, sometimes with little prior evaluation as to its’ effects.4 Modern technology such as drones and NASA’s Earth Orbiting System (IMINT) and high powered listening and long distance video recording devices (SIGINT) are all examples of modern intelligence collecting methods. SIGINT and IMINT operate very differently to HUMINT, as; (i) the intelligence collecting process is very different and can be done from afar, and (ii) the amount of data that can be collected is far greater in quantity. Table 1.1 details the main differentiating characteristics of the two, whilst also listing some of the benefits and drawbacks of both. 3 Richelson, Jeffrey T. The US Intelligence Community, 6th ed. (Westview: New York, 2012) Immerman, R. H. ‘Transforming Analysis: The Intelligence Community's Best Kept Secret.’ Intelligence and National Security (Vol 26, No. 2-3, 2011) 4 4 HUMINT Used primarily for specific informational requirements (OSI) Limited, condensed information Takes longer to collect Usually difficult to obtain Sources have short useful life span Source can be unreliable Low accuracy Danger of discovery Information can be acquired which is not available electronically or by imagery Takes less time to analyse SIGINT/IMINT Used primarily for general OVI Large tracts of information Can be collected quickly Usually highly accessible Sources are unlimited in life span Source usually reliable but can be prone to manipulation and deception Higher accuracy Done from afar, so if discovered, not too much of an issue Some information cannot be collected (e.g. paper based, human knowledge) Takes longer to analyse due to quantity of information Table 1.1 - Comparison of HUMINT with SIGINT/IMINT. Characteristics are in black font. Benefits are in green and drawbacks in red. As seen in Table 1.1, intelligence collection using SIGINT/IMINT can be seen to possess more advantages. However, analysis usually ignores the fact that both modern methods and HUMINT are used in tandem in a variety of intelligence collection operations, and that HUMINT is still somewhat central to intelligence agencies’ operations. Another factor with modern methods of intelligence collection is that it needs a multidisciplinary team based analysis because; (i) there is so much information collected, that it is impossible to be analysed by one actor, and (ii) that the variety and diversity of the information collected means that there can be no single expert on all fronts.5 The sheer quantity of information collected also means that there is a danger in making ‘intelligence operate faster than an analyst can analyse or verify.’ 6 Coupled with a perceived overreliance on SIGINT, this can lead to intelligence failures. But what is an intelligence failure? 5 Smith. Michael E. International Security: Politics, Policy, Prospects. (Palgrave Macmillan: London, 2010) George, R. ‘Intelligence and Strategy’, in J. Baylis, J. Wirtz et al, (eds), Strategy in the Contemporary World: An introduction to strategic studies, (OUP: Oxford, 2010) 6 5 The intelligence failure An intelligence failure is the causal component of a strategic surprise, which itself connotes a ‘significant negative impact on national interests.’7 The intelligence failure can usually be attributed solely to intelligence agencies and their role in the intelligence cycle, whilst a strategic surprise is the result of either, the intelligence failure meaning an incomplete threat assessment given to the policy/decision maker, or good intelligence not being acted upon by the decision maker. Since this study evaluates intelligence failures and the role played in the intelligence cycle by intelligence agencies in intelligence collection, we will not factor in too much the decision maker in our analysis. To understand what an intelligence failure is, we must understand its causes. There are multiple causes of an intelligence failure; (i) there is a collection failure, which means that the information is either lacking or contradictory. This can result from ineffective collection techniques or a lack of good direction to collect the information. (ii) There is a misinterpretation of the information collected because of a misconception of the opponent. Pre-conceived notions are dangerous in the intelligence business as one should not always expect the adversary to do what one expects him to do. (iii) There is deception on the part of enemy. This will occur in the collection stage, where the enemy uses deceptive information to place and relay information about himself which is misleading and wrong, with the aim of giving the opponent a distorted picture of himself. This, again, can lead to intelligence failures as the threat assessment is either underestimated, overestimated or outright wrong. (iv) The adversary can deny, through good operational security, the full picture of the threat assessment - defensive intelligence. (v) Compartmentation, whereby there is inadequate inter-agency sharing, meaning the pieces of the puzzle are not put together. The ‘need to know’ requirement and a separation of operational field cells can also be a factor in this. (vi) There is ineffective communication to the decision maker, and lastly, (vii) the decision maker himself is lacking.8 7 Ibid. Baylis, J. and Wirtz, J. et al. Strategy in the Contemporary World: An introduction to strategic studies, (OUP: Oxford, 2010) 8 6 Now that we have a basic understanding of what an intelligence failure is, we can move on to analyse if the modern methods of collection reduce the possibility of intelligence failures. The intelligence cycle However, before this, a short note on the intelligence cycle is necessary. Below is the simplified version; DIRECTION COLLECTION ANALYSIS DISSEMINATION Collection, though an important aspect, is only one part of the cycle, and intelligence failures can occur at any point in the cycle. This is a worthy point to note when assessing the probability of intelligence failings.9 Have modern methods reduced the probability? The title question can now be evaluated - has the use of these modern methods reduced the probability of intelligence failures? The simple answer is no. Thought it is not as straightforward as this. 9 See Hulnick, A.S. ‘What's wrong with the Intelligence Cycle.’ Intelligence and National Security (Vol. 21, No.6, December 2006) 7 Let us compare intelligence gathering before the widespread introduction of SIGINT and IMINT, the period from 1935-1985, and after its predominant use, from 1985 onwards. I have chosen these time periods, because although technological change had already come about from around the mid 1960’s, it cannot have been said to have fully taken off till around the mid 1980’s. During World War II and most of the Cold War, HUMINT gathering was the principle method used by both Western allies and the Soviet bloc, complemented with a lot of IMINT and some, if rather inaccurate, SIGINT.10 If we examine these time periods, we see many examples of colossal intelligence failures. The question is – can these be attributed to the collection phase of the intelligence cycle, or are they the result of other failings? Early failures could include the Third Reich’s failure to anticipate the correct landing site of the Allied invasion of Europe – a result of excellent deceptive intelligence employed by the Allies and the Nazi’s own failings of having pre-conceived notions of their enemy. Another failure at the dissemination stage can be said to be Stalin’s refusal to accept his own intelligence estimates of Operation Barbarossa. However, from a Western perspective, four intelligence failures stand out during this period - Pearl Harbor, the 1962 placing of Soviet missiles in Cuba, the 1973 Yom Kippur War, and the Argentine invasion of the Falklands in 1982. All these failures led to strategic surprises, compromising national security, however were not the result of a failure of intelligence collection. Warnings of Japanese plans to attack Pearl Harbor were relayed to naval high command as early as mid-1941, but were either lost in translation or never taken seriously. Plans for the placing of Soviet missiles in Cuba in 1962 were also reported to the Kennedy administration as early as 1961, albeit here it can be said that the intelligence was somewhat lacking. Though this is the one example where it can be said that HUMINT collection failed, and IMINT succeeded later, it is one of few. Intelligence in both the Yom Kippur war and the Falklands invasion pointed to the respective countries war plans, either with the amassing of Butler R. ‘Review of Intelligence on Weapons of Mass Destruction.’ Report of a Committee of Privy Counselors (The Stationary Office: London, 2004). 10 8 troops on the border or the blatant statements made by the regime available on open-source intelligence.1112 We see here a pattern emerging – one that points to failings in the analysis and dissemination stages of the intelligence cycle. Let us see if this pattern continues on after the widespread introduction of SIGINT/IMINT. Post 1985, there are three huge intelligence failings which come to mind – Saddam’s decision to invade Kuwait in 1990, the attacks on the US on 9/11, 2001 and the intelligence which reported that Iraq possessed weapons of mass destruction (WMD’s) in 2003. All three of these failings, once again, point not to a collection failure, but a failure either of the decision maker, or in the analysis phase. US intelligence monitored the build-up of Iraqi armored divisions along the Kuwaiti border, but Washington refused to act or dissuade Saddam. 13 The reports surrounding 9/11 are less concrete, but overwhelming evidence suggests that US intelligence agencies did have knowledge of an impending attack on the US by way of aircraft, and they were monitoring the hijacker cells. The daily intelligence brief to George W. Bush the day before says as much.14 This again, is a failure not of SIGINT/HUMINT collection but of improper analysis and ‘connecting the dots’ at the least, leaving aside conspiracy theories. The 2003 invasion of Iraq was justified on the pretext of Iraq possessing WMD’s - chemical and biological in nature. The intelligence for this supposition came primarily from a single source an Iraqi dissident who had ulterior motives.15 The failure to check, or corroborate his story and not use the multitude of resources available to the 16 dedicated US intelligence agencies, was either the result of a colossal mismanagement of intelligence analysis or willing blindness on the part of those concerned. All these are examples of the fact that it matters not what intelligence Shlaim, Avi. ‘Failures in National Intelligence Estimates: The Case of the Yom Kippur War.’ World Politics (Vol 28, Iss 03, April 1976) 12 Aldrich, R.J. GCHQ, (Harper Press: London, 2010) 13 Atran, Scott. ‘A Failure of Imagination: Intelligence, WMDs, and Virtual Jihad.’ Studies in Conflict & Terrorism (Vol 29, No.1, 2006) 14 Aldrich, R.J. ‘Transatlantic intelligence and security co-operation.’ International Affairs (Vol 80, No 4, 2004) 15 Jervis, Robert. Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War (Cornell University Press: New York, 2010) 11 9 collection method is used - no matter how good the methods – if there are failings in other parts of the intelligence cycle, intelligence failure will occur regardless. Conclusion In today’s world, the source of threat has shifted, from state actors, to non-state entities.16 The availability of newer methods of intelligence gathering has boosted agencies’ abilities to monitor and counter these threats. However, intelligence failures will occur nevertheless because there are misgivings in the analysis and dissemination phases. Though SIGINT and other newer methods of intelligence collection have the possibility of reducing the chances of intelligence failures, it inevitably depends on the analysis of said intelligence and how it is put into action. Drones and satellites do not decrease the likelihood of intelligence failures; they only make collection faster and easier. In this work, I have explained why it is dangerous for analysts to slip into the mistake of overreliance on technology, which is but a tool, and must be used carefully. For national institutions to have the ability to safeguard national interests, all aspects of the intelligence cycle must not be impeded and be allowed to function. Sherman Kent’s assertion still holds true.17 16 Treverton, Gregory F. Intelligence for an age of terror. (CUP: Cambridge, 2009) Taylor, S.A. ‘The Role of Intelligence in National Security’ in A. Collins, (ed), Contemporary Security Studies, (OUP: Oxford, 2010) 17 10 Bibliography Aldrich, R.J. ‘Transatlantic intelligence and security co-operation.’ International Affairs (Vol 80, No 4, 2004) Aldrich, R.J. GCHQ, (Harper Press: London, 2010) Atran, Scott. ‘A Failure of Imagination: Intelligence, WMDs, and Virtual Jihad.’ Studies in Conflict & Terrorism (Vol 29, No.1, 2006) Baylis, J. and Wirtz, J. et al. Strategy in the Contemporary World: An introduction to strategic studies, (OUP: Oxford, 2010) Butler R. ‘Review of Intelligence on Weapons of Mass Destruction.’ Report of a Committee of Privy Counselors (The Stationary Office: London, 2004). George, R. ‘Intelligence and Strategy’, in J. Baylis, J. Wirtz et al, (eds), Strategy in the Contemporary World: An introduction to strategic studies, (OUP: Oxford, 2010) Hulnick, A.S. ‘What's wrong with the Intelligence Cycle.’ Intelligence and National Security (Vol. 21, No.6, December 2006) Immerman, R. H. ‘Transforming Analysis: The Intelligence Community's Best Kept Secret.’ Intelligence and National Security (Vol 26, No. 2-3, 2011) Jervis, Robert. Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War (Cornell University Press: New York, 2010) Odom, W. E. ‘Intelligence Analysis.’ Intelligence and National Security (Vol 23, No.3, 2008) Richelson, Jeffrey T. The US Intelligence Community, 6th ed. (Westview: New York, 2012) Smith. Michael E. International Security: Politics, Policy, Prospects. (Palgrave Macmillan: London, 2010) Shlaim, Avi. ‘Failures in National Intelligence Estimates: The Case of the Yom Kippur War.’ World Politics (Vol 28, Iss 03, April 1976) Taylor, S.A. ‘The Role of Intelligence in National Security’ in A. Collins, (ed), Contemporary Security Studies, (OUP: Oxford, 2010) Treverton, Gregory F. Intelligence for an age of terror. (CUP: Cambridge, 2009) Cover design courtesy of MS Office.com Publications. All effort has been made to track the sources referred to in this article, however the author wishes to apologize if any breach of intellectual property law has unknowingly taken place.