2RHPZ
10-02-2004, 03:15 AM
DIMENSIONS OF INTELLIGENCE OPERATIONS
TEXT OF MAJ.GEN.SAMIR SINHA MEMORIAL LECTURE—2004 AT THE UNITED SERVICE INSTITUTION OF INDIA, NEW DELHI, ON APRIL 26, 2004
There is a clear distinction between an intelligence process and an intelligence operation. When one talks of an intelligence process, one has in mind the various stages in the evolution of intelligence---its collection, evaluation, analysis, assessment and follow-up action. The expression "intelligence operation" as used in the theme proposed to me for my address today restricts itself to the first stage in this entire scale of evolution--namely, the collection of intelligence, and tends to exclude from consideration the other stages in the evolution, which are as important as the collection, if not more.
2. In intelligence parlance, one uses the expression "intelligence operation" essentially with reference to two aspects of the functioning of an intelligence agency----its ability to collect secret, precise and timely intelligence of relevance to national security and to undertake covert actions, if need be in furtherance of national security objectives, if called upon to do so by the political leadership.
3. These two are the most sensitive aspects of the functioning of an agency, but not necessarily the most decisive aspects in terms of safeguarding national security. Analysis, assessment and follow-up action tend to be more decisive in this regard. A wealth of precise secret intelligence may not help in safeguarding national security if ill or inadequately analysed with unsatisfactory follow-up action. On the other hand, even incomplete intelligence, if well analysed and assessed and promptly and effectively acted upon, can often safeguard national security.
4.In a presentation of this nature, it would, therefore, be necessary to deal with the entire gamut of the intelligence evolution instead of restricting oneself to only one stage of it, namely, the collection operation. There are two ways of dealing with the subject--- in general terms by discussing the basic principles and concepts which govern the entire process or in specific terms by discussing some instances which are considered as landmarks in the evolution of the intelligence profession because of the lessons which they imparted to the cognoscenti, provided they had the desire and the ability to learn the right lessons. I propose to adopt the second in this talk.
5.All serious breaches of national security and all instances in which the national security managers, in the political leadership as well as in the policy-making bureaucracy, are taken by surprise, either by State or by non-State actors, have an intelligence-related cause and effect. For analysing this intelligence-related cause and effect, all instances of failures to safeguard national security can be attributed to one of the following five causes---zero intelligence, inadequate intelligence, inaccurate intelligence, excessive or an overload of often irrelevant intelligence and inadequate follow-up action on the available intelligence, whether adequate or inadequate.
ZERO INTELLIGENCE: FALKLANDS
6.A typical example of zero secret intelligence was the failure of the Secret Intelligence Service (SIS) of the UK, popularly known as MI-6, to collect intelligence regarding the plans of the Argentine Government to send its Armed Forces to occupy the Falkland Islands in 1982. Lord Franks, who constituted the one-man Commission, which enquired into this failure, brought on record the fact that the MI-6 had failed to detect the Argentine plans for the occupation, but held, at the same time, that it could not be held responsible for this failure because the British Government had not endowed it with the capability for the collection of such secret intelligence. Its proposal long before the war for the sanction of an intelligence collection post under diplomatic cover in the British Embassy in Buenos Aires had been rejected by the Foreign Office. In the light of this, the MI-6 was absolved of any responsibility for the failure.
7. After having done so, the Commission had pointed out that even if secret intelligence was not available, adequate open source intelligence in the form of media reports and despatches from the British Embassy in Buenos Aires was available, which, if it had been adequately analysed, would have served as a wake-up call to the Government and prompted it to rush the Navy to prevent the occupation of the islands.
8.The Falklands Enquiry highlighted certain important aspects of the various dimensions of the intelligence process:
* First, the importance of a physical presence of an intelligence agency in a targeted country, either under a diplomatic or a non-diplomatic cover, before it can be expected to collect secret intelligence of relevance to national security from that country.
* Second, the importance of a systematic monitoring of open source intelligence, whether adequate secret intelligence is available or not.
* Third, the importance of continuous and timely analysis of all intelligence having a bearing on a development, whether open or secret.
* Four, the importance of reverse analysis of a development, that is analysing a development not from your own perspective, but from the perspective of an adversary.
* And five, the importance of the analysis and assessment machinery existing and functioning independently without letting itself be influenced in its analysis and assessment by the pre-conceived ideas, conventional wisdom and political compulsions of the policy makers, whether they be of the political leadership or the policy-making bureaucracy.
9. A physical presence in the targeted country is the sine qua non of any successful intelligence collection operation. While it is for the political leadership and the policy-making bureaucracy to lay down the tasks of the intelligence agencies, it is for the agencies to decide what presence they would require, where and how for fulfilling those tasks. Very often, the intelligence agencies are denied this vital requirement due to one of two reasons:
* First, the Foreign Office or other policy-making Ministries with no responsibility for intelligence collection tend to arrogate to themselves the right to impose their non-professional judgement over the professional judgement of the intelligence agencies as to what collection presence they should have, where and how. The result: National security failures. The Falklands incident was only one such instance. There are many such instances strewn across the histories of the intelligence agencies of many countries, including India.
* Second, the disruption of diplomatic relations with a targeted country, which denies the Foreign Office as well as the intelligence agencies the benefits of a physical presence in that country. A typical example would be the difficulties faced by the US intelligence in Iraq. The non-existence of diplomatic relations between Iraq and the US since 1991 came in the way of the US intelligence being able to collect secret intelligence from Iraqi policy-making circles in Baghdad.
10. When an intelligence agency does not have a presence in the capital of a targeted country, it has other options such as trans-border and third country operations and exploitation of the presence of other countries and international organisations in the targeted country. That is what the US intelligence sought to do. It ran trans-border operations into Iraq from neighbouring countries such as Jordan, Saudi Arabia and Kuwait. It ran third-country operations from countries, which were not neighbours of Iraq, by recruiting agents from amongst anti-Saddam political exiles in the US and West Europe. It took advantage of the presence of UN inspection teams in Iraq by penetrating its intelligence officers into those teams.
11. It ultimately found that in the absence of an intelligence collection presence in Baghdad with the capability to penetrate Government, military and scientific establishments in the capital, alternate means such as those described above alone did not provide it with the required intelligence, whether it be relating to the existence of weapons of mass destruction (WMD) or the intentions and capabilities of the Iraqi Armed Forces or the attitude of the Iraqi population as a whole to the Saddam regime on the one hand and to the US-led attempts for a regime range on the other.
12. Excessive reliance for the collection of human intelligence (HUMINT) on third-country-based anti-Saddam political exiles led to the agencies being willfully misled by such elements which had their own axe to grind. The penetration of US intelligence officers into the UN inspection teams did not produce the required results by way of HUMINT collection.
13.Since the US-led occupation of Iraq in April last year, the US intelligence agencies, both civilian and military, have an ubiquitous presence in Iraq and, yet, their HUMINT collection capability remains as poor as ever as seen from the serious difficulties faced by the US Armed Forces in dealing with the post-April 2003 resistance movement and the activities of foreign jihadi terrorists, who have infiltrated from outside. This is due to the fact that the US intelligence is now required to collect HUMINT not about an adversary State and its military apparatus, which have collapsed, but about innumerable non-State actors who have taken to arms against the occupation forces and the fledgeling new State machinery being created by the occupation forces.
14. The Iraqi war and its sequel have already had two important lessons for intelligence professionals.
* First, an intelligence presence alone in the targeted country or area is not sufficient unless that presence has the required penetration capability.
* Second, it is more difficult to collect intelligence, particularly HUMINT, about a non-State adversary than about a State adversary.
15. All secret intelligence collection, whether it be HUMINT or technical intelligence (TECHINT), is the result of penetration. One collects HUMINT by penetrating the adversary's set-up through human sources. One collects TECHINT by penetrating through technical means. A well-established State, with a visible, concrete and identifiable structure, is easier to penetrate than an amorphous non-State adversary with no such structure. Traditional tradecraft used for the penetration of State adversaries may not serve the purpose of penetration of non-State actors. Moreover, while the penetration of a State actor poses no ethical problems, that of a non-State actor, particularly terrorist groups, poses ethical problems. Such ethical problems arise from the fact that a human agent used for penetrating a terrorist organisation may be called upon by its leadership to carry out an assassination in order to prove his or her credibility. Should the handling officer of the source condone such acts of criminality in order to be able to penetrate the terrorist organisation? It is a question which has remained without answers.
16.Penetration through technical means for the collection of technical intelligence (TECHINT) is an important component of all intelligence collection operations, whether in respect of State or non-State adversaries. Many successful preventive actions were attributable to successful technical penetrations. The US intelligence agencies have scored many tactical successes against Al Qaeda because of their successful penetration of its communications set-up, but they have not scored any strategic success because only HUMINT collected through successful penetration can facilitate a strategic break-through.
17. Moreover, when an adversary avoids the use of modern means of communications and observes effective communications security, penetration through technical means becomes difficult, if not impossible. That is what has been happening in Iraq. The resistance fighters have been avoiding communicating with each other through technical means and the US has till now not been able to clearly identify its adversaries and establish their organisational structure. With the result, there has been very little flow of either HUMINT or TECHINT.
18.The lessons: There is no substitute for good HUMINT. While some HUMINT of a peripheral and sometimes even of a preventive nature can come from the interrogation of arrested suspects and itinerant trans-border and observer sources, a substantial and sustained flow of HUMINT of good quality can come only through successful penetrations. Such operations, inter alia, require good linguistic capability in the handling officers, a good knowledge of the culture and psyche of the society targeted, identification of dissatisfied sections of that society which might be willing to co-operate with the handling officer and their intelligent exploitation. This is something to be achieved over a long period and cannot be done overnight as the US is learning in Iraq.
19. Flow of secret and valuable HUMINT through successful penetrations has to be an important yard-stick for evaluating the performance of intelligence agencies. In the USA and other countries, it has been laid down that the performance of the agencies would be evaluated on the basis of the secret HUMINT of relevance to national security supplied by them to the security forces, policy-makers and other consumers. Despite this, there has been no significant improvement in their performance, particularly in respect of non-State actors because on grounds of agent security, intelligence agencies have been resisting any external evaluation of their operational capability. It has remained largely an in-house affair, with no scope for a flow of objective criticism and ideas for improvement from outside the agencies. This is a matter, which needs careful examination, if HUMINT capability has to improve. This applies as much to India as to any other country.
20. The second lesson from the Falklands failure, which continues to have validity even today, is the importance of open source intelligence. In the look-out of the collection and assessment agencies for secret intelligence, a large volume of open intelligence on Argentina's possible intentions and plans was not given the attention it deserved until it was too late to act on it. This open intelligence came to the British Foreign Office from two sources--- the Argentinian and the US media and the despatches of the British diplomats posted in Argentina on the basis of their discussions with local Government officials and diplomats of other countries.
21. Since there was no secret intelligence from the MI-6 indicating the possibility of an Argentine invasion, open intelligence indicating such a possibility was not taken seriously either in the Foreign Office or in the Joint Intelligence Committee(JIC). The enquiry identified three major deficiencies--- the absence of a systematic dissemination and analysis of open intelligence having a bearing on national security; the absence of the practice of reverse analysis of available intelligence from the perspective of the adversary; and the absence of independent thinking of the Chairman of the JIC.
22. At the time of the Falklands invasion, an officer of the British Foreign Office, with a rank equivalent to that of an Additional Secretary in the Government of India, used to act concurrently as the JIC Chairman. It was said that he was influenced by the then prevailing wisdom in the Foreign Office that Argentina would bark, but never bite and any open source intelligence, which indicated otherwise, was not taken seriously by him. The Franks Commission recommended the strengthening of the analysis and assessment capability of the JIC and the appointment of an independent expert with easy access to the Prime Minister as the Chairman.
23. Despite this, things have not improved in the British JIC as one had seen by the less than objective manner in which it had functioned in the period preceding the invasion of Iraq and had apparently allowed itself to be influenced by Prime Minister Tony Blair in its analysis and assessment of intelligence relating to the presence of WMD in Iraq and the imminence of the threat from Iraq's chemical and biological warfare capability.
24. Analysis and assessment have also been the Achilles heel of the security apparatus in the US as was seen in the way the pre-9/11 reports relating to Al Qaeda's presence in the US homeland and the threat posed by it to the US homeland and Iraq's WMD capability were analysed and assessed by the National Security Council Secretariat (NSCS). Just as the JIC in the UK let itself be influenced by the Iraqi agenda of Blair, the NSCS too let itself be influenced in its analysis and assessment and in its identification of operational priorities by the Iraqi agenda of President Bush and his close neo-conservative advisers. The decision to go to war in Iraq was made not on the basis of an objective assessment of available intelligence relating to its WMD capability, its alleged links with Al Qaeda and the likely popularity of such an invasion with the Iraqi people, but on the basis of the ideological predilections of Bush and his neo-conservative advisers.
25. The decision to go to war was first made in their mind and only then the available open and secret intelligence reports were examined in order to choose those which can be used before US public opinion and the international community for justifying the decision. Dubious secret intelligence supplied by anti-Saddam political exiles indicating the imminence of a WMD threat was accepted and played up and credible open and semi-open intelligence from the media and the reports and observations of the UN inspectors, which seriously doubted or dismissed such claims, was rejected.
26. How good is the Indian analysis and assessment machinery? Has a similar thing happened in India? Or, can it happen in future? A study of the Lord Franks report by the Government of India in 1983 led to two decisions to strengthen the JIC. Firstly, to bifurcate the JIC to create two JICs---one to analyse and assess threats to internal security and the other to deal with external security. The need for a separate JIC to deal with internal security was felt because of the persistence of tribal insurgency in the North-East and its spread to Tripura. The second decision related to the appointment of officers of the Indian Police Service from the intelligence agencies as heads of the external JIC. The argument advanced by the intelligence agencies, which were the driving force behind this decision, in favour of it was that only professional intelligence officers could tone up the working of the JIC.
27. While the decision to bifurcate the JIC was reversed after the death of Mrs. Indira Gandhi in October 1984, the practice of appointing intelligence officers as heads of the re-unified JIC continued till the Kargil conflict in 1999. In 1998, a reported request by the Army that the officers of its intelligence directorate should also be considered for appointment as the JIC Chairman led to an impasse which continued for months, with the head of the Research & Analysis Wing (R&AW) being asked to hold additional charge as the JIC Chairman.
28. Successive Governments devalued the JIC to such an extent that senior officers of the intelligence producing and consuming agencies stopped attending its weekly meetings and deputed junior officers, often of the rank of Under-Secretaries, to attend them and the intelligence directorate of the Army even stopped disseminating the military intelligence collected by it to the JIC. The Kargil Review Committee (KRC), which enquired into intelligence and operational deficiencies preceding the Kargil conflict, had pointed out that during a period of two years preceding the Kargil conflict the Military Intelligence Directorate had hardly shared any military intelligence with the JIC.
29. The Special Task Force for the Revamping of the Intelligence Apparatus set up by the Government of India in 2000 of which I was a member, had made a statistical analysis of the number of reports shared by the R&AW, the IB and the Directorate-General of Military Intelligence (DGMI) with the JIC in the months preceding the Kargil conflict and found that while the R&AW and the IB continued to share their reports with the JIC, there was a steep drop in the number of reports shared by the DGMI. When a representative of the Army, who had interacted with the Task Force, was asked about the reasons for this decline, he replied: " The Army is the end-user of all military intelligence. It is, therefore, immaterial whether it shares its MI reports with the JIC or not."
30.Was Sri Lanka India's mini Iraq? Was the initiative for the Indian involvement there taken by the political leadership on its own or at the instance of the Army or the intelligence agencies? Did the JIC play any advisory role in the matter? It is difficult to answer these questions in the absence of authentic information about the decision-making process in matters relating to Sri Lanka. It is a pity no case study has been made of how the highly sensitive Sri Lanka operation was handled at various levels---at the political, operational and diplomatic. It would be a good case study of how not to handle a sensitive intelligence operation. Apparently, an example of too many cooks spoiling the broth.
31. The lack of professionalism in the manner in which we handle the important task of intelligence analysis and assessment became evident in our handling of the sequel to the Mumbai blasts of March,1993. Two separate and ad hoc groups came into existence in the MHA---one reporting to the then Home Minister and the other to the then Minister of State for Internal Security--- to monitor and assess all terrorism-related intelligence and co-ordinate the follow-up action, each often not knowing what the other was doing and with the JIC's role in the analysis and assessment of terrorism-related intelligence downgraded. In many instances, the then Chairman of the JIC was kept out of many of these meetings.
32. The creation of a National Security Council Secretariat (NSCS) in 1999 and the merging of the JIC with it were meant to correct the erratic functioning of the JIC before the Kargil conflict. Has this led to an improvement in our capability for analysis and assessment? Have we been able to develop an expertise in the analysis and assessment of intelligence relating to State and non-State adversaries? The Special Task Force on the Revamping of the Intelligence Apparatus had recommended a role for the NSCS in the co-ordination of the performance of the intelligence agencies. How effectively is this role being performed? Are any corrections called for? Now that it is five years since the new national security management infrastructure was set up, the time has come for an external evaluation of its performance by a group of independent experts.
33. A few words about the importance of open source intelligence before we move on to the next topic of inadequate intelligence. The Brown-Aspen Commission set up by the Clinton Administration in the US to look into the functioning of the intelligence agencies had underlined the importance of systematic monitoring of open intelligence for two reasons. First, if the intelligence agencies do not do so, they would not know what intelligence is already available from open sources and would be wasting time and money for collecting, through HUMINT operations, something which is already available. Second, any assessment based only on secret intelligence could be incomplete and misleading.
34. The Task Force on the Revamping of the Intelligence Apparatus had drawn the attention of our Government to the tremendous potential of the Internet as a source of open intelligence and stressed the importance of setting up a nodal agency for data mining from the Internet and sharing it with all agencies of the intelligence community as well as all traditional consumers of secret intelligence. This is an aspect which needs attention, if this has not already been acted upon.
35. The importance of reverse analysis as one of the analytical tools was underlined by the report of the in-house enquiry into the failure of the US intelligence to anticipate the decision of the BJP-led Government to have the Pokhran-II nuclear tests carried out. The report pointed out that no attempt was made by a group of experts in the State Department to put themselves in the place of the BJP and analyse what options they would have to strengthen their political position, which was wobbly, at that time. It was felt that they might have concluded that one good option available to strengthen their political position was to carry out the nuclear tests.
MUMBAI BLASTS & RAJIV'S ASSASSINATION
36.A typical Indian example of zero secret intelligence was the Mumbai explosions of March,1993. Neither the IB nor the R&AW had the least inkling of the preparations for the explosions, which had been going on for at least a month before they took place---- such as recruitment of volunteers for organising the explosions, their travel to Pakistan via Dubai for training and their return to Mumbai, the clandestine landing of explosives, hand-grenades and other weapons from Pakistan by Dawood's gang and their secret storage.
37. The intelligence agencies--- the IB in Mumbai and the R&AW in Dubai--- were totally clueless about the clandestine goings-on till the explosions took place. Interestingly, within a few hours of the explosions, the IB and the Mumbai Police managed to establish the identities of those responsible and how they had planned and carried out the explosions and the R&AW played a remarkable role in obtaining documentary evidence of the role of Pakistan's Inter-Services Intelligence (ISI) in their training and in the hunt for the perpetrators, many of whom were arrested in Kathmandu and elsewhere and brought to India largely on the basis of valuable secret intelligence collected by the R&AW.
38. The question, which should have been addressed, but was not, was: If the intelligence agencies had the remarkable operational capability for collecting secret intelligence after the explosions, how was it that this capability failed to anticipate the explosions and to prevent them by collecting intelligence about the preparations? A similar question had confronted us after the assassination of Rajiv Gandhi in May,1991. Unlike the Mumbai explosions, Rajiv's assassination was more a case of inadequate intelligence, like the Kargil conflict of 1999, than one of zero intelligence.
39.In 1989, when the V.P.Singh Government, after coming to office, had ordered an assessment of likely threats to Rajiv Gandhi as the leader of the Opposition, the R&AW had correctly assessed that the threats to him during his travels in North India and abroad would come mainly from the Sikh terrorists and that threats during his travels in South India would come mainly from the Sri Lankan Tamil terrorist elements with a base in Indian territory. Moreover, subsequently, the West German intelligence had twice passed on a warning that a Sri Lankan Tamil living in West Germany, who was frequently visiting Chennai, had the reputation of being an explosive expert.
40. Thus, the R&AW's analytical capability had proved to be remarkably prescient as a result of which it was the only agency which came out of the Verma Commission of Enquiry almost totally unscathed. During the investigation of the assassination, no evidence emerged to show that the visits of the suspected explosive expert from West Germany to Chennai were in any way connected with the preparations of the LTTE for the assassination. Despite this, in retrospect, one felt that all the intelligence agencies should have alerted their sources in Tamil Nadu and Sri Lanka and tried to find out whether the LTTE was up to anything sinister. They did alert their HUMINT sources, but they could not pick up any intelligence about the plans and preparations of the LTTE to carry out the assassination, and their TECHINT means were not effectively utilised before the assassination.
41.Remarkably, within a few days of the assassination, the intelligence agencies managed to establish the involvement of the LTTE and the TECHINT Division of the R&AW managed to track down from minute to minute the movement of the survivors among the perpetrators by intercepting their communications with their headquarters and repeatedly breaking the LTTE's code. This ultimately led to their being cornered in Bangalore and their subsequent suicide before they could be captured.
42. A question, which should have been addressed, but was not, was: If the TECHINT Division of the R&AW had such a remarkable capability for intercepting the LTTE's communications and breaking its code as was demonstrated after the assassination, why this capability failed to detect the arrival of the assassination team in Chennai and its subsequent movements and preparations? The Jain Commission was reported to have found that the IB did intercept some coded communications of the LTTE having a bearing on the assassination, but did not have the capability to break the code used at the time the messages were intercepted. It was reportedly broken much later after the assassination..
43. The position apparently was: The IB was intercepting more coded messages than the R&AW, but did not have the capability to break their code. The R&AW had the capability to break the code, but had not intercepted the messages which were intercepted by the IB. The IB did not inform the R&AW that it had some coded communications and ask it whether it could help to break them. The R&AW did not share with the IB the information that it had the capability to break the LTTE's code and tell it that it might be able to help in breaking any coded messages intercepted by it.
44. In his interaction with the Special Task Force on the Intelligence Apparatus, a senior official of the Govt. of India, who headed a Ministry which is one of the consumers of intelligence, mentioned that his own assessment was that the Indian intelligence agencies were excellent investigating agencies after a breach of national security had taken place, but had very limited capability for anticipating and preventing such breaches. The criticism is partly invalid, partly valid.
45.It is invalid in the sense that there were innumerable instances in which the intelligence agencies had played a stellar role in collecting secret preventive intelligence, but public knowledge of it had remained limited for valid operational reasons. Such intelligence was effectively used by the physical security agencies in some cases to prevent breaches of national security, but in some cases their failure to act on them led to serious breaches, despite specific intelligence being available. Some examples: The aborted attempt to assassinate Rajiv Gandhi at Rajghat in October 1986, the neutralisation of many terrorist leaders of Punjab and Jammu & Kashmir and the Purulia arms drop of 1995.
46.It is valid in the sense that, despite such good work, there are equally innumerable instances where the intelligence agencies had failed to anticipate and prevent breaches of national security. It is an unfortunate fact of life that the public would always judge the intelligence agencies by their failures, which are public knowledge, rather than by their successes of which it has limited or no knowledge.
47. While submitting his report, Justice Verma had suggested that the intelligence agencies should do an off-the-record introspection as to why they had failed to prevent the assassination of Rajiv. He apparently felt that there might be instances of commission and omission which they might not have shared with him, and wanted that they should at least meet and ruminate over them among themselves in order to draw lessons for the future. One introspection session was held under the leadership of the IB for about three hours, but nobody was prepared to concede that there was anything wrong with them. They held the introspection session not because they themselves, in a fit of remorse over their failure to prevent Rajiv's assassination, felt the need for it, but because Justice Verma had recommended it. It was a nam-ke-vaste introspection without any soul-searching.
TEXT OF MAJ.GEN.SAMIR SINHA MEMORIAL LECTURE—2004 AT THE UNITED SERVICE INSTITUTION OF INDIA, NEW DELHI, ON APRIL 26, 2004
There is a clear distinction between an intelligence process and an intelligence operation. When one talks of an intelligence process, one has in mind the various stages in the evolution of intelligence---its collection, evaluation, analysis, assessment and follow-up action. The expression "intelligence operation" as used in the theme proposed to me for my address today restricts itself to the first stage in this entire scale of evolution--namely, the collection of intelligence, and tends to exclude from consideration the other stages in the evolution, which are as important as the collection, if not more.
2. In intelligence parlance, one uses the expression "intelligence operation" essentially with reference to two aspects of the functioning of an intelligence agency----its ability to collect secret, precise and timely intelligence of relevance to national security and to undertake covert actions, if need be in furtherance of national security objectives, if called upon to do so by the political leadership.
3. These two are the most sensitive aspects of the functioning of an agency, but not necessarily the most decisive aspects in terms of safeguarding national security. Analysis, assessment and follow-up action tend to be more decisive in this regard. A wealth of precise secret intelligence may not help in safeguarding national security if ill or inadequately analysed with unsatisfactory follow-up action. On the other hand, even incomplete intelligence, if well analysed and assessed and promptly and effectively acted upon, can often safeguard national security.
4.In a presentation of this nature, it would, therefore, be necessary to deal with the entire gamut of the intelligence evolution instead of restricting oneself to only one stage of it, namely, the collection operation. There are two ways of dealing with the subject--- in general terms by discussing the basic principles and concepts which govern the entire process or in specific terms by discussing some instances which are considered as landmarks in the evolution of the intelligence profession because of the lessons which they imparted to the cognoscenti, provided they had the desire and the ability to learn the right lessons. I propose to adopt the second in this talk.
5.All serious breaches of national security and all instances in which the national security managers, in the political leadership as well as in the policy-making bureaucracy, are taken by surprise, either by State or by non-State actors, have an intelligence-related cause and effect. For analysing this intelligence-related cause and effect, all instances of failures to safeguard national security can be attributed to one of the following five causes---zero intelligence, inadequate intelligence, inaccurate intelligence, excessive or an overload of often irrelevant intelligence and inadequate follow-up action on the available intelligence, whether adequate or inadequate.
ZERO INTELLIGENCE: FALKLANDS
6.A typical example of zero secret intelligence was the failure of the Secret Intelligence Service (SIS) of the UK, popularly known as MI-6, to collect intelligence regarding the plans of the Argentine Government to send its Armed Forces to occupy the Falkland Islands in 1982. Lord Franks, who constituted the one-man Commission, which enquired into this failure, brought on record the fact that the MI-6 had failed to detect the Argentine plans for the occupation, but held, at the same time, that it could not be held responsible for this failure because the British Government had not endowed it with the capability for the collection of such secret intelligence. Its proposal long before the war for the sanction of an intelligence collection post under diplomatic cover in the British Embassy in Buenos Aires had been rejected by the Foreign Office. In the light of this, the MI-6 was absolved of any responsibility for the failure.
7. After having done so, the Commission had pointed out that even if secret intelligence was not available, adequate open source intelligence in the form of media reports and despatches from the British Embassy in Buenos Aires was available, which, if it had been adequately analysed, would have served as a wake-up call to the Government and prompted it to rush the Navy to prevent the occupation of the islands.
8.The Falklands Enquiry highlighted certain important aspects of the various dimensions of the intelligence process:
* First, the importance of a physical presence of an intelligence agency in a targeted country, either under a diplomatic or a non-diplomatic cover, before it can be expected to collect secret intelligence of relevance to national security from that country.
* Second, the importance of a systematic monitoring of open source intelligence, whether adequate secret intelligence is available or not.
* Third, the importance of continuous and timely analysis of all intelligence having a bearing on a development, whether open or secret.
* Four, the importance of reverse analysis of a development, that is analysing a development not from your own perspective, but from the perspective of an adversary.
* And five, the importance of the analysis and assessment machinery existing and functioning independently without letting itself be influenced in its analysis and assessment by the pre-conceived ideas, conventional wisdom and political compulsions of the policy makers, whether they be of the political leadership or the policy-making bureaucracy.
9. A physical presence in the targeted country is the sine qua non of any successful intelligence collection operation. While it is for the political leadership and the policy-making bureaucracy to lay down the tasks of the intelligence agencies, it is for the agencies to decide what presence they would require, where and how for fulfilling those tasks. Very often, the intelligence agencies are denied this vital requirement due to one of two reasons:
* First, the Foreign Office or other policy-making Ministries with no responsibility for intelligence collection tend to arrogate to themselves the right to impose their non-professional judgement over the professional judgement of the intelligence agencies as to what collection presence they should have, where and how. The result: National security failures. The Falklands incident was only one such instance. There are many such instances strewn across the histories of the intelligence agencies of many countries, including India.
* Second, the disruption of diplomatic relations with a targeted country, which denies the Foreign Office as well as the intelligence agencies the benefits of a physical presence in that country. A typical example would be the difficulties faced by the US intelligence in Iraq. The non-existence of diplomatic relations between Iraq and the US since 1991 came in the way of the US intelligence being able to collect secret intelligence from Iraqi policy-making circles in Baghdad.
10. When an intelligence agency does not have a presence in the capital of a targeted country, it has other options such as trans-border and third country operations and exploitation of the presence of other countries and international organisations in the targeted country. That is what the US intelligence sought to do. It ran trans-border operations into Iraq from neighbouring countries such as Jordan, Saudi Arabia and Kuwait. It ran third-country operations from countries, which were not neighbours of Iraq, by recruiting agents from amongst anti-Saddam political exiles in the US and West Europe. It took advantage of the presence of UN inspection teams in Iraq by penetrating its intelligence officers into those teams.
11. It ultimately found that in the absence of an intelligence collection presence in Baghdad with the capability to penetrate Government, military and scientific establishments in the capital, alternate means such as those described above alone did not provide it with the required intelligence, whether it be relating to the existence of weapons of mass destruction (WMD) or the intentions and capabilities of the Iraqi Armed Forces or the attitude of the Iraqi population as a whole to the Saddam regime on the one hand and to the US-led attempts for a regime range on the other.
12. Excessive reliance for the collection of human intelligence (HUMINT) on third-country-based anti-Saddam political exiles led to the agencies being willfully misled by such elements which had their own axe to grind. The penetration of US intelligence officers into the UN inspection teams did not produce the required results by way of HUMINT collection.
13.Since the US-led occupation of Iraq in April last year, the US intelligence agencies, both civilian and military, have an ubiquitous presence in Iraq and, yet, their HUMINT collection capability remains as poor as ever as seen from the serious difficulties faced by the US Armed Forces in dealing with the post-April 2003 resistance movement and the activities of foreign jihadi terrorists, who have infiltrated from outside. This is due to the fact that the US intelligence is now required to collect HUMINT not about an adversary State and its military apparatus, which have collapsed, but about innumerable non-State actors who have taken to arms against the occupation forces and the fledgeling new State machinery being created by the occupation forces.
14. The Iraqi war and its sequel have already had two important lessons for intelligence professionals.
* First, an intelligence presence alone in the targeted country or area is not sufficient unless that presence has the required penetration capability.
* Second, it is more difficult to collect intelligence, particularly HUMINT, about a non-State adversary than about a State adversary.
15. All secret intelligence collection, whether it be HUMINT or technical intelligence (TECHINT), is the result of penetration. One collects HUMINT by penetrating the adversary's set-up through human sources. One collects TECHINT by penetrating through technical means. A well-established State, with a visible, concrete and identifiable structure, is easier to penetrate than an amorphous non-State adversary with no such structure. Traditional tradecraft used for the penetration of State adversaries may not serve the purpose of penetration of non-State actors. Moreover, while the penetration of a State actor poses no ethical problems, that of a non-State actor, particularly terrorist groups, poses ethical problems. Such ethical problems arise from the fact that a human agent used for penetrating a terrorist organisation may be called upon by its leadership to carry out an assassination in order to prove his or her credibility. Should the handling officer of the source condone such acts of criminality in order to be able to penetrate the terrorist organisation? It is a question which has remained without answers.
16.Penetration through technical means for the collection of technical intelligence (TECHINT) is an important component of all intelligence collection operations, whether in respect of State or non-State adversaries. Many successful preventive actions were attributable to successful technical penetrations. The US intelligence agencies have scored many tactical successes against Al Qaeda because of their successful penetration of its communications set-up, but they have not scored any strategic success because only HUMINT collected through successful penetration can facilitate a strategic break-through.
17. Moreover, when an adversary avoids the use of modern means of communications and observes effective communications security, penetration through technical means becomes difficult, if not impossible. That is what has been happening in Iraq. The resistance fighters have been avoiding communicating with each other through technical means and the US has till now not been able to clearly identify its adversaries and establish their organisational structure. With the result, there has been very little flow of either HUMINT or TECHINT.
18.The lessons: There is no substitute for good HUMINT. While some HUMINT of a peripheral and sometimes even of a preventive nature can come from the interrogation of arrested suspects and itinerant trans-border and observer sources, a substantial and sustained flow of HUMINT of good quality can come only through successful penetrations. Such operations, inter alia, require good linguistic capability in the handling officers, a good knowledge of the culture and psyche of the society targeted, identification of dissatisfied sections of that society which might be willing to co-operate with the handling officer and their intelligent exploitation. This is something to be achieved over a long period and cannot be done overnight as the US is learning in Iraq.
19. Flow of secret and valuable HUMINT through successful penetrations has to be an important yard-stick for evaluating the performance of intelligence agencies. In the USA and other countries, it has been laid down that the performance of the agencies would be evaluated on the basis of the secret HUMINT of relevance to national security supplied by them to the security forces, policy-makers and other consumers. Despite this, there has been no significant improvement in their performance, particularly in respect of non-State actors because on grounds of agent security, intelligence agencies have been resisting any external evaluation of their operational capability. It has remained largely an in-house affair, with no scope for a flow of objective criticism and ideas for improvement from outside the agencies. This is a matter, which needs careful examination, if HUMINT capability has to improve. This applies as much to India as to any other country.
20. The second lesson from the Falklands failure, which continues to have validity even today, is the importance of open source intelligence. In the look-out of the collection and assessment agencies for secret intelligence, a large volume of open intelligence on Argentina's possible intentions and plans was not given the attention it deserved until it was too late to act on it. This open intelligence came to the British Foreign Office from two sources--- the Argentinian and the US media and the despatches of the British diplomats posted in Argentina on the basis of their discussions with local Government officials and diplomats of other countries.
21. Since there was no secret intelligence from the MI-6 indicating the possibility of an Argentine invasion, open intelligence indicating such a possibility was not taken seriously either in the Foreign Office or in the Joint Intelligence Committee(JIC). The enquiry identified three major deficiencies--- the absence of a systematic dissemination and analysis of open intelligence having a bearing on national security; the absence of the practice of reverse analysis of available intelligence from the perspective of the adversary; and the absence of independent thinking of the Chairman of the JIC.
22. At the time of the Falklands invasion, an officer of the British Foreign Office, with a rank equivalent to that of an Additional Secretary in the Government of India, used to act concurrently as the JIC Chairman. It was said that he was influenced by the then prevailing wisdom in the Foreign Office that Argentina would bark, but never bite and any open source intelligence, which indicated otherwise, was not taken seriously by him. The Franks Commission recommended the strengthening of the analysis and assessment capability of the JIC and the appointment of an independent expert with easy access to the Prime Minister as the Chairman.
23. Despite this, things have not improved in the British JIC as one had seen by the less than objective manner in which it had functioned in the period preceding the invasion of Iraq and had apparently allowed itself to be influenced by Prime Minister Tony Blair in its analysis and assessment of intelligence relating to the presence of WMD in Iraq and the imminence of the threat from Iraq's chemical and biological warfare capability.
24. Analysis and assessment have also been the Achilles heel of the security apparatus in the US as was seen in the way the pre-9/11 reports relating to Al Qaeda's presence in the US homeland and the threat posed by it to the US homeland and Iraq's WMD capability were analysed and assessed by the National Security Council Secretariat (NSCS). Just as the JIC in the UK let itself be influenced by the Iraqi agenda of Blair, the NSCS too let itself be influenced in its analysis and assessment and in its identification of operational priorities by the Iraqi agenda of President Bush and his close neo-conservative advisers. The decision to go to war in Iraq was made not on the basis of an objective assessment of available intelligence relating to its WMD capability, its alleged links with Al Qaeda and the likely popularity of such an invasion with the Iraqi people, but on the basis of the ideological predilections of Bush and his neo-conservative advisers.
25. The decision to go to war was first made in their mind and only then the available open and secret intelligence reports were examined in order to choose those which can be used before US public opinion and the international community for justifying the decision. Dubious secret intelligence supplied by anti-Saddam political exiles indicating the imminence of a WMD threat was accepted and played up and credible open and semi-open intelligence from the media and the reports and observations of the UN inspectors, which seriously doubted or dismissed such claims, was rejected.
26. How good is the Indian analysis and assessment machinery? Has a similar thing happened in India? Or, can it happen in future? A study of the Lord Franks report by the Government of India in 1983 led to two decisions to strengthen the JIC. Firstly, to bifurcate the JIC to create two JICs---one to analyse and assess threats to internal security and the other to deal with external security. The need for a separate JIC to deal with internal security was felt because of the persistence of tribal insurgency in the North-East and its spread to Tripura. The second decision related to the appointment of officers of the Indian Police Service from the intelligence agencies as heads of the external JIC. The argument advanced by the intelligence agencies, which were the driving force behind this decision, in favour of it was that only professional intelligence officers could tone up the working of the JIC.
27. While the decision to bifurcate the JIC was reversed after the death of Mrs. Indira Gandhi in October 1984, the practice of appointing intelligence officers as heads of the re-unified JIC continued till the Kargil conflict in 1999. In 1998, a reported request by the Army that the officers of its intelligence directorate should also be considered for appointment as the JIC Chairman led to an impasse which continued for months, with the head of the Research & Analysis Wing (R&AW) being asked to hold additional charge as the JIC Chairman.
28. Successive Governments devalued the JIC to such an extent that senior officers of the intelligence producing and consuming agencies stopped attending its weekly meetings and deputed junior officers, often of the rank of Under-Secretaries, to attend them and the intelligence directorate of the Army even stopped disseminating the military intelligence collected by it to the JIC. The Kargil Review Committee (KRC), which enquired into intelligence and operational deficiencies preceding the Kargil conflict, had pointed out that during a period of two years preceding the Kargil conflict the Military Intelligence Directorate had hardly shared any military intelligence with the JIC.
29. The Special Task Force for the Revamping of the Intelligence Apparatus set up by the Government of India in 2000 of which I was a member, had made a statistical analysis of the number of reports shared by the R&AW, the IB and the Directorate-General of Military Intelligence (DGMI) with the JIC in the months preceding the Kargil conflict and found that while the R&AW and the IB continued to share their reports with the JIC, there was a steep drop in the number of reports shared by the DGMI. When a representative of the Army, who had interacted with the Task Force, was asked about the reasons for this decline, he replied: " The Army is the end-user of all military intelligence. It is, therefore, immaterial whether it shares its MI reports with the JIC or not."
30.Was Sri Lanka India's mini Iraq? Was the initiative for the Indian involvement there taken by the political leadership on its own or at the instance of the Army or the intelligence agencies? Did the JIC play any advisory role in the matter? It is difficult to answer these questions in the absence of authentic information about the decision-making process in matters relating to Sri Lanka. It is a pity no case study has been made of how the highly sensitive Sri Lanka operation was handled at various levels---at the political, operational and diplomatic. It would be a good case study of how not to handle a sensitive intelligence operation. Apparently, an example of too many cooks spoiling the broth.
31. The lack of professionalism in the manner in which we handle the important task of intelligence analysis and assessment became evident in our handling of the sequel to the Mumbai blasts of March,1993. Two separate and ad hoc groups came into existence in the MHA---one reporting to the then Home Minister and the other to the then Minister of State for Internal Security--- to monitor and assess all terrorism-related intelligence and co-ordinate the follow-up action, each often not knowing what the other was doing and with the JIC's role in the analysis and assessment of terrorism-related intelligence downgraded. In many instances, the then Chairman of the JIC was kept out of many of these meetings.
32. The creation of a National Security Council Secretariat (NSCS) in 1999 and the merging of the JIC with it were meant to correct the erratic functioning of the JIC before the Kargil conflict. Has this led to an improvement in our capability for analysis and assessment? Have we been able to develop an expertise in the analysis and assessment of intelligence relating to State and non-State adversaries? The Special Task Force on the Revamping of the Intelligence Apparatus had recommended a role for the NSCS in the co-ordination of the performance of the intelligence agencies. How effectively is this role being performed? Are any corrections called for? Now that it is five years since the new national security management infrastructure was set up, the time has come for an external evaluation of its performance by a group of independent experts.
33. A few words about the importance of open source intelligence before we move on to the next topic of inadequate intelligence. The Brown-Aspen Commission set up by the Clinton Administration in the US to look into the functioning of the intelligence agencies had underlined the importance of systematic monitoring of open intelligence for two reasons. First, if the intelligence agencies do not do so, they would not know what intelligence is already available from open sources and would be wasting time and money for collecting, through HUMINT operations, something which is already available. Second, any assessment based only on secret intelligence could be incomplete and misleading.
34. The Task Force on the Revamping of the Intelligence Apparatus had drawn the attention of our Government to the tremendous potential of the Internet as a source of open intelligence and stressed the importance of setting up a nodal agency for data mining from the Internet and sharing it with all agencies of the intelligence community as well as all traditional consumers of secret intelligence. This is an aspect which needs attention, if this has not already been acted upon.
35. The importance of reverse analysis as one of the analytical tools was underlined by the report of the in-house enquiry into the failure of the US intelligence to anticipate the decision of the BJP-led Government to have the Pokhran-II nuclear tests carried out. The report pointed out that no attempt was made by a group of experts in the State Department to put themselves in the place of the BJP and analyse what options they would have to strengthen their political position, which was wobbly, at that time. It was felt that they might have concluded that one good option available to strengthen their political position was to carry out the nuclear tests.
MUMBAI BLASTS & RAJIV'S ASSASSINATION
36.A typical Indian example of zero secret intelligence was the Mumbai explosions of March,1993. Neither the IB nor the R&AW had the least inkling of the preparations for the explosions, which had been going on for at least a month before they took place---- such as recruitment of volunteers for organising the explosions, their travel to Pakistan via Dubai for training and their return to Mumbai, the clandestine landing of explosives, hand-grenades and other weapons from Pakistan by Dawood's gang and their secret storage.
37. The intelligence agencies--- the IB in Mumbai and the R&AW in Dubai--- were totally clueless about the clandestine goings-on till the explosions took place. Interestingly, within a few hours of the explosions, the IB and the Mumbai Police managed to establish the identities of those responsible and how they had planned and carried out the explosions and the R&AW played a remarkable role in obtaining documentary evidence of the role of Pakistan's Inter-Services Intelligence (ISI) in their training and in the hunt for the perpetrators, many of whom were arrested in Kathmandu and elsewhere and brought to India largely on the basis of valuable secret intelligence collected by the R&AW.
38. The question, which should have been addressed, but was not, was: If the intelligence agencies had the remarkable operational capability for collecting secret intelligence after the explosions, how was it that this capability failed to anticipate the explosions and to prevent them by collecting intelligence about the preparations? A similar question had confronted us after the assassination of Rajiv Gandhi in May,1991. Unlike the Mumbai explosions, Rajiv's assassination was more a case of inadequate intelligence, like the Kargil conflict of 1999, than one of zero intelligence.
39.In 1989, when the V.P.Singh Government, after coming to office, had ordered an assessment of likely threats to Rajiv Gandhi as the leader of the Opposition, the R&AW had correctly assessed that the threats to him during his travels in North India and abroad would come mainly from the Sikh terrorists and that threats during his travels in South India would come mainly from the Sri Lankan Tamil terrorist elements with a base in Indian territory. Moreover, subsequently, the West German intelligence had twice passed on a warning that a Sri Lankan Tamil living in West Germany, who was frequently visiting Chennai, had the reputation of being an explosive expert.
40. Thus, the R&AW's analytical capability had proved to be remarkably prescient as a result of which it was the only agency which came out of the Verma Commission of Enquiry almost totally unscathed. During the investigation of the assassination, no evidence emerged to show that the visits of the suspected explosive expert from West Germany to Chennai were in any way connected with the preparations of the LTTE for the assassination. Despite this, in retrospect, one felt that all the intelligence agencies should have alerted their sources in Tamil Nadu and Sri Lanka and tried to find out whether the LTTE was up to anything sinister. They did alert their HUMINT sources, but they could not pick up any intelligence about the plans and preparations of the LTTE to carry out the assassination, and their TECHINT means were not effectively utilised before the assassination.
41.Remarkably, within a few days of the assassination, the intelligence agencies managed to establish the involvement of the LTTE and the TECHINT Division of the R&AW managed to track down from minute to minute the movement of the survivors among the perpetrators by intercepting their communications with their headquarters and repeatedly breaking the LTTE's code. This ultimately led to their being cornered in Bangalore and their subsequent suicide before they could be captured.
42. A question, which should have been addressed, but was not, was: If the TECHINT Division of the R&AW had such a remarkable capability for intercepting the LTTE's communications and breaking its code as was demonstrated after the assassination, why this capability failed to detect the arrival of the assassination team in Chennai and its subsequent movements and preparations? The Jain Commission was reported to have found that the IB did intercept some coded communications of the LTTE having a bearing on the assassination, but did not have the capability to break the code used at the time the messages were intercepted. It was reportedly broken much later after the assassination..
43. The position apparently was: The IB was intercepting more coded messages than the R&AW, but did not have the capability to break their code. The R&AW had the capability to break the code, but had not intercepted the messages which were intercepted by the IB. The IB did not inform the R&AW that it had some coded communications and ask it whether it could help to break them. The R&AW did not share with the IB the information that it had the capability to break the LTTE's code and tell it that it might be able to help in breaking any coded messages intercepted by it.
44. In his interaction with the Special Task Force on the Intelligence Apparatus, a senior official of the Govt. of India, who headed a Ministry which is one of the consumers of intelligence, mentioned that his own assessment was that the Indian intelligence agencies were excellent investigating agencies after a breach of national security had taken place, but had very limited capability for anticipating and preventing such breaches. The criticism is partly invalid, partly valid.
45.It is invalid in the sense that there were innumerable instances in which the intelligence agencies had played a stellar role in collecting secret preventive intelligence, but public knowledge of it had remained limited for valid operational reasons. Such intelligence was effectively used by the physical security agencies in some cases to prevent breaches of national security, but in some cases their failure to act on them led to serious breaches, despite specific intelligence being available. Some examples: The aborted attempt to assassinate Rajiv Gandhi at Rajghat in October 1986, the neutralisation of many terrorist leaders of Punjab and Jammu & Kashmir and the Purulia arms drop of 1995.
46.It is valid in the sense that, despite such good work, there are equally innumerable instances where the intelligence agencies had failed to anticipate and prevent breaches of national security. It is an unfortunate fact of life that the public would always judge the intelligence agencies by their failures, which are public knowledge, rather than by their successes of which it has limited or no knowledge.
47. While submitting his report, Justice Verma had suggested that the intelligence agencies should do an off-the-record introspection as to why they had failed to prevent the assassination of Rajiv. He apparently felt that there might be instances of commission and omission which they might not have shared with him, and wanted that they should at least meet and ruminate over them among themselves in order to draw lessons for the future. One introspection session was held under the leadership of the IB for about three hours, but nobody was prepared to concede that there was anything wrong with them. They held the introspection session not because they themselves, in a fit of remorse over their failure to prevent Rajiv's assassination, felt the need for it, but because Justice Verma had recommended it. It was a nam-ke-vaste introspection without any soul-searching.