Intelligence and policy: A practitioner’s view

This essay by Johan Mostert contains some random notes on the delicate interaction between intelligence producer and the national consumer, specifically the cabinet. Special attention will be given to the nature of intelligence  advice. The author, as head of interdepartmental intelligence production, was intimately involved in the production of intelligence for a number of years. Comments are provided by a South African scholar of intelligence, Dr. Wilhelm Bernhardt.

Even though this essay was written a number of years ago, it contains material which may be of interest to historians as well as current practitioners. Many of the views express the thinking of the National Intelligence Service (NIS) in the previous intelligence dispensation. Dr. Wilhelm Bernhardt was recently requested to provide comments from an academic perspective. He provides valuable insight on the current legislation and White Paper on Intelligence.

Intelligence advice

Views on the role of intelligence in advising policy formulation or monitoring are often nebulous. There is general agreement in the literature on intelligence that the purpose of intelligence is to assist government in the formulation and execution of policy, but exactly how this function is to be discharged is unclear. These notes attempt to provide some points for consideation.

The traditional position was that intelligence and policy formulation are operating in different spheres and “never the twain shall meet”. It was held that intelligence should deliver its “objective” product to the policymaker and, in order not to be contaminated (lose its independence) by policy issues and debates, return to its splendid isolation, although within “hearing distance” from policy making to obtain feedback. It was up to the policy formulator to request additional intelligence if needed; the two functions were to operate separately to eliminate undue mutual influence.

This view has been challenged in recent reflections on the role of intelligence and the current approach is that intelligence and policy should not only be located closer to each other but that the two should exist in a constant interactive relationship. The respective functions should still retain their distinctive roles, but should cooperate more dynamically. It was foreseen that the realities of the different worlds may create tension but that it would not be an adversarial relationship but one where creative tension will lead to the formulation of the best possible course of action for government.

An earlier view of the role of intelligence was that intelligence should provide the decision-maker with a broad coherent picture of a security situation or event and leave it to him/her to derive meaning from the facts. A later approach was that intelligence should provide the decision-maker with an interpretation (assessment) or estimate of a situation with an indication of possible/probable future trends (forecasting). A more recent trend is that the purpose of intelligence is not to predict the future, but to help the policymaker shape the future. This may seem like a small modification but actually represents a significant change in the mindset of the analyst.

An intelligence veteran, Doug MacEachin, depicts the role of intelligence very aptly when he compares the role of intelligence to that of a scout in relation to a football coach. “The job of the scout is not to predict in advance the final score of the game, but to assess the strengths and weaknesses of the opponent so that the coach can devise a winning game plan. Then the scout sits in a booth with powerful binoculars, to report on specific vulnerabilities the coach can exploit.”

The role of intelligence in policy formulation is therefore to:

· inform government by providing a clear picture of the security dimension of the situation upon which solid decision-making can be based;

· warn government about the security obstacles in the way of achieving its objectives;

· predict possible/probable future developments (various scenarios); and

· advise government on opportunities to exploit the vulnerabilities of the adversary or competitor and the security consequences (results) of policy intervention in the area under consideration.

When advising government, intelligence should never be tempted to cross the boundary between its own unique function and that of policy formulation and execution. Even though intelligence co-operates closely with the policy developer and implementer, it retains its own integrity and rules of the game.

It remains true to its calling of advising on the security dimension of the affairs of state and acts in a real sense as conflict adviser to government. It is forever playing devil’s advocate to government; it represents the opposition; it informs government of the moves and counter-moves or potential counter-moves of the opposition (outlining the consequences of government’s intervention).

On matters of policy advice, intelligence never acts in a descriptive manner. It does not tell the policy formulator what to do; what plan to adopt to deal with issues or situations. It provides its best estimate on what the future may hold (enabling the policy formulator to think meaningfully about the future) and advises on security consequences that may result from government decisions. Intelligence provides advice, not prescriptions. Intelligence identify and defines the situation/problem; the policy-maker designs policies and strategies to deal with it.

There are real dangers involved when intelligence attempts to venture into the field of policy advice (advising on what to do). Objections to intelligence becoming part of policy prescription, are the following:

The subject matter of intelligence represents the entire range of the security spectrum (military, political, economic, societal, and environmental). There are government departments responsible for dealing with each aspect. Experience has shown that if intelligence prescribes policy options in areas where these departments have a line functional responsibility, they will take offence and object in the strongest terms.

Intelligence may be arrogant and persist and thereby win the battle, but the war will continue for a long time into the future.  Practical examples are available to show that intelligence has not succeeded in winning this war in the long run. It is also not in the interest of good governance for intelligence to create a situation whereby line functionaries are antagonised by the prescriptive behaviour of a department which does not have a policy formulation function.

If intelligence, despite the fact that it may be in the best position to prescribe in a specific situation, persists in being prescriptive to such an extent that it antagonises another department, it may in the long run become counter-productive. It may raise opposition simply for the sake of opposition. That is the way bureaucracies operate. Intelligence should be sensitive to that if it wants to operate effectively. It must be able to read the mind of own forces as well as it reads the opposition. This aspect seems somewhat trivial on the surface, but in real life has created debilitating problems in the past.

Intelligence may compromise the integrity of its product if it prescribes policy options which go awry. The question may well be asked by members of the policy formulation community whether the intelligence base can be trusted if the policy advice of intelligence did not work out.

Furthermore, if the policy advice did not succeed, the intelligence practitioner may (and experience has shown that it occurs in real life situations) be tempted to justify its policy prescription in stead of providing an objective or non-partisan evaluation of why things went wrong. This urge to justify one’s proposals is one of the strongest reasons why intelligence should not become involved in policy formulation. It creates a bias in the mind of the intelligence practitioner.

Just as the HUMINT collector can fall in love with his source, so a policy advisor can fall in love with his/her policy/strategy/plan. If an analyst becomes involved in policy prescription, he/she may “fall in love” with the plan and manipulate intelligence in such a manner that it supports the favoured policy option. This is a dangerous development.

The mind-set of an analyst differs from that of the policy advisor. The intelligence analysts must speak truth to power, no matter how unpalatable it is to the decision-maker. Intelligence focuses on identifying problems. Intelligence focuses on present realities and developments into the future. The policy formulator attempts to rescue the decision-maker from predicaments by providing advice on methods to deal with it in a pro-active manner. Policy formulation focuses on solutions. Policy focuses on a desired future and the means to achieve it. It is the creative tension between the two functions in the policy formulation process that guarantees the best results, but on condition that each maintains its own identity.

Some understood the previous Minister for Intelligence [minister Lindiwe Sisulu] to mean that intelligence should participate in the formulation of policy proposals and recommendations when she exhorted the intelligence community to “be at the centre of Government”. This is wrong. For intelligence to be in the centre of government, means to act as security (conflict) advisor to all sectors of Government. That means that intelligence should be present when all cluster committees (the centre of decision-making) of Government meet in order to provide the policy makers with a security perspective on whatever executive policies they are developing.

What happens if the line functional department clearly does not know how to properly handle a specific matter and the intelligence structures have the required expertise available? This happens frequently. Would it be expected of intelligence to sit back and say that the theory of intelligence does not allow intelligence to be policy prescriptive? This is obviously ridiculous. If the intelligence practitioner has done proper research on a particular topic and developed expertise in that field, his or her knowledge and insight must be utilised to deal with the matter. The question is: How should intelligence deal with it as intelligence and not as policy developer?

The correct route for intelligence is to adhere to the doctrine that intelligence provides intelligence advice, not policy advice. But what does it entail in practice? Intelligence advice implies more than merely the delivery of intelligence estimates in which opportunities based on the vulnerabilities of the opposition already serves as advisory pointers. Advice is something, a function, in addition to estimating. What then is intelligence advice when engaged in assisting policy formulation in addition to providing advice in the form of estimates?

In the first instance, as already pointed out above, intelligence has the obligation to advise the policy formulator about the possible/probable actions or counter-actions of the opposition/competitor. This means the vigorous participation of intelligence in the formulating process where intelligence acts as the “enemy”. A best option policy will emerge from the dialectical discourse between intelligence and policy-maker.

The second instance occurs where the decision-maker directly requests intelligence to provide the answer to the problem. “You have outlined the problem and know exactly what is going on, why don’t you tell us what to do?” is a frequent demand. Intelligence cannot escape situations like this. In such a case intelligence has to refer to the language of an advisor.

Instead of responding with a policy prescription, intelligence should reply by outlining the various options open to the decision-maker based on the trends outlined in the estimate, weigh them against each other, and advise him/her that given the prevailing circumstances, he/she may consider option A as the most feasible course of action to deal with the vulnerabilities of the opposition.

It is important to keep in mind that intelligence should concentrate on the security dimension, in other words the opposition in security context, because party political considerations may also come into play which the policy-maker must take into account but which is not the concern of intelligence if it does not have security implications. Intelligence has no business advising the decision-maker on non-security matters. Intelligence is therefore not telling the decision-maker what he/she should do (prescribing), but to advise about the feasibility of options.

Intelligence advice may at times sail close to the wind as far as policy prescription is concerned, but the important aspect is that it must not be seen to transgress on the line functional responsibilities of executive departments (which have policy advisory functions). Turf protection and subsequent battles are guaranteed outcomes in the long run. Intelligence must also take care that its advice is framed in language which does not show any bias.

A non-partisan approach must at all times be reflected in intelligence advice. It must also at all times be phrased in a manner which leaves the choice of option (the final decision) open to the policy advisor or decision-maker. Intelligence must not after a recommended act failed be accused of providing the wrong advice (policy advice). It will impact negatively on the status of the intelligence product. This is, however, not a way of copping out as intelligence is sometimes accused of doing. It is merely protecting the integrity of its line function.

It has happened on occasion that the intelligence document is simply provided with a new letterhead of an executive institution and forwarded to the authorities as a final recommendation. This should never happen. Intelligence must at all times phrase its documents in a manner befitting intelligence and be rewritten in a different (policy) language by the executive institution if it becomes a policy proposal or recommendation.

Exceptions to the above approach must be seen as rare events and should be avoided as far as possible. It has happened in the past and may happen in future that intelligence practitioners will be requested to provide policy advice or proposals because of their evident expertise. When he/she does that, it must be realised that they are doing it not as an intelligence operative. They must be fully aware that they are engaged in another discipline, which is not intelligence.

April 2009

Mr. Mostert was head of the Branch National Interpretation (BNI) of the Secretariat of the State Security Council (SSSC) 1987 - 1989

***

Comments on Intelligence and Policy by Dr Wilhelm Bernhardt

The relationship between intelligence and policy is complex, often controversial, deeply nuanced, and seemingly irretrievably entangled in uncertainty.

Interestingly, and perhaps purposefully, there is no reference in the National Strategic Intelligence Act, or any of its subsequent amendments, to policy, policymaking or policy formulation. Instead, the emphasis is primarily put on the identification of threats, the provision of warning and on the transmission of departmental intelligence. Note that although none of the afore-mentioned specifically references policy, although all of the afore-mentioned can theoretically include policy.  

Contrary to the relative silence of the Act on this matter, the preceding White Paper contains multiple references to the relationship between intelligence and policy. In this respect, the following quotes are quite instructive:

- Intelligence is a product “...supportive of the policy- and decision-making processes pertaining to the national goals of stability, security and development”.

- “Modern intelligence can thus be described as "organised policy related information"

- “The relationship between intelligence and policy-making is a dynamic, reciprocal one. Intelligence is but one tool in the successful implementation of domestic and foreign policy”.

- “...for the intelligence organisation to operate optimally, and to the benefit of the policy-maker, intelligence must be valued and nurtured as an instrument of policy.”

- For intelligence to be relevant it must “...provide the policy-makers, timeous, critical and sometimes unique information to warn them of potential risks and dangers. This allows the policy-makers to face the unknown and best reduce their uncertainty when critical decisions have to be made.”

- “...the intelligence community will have to interface with other institutions of society engaged in strategic research, in order to promote integrated analysis which is supportive of policy making.”

Based on the above, the authors of the White Paper therefore apparently clearly intended a relationship between the post-apartheid intelligence structures and policy, albeit within very specific parameters. See for example the clear delineation above of intelligence products that must only support policy and decision-making processes in as far as national goals pertaining to stability, security and development are concerned, and the reminder that intelligence is only relevant in the policy environment if it warns policy-makers about potential risks and dangers.

The inclusion of these parameters in the White Paper appears to reflect a concern about the potential misuse of intelligence by policymakers as policy, in the end, is and remains a political prerogative, with political objectives, and political gains. The objective of curbing the temptation to misuse intelligence for political purposes must therefore have weighed heavily on the minds of the original authors of the White Paper on Intelligence.  

In fact, in the section dealing with the “Basic Principles of Intelligence”, the White Paper unequivocally (and unfortunately without further explanation) quite strongly references the “Principle of the separation of intelligence from policy making”. In an apparent further measure to increase the distance between intelligence and policy, the White Paper also prescribed that the NICOC “will act as the key link between the intelligence community and policy-makers”, but that it will only advise the government on policy relating to the conduct of intelligence at national, regional and local levels” – in other words, not on policy in general, but on policy that affect the conduct of intelligence.

The article lists 4 functions of intelligence in relation to policy formulation: To inform, to warn, to predict and to advise. As long as the focus of such activity is on threats and potential threats arising from a national security perspective, one cannot disagree with these roles. In a constitutional democracy, the element of focus is therefore critically important when considering the relationship between intelligence and policy.

Intelligence should only bring security thinking to policy. This implies a narrow, and specific focus, with constant vigilance against overstepping. This caveat is especially relevant in the context of domestic intelligence, where opposition to policy can very easily be mistaken a threat, and advice to government “...on opportunities to exploit the vulnerabilities of an adversary or competitor” could very easily be construed as the provision of intelligence on any democratic opponent of the government-of-the-day. 

It is therefore also very dangerous, and actually unacceptable, to compare intelligence in an unqualified manner to “the opposition”, and to prescribe that it must inform government “...of the moves and counter-moves of the opposition”. Such actions are only possible once ‘the opposition” has clearly been qualified as a national security threat, by definition excluding political opponents following mom-securitised paths to oppose or frustrate policy. 

Intelligence can, and may never be, the government-of-the-day’s devil’s advocate when it comes to policy. It should not point out weaknesses in policy, nor engage in ‘dialectical discourse’ on the perceived strengths of any policy. It’s one and only function is to inform, warn, predict and advise on threats and potential threats, as per its legally assigned mandate.

To conclude: Should an intelligence analyst know government’s policy direction on any given matter by heart? To this question the answer can only be a resounding ‘yes’. Should the analyst warn, inform, predict and advise government on the security dimensions of its chosen course of action? Again the answer is ‘yes’. Should it do anything more than this? No, not under any circumstances. It is not the responsibility of an intelligence analyst to play any form of devil’s advocate, to point out weaknesses, or to tell government that it is pursuing the wrong, a weak or a poor policy, unless those issues have the potential, or are actually, contributing to the emergence of a national security threat. 

12 August 2024

Dr. Wilhelm Bernhardt is an independent researcher on intelligence

***

The article is due to appear in the next edition of the Nongqai.