The contents of this document are the sole responsibility of the author and do not necessarily represent the views of CCADD.



Published in “Intelligence and National Security”, Vol. 22, Issue 1, March 2007, and as Chapter 6 in “The New Protective State”, ed. Hennessy, pub. Continuum 2007]




Prolegomena to an Ethical Theory


Michael Quinlan


At various points in a career as a United Kingdom civil servant, working mostly in the defence field, I was a customer of secret intelligence, and after leaving full-time Government employment I carried out in 1994 for the Prime Minister of the day a study on certain aspects of it.  (Though the fact of this study soon became public knowledge, it was not until 2006 that limited elements of the report were released to public inspection at the National Archives.).  I had long taken an interest in the ethics of military conflict, and in 2000, at an Oxford conference on intelligence (later reflected in a book [1]) I ventured the suggestion that there was a case for developing a theory of “just intelligence”, on the analogy of “just war” theory.  I was subsequently challenged to contribute to such a task; and the present essay – by someone who is neither an intelligence professional nor a moral philosopher - offers accordingly some prolegomena, in the hope of providing stimulus to more thorough and expert addressal by others.  In the course of developing it I have become increasingly aware that the field is very far from being untilled; but there is work yet to be done, particularly in the climate of heightened concern about intelligence work in the context of the 2003 invasion of Iraq. 


Two basic questions present themselves at the outset of reflection.  First: why should we be concerned at all with ethics in this field? - is it not quintessentially one where Machiavelli and realpolitik have to rule?  There are aspects of the intelligence business, as practised by all major countries, that seem notably disreputable by the behavioural standards of normal human settings; what purchase can ethics expect to have?                   “ ‘Intelligence ethics’ is an oxymoron”, a long-serving officer from the US Central Intelligence Agency was once quoted as saying [2].  In the discussion at the Oxford conference one hardened intelligence professional – not from Britain or the United States – said of ethics in this field, with a dismissive wave of the hand, “I leave all that to the clerics”.  But that surely will not do.  By contrast, Admiral Stansfield Turner, as Head of the US Central Intelligence Agency in the 1980s, said  “There is one overall test of the ethics of human intelligence activities.  That is whether those approving them feel they could defend their actions before the public if the actions became public” [3].  That is closer to the mark, if somewhat subjective.  The fact is that inescapably and inherently (to adapt Aristotle) the human being is an ethical animal – that is part of what it means to be human.  We can no more step outside ethics than we can opt out of the force of gravity.  There is no area of human activity, whether public or private, collective or individual, that has an a priori entitlement to require the moralist to be silent.  If the effective practice of intelligence raises awkward ethical questions, we are obliged ultimately to face them.


A consideration at a less elevated level, from practical expediency, points in the same direction.  In most major Western countries over the last couple of decades, and especially since the 2003 Iraq episode, the business of intelligence has for a mix of reasons emerged from the shadows into open view far more fully than was customary in the past, notably in the United Kingdom.  Our publics, thus increasingly aware of the activity, would not long remain content that what is undertaken in their name should be exempt from moral standard, constraint and scrutiny. 


The second basic question comes at the matter from a different angle.  Why should intelligence need any distinctive ethical consideration, any more than do other governmental activities like diplomacy, bureaucratic administration or military procurement?  The answer lies in the characteristic already noted: that its effective practice seems unavoidably to entail doing some things that are plainly and seriously contrary to the moral rules accepted as governing most human activity.


The Just War Paradigm


The “Stansfield Turner” criterion, taken in isolation, lacks objective anchorage.  We must look for something more systematic.  In the search for that the paradigm of the Just War tradition [4] suggests a starting-point.  The thinkers who developed that tradition were wrestling with the harsh tension between two truths about the activity of war both of which seemed inescapable.  The first was that it entailed doing things – above all, killing people – which in any ordinary context were gravely wrong.  The second was that amid the bitter realities of human existence it surely did not make practical sense – and therefore not moral sense – to demand that the entire activity of war be dismissed by pacifist anathema insisting that the virtuous abstain from it.  What emerged from the wrestling was the concept, elaborated over many centuries, of two levels of moral discipline to govern and limit the activity of war, within the broader context of the duty of public authorities for sustaining the well-being of their people - a goal that remains in itself, especially in relation to tolerably well-ordered polities, a proper ethical good; national interest (to use a modern term) is not an amoral concept.  One level of discipline concerned the criteria that ought to be met if war was to be justly undertaken – jus ad bellum.  The other concerned the constraints and prohibitions that ought to be observed in the conduct of war once entered into – jus in bello.


It is unnecessary here to go into the detailed content of either set of criteria. 

The point for the present analysis is that this seems a natural way to approach the ethical analysis of any activity that poses,  prima facie, considerations in awkward tension, through purporting to require or allow conduct towards others that would be reprehensible in most ordinary situations.  The rest of this essay seeks to explore how far and in what ways the activity of intelligence as practised by the state (issues about information-gathering by the media or other commercial organisations are not addressed here) has to be viewed as presenting major ethical considerations in apparent opposition to one another; and, if so, what we might begin to say about the conditions under which it may properly be engaged in – jus ad intelligentiam, as it were – and about the limitations that ought to be observed in carrying it out even when it is properly undertaken – jus in intelligentia.


Two general difficulties about examining the subject need to be acknowledged.  Firstly, far less is dependably known about what is really done in the business of intelligence than in the business of war.  There are mountains of vivid fiction, a certain amount of shrewd conjecture and hint, and some wary memoir-writing, but the vast majority of citizens do not know and cannot readily find out in any specific, comprehensive and dependable way precisely what intelligence professionals do in concrete day-to-day operational terms, in the way and degree that we know or can find out what it is that soldiers do.  Notably authentic glimpses of some aspects were provided in 2004 and 2005 through after-the-event inquiries in the United States and the United Kingdom into the role played by intelligence in relation to the 2003 Iraq war.  The report   of the Committee of Privy Counsellors chaired by Lord Butler of Brockwell [5] is a vivid United Kingdom example.  But these were special and limited instances.


The second general difficulty is partly related to the first.  There might be real disadvantage – not just presentational discomfort – in having governments make to potential adversaries and wrong-doers a public present of extensive and detailed knowledge about exactly what public authorities will and will not be prepared to do in gathering information.  There is again a partial analogy here with the use of military force – consider, for example, the matter of nuclear-weapon  targeting for deterrence, on which governments have scarcely ever been willing to go beyond making, at most, very general statements rather than spell out exact limits to the action they would regard as tolerable in extremis.  And the underlying consideration seems if anything even more cogent in respect of intelligence than of war.


All that noted, we can with reasonable confidence say a number of descriptive things about intelligence work.  First, much of it consists simply of carefully collating and analysing material that is available to anyone with the resources and inclination to gather it, especially in the modern world of enormous information flows.  No moral problem arises there.  Second, within the category of intelligence information that is not available more or less openly, a great deal is collected by eavesdropping of one sort or another – picking up communications not meant to be heard by outsiders, breaking codes that are intended to conceal content, and the like, including also watching activities which the actor would have preferred to keep unobserved.  It is not obvious that this is in itself seriously wrong in moral terms; in ordinary life it may often be a breach of courtesy or social convention, but scarcely a grave evil.  It may in some circumstances be imprudent to eavesdrop – for example, while it might be highly interesting to install listening devices in the offices of high functionaries of friendly countries, the penalties of being found out doing so might well greatly outweigh the advantages of knowing whatever might be learned.  But that is essentially a calculation about expediency, not moral value.  Eavesdropping by intelligence may nevertheless sometimes entail breaking the law, if not of the eavesdropper’s country then at least of the target country, and perhaps international law, as in Cold War penetration of territorial waters or airspace by intelligence-gathering vehicles; and infringement of law is in itself prima facie in need of special justification.


In addition, at least some eavesdropping is in itself, and almost irrespective of the method employed, constrained by particular agreements voluntarily entered into between states; and it may be that it would be in the general interest to widen the ambit of such agreements [6].  Whether or not there is such a widening, any deliberate breach of such commitments must import moral as well as prudential considerations; but it is then primarily the fact of breaching agreement rather than the inherent nature of eavesdropping that would raise the moral considerations. 


But third, it can be no secret that some of what is done in intelligence-gathering, above all in the category commonly known as HUMINT (“human intelligence”) entails a good deal more than eavesdropping.  This essay does not address what is commonly called “covert action” in the sense of physical intervention to disrupt or prevent, not merely to discover, what an adversary does; that raises other considerations.  Even short of that, however, HUMINT and other comparable intelligence business have to involve, just for example, readiness to tell untruths about who and what one is.  Conventional fictions like the Ruritanian Embassy presenting as its Deputy Assistant Cultural Attache a tough-looking personage of no obvious aesthetic sensibility present no problem, but covert operators have to be prepared to tell untruths seriously intended to mislead public officials like immigration officers or policemen who customarily have a particular right to be told the truth.  Another example might be stealing the documents of another government; and a further and harsher one might be turning a blind eye at least temporarily to serious crimes being committed – perhaps even, still more disquietingly, standing by when one could have prevented them - by individuals with whom one is interacting, in order to remain deceptively in their confidence for larger purposes like penetrating a murderous terrorist organisation.  (A partial parallel is the occasional acceptance by British authorities in World War Two, so it is understood, of losses or reverses which the knowledge gleaned by the “ULTRA” code-breaking success could have been exploited to prevent, but only at the cost of forfeiting greater long-term benefits expected from continuing to conceal from the enemy that his signals traffic was being read.)  A yet further instance is that of inducing individuals in other countries to breach the formal duties and loyalties of their public functions or their citizenship, and thereafter protecting them in that breaching.  Professional insiders could undoubtedly construct a longer catalogue, but the examples noted above suffice for the present analysis.


This evident reality, of doing things that in normal settings would merit condemnation, is one component of the moral tension that besets the activity of intelligence; it is the analogue, in the just-war comparison, of the fact that engaging in war entails readiness to kill.  In the just-war context the other component of the tension is that it seems repugnant to practical and therefore to moral common sense to hold that one must therefore never engage in war even to stop an Attila or a Hitler.  The analogue of this in the intelligence context would need to assert that it is similarly repugnant to practical and moral commonsense to hold that we must never engage in intelligence activity of the kinds just exemplified, even to thwart murderous terrorists or to strengthen our ability to resist wicked aggression.  Governments throughout the world have plainly been prepared to make that assertion.  The next stage in analysis is to consider what reasons might warrant it.


We need not attempt a comprehensive inventory – which might be long [7] - of situations or objectives in regard to which it would be unreasonable to dismiss clandestine intelligence-gathering as immoral, or as conferring too little distinctive benefit to warrant accepting ethical drawbacks.  (The adjective “clandestine” is hereafter used as shorthand for intelligence collection that has recourse to methods breaching normal ethical rules.  Some of what has been described above as eavesdropping and judged to be in itself morally unproblematic is also clandestine in the normal sense of the word, but that is not what will be being referred to.)  Two categories provide illustration.  The first concerns terrorism.  It would be absurd to disqualify wholesale the use of mendacious deception to penetrate sinister organisations like al-Qaeda now, or in Northern Ireland’s past the Provisional Irish Republican Army or its “Loyalist” counterparts, so as to enhance the chances of preventing lethal violence.   It by no means follows that absolutely anything goes in achieving and sustaining such penetration, but wholesale prohibition cannot be right.  The other illustration is espionage against enemies in time of war, as by the United Kingdom against Nazi Germany.  That needs no explanatory comment.


The foregoing paragraphs sketch a framework similar to that underlying just-war reasoning.  They portray an activity which at least in some respects cannot be conducted effectively without cutting across normal moral expectations, but which is essential for public purposes that seem plainly of compelling moral necessity and rightness.  In just-war analysis, however, that does not mean that we can morally engage in any war we like, and then fight it in any way we like; any exemptions from “normal” ethical behaviour have to be justified and limited.  Similarly, we cannot engage in the particular class of intelligence activity here labelled “clandestine” against any target we like, or use any methods we like. 


Invoking the concept of “moral commonsense” does not pre-suppose a purely utilitarian or consequentialist philosophy of public ethics.  The just-war tradition does not contend that, in war, good aims and expectations of good eventual outcomes justify any means.  In accepting that there is an entitlement if necessary to kill enemy soldiers, it holds that by the nature of their role such soldiers forfeit the normal human right not to be killed; and at the same time it typically recognises some deontological imperatives – that is, that there are some things that absolutely ought not to be done even to facilitate victory over an appalling aggressor.  The application of the paradigm to “just intelligence” must be ready to acknowledge comparable dimensions both of entitlement and of constraint.  Entitlement might rest, in the broadest terms, on a concept that we may legitimately collect information, even – perhaps especially? – in the face of attempts to withhold it, about activities that might lead to unjust injury to those for whom we have responsibility.  Constraint would accept that there are some methods of collection that must never be used, however severe we may judge the possible injury to be, and some that must not be used disproportionately – that is, in circumstances where the breach of normal moral rules is more severe than the importance of the legitimate objective reasonably warrants..


Against that background, issues arise about what sorts of limit might be appropriate in the selection of targets for clandestine intelligence-gathering – jus ad intelligentiam – and then in the choice of methods for such gathering – jus in intelligentia.




Jus Ad Intelligentiam


There is a very broad potential spectrum of things about which governments would like to get knowledge through their intelligence services, in the widest sense of finding out or confirming things which it would be helpful to them to know but which the targets – whether states or other entities – are either unwilling or unable to disclose promptly and dependably, and which cannot be discovered as promptly and dependably, or perhaps at all, by other means.  At one extreme of the spectrum there might be getting solid warning of terrorist plans to blow up Parliament; at the other, finding out what is the “bottom line” of the Ruritanian government in an impending negotiation about tariffs on trade in cabbages.  Clandestine intelligence-gathering would clearly be justified on the former; but it seems hard, as a matter of proportionate judgment, to maintain that it would be so on the latter.  I consider later a subsidiary issue about whether the moral situation is altered if we have grounds for believing that the Ruritanians are themselves engaged in clandestine intelligence-gathering about our own “bottom line”.  That particular point aside, however, the question is where along the spectrum the line of prohibition or limitation ought to be drawn.  For added complication, is it a single line or a series of lines, with a considerable array of clandestine methods morally allowable at the top end of the threat spectrum – that is, the end exemplified by major terrorism – but perhaps only a more modest and restricted set at intermediate points?  Many subject-matters may lie between the extremes.  The development of military capability by states whose long-term peaceful intentions towards us or our friends we believe we cannot take wholly on trust?  The faithful observance of arms control agreements like the 1968 Nuclear Non-Proliferation Treaty or the 1972 Biological Weapons Convention, or of mandatory Security Council resolutions?  Savage internal repression by tyrannical regimes?  Narcotics traffic?  International organised crime of other sorts?  Illegal immigration networks?  Violent animal-rights protest?  Large-scale and deliberate breach of agreed rules on trade?


It is surely evident that dividing lines ought to be drawn somewhere along the spectrum of iniquity to be countered and proper interest to be served; but it is much harder to formulate clear and precise proposals for where or how to draw them.  A first approximation might suggest that we are entitled to use clandestine means (still in the special sense of “clandestine” explained earlier)  if other means do not seem likely to suffice - that remains an important qualification, analogous to the Just War criterion of “last resort” - in order to collect information that we reasonably believe may help us to forestall, counter or alleviate actions that would be seriously damaging to peaceful and constitutional order, or to the lives or the major human rights of our own people or of others towards whom we accept significant responsibilities.


There is a great deal packed into that first approximation; and the application of some of its elements would be highly judgmental – just how damaging is “seriously damaging”, for example?  Al-Qaeda’s plans, Yes; those Ruritanian cabbages, surely No; but what about action that could inflict grave economic damage leading to real hardship to our citizens, like attempts to disrupt flows or distort prices in the international oil market?  The economic field cannot be totally excluded.  It is inescapable that any general principle, however skilfully shaped, will always leave a large middle ground within which there have to be judgments made in all the particular circumstances, with ample room accordingly for disagreement and dispute, whether in good faith or in bad, about on which side of the permissibility boundary particular intelligence targets fall.  That is in the nature of practical affairs amid the manifold complexities of human life.  Similar judgmental uncertainties often beset the application of just-war concepts; but those concepts, honestly applied, can nevertheless often lead to clear and important conclusions.  That can be equally true of just intelligence.


Jus In Intelligentia


If clandestine intelligence-gathering is to be conducted effectively, actions like (for example) agent-runners telling untruths about who they are or what they are doing are unavoidable.  It seems unavoidable also that they will have to be ready to exploit the willingness of individuals from the target country, or in whatever else is the target field, to act in breach of normal loyalties.  But are there limits to what they should be prepared to do in order to generate or exploit that willingness?  Ideological persuasion?  Bribery, in one form or another?   So far, perhaps reasonable.  But blackmail?  fostering narcotics addiction? entrapment?  threat to family?  - surely not.


Once more, no ready formula suggest itself for where the line should be drawn between the permissible and the impermissible, especially since, as implied earlier, it might be appropriate to draw the line in different places according to the gravity, and perhaps also the imminence and probability, of the harm we are seeking to forestall or diminish.  Such a concept would be the analogue of the just-war criterion of proportionality.  Broadly, however, the line of prohibition might relate to whether serious coercive violence – or its near-equivalent, as in blackmail – is done to individuals whom we are not entitled to harm.  The point of that last qualification – “whom we are not entitled to harm” – is that in, for example, the Second World War it would surely have been legitimate to kill an enemy sentry in the course of a breaking-and-entering operation to get crucial operational information.  That would be a justifiable warlike action; but it would scarcely be allowable in peacetime, though circumstances approaching those of wartime might arise over imminent terrorism


There then arises the difficult matter of interrogation, which is often an aspect of intelligence-gathering and has been the focus of especial public debate and concern in recent years.  What is legitimate?  Arbitrary and sometimes covert imprisonment, as in Guantanamo Bay?  - not of that scale and duration, many (including this writer) would say.  More awkward still, what is to be said about torture, and what exactly is to be classified as that?  Little respect worldwide has been accorded to, or merited by, attempts made within the Administration of President George W. Bush to re-define “torture” away from its natural meaning and from the definition in the 1984 UN Convention Against Torture, to which the United States is a party, so as to encompass only extreme actions at the top end of the term’s range as normally understood.  It is at least more honest to claim, as others in that Administration or supporting it have occasionally done, that torture may be justified (subject perhaps to some procedural safeguards) if the end is pressing and important enough.  But though it is possible to imagine elaborate and far-out scenarios in which the temptation might seem very cogent, that claim ought to be firmly rejected for reasons both of ultimate moral principle and of pragmatism (including the grave “slippery-slope” danger).  Even within a more moderate ethical calculus, however, and quite aside from considerations about the frequent unreliability of information gained under torture and also about the damage which use of such methods entails to any prospect of subsequent criminal prosecution, there are questions of definition, as was seen in the 1970s when the United Kingdom government changed its practices to conform with the finding of the European Court of Human Rights against some of the methods which had occasionally been used in the counter-terrorist campaign in Northern Ireland.  It would be absurd to say that interrogation, perhaps of people whose malignity is incontestable, must be conducted entirely without pressure – in a comfortable armchair, as it were, with a cup of tea and a biscuit every hour and no harsh expression or frightening tone.  Again, some general guideline is needed to anchor judgment.  The core of the formulation in the 1984 Convention (tighter than United States domestic law appears to be) seems well framed for the purpose – “the intentional infliction of severe physical or mental pain or suffering to obtain information or a confession”. 


There is however a yet further issue in this zone: the matter of what might be called “torture at second hand”.  Media accounts have exposed a practice termed “extraordinary rendition”, meaning handing over terrorist suspects to friendly countries whom there is reason to think less scrupulous than the hander-over about the methods of interrogation used.  That would seem plainly wrong as a deliberate action, and it is indeed prohibited by the 1984 Convention.  But another awkward question arises: even where the handover of individuals is not in question, what should be the moral evaluation of continuing intelligence cooperation and information exchange, for example over international terrorism, with countries known to use methods which we ourselves regard as wrongful?  If we come into possession of particular information that is operationally important for the protection of our people we cannot expunge it from our minds or artificially pretend not to know it even if we believe or suspect it to have been wrongfully acquired; but a systematic and ongoing relationship – to sustain which, moreover, we may be expected to divulge information about individuals as well as to receive it – is a different matter.  It is not easy to find a ready answer to this problem, which is the more awkward in that it may sometimes in practice be precisely from countries where the risk of brutal interrogation may be high that information about terrorism is most needed. 


There is some similarity here with the problem noted earlier, of how we are to justify continuing to work, in our penetration of terrorist or other criminal organisations, with individuals whom we know to be engaged in evildoing.  Once more, any purist demand for perfectly clean hands, both directly and at one remove or more, is not easy to square with practical realities – realities, moreover, which are becoming increasingly the normal context of intelligence work as the prime working environment shifts away from classical inter-state conflict.


We might consider now the “Everybody does it” argument – the right of retaliation, from another standpoint.  If we have reason to believe that others are collecting intelligence against us for purposes or by methods which we ourselves would prefer to rule out, are we thereby dispensed from our own obligations?  The simple answer to this is “No”, or at least “Not entirely”.  If the obligations are founded not on basic moral principle but on particular agreement or received international law, then indeed serious breach of the bargain by other parties may properly remove or diminish their entitlement to remain protected by the agreement or law.  To take again an analogy from the context of war, it is likely that some limited use of chemical weapons would not always or in all circumstances have been contrary to moral principle on the discriminate and proportionate use of military force, but under the 1925 Geneva Protocol states agreed to abstain from it entirely.  Many of them however attached a rider to the effect that if an adversary breached the Protocol the right to retaliate was reserved.  That was justifiable; but there would still, in any such use, have been a moral duty to continue to abide by the more basic ethical constraints – not to retaliate against non-combatants or with a severity disproportionate to proper military objectives, whatever the adversary might have done.  Similarly, if there is an agreement with the authorities of Country X that we will not collect intelligence covertly from one another, but we then find that Country X is in fact doing so, we would be entitled in retaliation to eavesdrop (rather as during the Cold War there seems to have been almost a tacit agreement, one or two special episodes apart, not to make much public fuss about the intrusive intelligence-gathering operations which both sides conducted in breach of strict international law).  But we would not be entitled to extract information by entrapment, blackmail or torture even if Country X is guilty of such methods.


Using Intelligence


The discussion so far has operated within the broad structural paradigm of just-war reasoning.  In at least one respect, however, intelligence calls for a wider paradigm.  Beyond the activity of collecting intelligence there lies that of using it, and distinctive ethical questions can arise there too.  This is not a matter only of whether the intelligence product is culpably distorted (whether by provider or recipient) to fit policy preference, or its limitations and uncertainties brushed aside in order to justify or at least not impede decisions desired on other grounds – issues such as those which the Butler Committee and its counterpart in the United States examined over the 2003 invasion of Iraq.  Most of the moral questions in this area are general ones bearing upon any use of evidence, not peculiar to that collected by secret intelligence effort.  But where, as is usually the case with secret intelligence, the evidence is for good reason not made available openly for all hearers to probe and test, a special moral responsibility lies upon intelligence authorities and their customers not to claim wider knowledge or greater certainty than is genuinely warranted – not to assert, for example, that evidence is “extensive, detailed and authoritative” [8] when its true quality is such that objective evaluation of the same evidence by a broadly-composed committee of inquiry can lead its chairman to conclude that it is in reality “very thin” [9]. 


There are moral issues also about the use of secret intelligence when it is brought to bear in ways that touch seriously upon the rights of individuals.  It would normally be wrong for public authorities to use to the disadvantage of individuals information which those individuals had no opportunity to test or rebut, and of which they might even be wholly unaware.  Yet the product of intelligence is occasionally used in such ways, for example to trigger intrusion into normal privacy, to refuse or limit public employment through processes like vetting, or more recently and controversially in decisions to restrict liberty or withdraw right of residence.  Moreover, such actions by government may often rest upon judgments of probability, not upon proof of the standard that court proceedings in criminal justice customarily require.  Given the entitlement of our societies to be protected so far as possible from grave risks, we cannot reasonably say that the product of intelligence must absolutely never be so used where we perceive pressing danger from traitors or terrorists, or during the Cold War when Western countries honestly believed that this was necessary in order to guard against grave potential threat from the Soviet Union.  Once more there is no escape from weighing conflicting considerations, rather than resting on some neat and comprehensive rule.  The weighing must however have an ethical dimension and take into account, alongside whatever are the proper benefits that we believe can be secured in no other way, the limitations and uncertainties of secret intelligence as well as the disagreeable (and ultimately even corrupting) subtractions which its use in such circumstances entails from the customary and proper standards of open democratic societies, and in the long term from the valuable wider respect in which such societies seek to be held.  The bar ought to be set high.




There are further aspects of the topic on which this essay has not attempted to touch, such as the proper relationship of intelligence activity with one’s own domestic law.  (Should the duty of obedience to that law be regarded as absolute, or as open to be qualified in extreme circumstances by considerations of proportionality?)   In brief, however, secret intelligence-gathering is a valuable, sometimes even an indispensable, tool in the ability of public authorities to fulfil their duties of protection towards those for whose well-being they bear responsibility.  Its effective conduct must sometimes require that action be taken which ought in most contexts to be regarded as morally wrong.  In face of that tension we cannot say that morality must simply be set aside; we have to identify some conceptual structure for legitimating and disciplining the activity.  An ethical framework is needed in at least two parts: the first for delineating the purposes important enough to justify some stepping outside normal ethical expectation, and the second to limit how far that stepping outside may reach and what forms it may take.


It would be absurd and offensive to suggest that intelligence professionals in countries like those of the West do not already have ethical concepts which they bring to bear on what they do.  For reasons sketched earlier, it would moreover be unrealistic to expect to frame an open and explicit code in specific terms to govern the entire activity.  There would however be merit – not least for public confidence and support – in seeking to develop a wider and more systematic understanding of principles than seems yet to have been generally established and recognised on either side of the Atlantic.


This essay is developed from the author’s Annual Lecture to the Centre for Intelligence Studies, Department of International Relations, University of Wales, Aberystwyth, in November 2005.




1.”Agents for Change”, ed. Harold Shukman, pub. St. Ermin’s Press 2000, pp.61-71

2. Duane R. Clarridge, quoted in the New York Times 2 February 2006

3. “Security and Democracy”, Admiral Stansfield Turner, pub. Sedgwick & Jackson 1986

4. For a brief overview see “Justifying War”, Michael Quinlan, in Australian Journal of International Affairs Vol.58 No.1, March 2004, pp. 7-15

5. Review of Intelligence on Weapons of Mass Destruction: Report of a Committee of Privy Counsellors, HC 808, HM Stationery Office 14 July 2004

6. See discussion by Michael Herman in “Understanding Intelligence in the Twenty-First     Century”, ed. Scott & Jackson, pub. Routledge 2004, pp.187-192

7. For a preliminary survey see “Agents for Change”, op.cit., pp.62-65

8. Prime Minister Blair on Iraqi possession of weapons of mass destruction, House of Commons Official Report 24 September 2002, column 3

9. Lord Butler of Brockwell, House of Lords Official Report 7 September 2004, column 463