A Little Intelligence ...

... about Intelligence.

I was very pleased earlier this week to discover from the Substackometer that I now have a total of over 10,000 subscribers and “followers.” There is a distinction between them known to Substack, but it amounts to essentially the same thing. It looks as though my essays have an average of about 12,000 readers, together with those who read them in translation, and those who read them in email form, passed on by others.This is so much more than I was expecting or hoping for that it’s actually quite hard to explain how much more it actually is. Thank you to all those who subscribe and “follow,” especially those who buy coffees and throw occasional coins into the bowl.

Meanwhile, these essays will always be free, but you can continue to support my work by liking and commenting, and most of all by passing the essays on to others, and passing the links to other sites that you frequent. If you would like to take out a paid subscription I won’t stand in your way, (I’d be very honoured in fact) but I can’t promise you anything in return except a warm feeling of virtue.

I have also set up a Buy Me A Coffee page, which you can find here.☕️ Thanks to all of those who have recently contributed.

And as always, thanks to others who tirelessly supply translation in their languages. Maria José Tormo is posting Spanish translations on her site here, and Marco Zeloni is also posting Italian translations on a site here. And many of my articles are now on line on the site Italia e il Mondo: you can find them here. I am always grateful to those who post occasional translations and summaries into other languages just as long as you credit the original and let me know. And now, onward and (we hope) upward;

*****************************************

In my last essay, I touched on the topical issue of the existence or otherwise of an “Iranian nuclear programme” and what intelligence sources were supposed to be saying or not saying about it. I used it as an example of a case which is inherently complicated, where any judgement has to be hedged around with nuances, and where the political leadership and the media, seldom bothering to try to understand such complexities, want simple answers that are often not available.

Intelligence has been in the news this week as well, after the CIA’s interesting and quite damning nostra culpa over its involvement in the “Russiagate” hoax, and the many professional errors that were made. In fact, Intelligence as a subject is very rarely out of the news these days, and even less rarely out of the breathless pronouncements of pundits writing about current conflicts and crises.

And yet the quality of information you find in the popular media is in general extremely low. I don’t mean just errors of fact, of which there are plenty, but even just a lack of basic awareness of what Intelligence is all about, and how it has historically worked: information which is not hard to find if you care to look for it. But there is probably no subject of such importance in a democracy which is so poorly understood, yet so frequently pontificated about, drawing overwhelmingly on popular culture stereotypes presenting intelligence services as Hollywood heroes or Hollywood villains, according to taste, and in both cases crediting them with supernatural powers of omniscience and omnipotence that they don’t have.

This is curious at first sight, since between the end of the Cold War and the generalisation of the Internet, information about Intelligence has never been so available as now. Most agencies, in the West at least, have their own websites, and recruit staff openly, as does the Russian SVR. Senior figures from these agencies speak on the record, some write books and even novels on retirement. Intelligence Studies is a modest but lively academic discipline with its own journals and conferences, and the subject is taught at various levels in a number of western universities. The CIA has a massive online Centre for the Study of Intelligence, a huge repository of historical documents and studies, with an online journal. Many countries, like Canada, have national academic associations for intelligence studies. Many official or semi-official histories of intelligence agencies have been published, as well as official enquiries into intelligence scandals, such as the Butler Report into intelligence failures in Britain before the second Iraq War, and even government responses to them. So anyone wanting to find out basic facts (like the real names of “MI5” and “MI6” for example) can do so quickly, and anyone who wants to know how intelligence agencies are structured, what they do and the problems that management of Intelligence poses in a democratic society has a mass of material to work with.

But in general, they don’t want to know. There are a number of reasons for this, and some are straightforward: the amount of work needed to acquire a tolerable understanding of the subject is substantial and off-putting, for example. But a more important reason takes us to John Le Carré’s attributed remark that the intelligence services are a kind of X-ray of the soul of the nation. Le Carré was thinking, to be sure, about how national cultures shape the way that intelligence services function, which is a point I’ll return to, but there is a wider issue also, in the way that Intelligence is conceived in different cultures. Intelligence seems to function as a blank screen onto which different fantasies and fears are projected, often having only a passing relationship with reality: a few very different examples must suffice.

In the Arab world, intelligence services are feared, even more than the wider security sector, and interest in the Mukharabat is strongly discouraged: you stay out of their way. In France, the services have a rather romantic, derring-do image, itself related the pride the French have historically had for their military, and which is widely shared across the political spectrum. In many other countries (post-Communist states for example) the intelligence services are seen as corrupt and politicised, whilst in still others the subject isn’t mentioned in polite conversation. (“In our country these things are just not discussed” as a Swedish official said to me some years ago.) In Anglo-Saxon countries however, and especially under the influence of US popular culture, there is an entire virtual construct, largely unmoored from reality, derived from popular fiction since John Buchan, from Hollywood thrillers and treatments of real historical events such as Watergate, from sensationalist reporting, and from the mutual interaction of all of these elements Thus, the value of any writing about Intelligence in the West today is primarily judged, not by its authority and persuasiveness, but by how closely it adheres to popular cultural stereotypes. As psychologists have known for ages, the endless repetition of ideas and memes, whether accurate or not, eventually convinces people that they are true. Anyway, it’s all much easier and more fun than doing real research.

So it struck me, in furtherance of the principle that these essays should be helpful, that it might be useful to recall the basic outlines of what Intelligence is all about, and to summarise them here. My aim is the very modest one of trying to assist people in better understanding and making sense of what appears in the media and in government pronouncements, bearing in mind that between ignorance, bias, fantasy and the deliberate intention to mislead, it’s easy to get lost completely. This is not a subject where I consider myself an expert, though like anyone who hangs around in government long enough dealing with international affairs and security I’ve had some exposure to the issues, but in any case I’m concerned here with the 10,000 metre view of the subject that anyone who’s been in government will be familiar with. There are no secrets given away here: I don’t think I know any, actually.

First of all, though, it’s fair to concede that the more responsible and objective sources I mentioned above do have their limitations. It’s not just that they tend to be unexciting and scholarly, they also, by definition, leave a lot out. After all, the essence of Intelligence is secrecy, and it is of little value if the target knows what you have collected, and how you have collected it. So the CIA doesn’t hold a press conference to explain that it has recruited a new source in the Kremlin, any more than the Chinese Ministry of Security publicly announces that it has managed to insert a backdoor in a new computer chip. Even historical information about what are described as “sources and methods” may be too sensitive to print.

More importantly perhaps, the sizeable amount of reliable material now available is generally written from a narrowly western-Liberal perspective, and often from an even narrower Anglo-Saxon one. In studies of intelligence ethics, for example, a field which which has blossomed in recent years, practitioners are almost all from Anglo-Saxon nations, and the studies, interesting as some of them are, tend to be explicitly concerned with how intelligence agencies of western powers should behave. (For example the idea of Just Intelligence, much discussed a few years ago, was essentially an offshoot of Just War theory, and, like the latter, essentially incomprehensible outside a very narrow ethical and political framework.) Likewise, many books and articles on Intelligence are by western lawyers and political scientists, preoccupied— obsessed, even—with the problems that they see in fitting the rough beast of Intelligence into the constraints of a modern liberal democratic society, and consequently fixating on legal and political “controls.” Similarly, studies of intelligence services in political transitions, of which there are now many, tend limit themselves to awarding marks out of ten for how far they imitate the ideal form of western intelligence arrangements, rather than whether they are any good at their jobs. It’s hard to imagine that an Iranian or a Chinese intelligence official could draw anything useful from this material, and it’s a shame that there is very little theoretical or descriptive work on Intelligence from outside the West.

The final caveat is an epistemological one. There is a widespread assumption that that the “truth value” of a piece of Intelligence, to borrow a mathematical concept, must necessarily be high. This impression is reinforced by the necessary secrecy of its acquisition and treatment and the limited number of people allowed to see it. Yet the “truth value” of a piece of Intelligence is not necessarily greater than information coming from elsewhere: it depends heavily on the subject, the sensitivity of the question, the way the information is collected, the reliability of the source, the degree to which the information confirms (or doesn’t) to other information available, and many other things. A technical report from an IRGC officer to his hierarchy about a missile test in Iran may have a high truth value, albeit in a limited context, whereas the Foreign Minister’s alleged comments about upcoming trade negotiations, reported at third-hand by a source in an Embassy, may have a much lower truth-value. And then history shows, surprisingly enough, that human sources can be mistaken, confused or may even be making up information in the hope of earning money.

Thus, the idea of “proof” in Intelligence issues, except in very special contexts, is an epistemological category error. “Proof” is something that exists in debates in science or the law, where there are distinct and accepted sets of rules for determining “truth,” and a means of judging who has been able to demonstrate it. With Intelligence, the best that you can hope for in most cases is a sufficiently strong presumption that you can act upon it. Needless to say, just as one group will accuse you of acting or speaking “without proof” if you make one decision, so other groups will accuse you of “ignoring the evidence” if it later emerges that you should have made another. But this is all part of the wider political game surrounding Intelligence.

Intelligence is generally collected for a reason, and reasons generally involve decision-taking at some point. Take a simple and typical case. Your country has a difficult relationship with its neighbour, and there are persistent allegations in the media that they are funding and training separatist militants in the border region. Your neighbour firmly denies this. Another regional power has managed to persuade the two Presidents to meet for talks to calm the situation. At that moment, photographs of dead bodies in the uniforms of your neighbours’ military are circulating, allegedly killed in anti-guerilla operations, and there are various anonymous social media accounts allegedly by militants recording training in your neighbour’s country. The Opposition is demanding that the talks be cancelled: your neighbour’s government is alleging a “false flag” operation designed to sabotage the talks. The President wants to know what to do.

Assuming your government has some kind of central analytical staff, they will commission an assessment. What it says is that there is some evidence, but not a lot, of your neighbour’s involvement. Some captured militants claim to have been trained by foreigners, and some weapons used by your neighbour’s military have been recovered, but these weapons are also available from other sources. A human source in your neighbour’s Army claims to have heard that senior figures from the militant group have been received at a high level in Army HQ. Your Ambassador chips in with the observation that, according to one of the Embassy’s most trusted contacts, factions in the Army may be operating independently of political control in this case. Just another day in the confusing world of intelligence assessments.

So far, I have used examples you might think of as “classic,” drawn from the security area. But the logic of intelligence collection and assessment is not restricted to any particular domain. We can think of Intelligence as a particular type of information: by definition, all Intelligence is information, but not all information is Intelligence. That qualification is reserved for information collected surreptitiously, so the first requirement is to define circumstances where the cost, time and potential risk of collecting and analysing it are justified. Clearly, there will be a threshold beneath which the extra insights that might be gained are not worth the effort that needs to be invested. There is a whole range of subjects where information is openly available and not contentious, where governments exchange information freely, or where the government already has access to all the information it might need for some domestic initiative. If you are planning an educational exchange programme with another country, for example, little if any information will be hidden by either side. (Of course if you think the other country might be using this as an intelligence-gathering opportunity, then different rules apply, as we’ll see.)

The use of intelligence assets is therefore the exception, and it doesn’t always follow that the obvious subjects (defence, security, foreign affairs) are the ones that your intelligence services should take an interest in: it depends on the overall priorities of you government. You may live in a region which is basically quiet and stable, but where transnational organised crime (TOC) is a problem. In that case, not only will a lot of your intelligence effort be targeted against TOC, but the very structure of your intelligence community will reflect this priority, and since TOC is by definition international, your intelligence services are likely to have substantial contacts with other countries in the region and elsewhere, and with organisations like UNODC. But on the other hand you may be a small country in a resource-rich area, and you are primarily interested in economic issues, such as the investment plans that major states and transnational industries might have in your region. It all depends.

I’ll take an imaginary but realistic example, and like all of these examples it will be from outside the Anglo-Saxon bubble. Let’s assume that Russia is hosting a BRICS summit at the end of this year. Now a host state generally has two priorities in such circumstances. One is that the summit is seen to be a success, and so reflects well on the organisers. The other is that initiatives by the host state (since there will generally be at least one such) should make progress at the Summit. Preparations for the Summit will already have begun, and many of them are concerned with establishing the baseline: what do we want? what can we accept? who will take which position? who will be for/against our initiatives? what is it reasonable to expect? To do this, you obviously have to know something about what your guests want, hope for and will accept. Some of this (communiqué or declaration drafting, for example, which begins some time ahead of the actual meeting) will be overt, and conducted by embassies and ad hoc meetings. The Russian embassies in the various countries will then make contact with their host governments, compare ideas, lobby and receive lobbying, as part of the process of constructing a programme for a successful meeting.

But there will be key issues where there is still uncertainty. The Chinese system is not easy to deal with, and is famously complex and opaque. So the Russians will target intelligence assets on the Chinese to discover where their red lines are, where they hope to make progress, what their assumptions are about Russia’s own objectives and so forth. I have no idea what technical capability the Russians have against the Chinese, but the rule of thumb is to take the easiest way in, so they may attack the Chinese Embassy in Pretoria, for example. They will also be attacking the Indians. In turn, the Chinese will be doing the same to the Russians and to the Indians as well. So from the Russian perspective, when Mr Putin comes to actually chair the meeting, he may know, for example, that a reported Chinese red-line is in fact just negotiating fat, and that if he presses hard they will give way. He may also know that an Indian initiative to be tabled at the Summit is controversial among major government figures, and that a small push-back will be enough to get them to withdraw it.

But wait, I hear you say. China and Russia are allies, aren’t they? Surely they wouldn’t spy on each other? That would be ungentlemanly. Well, the truth is they do, even if it is. In fact, broadly speaking, everybody spies on everybody else, and in general this is accepted as part of the rules of the game, subject to qualifications I will touch on later. Of all of the characteristics of the intelligence world, this is probably the most difficult to wrap your neurones around, but it is fundamental to the understanding of how things actually work.

In turn, this derives from the way the international system functions. Most people have one of two vague ideas about this in their heads. One is the broadly realist paradigm of states in eternal conflict for power and influence; the other is of states with long-term, if not eternal, “friends” and “enemies.” In practice, most people believe both of these things, sometime on alternate days, sometimes at the same time. The reality is simple to explain, although its application can be confusing. Consider the world, as I’ve said before, as a gigantic series of Venn diagrams. Each circle represents the interests of a country in a particular subject, and will overlap with the interests of other countries in certain cases. In general, there is nothing as simple as “the” national interest, and with some countries you can have a common interest in one area, and be completely opposed in another. Where interests overlap you may cooperate, or simply decide not to get in each other’s way. In certain cases, this may mean that you share Intelligence with nations on one subject even as they are a major intelligence target in other areas.

In the case above, China and Russia have all sorts of relationships, some positive, some neutral, some conflictual. They collaborate on some technological questions, and are at odds on others. But even when they are cooperating, they will use their intelligence capabilities to find out what the other party is up to, and what they want. Both sides understand that this is how the game is played. Now of course all intelligence collection involves risks of some kind or another. The organised hypocrisy that surrounds the subject means that for most of the time that doesn’t matter but, if for example word gets out publicly that one or the other has made a successful technical attack, then although the attacking power will always deny such scandalous slanders, there will usually be real-world consequences of one kind or another, even if they are not made public.

Sometimes, overlapping interests make even quite surprising types of cooperation possible. For example, in Syria, the US and other western powers were politically committed to getting rid of Assad, whereas the Russians thought it was in their interests that he should survive. Yet it was in the interests of neither that their forces should come into conflict, so deconfliction arrangements were introduced. Nor was it in the interests of either that the Islamic State should prosper, so there were reported intelligence-sharing arrangements between the two countries, and probably the British and French as well.

In turn, this reflects the very pragmatic, often brutal, way in which intelligence, and more widely influence, are traded between states. International relations as a whole is very Darwinian, just not in the sense that Realists imagine. Rather, your influence depends on what you have to offer that others need. Nobody is interested in your moral standing or your glorious history, but rather in what you can put on the table. You can also have a power of nuisance that means other countries have to take account of you: Algeria and the Sahel is a good example.

This often puts small nations in a powerful position. During the Cold War, for example, Swedish attention was almost exclusively focused on the Soviet Union, and especially on the organisation and deployment of Soviet forces in their region. Under secret agreements with NATO, that made them de facto members, intelligence was certainly passed to others including the US: what was bought with it precisely will never be known, but it was probably substantial. It’s likely that much the same thing is happening with Russia today. Similarly, one of the major priorities of Austrian intelligence at the same period was the stability of the ex-Yugoslavia, and in 1991 they were an important source of information for other western states. Finally, the Australians have had for many years a special interest in their own region, notably Indonesia, and have had an Over-the-Horizon radar capability (the Jindalee system) which monitors aircraft movements thousands of kilometres away.

This kind of thing is important because, in reality, even large states lack the capacity to do everything. Language is often a major limiting factor, and, if you have had to learn a language as an adult, that won’t surprise you. Languages like Japanese, Chinese, Arabic and Russian take years of full-time study to master, and there will still be large areas of technical vocabulary to be added before you can work as an analyst. And even then, you are unlikely to be able to do more than survive in such societies: finding and developing human sources is another kind of problem entirely. And of course agencies never know what’s coming next. At the end of the Cold War, most countries had lots of Russian linguists. Quickly, they started to recruit and train Arabists, only to have the war in Yugoslavia break out. At that point the British had, as I recall, fewer than half-a-dozen reasonably fluent speakers of Serbo-Croat throughout the entire government. Quite a few countries had none at all, so that much potential intelligence about what the factions were up to simply could not be used, even if it was collected. But in turn, this demand gave way to the need for speakers of Albanian (Shiq) during the Kosovo crisis, followed by speakers of Pashtun and Dari as a result of the Afghanistan adventure. Then Syria and Libya brought back the requirement for Arabic, just as Ukraine was heating up. And let’s not even go into the question of dialects: what is known as Modern Standard Arabic is taught in western universities, and most Arabaphones should be able to read it, but it’s very different from the Arabic ordinarily spoken, which is often littered with loan-words from the past (Turkish, French, Italian) and these days from English as well.

Won’t technology save us? Well, up to a point. Automatic translation can be extremely useful for documents, in at least giving you, perhaps, ninety per cent of the sense, but is unlikely by itself ever to produce results you can use to take decisions. And as for voice, most people who communicate sensitive information know how to use speech codes and slang. Few automatic translators could cope with Respect, reuf! la gonzesse a le kertru du toubib, gare aux keufs. (Roughly, Hello my brother, the lady is bringing the merchandise from the medical facility. but look out for the Police!), especially with a strong Maghrebin or West African accent. And few automatic transcription systems can cope with accents anyway, as is hilariously demonstrated every day by YouTube. You can use software increasingly to sort through intercepted phone calls and identify key words, but in a sense, that’s the easy bit, because almost by definition, you will get huge numbers of false positives. And even then, “meet me at the first fallback rendezvous half an hour later than normal on the usual day” doesn’t mean much in any language. Intelligence analysis is an extremely high-context activity, where, unless you know the background, an individual intercept, or a single human intelligence report, can be essentially meaningless.

However, the more important point here is political, in the degree of control and influence that some states can acquire through Intelligence. For example, Korean (Hangul) is a difficult language to learn, and essentially unusable outside Korea. So whilst there are various technical means of gathering information about North Korean military programmes, for anything more detailed, and anything at all about the political and military situation in Pyongyang, the US, and for that matter other western countries, are entirely dependent on the National Intelligence Service of South Korea, which has its own agenda and of course reflects that of its government. It is very reluctant, for example, to grant other nations access to North Korean defectors. Likewise, the degree of reliance of the US system on Israel for its Intelligence on Arab states and Iran is proverbial, and it has become clear since 2021 that the CIA was hopelessly manipulated by the Pakistani Inter-Services Intelligence throughout the Afghan crisis. That said, the opposite is also true: Intelligence can also be traded for other political or economic favours, or simply used to influence the thinking of other governments.

Intelligence is therefore a pervasive tool of policy in all areas. This doesn’t mean its use is automatic: as I’ve indicated, it’s a cost-benefit analysis in part, and depends also on the consequences if things go wrong. But the easiest rule of thumb is that if something is both cost-effective and potentially useful, then an intelligence agency is very probably doing it. Because why wouldn’t they? If I were the President of China, and I found out that the Ministry of State Security (MOSS) was not busy installing back doors in computer products, not making use of overseas students and workers to steal secrets, and not spying on the Chinese diaspora, then I would want to know what the hell they were doing, to earn their salaries: playing video-games? Which is to say that where there are well-known and well-tried techniques of gathering Intelligence and manipulating people, it’s wise to assume that every national agency that has the opportunity to use them is doing so.

This isn’t all high-profile, short-term stuff, either. Some countries take a very long-term view, and encourage their operatives to identify people who, one day, might be of use to them in certain circumstances. The old NKVD, for example, recruited university students from the British establishment in the 1930s in the hope that they would, years in the future, be influential and useful. It would not be surprising if the Chinese, whose approach is similarly long-term, were doing something similar: indeed, I would be surprised if they were not. But that’s the way the game is played.

Finally, international and multinational organisations are also a magnet for intelligence officers of the world: heaven alone knows how many are working under cover in the UN, for example. One reason for this interest is obviously to find out what sensitive parts of such organisations are up to (the IAEA s a topical example) but more generally, such organisations are target-rich environments for identifying future sources. They are full of people living an expatriate life who may well be lonely, but are also becoming accustomed to a standard of living they will not be able to enjoy once they return home. A sympathetic and friendly “diplomat” may well be able to help them with both problems, in return for “small favours.” Other reliable targets are interpreters and translators with families back home. It doesn’t even have to be overt pressure: do us a few favours, and we’ll get your brother that job in the University he so much wants.

So far, I have sketched out, very simply, what it is that intelligence agencies spend most of their time doing: surreptitious collection and analysis of information believed to be important. I’ve largely discussed foreign Intelligence, because the domestic variety has other issues as well, and there isn’t the space for them here. But I haven’t said anything about the image of intelligence agencies found in the popular media: assassinations, “false flags”, overthrowing governments, manipulating journalists, fixing elections, training terrorist groups and so forth. What about that?

Well, the first thing to say is that generalisation is dangerous. Whilst Intelligence organisations share common characteristics, the environment in which they work, and so the tasks they carry out, are very varied: we can look quickly at a few general types. One is a service intended to keep a regime in power. Here there is generally an ideological component. The old KGB was theoretically a Department of the Communist Party, not the government, although in practice it was treated as a Ministry. The Chinese MOSS began the same way, although it technically reports to the State Council. And in Iran, the Ministry of Intelligence (and almost certainly parts of the Revolutionary Guard Corps) are clearly servants of the Islamic Republic, not the country. Here, the first target is the population of the country, or at least those who might want to live under a different system. There are also expatriate enemies to counter. Such agencies naturally place a lot of emphasis on what would elsewhere be Police work (MOSS personnel have powers of arrest and the organisation has its own prisons.) They are also the first choice for harrying, and if necessary assassinating, dissidents abroad.

Similar but not identical are the Intelligence services which maintain regimes (often personality-based) in power. This was the case in Iraq under Saddam Hussein, in Syria under Assad and Libya under Gaddafi. It is arguably the case in Algeria where the regime has lost whatever ideological beliefs it once had, and in Rwanda, where the Intelligence service essentially keeps Kagame in power by murdering anyone who gets in the way. A characteristic of both the Iraqi and Syrian regimes was the multiplicity of agencies (seven were counted in Iraq) who were given deliberately overlapping mandates and encouraged to spy on each other to maintain the regime in power. Such organisations can be extremely powerful. Between 1990 and 2005, for example, Lebanon was essentially run by Syrian Military Intelligence: the Director had a special entrance into the Serail, the Prime Minister’s office, which he used every evening to come and give the Prime Minister his instructions. And the feared Direction du Renseignement et de la Sécurité in Algeria was often seen as the effective government of the country, until its dissolution a decade ago.

All of the agencies mentioned in the last paragraph are or were under the control of the military, and indeed, this is how Intelligence agencies started, and how most people still think of them. In the nineteenth century state secrets were essentially military, related to mobilisation and war production. When the French organised their General Staff after the defeat by the Prussians in 1871, they created the Deuxième bureau, which dealt with military intelligence issues. Other countries followed, and today headquarters organised along western lines all have what is known as the “J2” function. (J1 is organisation, J3 is Operations, J4 Logistics, and so on.)

In the French case, the overseas intelligence organisation remained heavily militarised for a long time, and the DGSE is still under the control of the Defence Minister (but not part of the Ministry.) Although the DGSE has rapidly been civilianising (most of its analysts, apparently are now civilians), its ethos is still heavily influenced by a history of clandestine military operations, a tradition going back at least to the era of the Free French in London. Its Service action is only tenuously linked to Intelligence: its principal tasks are operational, including kidnapping and assassination. (The SA was responsible for the Rainbow Warrior fiasco in 1985.) By contrast, the British decided as long ago as the 1920s that Intelligence was too important to be left to the military. The Directorate of Military Intelligence in the War Office lost all its non-military functions, which were transferred to newly-created civilian agencies. Essentially the same model has been followed by other Commonwealth countries.

This could go on for pages, and the sociology of intelligence organisations is a fascinating subject, for me at least. But I just want to emphasise again that the organisation and tasks of intelligence agencies reflect the history and culture of their country, and probably no two have the same exact structure or list of functions. That said, they have generic operational skills that governments often find useful. Making unofficial contact with rebel or organised crime groups or conducting hostage negotiations, for example, is not something that you can or would ask accredited diplomats to do. So it’s apparently the case that when the British government started to open contacts with the African National Congress in the 1980s, it turned to the intelligence services first. This is typical.

Depending on the context, then, “intelligence agencies” can be anything from the secret arm of the State, at one extreme, to small analytical organisations attached to the Foreign or Defence Ministries at the other. It’s important to understand this as you read garish accounts of the alleged activities of such agencies in the media. The only general rule, I think, is that the larger and more powerful an intelligence agency is, and the more independence it has, the more it risks moving away from its core tasks of gathering and analysing intelligence.

Which I suppose brings us back to the CIA, where we started. It’s ironic that an agency that publishes an overwhelming amount of material itself, and has been written about at such inordinate length, remains so much of a mystery: or rather, it’s treated as such by people who find real life too boring. All one can say, I think, is that it is part of the US system, and that system is fragmented, conflictual and personalised, such that each organisation aspires to expand into the area of others. It’s often alleged, for example, that the CIA has its own foreign policy, and there is some evidence from the past that this is so. The Agency has a history of ignoring, or at least creatively interpreting, the wishes of the government, and its very size, its budget, and the extent of its resources mean that doing so is always a temptation. To this has to be added the heritage from the wartime Office of Strategic Services, generally taken to be its parent organisation. OSS, from what we can gather, was a pretty amateurish organisation, at least when it started, and many of its senior personnel had no Intelligence background. It was also heavily militarised, and much of its work consisted of thud-and-blunder operations of questionable ultimate value, and which obscured the much more useful work of gathering and processing Intelligence.

The CIA is the inheritor of these traditions, and for much of its history there was a tension between the Directorate of Analysis, which was well-respected and quite sensible in its judgements, and the Directorate of Operations, which had a tendency to play Cowboys and Indians around the world, often with disastrous results. According to those who have worked there recently, this tendency was enormously strengthened after 2001, to the point that the military and political activities of the Operations side threatened to get out of control. This is perhaps an extreme case of one of the basic problems with intelligence agencies: the political leadership becomes so dazzled by the promises of what they can accomplish that it loses its marbles.

That’s a very brief and superficial canter over the field of Intelligence, based for the most part on general knowledge And it’s not as though important questions aren’t immediately raised. For example, how should intelligence services be targeted in a democracy, what methods should they be allowed to use, and who decides? Or how do we deal with the acknowledged threat from foreign intelligence services without harming the interests of expatriates or immigrants from those same countries?

But actually, people aren’t much interested in such questions. As I suggested at the start, intelligence services are a kind of blank screen onto which our fears and fantasies are projected, enabling us, according to taste, to feel admiration or a sense of moral superiority towards them. And so people speak with utter confidence about subjects on which they are largely ignorant, since they have no interest in educating themselves. Many Internet sites, and even more commenters on those sites, offer learned discourses, throwing around words like “asset,” “intel” or “spook” in an attempt to pass themselves of as experts, and recently, I see, terms like “spook adjacent” have started to appear, whatever that is supposed to mean. So the late Mr Epstein is confidently pronounced to be a CIA agent, a CIA “asset,” whatever that is, a Mossad ditto, a double agent of some kind, an independent blackmailer, murdered by the CIA, murdered by the Mossad, murdered by the Russians, and half a dozen other theories without a scrap of evidence, put forward by people who don’t know the difference, for example, between the FSB and the SVR. But it’s all good fun.

This is not good for democracy, whatever it may do for internet clicks, and it transforms what are actually serious questions of the control and tasking of intelligence agencies into a kind of competitive fantasy-writing competition where the more outrageous the accusation, the more clicks you receive. The irony is that Intelligence is a fascinating subject anyway in real life, and raises all sorts of interesting moral, political and practical issues, some of which I’ve touched upon lightly here, and about which there is much more to be said. There is an obvious advantage, after all, to discussing Intelligence issues with the application of, well, a bit of natural intelligence, rather than Hollywood conventions. Let me know if you would be interested, and in the meantime, when reading about Intelligence, don’t forget what the Dao de Jing says:

Those who know do not speak.

Those who speak do not know.