15 July 2001. Thanks to Anonymous.
Source: http://www.cer.org.uk/n5publicatio/cerwp4.pdf (209KB)


Working Paper

Intimate relations

Can Britain play a leading role in European defence -- and keep its special links to US intelligence?

Charles Grant

Charles Grant is director of the Centre for European Reform and a former defence editor of The Economist.

April 2001

Centre for European Reform
29 Tufton Street London SW1P 3QL

A problem which must be addressed

One of the most constant features of the geopolitical landscape is the special relationship between London and Washington on intelligence matters.1 One of the most rapidly changing and unpredictable elements of that landscape is the emergence of a European Common Foreign and Security Policy (CFSP). This paper examines whether, and to what degree, these two phenomena may be compatible.


1 The author wishes to thank the many people who have offered comments and criticisms of earlier drafts. Most of them would wish to remain anonymous. He is grateful to Francois Heisbourg and to Nicole Gnesotto for suggesting that he write this paper. An earlier version was commissioned by the WEU Institute for Security Studies. 

Many British officials involved in defence and foreign policy are relaxed about the question raised in the title of this paper. They assume that Britain can continue to have its cake and eat it - enjoying privileged access to US intelligence, while counting as much as any country in the embryonic CFSP. They argue that "firewalls" within the British government allow the British to keep a foot in both camps: the US will hand over certain reports on the understanding that Britain's European allies will not get to see them, while at the same time Britain can exchange other material with its European partners.

But some continental officials are convinced that if Europe becomes a significant player in foreign and defence policy, Britain will eventually have to confront a painful strategic dilemma. One French official argues that Britain will not be able to play a leading role in the EU unless it jettisons the special intelligence links to the US: "Britain must choose Europe or betray it."

That assessment is over-dramatic and, in the opinion of this author, false. But the British are too insouciant. For if the CFSP proves a successful enterprise, the special relationship will start to create difficulties. Since the formation of foreign policy depends, at least in part, on intelligence assessments, the fact that EU countries receive different and divergent assessments must make it harder for them to forge common policies.

Intelligence may not often be the determinant factor in the making of foreign policy. But sometimes it does matter, particularly in the shaping of policy towards countries with closed societies, such as the so-called rogue states; in an open society, one can usually find out what is going on in through monitoring the media. And intelligence is hugely important for the successful conduct of military operations. Thus Europe's embryonic foreign policy and the projected European Security and Defence Policy (ESDP) will be handicapped unless there is a high degree of intelligence sharing among EU governments,

Britain's intimate connections to the US may make it harder for the Europeans to share intelligence among themselves - because Britain may be less interested in intra-European sharing, and because its EU partners may trust Britain less. Equally, if the Americans believe that Britain has developed special links with its European partners, and that it is part of a European enterprise that is challenging American power, they may become wary of sharing with the British.

Britain has a clear national interest in encouraging the development of a European intelligence capability, as a means towards a more effective CFSP; but also in preserving its special access to US intelligence. The point of this paper is to suggest how those objectives can be reconciled.

The paper examines the nature of the special relationship; the extent of intelligence co-operation among Europeans; the controversy over the Anglo-Saxon countries' signals intelligence network, known as Echelon; the argument over whether Europe should have its own spy satellites; and the significance of intelligence in the formation of European countries' foreign policy. Finally, the paper makes some suggestions on how the Europeans could deepen their cooperation on intelligence, in ways that need not damage the special UK-US relationship.

The special relationship

Relations between Britain and America are very special in at least three areas:

Each of the British intelligence services has a liaison office, staffed by senior officers, in the US. These offices obtain material from the US services and supply British intelligence to them. There are also British officers seconded to US agencies at an operational level, and vice versa. No other European or Asian country has such intimate relations with the US agencies.

British-American co-operation on human intelligence usually involves exchanges of intelligence assessments, rather than joint operations. The difference in styles of the SIS and the CIA - the former stressing the use of agents, the latter devoting more resources to sophisticated technology, the processing of information and analysis - means that it is not easy for them to work together on operations.

Signals intelligence is the most special part of the special relationship - and has been ever since 1941, when American and British code-breakers started to work together at Bletchley Park. Britain's GCHQ and America's NSA exchange many dozens of staff with each other. Each organisation takes responsibility for certain parts of the world. The British have listening posts in places like Cyprus, where the US has none, so the Americans regard the British contribution as very useful. But in "sigint", as in other forms of intelligence, the British services have no doubt that they get more out of these sharing arrangements than they contribute. So they are strongly wedded to the special relationship.

Australia, Britain, Canada, New Zealand and the US, bound together by various intelligence-sharing agreements that date back to 1948, reveal more to each other than to other allies. This intelligence sharing among the five Anglo-Saxon countries is institutionalised at the very heart of the British system of government. The Joint Intelligence Committee is the body in the Cabinet Office which sets goals for the UK agencies; sifts and evaluates their output; and presents summaries to the prime minister. Most other countries do not have an equivalent of the JIC, with the result that their intelligence agencies tend to be less well co-ordinated. There are two categories of JIC meeting: those at which the Anglo-Saxon allies are represented; and those at which only Britons are in the room. Britain's European allies do not attend any sort of JIC meeting.

The British and American intelligence establishments are bound together not only by practical co-operation, but also by a common approach to the use of intelligence. According to senior figures in the UK and US governments, intelligence has more influence on their foreign policies than it has on the policies of continental European governments. The reason, they say, is that the Anglo-Saxons use intelligence in an empirical way: it is about gathering facts, and if the facts are significant, the policies may get changed. The view in London and Washington is that the French and other continentals, being essentially deductive in their thinking, develop sophisticated analyses and policies and then draw on intelligence to support them; but that they seldom allow intelligence to shift policy.

Is that self-congratulatory British-American analysis true? According to one senior French official, the conclusion, that intelligence is less influential in France, is correct, but not because the French are so Cartesian that they ignore facts. The reasons, he says, are social, historical, and bureaucratic. "In France there has been less investment in intelligence capabilities, and a lower grade of people choose to work in intelligence, which is seen as something dirty. There is no bureaucratic system for diffusing assessments to the key branches of government. " The result, he says, is that decision-makers do not have a lot of confidence in what the intelligence services provide.

Thus the common ground between the British and American intelligence services is extensive. It is inconceivable that a British government would ever wish to abandon the special relationship. So the key questions are whether, and how that special relationship can be made to fit with Europe's emerging CFSP.

Intelligence sharing in Europe

There is a large amount of intelligence sharing among European governments. Some of this sharing is multilateral, within NATO and the Western European Union (a rather sleepy organisation which has acted as the EU's defence club2). However, governments are generally reluctant to circulate the highest-grade material within multinational organisations, because too many people are liable to see it. They tend to be more willing to share sensitive material bilaterally.


2 The WEU has ten full members: the EU's 15 countries, minus Austria, Denmark, Finland, Ireland and Sweden. But it often meets with those five, plus the six European countries in NATO but not the EU. And sometimes the East European countries which have applied to join the EU are also present.

It is important to distinguish between the raw data of intelligence - reports from agents, transcripts of wire-taps or satellite photos - and the assessments based on the data. Governments are naturally more relaxed about sharing analysis than source material. For example, if Britain passed on a report from an agent in Iraq, it could endanger his or her life; passing on the assessment of that report need not.

Even assessments, however, are often regarded as highly sensitive. If one government studies several of another government's assessments carefully, it may be able to guess the sources, and it will certainly gain some insight into the other government's intelligence capabilities. Thus a country with sophisticated intelligence networks is unlikely to want to share high-grade assessments with another country unless it thinks it will get a good "trade" in return. On the other hand, one government's intelligence is more likely to influence another government if it is passed on in a relatively raw state: a photo of a missile silo is more potent than a report saying "there are missile silos".

It is also worth distinguishing between political intelligence, which is relevant to decision-making at the highest levels of government; and military intelligence. The latter can be "strategic", concerned, for example, with a country's weapons programmes, or its defence industrial base; or "tactical", information that is relevant to a military operation. Governments tend to be more willing to share tactical intelligence than the political or strategic sort, particularly with allies who are engaged in a common military enterprise.

For example, in June 1999, just after Milosevic agreed to withdraw from Kosovo, the Russian army despatched 200 of its peacekeeping troops from Bosnia towards Pristina airport. This was the first move in a plan which, if it had succeeded, would have led to thousands of Russian troops flying into Pristina and partitioning Kosovo. The Americans discovered the troop movement as soon as it started, through signals intelligence, and informed their NATO allies immediately. The allies therefore knew about the troop movement before the Russian foreign ministry.

All governments are inherently reluctant to share even military intelligence, especially within multilateral bodies such as NATO. The American, French, British and German intelligence services are among those that provide reports to NATO, but they are doctored so that references to sources or sensitive pieces of information are removed. America's allies have long complained that it is particularly mean with its intelligence; for example it refused, until quite recently, to let NATO allies see satellite photos. However. the US has become more generous in recent years, perhaps because of the growing availability of imagery from commercial satellites.

Any multinational organisation is by definition leaky. Throughout the Bosnian war, NATO secrets were ending up in Bosnian Serb hands. In November 1998 a French officer working within NATO, Pierre-Henri Bunel, was found to have passed NATO's target plans for Kosovo to a Yugoslav diplomat in Brussels. And in March 2000 it emerged that, at the start of the bombing campaign against Serbia which had begun a year earlier, 600 people within NATO had had access to the flight plans of the NATO bombers. That may explain why the Serbs evacuated so many prime targets in Kosovo a few hours before the bombs struck.

The Western European Union, which will soon be folded into the EU's Council of Ministers secretariat, has a small unit that gathers and analyses intelligence from its member-governments.3 The WEU also has its own "satellite centre" at Torrejón in Spain. This processes information from commercial satellites and the two Helios 1 spy satellites (which belong to France, Italy and Spain).


3 The WEU's intelligence capabilities are discussed in "Towards a European intelligence policy", edited by Alessandro Politi and published in 1998 by the WEU Institute for Security Studies.

The EU's growing involvement in the fight against terrorism, drug trafficking and organised crime has led to a growth in intelligence sharing among domestic agencies. Britain's Security Service (also known as MI5), France's DST, Germany's BFV and the other domestic agencies exchange information in the so-called Club of Berne.

Most EU countries have bilateral arrangements for sharing intelligence with each other. The continental countries also have bilateral relationships with the Americans, which are often productive, although not as intense as the UK-US relationship. For example in the mid- 1980s President Mitterrand gave President Reagan information gleaned from a highly-placed Soviet source, known as "Farewell"; the two countries exploited the information jointly and 100 Soviet agents were expelled from France.

France can be difficult for other governments to deal with, because the half-dozen French agencies seldom tell each other what they are doing. France lacks the equivalent of a JIC. It does have a Comité Interministériel du Renseignement, which sets priorities for the various services, but there is no central system for consolidation and analysis of intelligence. It would be hard to establish a French JIC, because both the prime minister and the president would wish to be in charge.

However, the British and French intelligence agencies sometimes work very closely together. This is a tradition which - according to some - stretches back to World War II, when the British Special Operations Executive supported the French résistance. In the 1980s, the French services helped the British to intercept boats that were running Libyan guns to the IRA - even though the initial tip-offs had come from the Americans.

France also proved helpful during the Falklands conflict, when President Mitterrand directed the French intelligence services to support Britain. They helped to track the movements of an Argentine ship that sailed close to France and Spain, and was suspected of trying to obtain French-made Exocet missiles. They helped to monitor a threat from Argentine special forces, who were thought to be planning operations in continental Europe, possibly in Gibraltar. And when France was due to deliver Exocets to a neighbour of Argentina - a country which, according to intelligence reports, was prepared to pass on the missiles to the Argentines - Mitterrand blocked the sale.

According to SIS sources, it is not necessarily true that Britain shares a higher quality of human intelligence with the Americans than with the French; it is the quantity, rather than the quality of the UK-US "humint" trade that is unique. "Personal ties between the SIS and the DGSE are sometimes closer than between the SIS and the CIA," says one source. "Ties between the SIS and the [domestic] DST are particularly warm, and they sometimes conduct joint operations on sensitive subjects. "

Co-operation between GCHQ and the French signals intelligence organisation, which is part of the DGSE, is much less intense. This is because French " sigint" priorities - more focused on France itself and the Francophone world - are different to those of GCHQ. It is also because of GCHQ's close ties to the NSA. Nevertheless selective releases of French signals intelligence proved useful to Britain during the Falklands war. And there has been good co-operation on counter-terrorist work.

Germany's allies sometimes worry that its intelligence services, particularly the BND (the external service) may still be penetrated by Russian agents. They are therefore reluctant to show the German services some sensitive material. Germany's allies are also quite rude about the quality of German intelligence. However, SIS sources say that, after the Americans, their biggest "trade" - in terms of the quantity of human intelligence - is with the Germans.

Germany, like France, has a problem with co-ordination. Its foreign intelligence service, the BND, which also covers signals intelligence, is based in Munich. The domestic BFV is based in Cologne, while each of the 16 Länder has its own intelligence organisation. However, parts of the BND and the BFV are moving to Berlin.

When the Franco-German relationship is working well at the highest levels, the two countries' intelligence services are likely to work closely together, and sometimes to engage in joint operations. But there are no institutional structures that promote a permanent special relationship between the French and German intelligence services.

The services of the smaller countries sometimes provide useful information to those of the larger countries - in hope of making trades. The SIS has at various times worked closely with, among others, the Austrian, Finnish, Danish and Swedish services. The Finns, for example, had particularly good contacts with the KGB, the fruits of which were sometimes passed to the SIS.


There is sometimes quiet and effective co-operation between French and American intelligence agencies, particularly on counter-terrorism. Nevertheless relations between the French and the Americans are often strained. Ever since the early 1980s - when the DGSE was found to have bugged those travelling on Air France Concordes, including American businessmen - the US has complained about France's emphasis on industrial espionage in America. Some Americans express the worry that if they pass an intelligence assessment to France, there is a risk of it ending up in Iran or Iraq.

Both the French and the Americans have targeted each other in what some have described as an "intelligence war" - and both sides have engaged in tit-for-tat expulsions. The Americans were particularly annoyed in 1995, when Charles Pasqua, a Gaullist interior minister, made a public issue of expelling a group of CIA officers. The French argue that they had no choice: those officers had been caught bribing a senior member of the prime minister's office to supply information on the French position on the GATT trade talks, and did not respond to quiet hints that they should leave,

The French are particularly exercised about "Echelon", a signals intelligence network among the five Anglo-Saxon countries that was established in the 1980s. It is alleged that Echelon can record, through a network of listening stations on the ground and in space, any normal phone call, fax or email. The use of certain key words, apparently, triggers an analysis of the communication. One of the rules of Echelon is that the five countries do not listen to each other. Thus senior officials in the US government say that they see reports of intercepts from all the European countries bar the UK.

A report commissioned by the European Parliament and published in 19984 claims to describe the Echelon network in detail. The publication of a revised version of that report led Le Monde to lead with a story about Echelon in February this year5. The headline ran:



4 , An appraisal of the technologies of political control", European Parliament, Civil Liberties Committee, September 1998.

5 Le Monde, 23.2.2000.

This front-page lead was followed by two pages of detailed analysis. Le Monde claimed that information supplied by the NSA to American companies had allowed them to beat their European rivals to lucrative contracts. Taking as its source the European Parliament report, Le Monde alleged that in 1995 Airbus lost a battle with McDonnell Douglas for a Saudi Arabian contract - after its offer was overheard and passed to the Americans. It also alleged that Thomson lost a $1.4 billion Brazilian radar contract to Raytheon in similar circumstances. The London Times quoted George Sarre, a left-wing French deputy, as saying that "the participation of the UK in spying on its European partners for and with the US raises serious and legitimate concerns in that it creates a particularly acute conflict of interest with the EU. "

Responding to these stories of commercial espionage - for which the European Parliament report offers very little evidence - American sources claimed that the US agencies have never supplied intelligence to American companies for their advantage. But they admitted that agencies have passed evidence of attempted bribery to foreign governments. James Woolsey, a former director of the CIA, wrote a colourful article in the Wall Street Journal, entitled "Why America Spies on its Allies: because they bribe".6


6 Wall Street Journal, 22.3.2000

"When we have caught you at it ... we haven't said a word to the US companies in the competition," Woolsey wrote. "Instead we go to the government you are bribing and tell its officials we don't take kindly to such corruption. They often respond by giving the most meritorious bid (sometimes American, sometimes not) part of the contract. " Woolsey went on to say that the US would also conduct economic espionage for the purpose of monitoring transfers of dual-use technologies, such as supercomputers and chemicals which could be used in the manufacture of weapons of mass destruction; and in order to keep an eye on countries subject to sanctions, such as Serbia and Iraq.

In these "sigint" games, the French are not necessarily innocent. Anglo-Saxon sources point out that France is busy collecting signals intelligence - even if its more limited resources, and in particular its lack, of signals intelligence satellites (the Echelon network uses 120 satellites of various sorts, according to the European Parliament report), mean that its networks are not so comprehensive or omnipresent. For example, France has 15 listening stations, in places such as French Guyana, Nouvelle Calédonie, La Réunion and Djibouti. According to a French intelligence newsletter7, the NSA has given a lot of technical assistance to the DGSE in its efforts to develop a listening network, and senior officials of the two agencies meet every six months. This may explain why the French government was rather silent after the European Parliament published its report.


7 Le Monde du Reseignement, 16.3.2000

"We assume the French are listening to us when we use open phones," says a highly-placed American defence official. "If their listening systems are not so high-tech as ours, they'll certainly be trying to develop ones that are. " An American intelligence officer adds: "What is in the open air is fair game. But they [the French] shouldn't worry. The increasing use of both encryption and fibre-optic cables [which cannot easily be tapped] will solve their problem. "

British officials add that Echelon is not nearly so powerful as libertarian journalists, French officials or Euro-MPs claim, and that it is administratively and technologically impossible to keep tabs on the huge explosion of communications traffic. They stress that the 1985 Interception and Communications Act allows interception to be authorised only on grounds of national security, the prevention and detection of serious crime and safeguarding the nation's economic well-being (which it defines by means of examples, including the identification of warnings of threats to the supply of energy, commodities and raw materials on which the UK is particularly dependent; and better government understanding of events and trends which could have a serious effect on the UK economy). They point out that Echelon is hugely useful in combating terrorism and the proliferation of weapons of mass destruction.

This author's French friends have told him that this paper should contain much more about Echelon; his British friends have said that it should contain less. As far as this author is concerned, the precise capabilities of Echelon are less important that the fact that it remains a symbol of the continuing mistrust between, one the one hand, France and some other European countries, and on the other, the Anglo-Saxon nations.

These Franco-American spats on intelligence are undoubtedly one reason why the broader relationship between those two countries is currently so troubled. And so long as this problem persists, the Americans will not want the British to share the fruits of US-UK intelligence co-operation with the French.

Satellite battles

The EU's plans to develop the capacity to deploy military force, launched by Tony Blair and Jacques Chirac at St Malo in December 1998, have been described in detail elsewhere.8 The British and French governments agree that this initiative must involve not only new institutional arrangements, such as the merger of the WEU and the EU, but also enhanced military capabilities.


8 See the author's "European defence post-Kosovo", CER, June 1999.

The French have always emphasised that the EU must be capable of running "autonomous" military missions, even when NATO does not offer support, and that autonomous missions require an EU intelligence capability. The British accept the logic of the French position, without much enthusiasm. So the St Malo declaration said that when NATO as a whole was not engaged, "the Union must be given appropriate structures and a capacity for analysis of situations, sources of intelligence and a capability for relevant strategic planning, without unnecessary duplication."

There is, nevertheless, a divergence of thinking between the British and the French on how to interpret those words. The French believe that Europe should develop its own network of military intelligence satellites. The British say that "sources" really means access to intelligence, which could come from the US or from intra-European sharing. They point to the US administration's publicly-declared commitment that it will be involved in any major security crisis in Europe. That means, as far as the British are concerned, that the US will provide intelligence when Europe really needs it; and that the huge cost of building European satellites, just for those minor operations when the US is absent, is not worth it. Of course, it is easy for the British to take that line when they enjoy privileged access to data from US satellites.

The French believe that Europe will not be capable of having an independent foreign policy so long as it is dependent on the US for satellite intelligence. They argue that Europe cannot always rely on the Americans to provide data from their own spy satellites - because the American satellites may be busy, dealing with crises in other parts of the world; because some of them may be faulty; or because the US may pass on low-grade or misleading intelligence.

The French like to cite an incident in September 1996. President launched a salvo of cruise missiles against Iraq, in retaliation for Clinton Saddam Hussein allegedly moving a division of the Republican guard into Iraq's Kurdish area. France claimed that imagery from its Helios 1A satellite showed the troop movements to be insubstantial. It therefore refused to support the American strikes. One American intelligence officer responds thus: "The French are wrong on that, there were more tanks than Helios 1A showed: the frame of its images is not wide enough."

Some British officials assume that the French obsession with satellites is driven, in part, by industrial policy. "It is all about getting the Germans and the other Europeans to subsidise French aerospace companies", says one. Other British officials accept that, in an ideal world, it would be nice for Europe to have its own satellites. But they argue that, given the pressure on defence budgets everywhere, there are many other more urgent priorities - such as transport planes, battlefield communications equipment and friend-or-foe identification systems.

The British are also dismissive of the performance of France's two Helios 1 satellites, pointing out that their putative one-metre resolution is no better than what is available from commercial satellites. America's military satellites are much more powerful. "If the EU tried to replicate what we get from the US or what is available to the EU via NATO, it would be very expensive and of lower quality, " says a British official. The British pay about £1 million a year towards the running of the WEU satellite centre, but complain that during the 1999 Kosovo conflict its output was slow in coming and of poor quality; the French retort that in cloudy weather - such as that which hung over Kosovo for much of that conflict - even the most powerful optical satellites are of little help. One German general shares the scepticism of the British, claiming that "the staff at Torrejón admit that the images from the best commercial satellites are better than those from Helios 1, and they cost only $100 a frame.

The French maintain that Helios 2, to be launched within a few years, will be a more powerful satellite. And they claim that the problem with the highest-quality commercial imagery is that the US retains "shutter control" - that is, the Secretaries of Defence and State have the legal right to prevent the commercial firms from selling imagery to specified countries or organisations. That argument is somewhat weakened by the fact that non-American commercial satellite companies are planning to sell one-metre resolution imagery within a few years.9


9 According to the Defense News of 3.4.2000, by 2003 there will be 14 commercial satellites capable of producing one-metre imagery. One will be Russian, one Indian and one South Korean.

Nevertheless, most French defence analysts regard the roughly 2 billion euro spent on Helios 1, and the further 2 billion being spent on Helios 2, as a good investment. "Helios gives us a level of comfort, so that the Americans cannot provide us with nothing, or with misleading information, " says an analyst who worked at a high level in the French defence ministry in the 1980s. "We have just enough capability to keep the US on the straight and narrow, " Because of this, he believes, the Americans are more forthcoming than they were ten or 15 years ago. "Then, the US either gave us no information, or just sent summaries, or sent an envoy. I remember an American coming to Paris and virtually saying to us, 'take a good look, children, because I cannot trust you to keep these photos'."

Many other European governments are sympathetic to the French position - but not so sympathetic that they will invest in Helios. In the mid- 1990s France persuaded Germany to support the Helios programme. France said that in return it would invest in Horus, a cloud-piercing radar satellite that Germany wanted to build (America's own millimetre wavelength synthetic aperture radar satellites are capable of identifying vehicles through cloud). President Clinton tried to scupper these plans by sending John Deutsch, a senior defence official, to Germany. Deutsch sought to persuade Chancellor Kohl to buy an off-the-shelf Lockheed spy satellite rather than to invest in Helios.

"I asked Deutsch if we would have unlimited access to what the satellite produced, " recalls a Kohl adviser. "He said no, they would pre-select the photos. Yet the French said we could have everything from Helios." So Kohl decided to support the French plans. A few years later, however, budgetary constraints forced Germany to pull out of Helios and to abandon plans for Horus.

This year German policy appears to be shifting once again. The German defence ministry has complained about the quality of satellite imagery that the US provided during the Kosovo conflict and in its aftermath. The complaint is that, on three occasions, the US provided inadequate or misleading material that was relevant to the well-being of German forces on the ground. For their part, the Americans griped after the Kosovo conflict that it was hard to share tactical intelligence with NATO partners who were not equipped with secure, digital, broadband communication systems. Be that as it may, at the time of writing (April 2000) there is talk in Berlin of the government investing in a German radar satellite called SAR/LUPE, that would use relatively cheap off-the-shelf technologies.

In Whitehall, some of the younger officials involved in security policy express a certain sympathy for the French viewpoint on satellites. They believe that autonomous EU actions would benefit from autonomous EU sources of intelligence. One says that if the EU embarked on a rescue mission in a place such as North Africa, it would be useful for the EU to have its own satellite photos. He believes that the St Malo declaration's reference to the EU being able to conduct "autonomous" missions has shifted the satellite argument a few notches in France's favour. "Some EU intelligence would be good, so that we do not depend solely on the US, as long as it is available to NATO and it can be discussed within NATO," he says. "Then you won't get NATO-EU rivalry Both organisations should have access to each others' streams of intelligence. But this official believes, like most others in Whitehall, that there are more urgent things to spend money on than satellites.

Many Britons point to the Falklands war as an example of how, when the chip are down, the Americans can be counted on to help Britain. It is certainly true that Britain would have found it much harder to re-conquer the Falklands without American intelligence. Yet it is often forgotten that for a month after the Argentine invasion of March 2 , 1982, while a diplomatic solution seemed possible, American help was limited.

During that month, according to a former British DIS officer, America did no pass on high-quality satellite photos. "The Americans said there were 'technical' problems with the satellites, during Al Haig's shuttle diplomacy, " recalls the officer. General Haig tried to negotiate a compromise package that would have allowed the Argentines to withdraw in a face-saving manner. "The US gave us the good photos only after Argentina rejected Haig's compromise. If Argentina had accepted that compromise, and Britain had rejected it, I doubt the Americans would have wanted to help us. In the final analysis they will always do what is good for the US - and therein lies the core of the UK's problem."

Whether Europe really needs its own satellites depends, to a large extent, on the US. If, through their words and their actions, the Americans can convince the Europeans that they will always provide them with the right photos at the right time, the British argument - that European satellites are an unnecessary expense - becomes compelling. But to judge from the German reaction to American conduct during the Kosovo war, the Americans have still got some work to do.

Does intelligence matter?

The special relationship runs in the blood of senior British officials, and they have no intention of doing anything that would jeopardise it. But many of them are also good Europeans, committed to implementing Tony Blair's scheme for a European defence capability. They see no incompatibility between these two fundamental principles of British foreign policy. "If the European Security and Defence Policy (ESDP) succeeded, I doubt that the US would want to damage the special relationship [by cutting back on intelligence links]," says a Foreign Office man renowned for both his staunch Atlanticism and his pro-European views. "Our European partners should see it as an asset that we have access to US intelligence. "

Maybe they should. But many of those partners regard that special relationship as yet another indication, alongside the opt outs from Euroland and the Schengen agreement, that Britain is less than fully committed to the European Union. Notwithstanding the insouciance of British officials, the special relationship does create potential problems for the CFSP. Intelligence is an influence on the formation of foreign policy. So the fact that EU governments receive different intelligence assessments must - other things being equal - make it harder for them to achieve common foreign policies.

There are some in Britain's intelligence community who worry that, in the long run, Britain will not be able to ride the Atlanticist and European horses at the same time. "If political integration proceeds in Europe, Britain will have t choose between the US and the EU, " says a retired agent. "The French will force us to choose." And according to one senior State Department official, a successful ESDP could lead some Americans to question the special relationship with Britain.

But these gentlemen should probably not fret too much, at least for now. For one thing, ESDP is still on the drawing board. The kind of problems they envisage are problems of success. The Europeans are still some way from achieving an effective and coherent CFSP; they have several years to think about how that could be reconciled with the Anglo-Saxons' special relationship.

For another thing, intelligence is only one of many factors that determine foreign policy. The crucial question is, how important a factor? In general intelligence matters most for policy towards countries with closed societies. Thus the US needed spy planes to discover that Cuba was installing Soviet missiles in 1962. And over the past 20 years, the number of repressive countries has greatly diminished.

Those who stress the importance of intelligence point to the example of Iran. Britain and America have often taken a tougher line on Iran than have France, Germany and Italy (although at the time of writing Britain is quite close to its European partners: the Europeans take a rosier view of the prospects of Iran's reformists than do the Americans). At various times in the 1990s, US and UK intelligence suspected Iran of links to international terrorism, of seeking to build nuclear weapons, of constructing ballistic missiles and of planning to disrupt the Middle East peace process.

According to several senior British officials, different perceptions in London and Bonn can be explained, in part, by the British and German governments not having access to the same intelligence. But if that is the case, why did not the British pass material to the Germans, in order to influence their views? This question elicits three responses from British officials. First, "we did pass on intelligence, but only every now and then, and obviously in order to try and change German policy. Because they were not getting a regular stream of intelligence from us, they took these specific reports less seriously than they might have. " Second, "some of our intelligence was simply too sensitive to pass on". And third, some of the material handed to Germany was then handed on to Iran, "an experience that put the damper on further exchanges."

One American official shares the view that the western countries' difference over Iran are explained, in part, by their different sources of intelligence. And he cites ballistic missiles as another area where policy divergences are partially explained by intelligence divergences. Western governments disagree, first, on the severity of the threat posed by ballistic missiles; and second, on the right policy response to that threat. America has generally taken the threat more seriously than its European allies. America has also favoured a more robust response to the threat; hence its current plans for a system of National Missile Defense (NMD), designed to shoot down ballistic missiles from rogue states.

In February 2000 officials in Washington DC told this author first, that Britain was becoming more supportive of the American desire for NMD (a point disputed by some in London); and second, that one reason for the alleged British policy shift was that it had seen the same intelligence on North Korea, Iran and Iraq.

However, other highly-placed officials in London and Washington play down the role of intelligence in the making foreign policy. "Britain and France have very similar assessments of what is happening in Iraq, but still have very different policies", says a British official. France has often taken a different line on Iraq because of its commercial interests, its genuinely different analysis of how best to handle Saddam Hussein and - perhaps - its desire to stand up to American dominance in some parts of the world.

National traditions have a big impact on foreign policy: given the same assessment of a particular country, the Anglo-Saxons will be inclined to take a tougher line than the Germans. That factor may count for as much as intelligence in explaining British-German differences on Iran. At one point, for example, German policy was heavily influenced by the fact that a German businessman was held prisoner in Iran.

On ballistic missiles, too, plenty of things other than intelligence determine policy. One British official contends that US intelligence reports have not had any major impact on the fact that, "for the past 40 years a minority of Americans has wanted to build a high-tech system of protection against ballistic missiles - whether the threat has been non-existent, moderate or serious.

Similarly, one American official contradicts the above-mentioned views of his compatriots, arguing that the recent (and tentative) British-American rapprochement on the desirability of NMD does not stem primarily from intelligence-sharing. He did not specify what had caused it, but the British defence establishment certainly feels inherently uncomfortable about opposing an American policy - especially when (as with NMD) America appears bent on pursuing that policy regardless of its allies' views - and is therefore eager to find reasons not to oppose it. The arrival of a new defence secretary, Geoff Hoon, has also influenced the evolution of British policy on NMD.

Having heard contrasting views among those who make foreign policy, this author reckons that intelligence is not often a crucial factor in its construction. Domestic party politics, economic interests, personal relations with other foreign leaders, shifts of public opinion and reports in the media often count for more. To a large degree it was the TV pictures of Albanian refugees - seen by people in every NATO country - that made it fairly easy for NATO governments to garner public support for their campaign of bombing Serbia. The Russian people saw no such pictures and thus favoured a very different foreign policy.

One of Helmut Kohl's former diplomatic advisers recalls that sometime intelligence was a help, "for example when [through a combination of reports from German, US and Israeli intelligence] we learned of Iran's missile programme." But he says that "during 10 years with the Chancellor, I do not recall that we made any decisions of life and death on account of intelligence reports. "

Sir Michael Alexander, a former British ambassador to NATO, would agree. In a speech given early in 2000 to the Diplomatic Academy in London, he said that in all the years that he was a private secretary to senior politicians such as Alec Douglas-Hume, Jim Callaghan and Margaret Thatcher, "I'm not sure I can remember any occasion where, in dealing with the fundamental objectives of government policy overseas, inputs from the intelligence community had a decisive impact on government thinking. In all three cases - but most obvious in the case of Mrs Thatcher - basic political instincts were far more important.

Building a European intelligence capability

And yet, even if Sir Michael is broadly right, the special relationship still pose problems for the EU's foreign and defence policies. For one thing, intelligence reports sometimes do impact policy, for example towards the so-called rogue states. For another, they have a major bearing on the success of EU military operations. Furthermore, intelligence links, or the lack of them, have psychological impact on relations among governments - nurturing or damaging the trust that politicians from different countries feel for each other. This year's furor over Echelon created bad feelings about the British, and not only in France. Without mutual trust among Europe's major powers, the CFSP will not get very far.

So Britain, its European partners and the US should recognise that there is a problem, rather than ignore it. In the interests of an effective European CFSP and a healthy transatlantic relationship, they need to search for ways of building up the EU's intelligence-sharing capability - but ways that also respect Britain's special relationship with the US. Most of the EU's intelligence capability will continue to be based on national sources. But the EU does need its own institutional arrangements for making use of the various national intelligence streams.

Improve the security arrangements within the EU's CFSP machinery.

Evidently, the governments of NATO and the EU should seek to share as much intelligence with each other as is compatible with their national security. Because the EU has a reputation for being a leaky organisation, and because some of its members are non-aligned, NATO is reluctant to pass intelligence to the EU. So the EU must ensure that the staff working on foreign and defence policy within the Council of Ministers secretariat, who report to High Representative Javier Solana, are situated within a special, secure building. There will need to be rigorous vetting procedures, stringent codes of conduct on access to documents and buildings, and harsh penalties for security breaches.

Encouragingly, in March 2000 Solana appointed a British brigadier with an intelligence background to head his military staff; the appointment implies that the EU's military staff will take the need for security very seriously. On visits to Solana and his team of advisers, this author has been impressed by their commitment to the creation of a culture of security. They are well aware that governments will not pass on high-quality intelligence unless they succeed. And they will find it hard to make a good job of guiding and co-ordinating EU foreign and defence policy unless they receive such intelligence.

When the EU embarks on a military operation that is supported by NATO (as would normally be the case), NATO's Deputy Supreme Allied Commander, who is always a European and who would normally be responsible for running such a mission, should pass relevant NATO intelligence assessments to the High Representative's military staff. These assessments should be made available to representatives of non-aligned EU members and to any non-NATO, non-EU country that contributes to the mission.

The EU's High Representative needs a powerful intelligence assessment capability.

Javier Solana already has a policy unit, consisting of some two dozen diplomats, who are specifically charged (in the 6th declaration attached to the Amsterdam treaty) with channelling confidential material from their own countries to the High Representative. The embryonic military staff will play a similar role for military intelligence. NATO has an Intelligence Board, which brings together the heads of each member's military intelligence for regular meetings. Solana's team should host similar meetings for the heads of military intelligence from the EU countries, to encourage the sharing of military intelligence among them. Solana should also establish a unit that focuses on analysing open sources of intelligence - for example, information that is available on websites - which are often more useful than secret sources.

In the long run the High Representative will probably need a more extensive capacity for analysing the intelligence that comes from the member-states (and whatever comes from the Torrejón satellite centre). There may be a case for establishing some kind of EU-level JIC, made up of senior intelligence figures from each member-state. One advantage of having senior figures meet together in Brussels is that it might encourage their governments to provide a higher grade of intelligence to Solana's team. And there might also be intangible benefits from the senior figures working with each other on a regular basis; Coreper, the committee of EU ambassadors, works effectively because its members know each other well, and it may be a good model.

This Euro-JIC would seek to co-ordinate the tasking of the various national agencies, so that they did not pursue divergent lines and priorities; but it would not have the power to order them to change their objectives. The committee would filter and analyse the various national assessments, in order to produce common assessments for the foreign ministers and the High Representative.

One of the difficulties in creating a Euro-JIC is that not every country has a single figure, such as the British official who chairs the Cabinet Office JIC, to represent it. Hopefully, the process of establishing this EU committee would encourage the French, Germans and others to get a move on with modernising their own intelligence structures.

In intelligence, as in other aspects of the CFSP, the reality is that large countries count for more than small countries. There may need to be an informal committee in which the intelligence representatives of the EU's large countries can meet together. However, given the reluctance of agencies to share the most sensitive information with multinational bodies - even within relatively small committees -individual agencies may wish to develop their own "bilateral" relationships with the High Representative and his chief aides.

The Americans should be associated with the EU's intelligence unit.

America's intelligence capabilities are significantly greater than the combined capabilities of the Europeans. The EU therefore has a strong interest in getting access to as much US intelligence as possible. There is a risk that autonomous EU military missions could be hampered by a lack of access to US assessments.

The Americans, of course, will be reluctant to share. But there is an argument which might persuade them to share more than they do today. The Americans understand that Europe is developing, albeit at a snail's pace, a more coherent common foreign policy. They do not want the CFSP to evolve in directions that oppose US interests. The sharing of intelligence would, other things being equal encourage US and EU policies to converge rather than diverge. The US should therefore share as much as it can without endangering its national security.

The best way of getting the Americans to share information is to make them feel included and involved in Europe's foreign policy machinery. They need to lea that the whole CFSP is not a plot to undermine NATO and/or American global influence. So they should attend some meetings of the Euro-JIC, just as they attend some meetings of the UK JIC. There will inevitably be times when EU and American foreign policies diverge; but it is better that such divergences should occur when both sides have a good understanding of the others' position. In return for US representation in the CFSP machinery, the EU should insist on having equivalent representation in the National Security Council.

The NATO countries should agree on a code of conduct for economic espionage.

If the Americans could be convinced that France was trustworthy, they would be much happier about sharing intelligence with continental European allies and with the EU as an organisation. The Americans and the French should agree on a set of non-binding rules on what is allowed, and what is not allowed, in the field of industrial espionage. The reality is that they will continue to spy on each other. But if they agreed to follow a set of mutually-agreed rules, these activities would be much less damaging. Such rules would obviously be easier to accept if they applied to the whole of NATO, rather than just to France and America.

Most developed countries have signed up to an OECD convention on corruption, that is intended, among other things, to limit the use of bribery to win commercial contracts. It is probable that many NATO countries will spy on each other to ensure that their competitors do not break the rules. This may be the best way of ensuring that the convention is respected. Thus such "defensive industrial spying" should probably be authorised by the code of conduct.

The EU countries should concert their efforts to develop capabilities in unmanned aerial vehicles.

Satellites tend to be useful for big operations and big threats - when NATO, rather than the EU, is likely to be the principal organisation involved. Europe should place a higher emphasis on developing autonomous capabilities in the less expensive field of aerial reconnaissance. During the Kosovo conflict, the prevalence of thick cloud meant that some unmanned aerial vehicles (UAVs) took more useful photos than did spy satellites. At the moment several European countries are spending money on developing their own, independent UAV programmes. If they pooled their efforts they could create an autonomous EU capability that would provide battlefield intelligence to its task forces, to its governments and also to NATO. Germany has proven expertise in this area and is well-placed to take a lead.

The EU should make greater use of commercial satellite imagery.

The EU's satellite centre at Torrejón should emphasise the analysis of photos from commercial, rather than bespoke military satellites. Torrejón should become a centre of excellence for the interpretation of satellite photos; the skill and expertise required for such analysis is as important as the quality of the photos themselves. In order to deal with the - extremely hypothetical - problem that the Americans might cut off European access to commercial imagery, the EU should insist on the US making written commitments never to do so. The Europeans should also urge the Americans to provide more material from their military intelligence satellites. Those European countries which can find the money to build their own spy satellites should be encouraged to do so, especially if the US is not forthcoming with imagery from its satellites - but only if they do not skimp on investments in other, more urgent military capabilities.

*                         *                         *                         *

These suggestions, if implemented, would give the EU its own intelligence capability - albeit one based mainly on national sources - and thus strengthen European foreign and defence policy. There is no reason why a more intense intra-European co-operation on intelligence need threaten the special UK-US relationship. The Europeans would normally be sharing assessments, rather than raw data. Evidently, the UK would not share those reports which derived from the special relationship and which the US asked it not to share.

Nor is it the case that an EU intelligence capability would be harmful to EU-US relations in general. In fact several mainstream US defence analysts argue that such a capability would have a positive impact on transatlantic relations. One reason they take this line is that they - like many people in America - are frustrated by the Europeans' apparent lack of interest in the proliferation of nuclear, chemical and biological weapons.

Thus Kori Shake, a professor at the Institute for National Strategic Studies In Washington DC, argues that the wide chasm in intelligence capabilities between the US and its European allies has resulted in them mistrusting the US. "Europeans do not believe US assessments of the threat, whether about proliferation [of weapons of mass destruction] or evidence of Sudanese complicity in attacks on US embassies in the summer of 1998: better intelligence would allow America's European allies to replicate and validate the factual basis of US assessments.10 (According to US intelligence sources, the material that linked the Khartoum pharmaceuticals factory destroyed by cruise missiles in 1998 to Bin Laden was too sensitive to pass on.) Shake thinks that while disagreements over interpretation would certainly occur, as they do between the various US agencies, better European intelligence will - other things being equal - help to bring about a convergence of European and American views on security threats.


10 Kori Shake, forthcoming article, Journal of Strategic Studies, Washington DC.

Shake urges the US to encourage not only better national intelligence capabilities in Europe, but also an enhanced role for the EU itself in collecting and assessing intelligence. She would even welcome European spy satellites. Although US government policy is opposed to the EU duplicating what NATO does, Shake thinks that such duplication is "a risk worth accepting, because the current distribution of power in NATO on intelligence issues is not conducive to cooperative policies or beneficial to either European or American interests. The US should be bold enough to want allies out of strength, even at the risk of duplicating valued and valuable assets, rather than retaining allies because of their weakness."

She is right. A Europe that has a stronger CFSP, including an intelligence component, will be a more useful partner to the US. As for Britain, it should be able to have its cake and eat it. To be sure, its special links to the US will often cause embarrassment, and may make some continentals trust it less. But the British should be capable of turning the special situation to their advantage. They should explain to the Americans why an effective CFSP is in US interests, and persuade them to share more intelligence with European partners. Equally, the British should work with other Europeans to ensure that the CFSP does not evolve in anti-American directions.

That objective will be easier to fulfil if the US and France can improve the troubled state of their relationship. The construction of a successful ESDP could, in itself, help to soothe Franco-American tensions. For the ESDP Will never work well until the French believe that that the Americans are prepared to live with, and even welcome, a more coherent European security policy; and until the Americans believe that France is not trying to undermine NATO. A greater degree of confidence between these two countries would do a great deal to encourage the sharing of intelligence among NATO's principal powers. The more that such sharing occurs, the easier it will be for Britain to enjoy its special relationship with the US while at the same time playing a leading role in the construction of European foreign and defence policy.

Charles Grant April 2001