For Queen and Company: The Role of Intelligence in the UK's Arms Trade



This article analyses the role that the UK intelligence services (particularly Secret Intelligence Service [SIS or MI6], the Defence Intelligence Staff [DIS], Government Communication Headquarters [GCHQ] and associated agencies) play in the legal UK arms trade. The article shows that intelligence has been used in support of British-based private commercial businesses, and occasionally in providing intelligence on the negotiating positions of rival manufacturers. This raises important questions about the role of the state in the private sphere, particularly the use of a large number of government assets in support of private interests and the elision of British government interests with those of a section of the manufacturing industry. This article also challenges existing conceptions of how the UK's intelligence agencies operate and relate to their customers. Conventional typologies of UK intelligence have emphasised the importance of the ‘central machinery’, highlighting the Joint Intelligence Committee as the focal point of intelligence tasking and analysis in the UK. However, in this case the intelligence support provided to the sale of military equipment suggests a range of parallel practices that are much more decentralised and often informal. This research therefore suggests that our conception of the UK intelligence architecture requires some reassessment.

This article analyses the role that the UK intelligence services (particularly Secret Intelligence Service [SIS or MI6], the Defence Intelligence Staff [DIS], Government Communication Headquarters [GCHQ] and associated agencies) play in the legal UK arms trade. Aside from research by Mark Phythian and Davina Miller, the role of intelligence in the UK arms trade has been left relatively unexplored (Miller, 1996; Phythian, 2001). In contrast to previous work in this area, this article seeks to challenge existing conceptions of how the UK's intelligence agencies operate and relate to their customers. Conventional typologies of UK intelligence have emphasised the importance of the ‘central machinery’, highlighting the Joint Intelligence Committee (JIC) as the focal point of intelligence tasking and analysis in the UK (Cabinet Office, 2001, pp. 15–20; 2005, pp. 17–25; Cradock, 2002). However, the case of intelligence support to arms sales suggests a range of parallel practices that are much more decentralised and often informal. This research argues that our conception of the UK intelligence architecture stands in need of some reassessment, since here the flow and dissemination of information and dynamics within the intelligence cycle are quite different to what are perceived as ‘normal’ UK intelligence processes. The evidence suggests that intelligence relating to the arms trade is heavily ‘stovepiped’ and circumvents the normal process of intelligence, being fed directly and relatively ‘raw’ to selected consumers, including officials, arms manufacturers and politicians within the governing party. ‘Stovepiping’ is an intelligence term for the process by which a request for data or a piece of intelligence that would normally be passed through successive centralised verifications or bureaucratic levels is instead taken directly to the relevant officer or consumer.

Moreover, this article also advances a wider proposition. The peculiar culture of the UK intelligence system is notoriously ‘analysis-lite’, as the Butler Report (Butler, 2004) which reviewed intelligence on Iraq and other states’ weapons of mass destruction revealed, and indeed lacks a professional analytical service. This lends itself to the stovepiping of raw intelligence to specialist consumers and therefore suggests that intelligence scholars need to adopt a pluralist understanding of the intelligence machinery. Anecdotal evidence suggests that, as with other niche subjects, such as intelligence in Northern Ireland, there are a great number of informal and local intelligence systems that remain tangential to the central machinery of intelligence in the UK. The local intelligence systems often have unique cycles and are relatively self-contained. The major exception to this format is where the government labels a particular arms transfer as monetarily or strategically significant. In these circumstances, discussed in detail below, separate requirements are established by the central machinery to support the smooth transition of the sale of military or dual-use equipment.

This research also illuminates the extent to which intelligence has helped facilitate the sale of UK-developed and manufactured arms to third countries, often via commercial agents, and has helped in protecting those markets from the manufacturers of rival exporting countries through covert actions. These activities have mainly been executed by the Defence Export Services Organisation (DESO) and Defence Attachés (DAs), who use intelligence to identify potential export markets and who also use their respective positions within the UK Ministry of Defence (MoD) to shape the ministry's procurement practices for the benefit of the export trade. Meanwhile, the available evidence suggests that a small and informal effort is made to monitor the end-use of legitimately transferred equipment which has led to some curtailment of trading. The extent to which intelligence and ‘ordinary’ government converge to assist the UK arms trade in securing commercial success abroad lends weight to the general argument put forward by both Peter Gill and Philip Davies that the use of intelligence has become a mainstream function of British government (Davies, 2004; Gill, 1994). Certainly, after 1989, the end of the Cold War allowed spare capacity to be focused upon a wider range of issues, and the lower security concerns associated with this sort of intelligence allowed for greater distribution and the development of a broader customer base in Whitehall.

The article is organised into sections that deliberately mirror the structure of a transfer of military or dual-use equipment. This format is employed to illustrate where and how the use of intelligence impacts on arms transfers. After discussing the methodological approach employed here and offering a working definition of intelligence, the pre-licensing (F680) stage of arms transfers is examined in some detail as it is markedly under-explored in the academic literature. Following this, the full licensing and sales stage is explored, together with the full role of the DESO and Defence Attachés in facilitating the transfer of materials. In the final section the role of intelligence in end-user monitoring is discussed.

There are three main approaches to the study of intelligence: the historical, political and critical studies schools. The historical approach seeks to explain the relationship between the agencies and government policy-making. The political science approach seeks to explain at which bureaucratic level analysis was made, why this was the case and similarly why a particular output was produced (Scott and Jackson, 2004, p. 142). The critical studies approach focuses on intelligence as a tool of oppression. Each of these approaches has its own assumptions and readings of the political process and culture that provide context for intelligence activity. Contemporary intelligence studies have developed predominantly within historical disciplinary approaches and have therefore focused on archival work rather than elite interviews, which are the mainstay of this research.

A reliance on elite interviews raises some notable methodological problems, which are more acute in the sub-field of intelligence studies, although only marginally more difficult than securing elite interviews within the mainstream of Whitehall (Dorril, 2000; Dover, 2005). A considerable sector of intelligence studies research relates to issues and time periods that afford a profitable exploration of document records held in archives like the National Archives, for example. These documentary records, as with interview evidence, should however be viewed with some scepticism by the researcher; some government papers kept at the National Archives have been subject to ‘sanitisation’ to remove evidence of UK intelligence involvement in the cases they discuss (Aldrich, 2001, p. 6; Davies, 2001, pp. 73–4). Supplementary documentary evidence that might be shown to the researcher, such as Cabinet Committee or Departmental minutes, does not often reveal individual positions or areas of tension between politicians and officials and therefore misses some of the rich tapestry of bureaucratic politics. The paucity of the official record is particularly clear in the case of JIC papers, which explicitly aim to absorb individual positions, while constructing a consensus that can be employed in government policy formulation (Herman, 2001; interview 04IS). It is therefore necessary, as other scholars have done, to use elite interview data as a means to supplement sparse and deliberately obtuse official documentary evidence (Hennessy, 2002; Scott and Jackson, 2004, p. 153; Smith, 1996).

Constructing a usable set of intelligence sources is also problematic – the availability of research subjects is highly restricted and developing new avenues of inquiry often dependent on a ‘snowballing’ strategy – that is, each respondent recommends one or two additional points of contact (Davies, 2001, p. 76; Scott and Jackson, 2004, p. 140). All 26 interviewees in this research have been given a random serial number and the comments attributed to them are referenced accordingly (following Davies, 2001, p. 77). Interviewees were contacted more than once, between March and November 2005, to ensure the accuracy of the comments attributed to them. In limited circumstances alternative primary sources were not available, but wherever possible interview evidence has been triangulated with additional primary and documentary evidence to provide corroborated accounts. In particular, this was done by triangulating interview evidence against the Butler (2004) and Scott (1996) Reports, which represent important contributions to our understanding of the government's role in the arms trade and the machinery and operation of British intelligence services. The Scott Report (officially titled ‘The Inquiry into Exports of Defence Equipment and Dual Use Goods to Iraq’) was tasked with establishing whether all relevant parts of the British government had followed the agreed government policy on exports to Iraq and also to report on related prosecution decisions, mainly those in the Matrix-Churchill trial.

Understanding Intelligence

Popular conceptions of intelligence are dominated by images of secret sources, secret methods and glamorous agents. Academic conceptions of intelligence incorporate a far broader spectrum of activities and information. Sherman Kent in his classic work defined intelligence as a type of ‘knowledge’ (Kent, 1965). Andrew Rathmell extends this definition beyond an activity that is designed to serve a state in its foreign and military spheres to a more general production of knowledge, ‘not just any knowledge, but targeted, actionable and predictive knowledge for specific consumers’ which also includes secret sources (Rathmell, 2002, p. 88). Distinguishing the ordinary production of information by governments from ‘intelligence’ might seem a matter of semantics. Michael Herman sought to dodge this pitfall by proposing that only those departments labelled ‘intelligence’ can be seen to produce ‘government intelligence’ (Herman, 1998, pp. 1–2). The difficulty with Herman's definition is that it underestimates the extent to which information from open sources feeds into the intelligence services’ own assessment processes and more broadly into domestic policy formulation. Abram Shulsky's suggestion that intelligence is more a question of analysis and assessment seems therefore to be a sensible one (Shulsky, 1993). The arms trade case study provides further evidence of a shift away from secret intelligence to information and assessments being made for commercial and trade reasons using both secret and open sources. In the case under consideration here, this information is often open source, but highly specialised, and is piped into the principal conduit of DESO.

The operationalisation of knowledge production includes many activities which Kent characterised as knowledge (the research process), organisation (the institution) and activity (the process of estimating) (Kent, 1965; Scott and Jackson, 2004, p. 141). Douglas Dearth reworked this typology into ‘Process, Profession and Politics’ (Dearth and Goodden, 1995). Process encapsulates the intelligence cycle and analysis for the institutional customers. Profession includes organisational structures, environment and ways of working – a branch of intelligence studies that Davies and Gill have analysed to provide public administration explanations for intelligence (Davies, 2000; Gill, 1994). The politics element of Dearth's typology alludes to the influence that agency (individual, institutional or overtly political) brings to the process of formulating an intelligence assessment (Dearth and Goodden, 1995, pp. 97–115; Scott, 2004, p. 331).

Defining intelligence agencies as the providers of targeted information is a somewhat dry and all-encompassing definition. David Kahn has described intelligence as serving the government and the military and as a means to ‘optimise one's resources’ (Kahn, 2001, p. 85). Herman sees intelligence as a form of state power in its own right, while John Ferris argues that intelligence is a means by which to guide the use of state power (Ferris, 2003, p. 308; Herman, 1996). Both of these readings place the emphasis on state control of the intelligence services and thus the political use of secret services is central to any debate. The arms trade case study not only impacts on foreign policy but also on economic policy and the role of covert and clandestine operations by the intelligence services – both of which are under-explored in the available literatures because of problems with access to information (Anderson, 1998/9; Godson, 1995, p. xii; Riordan, 2002, p. 1). A focus on covert activity and economic diplomacy would reinforce Ferris's view of intelligence as a means by which to execute government policy, although its inclusion in the range of activities covered by intelligence is contested (Ferris, 2003, p. 308). There is no public definition of covert action from the British government beyond the Intelligence Services Act (1994) which mandates the UK's Secret Intelligence Service or ‘MI6’ to engage in ‘other tasks’ beyond espionage without providing precise details – chapter 13, section 7.1 gives two classes of function that might shed light on this role: ‘(1) to obtain and provide information relating to the actions or intentions of persons outside the British Islands; and (2) to perform other tasks relating to the actions or intentions of such persons’ which includes ‘disruptive action’ (Intelligence Services Act, 1994). Shulsky's definition is when one government pursues its foreign policy objectives by conducting some secret activity to influence the behaviour of a foreign government or events in a foreign country in a way that is not attributable to the sponsoring state (Riordan, 2002, p. 2; Shulsky, 1993, p. 83). In a trade worth many billions of pounds a year and in an anarchic international arms trade system the benefits to be accrued from using intelligence assets in support of arms sales far outweigh the costs of being caught in the first instance but these costs increase once the initial sale is made and a bilateral trade pattern exists (Riordan, 2002, pp. 6–8).

UK intelligence services operate in a market system and respond to ‘intelligence requirements’– that is, information requested by a particular department to fill a gap in their knowledge rather than, as in the US system, information being gathered to generate a particular change in policy. The UK Joint Intelligence Committee (JIC) serves as the key coordinating body in tasking and sequencing the intelligence services. The Cabinet Office states that the JIC is ‘responsible for providing Ministers and senior officials with co-ordinated inter-departmental intelligence assessments … in the fields of security, defence and foreign affairs’ (Cabinet Office, 2005, p. 20). The Chairman of the JIC is a civil servant of standing with experience of dealing with cabinet ministers and also acts as the Head of the Intelligence and Security Secretariat (ISS) in the Cabinet Office (Cabinet Office, 2005, p. 21). Other members of the JIC include senior Foreign and Commonwealth Office, Ministry of Defence, Home Office, Department of Trade and Industry, Treasury and Cabinet Office officials as well as the heads of the three intelligence agencies (Cabinet Office, 2005, p. 20). On an annual basis the JIC constructs the general direction of requirements for the year (interview 04IS; interview 03IS). This system of requesting or ‘tasking’ intelligence puts the emphasis within the intelligence cycle on the consumer – that is the person or agency requiring the information. The JIC is supported by the ‘Assessments Staff’ who are a multidisciplinary group of experts who draft assessments of situations, capabilities and intentions of targets which are then subject to scrutiny from the ‘Current Intelligence Groups’ (Cabinet Office, 2005, p. 21). As will be demonstrated later this process is less of a cycle and more of an iterative process between the consumer and the producing agency (Davies, 2004, p. 17; Dorril, 2000).

The British government's use of intelligence in the arms trade, where it occurs, has clear impacts upon foreign policy, defence policy and also major commercial operations (interview 03IS; interview 23IS). Despite this overtly political dimension to intelligence the depoliticised central organisation for the evaluation of intelligence in the UK is the Joint Intelligence Committee, which exists to provide policy-makers with a consensus opinion about ‘required’ intelligence and also to illuminate any uncertainties around a particular assessment. Generally speaking, the JIC has been regarded as an objective provider of assessment and intelligence products. However, periodic reforms in terms of its structure and leadership, most recently following the Butler Inquiry, suggest that there have been problems (Butler, 2004, pp. 146–7). A substantial body of academic literature has already explored one of the main dilemmas here: the close proximity between intelligence officials and politicians is not conducive to objectivity, yet isolating them may result in intelligence that is irrelevant to the concerns of decision-makers (Coates, 2004; Handel, 1987, pp. 187–220; Phythian, 2005, pp. 660–1; Scott and Jackson, 2004, p. 150). Sir Roderic Braithwaite, a former Chairman of the JIC, has said that officers of the JIC ‘should be divorced from the pressures of both intelligence gathering and of operational decision-making’ (Braithwaite, 2003). More recently, the Butler Report has called for an ‘end of career’ chair for the JIC who is immune to political pressure (Butler, 2004, p. 159).

Pre-licensing of Exports

The arms trade arouses great passions among campaigners who argue the moral and economic cases against such trade, and industry insiders who advocate the political, strategic and counter-economic case for continued export activities. The industry commentator Joe Roeber, whose book on corruption in the arms trade was too contentious to go into print, suggests that the trade is the ‘most corrupt legal sector of the economy’ (Roeber, 2005, p. 54). Roeber also contends that in the late 1990s the Central Intelligence Agency (CIA) estimated that the arms trade accounted for 40–45 per cent of the total corruption in world trade despite only amounting to less than half a per cent of the total trade (Roeber, 2005). However, British arms exports have continued to prove highly lucrative for the UK Treasury. In 2004 arms exports reached a five-year trading high, according to figures released by DESO, exports totalling US$8.2 billion, up $400 million from 2003. In terms of market share this translates to a 20 per cent share of the world market in 2004, putting the UK second behind the United States (DESO, 2006; Isenberg, 2005).

A pro-trade preference is a strong element of the UK government's external relations policy generally. Within the arms trade this manifests itself more strongly through DESO whose raison d’être is to support the transfer of military and dual-use equipment (Norton-Taylor et al., 1996, p. 15). One interviewee characterised this preference as being a free-market philosophy under which the default is to sell as much material as possible but within a framework that exerts control over exports within given criteria (interview 23IS). The case of Saudi Arabia provides strong evidence for this claim through the long-established and important links to British arms manufacturers. Evidence of the strength of this trade link is shown by the answer to the Freedom of Information Act query placed by The Guardian newspaper which revealed that 161 of the 600 employees at DESO were assigned to the ‘Saudi Armed Forces Project’, which facilitates arms sales to Saudi Arabia (Leigh and Evans, 2005). This close relationship highlights the extent to which the British government's interests have elided with those of UK arms manufacturers. The use of intelligence, in this context, falls within Shulsky's concept of action that supports government activity, as outlined above.

The effect of a pro-sale operating principle is marked for UK embassy officials and for DAs attached to those embassies (Phythian, 2001). DAs have a quasi-intelligence role, but sit outside the formal structures of the intelligence agencies (Andrew, 1997, pp. 391–2). They are not expected to get involved in any covert intelligence activities but are expected, in the course of their duties, to collect openly available military information and this fits within Kahn's conception of intelligence serving government or military ends (Steele, 2004, p. 282). They spend a great deal of time liaising with the host armed forces and arms companies and are therefore an invaluable source of military information to those wishing to access their knowledge (Clarke, 2000, p. 730). The success of an Ambassador's period of tenure is partly judged upon whether they have assisted in securing a significant quantity of export trade, including arms sales, for UK companies (interview 24IS). A British company for these purposes is any company which brings or maintains employment in the UK (interview 03IS; interview 23IS). Former British Ambassador to the US, Christopher Meyer, illustrates the reciprocity of such relationships having been invited to join the board of UK arms manufacturer GKN (Barnett, 2005; Meyer, 2005). Moreover, for British diplomats seeking to secure rapid entry to the higher echelons of the diplomatic community a successful spell in a trade division of a UK embassy is essential, demonstrating the importance of trade to the overall British foreign policy effort (interview 24IS; interview 28IS). In these circumstances the stovepiped use of government intelligence, information and assets in support of trade is entirely understandable. This is especially so in the light of repeated reviews of the structure and costs of the diplomatic service which have also led to an increasing emphasis on export assistance to UK companies.

UK embassies support and facilitate arms sales. Embassies act as an essential marketing tool for the arms companies and assist in resolving contractual glitches. DAs are the arms companies’ principal point of contact in the country they hope to sell to; the DA reports directly to the Ambassador in weekly (and daily – depending on the progress of the sale) review meetings (interview 24IS). Between 25 and 40 per cent of a DA's time is spent on facilitating arms sales – including being directly tasked by DESO (something which is unique to the UK system) – this suggests arms sales are a significant part of the DA's role. The proximity of the manufacturer, MoD, DESO and DAs is partially illuminated through the MoD Defence Attaché committee that decides on which countries should receive UK DAs. This committee has DESO representation and one of the explicit criteria for a DA being sent to a country is the possibility of creating or further opening up a market for UK arms exports (interview 24IS; interview 36IS). DESO's intervention fits within Ferris's notion of intelligence as a means to guide state power; in this instance through helping to bring a particular country under the soft-intelligence scrutiny of DAs, guiding this state power on to very localised issues, often using information that has not been exposed to the centralised intelligence machinery, while also ensuring that MI6, GCHQ and DIS focus on the accumulation and analysis of this information to assist in granting or refusing licences.

Product Design and Development

The first stage in any arms transfer is the design and development of the product but latterly it might also be the sale of old stock. Finding suitable products for the markets open to UK arms manufacturers is a key challenge in maintaining healthy order books. Accordingly, privileged knowledge of the requirements of external militaries is invaluable information to arms companies and is available from government sources, including DAs. The available evidence suggests that no UK government agency, including the intelligence services, routinely plays a commercial marketing role for arms companies. However, they are occasionally used when the government is uncertain of its investment in a project and are therefore tasked to conduct intelligence-led assessments of a project's commercial viability. This happened notably in the case of the Merlin HM K1 helicopter produced by GKN Westland in the late 1980s and involved the central intelligence machinery (interview 03IS). The evidence points towards the process being mostly weighted in the opposite direction, with arms companies creating equipment and systems and then making a great effort through lobbying and back channels to influence government policy into requiring these products (Tigner, 2004b).

The terrorist attacks on New York, Madrid and London since 2001 have opened up new commercial possibilities for arms companies in the $100 billion ‘homeland security’ market (Shaheen, 2006). This includes technical support for data collection and analysis by overseas police and security services. Again, there is no evidence that the security services have been placing requirements on the arms companies to create products suitable for homeland security. Rather, the arms manufacturers have been operating in a highly permissive environment in which technical innovations are openly welcomed by the government (interview 14IS; interview 25IS). There is some evidence that a stovepiped ‘conversation’ between companies and intelligence officials is taking place outside the centralised intelligence machinery that is informing technical advances in this area, and which demonstrates the importance of informal networks in guiding these developments. The political salience of ‘homeland security’ has allowed companies to push cutting-edge technologies enthusiastically to national governments and through European Commission committees. Involvement with the intelligence services (MI5, Defence Science and Technology Laboratory [DSTL] and MI6) is limited to seeking advice on the applicability of certain pieces of equipment to the particular security problem (interview 26IS; interview 30IS). In this respect, the intelligence services are being accessed on an ad hoc basis for privileged information that helps shape government and business preferences, and that has already been gathered and analysed, rather than in placing fresh ‘requirements’ on them for information. Manufacturers are therefore going outside the formal centralised intelligence cycle and are making use of formally requisitioned information, through informal networks of contacts – the intelligence version of recycling.

‘F680’ and ‘Ratings’ Process

Defence equipment manufacturers almost invariably approach the government before commencing research or manufacturing work on a particular product to ensure that the start-up costs for either of these activities are not wasted through a straightforward refusal (interview 28IS). There are two formal processes whereby manufacturers make pre-licence inquiries. One is known as the ‘Ratings’ process and the other as the ‘F680’ process (interview 18IS; interview 36IS; Scott, 1996, C2.8–11, C2.29–31), and the assessments of the Defence Intelligence Staff (DIS) play a role in the replies companies receive (interview 18IS; interview 26IS; Scott, 1996, K7.2). This in itself is contested among the government officials that were interviewed; ranging from the view that there was no intelligence service involvement until products had actually been manufactured, to the view that DAs and DESO actively help in identifying products for markets (interview 36IS; interview 28IS; interview 27IS; interview 24IS). The balance of the available evidence is that the intelligence services play a marketing and sales facilitation role for arms manufacturers largely through recycled and privileged information, delivered in a decentralised and stovepiped fashion. In notable but rare cases defence ministers and the prime minister have commissioned bespoke intelligence to ascertain the commercial validity of a project or to assist UK firms to outmanoeuvre foreign competitors in negotiations, bringing the agencies’ role within the centralised intelligence machinery (interview 03IS; interview 04IS).

The F680 process is run by the Ministry of Defence, more precisely, the Directorate of Export Services Policy (DESP) within DESO (Scott, 1996, C2.29–31). The process obliges ‘List X’ companies like BAE Systems, GKN and MBDA – who handle material with a classification of ‘confidential’ or above – to apply for a licence before they promote their products. Materials requiring listing are provided in the ‘Manual of Protective Security’, which is issued by the Cabinet Office and often contains sensitive information about commercial agents in the country, political figures and general assessments about the stability or political situation within a particular region (interview 04IS; interview 23IS). There is reciprocity of information between the government and the arms manufacturers with commercially sensitive information being passed to UK government officials under a duty of confidentiality, but outside a formal intelligence-sharing arrangement (interview 23IS; Scott, 1996, E10.30–3). Non-List X companies (who can handle ‘unclassified’ or ‘restricted’ material) are not obliged to seek F680 clearance for promotions, but may need to have an F680 clearance to pursue full licences from the Department of Trade and Industry (DTI) (DESO, 2005b). The process helps prevent unauthorised disclosure of protective market (classified) assets and it gives companies an indication of which markets may provide viable export opportunities for their products while also potentially speeding up the assessment of any eventual export licence application made through the DTI (DESO, 2005b). The process can, however, be used to license arms transfers that occur outside the UK. The F680 process involves the submission of a written application to the MoD using the F680 form. On receipt of the form, officials within DESP circulate the contents to ‘MoD Advisers’– which include the intelligence assessments of the DIS and DSTL (interview 36IS; interview 26IS; interview 18IS).

Since 1999 F680 forms have also been circulated to the Foreign and Commonwealth Office (FCO) and Department for International Development (DFID) for their assessments. The FCO circulates these forms to their Counter Proliferation, Country Officers and Human Rights Departments as well as MI6 (interview 36IS). DIS, DSTL and general FCO assessments can involve requests for new intelligence but the majority of information used in assessments has been collected under different requirements, showing a reactive approach to intelligence assessments at the pre-licensing stage and placing an emphasis on the quality of inter-agency information sharing and ad hoc networks of officials (interview 36IS; interview 05IS). At the full licensing stage the use of intelligence is far more proactive with a greater need for new information, but still often outside the normal, centralised ‘intelligence cycle’.

The departments involved in the F680 licensing process have lists of destinations they believe are problematic. The countries on their lists reflect the differing priorities and worries within the departments, and often replicate the stereotypes of particular departments (interview 36IS; interview 10IS). For example, the Treasury has concerns about whether a client government will pay, the DFID is concerned about the impact of a sale on development issues and future regional or national stability, the MoD on the implications for UK interests at home and abroad and the FCO on the implications for the stability of regions (interview 36IS; interview 13IS). The role of the intelligence services, in contrast to the DTI, MoD, FCO and DFID who implement the controls over the transfer of arms and dual-use technology, is geared towards facilitating sales (identifying markets, individuals with influence within purchasing networks, etc).

Contentious issues with intelligence produced as part of the F680 process theoretically go to the JIC for its consideration. A former senior official within the JIC said that during his tenure no issues relating to the arms trade had gone through the committee. The explanation given for this was that the pre-licensing process is deemed to be an MoD-led initiative and therefore the MoD resolves its intelligence conflicts in-house (interview 05IS). While this is somewhat surprising the JIC's involvement is also likely to be dependent on the type of arms deals in the pipeline at any given time, their economic and strategic significance and whether there is government support for these transfers (interview 28IS). The use of intelligence in the legitimate arms trade goes outside the centralised cycle of ‘requirements’ and ‘production’ and is ‘stovepiped’ into an iterative discussion between the intelligence agencies, the relevant government departments and the arms companies which hold a prominent insider role.

The DTI runs an independent process to F680 known as ‘Ratings’. This process decides whether a licence is required to export particular goods through the Export Control Organisation. Inquiries are made through the Export Control Organisation's Technical Assessment Unit, which comprises a team of engineers and scientists – all of whom have worked for arms manufacturers – to advise exporters on technical issues; a structural position that belies the proximity of the government to arms manufacturers (DTI 2005a; interview 23IS). This is a change from the situation outlined by Sir Richard Scott, which highlighted the lack of expertise within the technical units at the DTI (Scott, 1996, C2.8–11). The post-Scott arrangement is a powerful commercial tool and provides a partial explanation of why UK arms manufacturers are such successful exporters (interview 39IS).

Contract Negotiation and Sales

Once the manufacturers have received F680 clearance they are free to engage in marketing activities to sell their products. Arms manufacturers are able to call on the full machinery of the state to support these efforts – depending on conditions that will be outlined. This section breaks down the marketing and sales cycle and identifies the role of government actors such as embassy staff, DAs, MoD officials and intelligence officers within it to suggest that stovepiped intelligence is used by informal networks of officials to support the legal sale of military equipment.

First Contact: Manufacturers and Customers

DESO's role is to provide support for defence sales, which includes research and providing negotiating assistance for manufacturers (interview 24IS; interview 36IS; Scott, 1996, C2.22–3). DESO provides this support through its officials in London and in embassies globally as well as through the Defence Attaché system (Scott, 1996, C2.25). DAs are very well placed to provide ‘privileged’ information to arms manufacturers in an ad hoc and stovepiped manner as they are routinely invited to briefings, as well as in the course of their duties networking with senior officials in their host nation's military. DESO regularly tasks DAs with providing various kinds of logistical, political or knowledge-based assistance to manufacturers – which is in marked contrast to other governments. The DA's assistance feeds into every stage of the marketing and sales process – making the DA the British government's person on the ground in these sales.

Inquiry and Introduction

Unlike competitor nations, British arms manufacturers are able to contact DAs directly, establishing the ‘stovepipe’ link between the provider and consumer of intelligence and privileged information. DESO will provide, through its ‘Partners Network’, contact details of the relevant DA and embassy staff that can assist throughout the life of the negotiation (DESO, 2005a). The British government makes no charge for the DA's assistance in trying to secure contracts – this is in contrast to the charges made by embassies for their assistance to other industries (interview 24IS). At this stage the manufacturer is likely to be looking for ‘first indications market research’, which establishes whether there is any demand in the host government for such items and potential local agents. The company itself will conduct a full market survey using in-house specialists at a later date. Access to this type of privileged information through the DA is key to manufacturers establishing or improving their presence in a particular market. This information is not covertly gathered – it will come across the DA's desk as a matter of routine and through their liaison activities with the host military and defence industry representatives. It is therefore ‘privileged information’ which demarcates it as a particular type of information available to the arms manufacturers, disseminated in a decentralised way that is a result of the UK government's policy to facilitate sales. For arms manufacturers to provide or gather information for a future sales pitch does not require an F680 licence (DESO, 2005). In practice it is in the DA's discretion, as a gatekeeper within this process, whether they introduce the representatives of manufacturers to government and industry officials in the host country with or without an F680 licence – a sensitive decision given the thin line between marketing and provision of information within a structural framework of the DA being responsible to the Ambassador (success of posting partly dependent on trade figures) and DESO, whose role is to support arms sales (interview 14IS; interview 24IS).

Having received first indications marketing and been introduced to agents and procurement officials the manufacturer takes steps to provide them with a corporate presentation. Information on these officials and agents will have been collated locally by embassy officials and might also have been subject to general or centralised information trawls by MI6, DIS and GCHQ – depending on the character and positioning of the person in question (interview 05IS; interview 24IS; Scott, 1996, C2.26). These presentations are discreet and are held without publicity. The DA will nearly always be in attendance at these presentations, as a representative of the UK government, and will often be in full dress uniform (interview 24IS). This emphasises the UK government's backing of the product and also allows the DA to pass on convincing accounts of how the equipment has been successfully used by the UK's armed forces (interview 24IS; interview 18IS). Anecdotally, this is often a persuasive factor in any sales pitch – there is a badge of credibility carried by British officers because of the perceived quality of UK forces on active service and it demonstrates an elision of interests between the British government and arms manufacturers. Moreover, it provides a direct link through which British government support for a manufacturer can be made explicit to a foreign procurement official.

The DA further assists senior sales officials of the manufacturer in their visit to the local agents and government officials (interview 19IS). Particularly, the DA is able to act to resolve problems that emerge in the early stages of any negotiation. Again, this is through the use of privileged access to secure ‘privileged information’– the DA is able to contact the agent or official through their official position in the embassy and then to pass on this information with accompanying advice on how to resolve it to the manufacturer in a decentralised manner (interview 24IS; interview 19IS). All of these processes occur locally, informally and on an ad hoc basis, emphasising the importance of individual personalities on the impact of these efforts. This particular role goes beyond what a DA is officially tasked todo for an arms manufacturer, but lies within the government's ‘can-do’ attitude towards arms transfers.

Official Receptions

UK embassies often host receptions in their residences for arms manufacturers and their potential customers. While the manufacturers pay for the food, drink and service at the reception, the embassy is all but formally prohibited from refusing a request from a company to host one (interview 24IS). The only viable reason an embassy can refuse to host a reception is if the local agents are known to the embassy, through local knowledge or centrally collected intelligence, to be engaged in criminal activities or in activities that would have negative diplomatic repercussions (interview 03IS; interview 28IS). The primary motivation for hosting such an event is to give the manufacturers an overt ‘kitemark’ of British government support. Such events also serve an information-gathering purpose – in soft terms as a means by which to network locally and illuminate matrices of influence and business. Of course, such ‘soft’ methods do not preclude the use of central intelligence assets – such as GCHQ to intercept communications or with human intelligence to reveal negotiating positions within rival companies or the client government, although this occurs only in a few notable cases (interview 05IS; interview 27IS). The Scott Report made it clear that there is a ‘close relationship’ between MI6 and FCO officials on export licence issues, which fits into the patterns described by this article (Scott, 1996, C2.49). The reception is therefore an important forum in which manufacturers and the government secure contracts and improve their collective knowledge of industry and actors in the host country, but this information is often retained locally rather than being disseminated through central intelligence channels, emphasising the stovepiped intelligence arrangements for arms sales (interview 03IS).

If there are problems with a set of negotiations the DA and the embassy can help in two further ways. First, the Ambassador can write to the Defence Minister or relevant Defence Ministry official within the host country, stressing the positive elements in the deal – although in practice the DA drafts the correspondence (interview 24IS). This is a significant action from the Ambassador, as it provides a direct, government-sanctioned intervention in a commercial transaction. Second, the DA can arrange an official visit from a high-ranking official from DESO or a Junior Defence Minister. In high-profile instances such as negotiations with the Saudi government, the Secretary of State for Defence and even the prime minister will be involved in official visits, emphasising the proximity between the government and arms manufacturers. For visits of a DESO official or junior minister, DESO in London will take the lead role in organising the visit and the itinerary (interview 36IS). Because of DESO's close day-to-day relationship with the large UK-based arms manufacturers such as BAE Systems, they are able to provide high-level support and advice to both the manufacturers and the embassy staff that accompany official visitors. Circumstantial evidence of this close relationship can be seen with the current head of DESO, Alan Garwood, having previously been the head of sales at BAE Systems, and his predecessor, Charles Masefield, who worked for Vickers before running DESO to then go on to become Vice-Chair of BAE Systems after leaving DESO in 1998. This interchange of government and commercial posts effectively creates a state–private network, which would then help to support Ferris's assertion that intelligence assists government programmes.

Ultimately, it is the job of the DA to assist the manufacturer – in receipt of an F680 and then full licence – to sell arms or dual-use technologies to third countries. The DA executes this function through the use of privileged access and privileged information. Very little information the DA is tasked to collate has to be collected covertly, although some of the information used to support arms sales is collected and analysed by DIS, in particular (Scott, 1996, K7.6). However, without the DA's structural position and privileged access the material would not be easily available and might therefore need more advanced centralised and less transparent collection methods. The DA's principal role is to coordinate the local pipeline of information between the manufacturer and the customer. The official rationale for this role comes from the enterprise imperative of embassies which need to secure export deals for UK companies. The DA uses a mixture of intelligence-gathering methods to help secure this outcome and this mix is dependent on the importance of the contract. The overriding conclusion that can be drawn from the DA's role is that they engage in little covert activity in support of arms transfers but do engage in a substantial effort with recycled intelligence, privileged access, information and influence to support the UK arms export trade, demonstrating the stovepiped, localised and ad hoc nature of the intelligence effort.

The Exception: Strategically or Financially Important Deals

The exception to the general rule that intelligence assets used in arms transfers are recycled from other requirements, stovepiped, or are the sort of information that ‘falls into one's lap’, are deals which are strategically or financially very important (interview 05IS; interview 18IS). The Scott Report demonstrated the intelligence assets the UK government is prepared to use in support of a strategically sensitive and important transfer of arms (Scott, 1996, K7.1–3). The recent examples of high-level political support being afforded to BAE Systems’ successful attempts to sell the Saudi Arabian government Eurofighter aircraft in a deal worth about £40 billion, and the parallel attempts to sell military aircraft and small arms to India and Pakistan, respectively, during their nuclear stand-off in 2002, provide evidence of the full machinery of government supporting these transfers (Norton-Taylor and MacAskill, 2002).

Because the Saudi Arabian example is contemporaneous the evidence is largely circumstantial; however, through official and secondary source reports it is clear that BAE Systems’ marketing effort to the Saudi Arabian government has been supported at the very highest level by the prime minister and the Secretary of State for Defence, both of whom have made personal trips to Saudi in pursuit of this contract (Leigh and MacAskill, 2005). Previous British prime ministers including Ted Heath, James Callaghan, Margaret Thatcher and John Major, as well as members of the British royal family, have all intervened in large Saudi arms deals; the latter option is unavailable to French competitors due to their republican constitution. This can be seen clearly in the example of the ‘Al-Yamamah’ deal in which the Saudis bought a fleet of Tornado aircraft (Financial Times, 9 July 1988; interview 03IS). Michael Turner, BAE Systems’ Chief Executive, stated his recent aims clearly: ‘The objective is to get the Typhoon into Saudi Arabia. We’ve had 43 billion pounds from Al Yamamah over the last 20 years and there could be another 40 billion pounds’ (O'Connell, 2006). This deal has been complicated partly by reports that the Saudi Arabian government had agreed to buy 90 Rafale aircraft from French-based Dassault Aviation in April 2005, and also by Saudi demands that the Serious Fraud Office (SFO) investigation into BAE Systems be dropped as part of the deal – a deal with a private British company, not the British government. The SFO investigation has looked at the legality of $2 billion of commissions allegedly made to five agents in Saudi Arabia (Isenberg, 2005).

The Saudi case study has all the prima facie elements to make it a good example of where the centralised intelligence cycle applies, and not the stovepiped system of ad hoc and recycled intelligence information, applied to routine arms sales. This is an example where the British government has a clear interest in supporting a financially lucrative and strategically important transfer of arms and dual-use technologies. What is more, the Saudi market represents a key area in which the British government has historically worked to suppress the competition presented by French defence companies, partly because of the economic loss this would present, but also because of the number of human intelligence sources that can be placed in Saudi Arabia as a result (interview 03IS). A great deal of government and intelligence activity is consumed by Saudi Arabia – as previously stated 161 out of the 600 staff at DESO work on the Saudi desk and with the instability in the Middle East intelligence efforts are heavily focused on this area. The counter to this comes from Scott, who noted an absence of intelligence on Saudi Arabia in the 1980s, although the evidence base for this view is not clear (Scott, 1996, E6.1). It would be perverse and surprising therefore if the full machinery of the state, including centralised intelligence, were not to be utilised in support of transfers that give the UK a balance of payments surplus and political leverage over events in the Middle East.

Licensing Stage

Once a contract is signed between a manufacturer and their customer a full export licence is sought by the manufacturer. Eight criteria are used by the DTI to decide whether to grant a full licence to export military or dual-use equipment, with the information for the criteria being provided by manufacturers and triangulated against open and secret information provided by government departments and intelligence agencies. These politically but not legally binding criteria are: respect for the UK's international commitments and obligations (UN and EU sanctions); respect of human rights and fundamental freedoms in the country of final destination; not exacerbating tensions in the country of final destination; preservation of regional peace, security and stability; the national security of the UK, of territories whose external relations are the UK's responsibility and of allies, EU member states and other friendly countries; the behaviour of the buyer country with regard to the international community, concerning in particular its attitude to terrorism, the nature of its alliances and respect for international law; the existence of a risk that the equipment will be diverted within the buyer country or re-exported under undesirable conditions; and the compatibility of the arms exports with the technical and economic capacity of the recipient country (Foreign and Commonwealth Office, 2000).

All licence decisions have to be made with reference to the eight criteria (interview 12IS; interview 23IS; interview 36IS). The DTI, as the lead department with DFID, MoD and FCO as partner departments, uses a system known as the ‘Smart Front End’ to streamline licence applications (interview 10IS; interview 23IS). This bureaucratic triage allows straight refusals and acceptances to be made far more quickly than previously and therefore brings out the contentious cases for further investigation and discussion. There is, however, a check in place so that all decisions taken by the daily ‘Smart Front End Committee’ can be reviewed by senior officials (interview 23IS). Decisions by this committee are taken on the basis of unanimity; where consensus cannot be reached, discussions go to successively higher levels of officials (outside a formal committee structure) until they reach the political sphere with junior ministers, secretaries of state and eventually the prime minister who, like officials, are politically but not legally bound to make their judgements with reference to the eight criteria (interview 10IS; interview 14IS). FCO documents covering arms export criteria suggest that a number of sources are used by the relevant government committees to form a view, one of these being ‘intelligence and information from open sources’, an official acknowledgement that central intelligence assets play a role in the granting of licences (Foreign and Commonwealth Office, 2000). How intelligence impacts on this process is dependent on the case. If, for example, DFID makes an objection on the grounds of regional instability, their information comes from their officials on the ground, non-governmental organisations (NGOs) and MI6 (interview 10IS). An MoD refusal, on the other hand, will usually be made on technical grounds and thus be driven by DIS or DSTL assessments (interview 18IS; interview 30IS; Butler, 2004, p. 110). Importantly, the intelligence input into granting licences comes from the central intelligence machinery through established procedures, while intelligence supporting sales is stovepiped. The main difference between the two practices is that stovepiped intelligence lends itself to greater flexibility and responsiveness to the needs of the consumer and therefore is an effective means by which to provide information.

The Scott Report provided a dramatic illustration of the role of government in the transfer of arms and dual-use technology. The use of Public Immunity Certificates by the government to try and hide its role caused a furore, with ministers eventually being forced to admit their participation in a secret foreign policy designed to enable British arms manufacturers to profit from an approach of supplying weapons to both sides in the Iran–Iraq war to prevent either side from prevailing (Scott, 1996, D1.56). For obvious reasons this policy remained secret while official policy revolved around the ‘Howe Guidelines’ and a restrictive policy concerning the transfer of ‘non-lethal’ defence equipment to Iran and Iraq (Scott, 1996, D1.10). The ‘Howe Guidelines’ were given in a written parliamentary answer by the then Secretary of State for the Foreign and Commonwealth Office in November 1984 and covered the following four areas: (1) that the British government should continue to refuse to supply lethal equipment to Iran or Iraq; (2) that being mindful to the precedent item (1) sets that British firms should try and fulfil existing contracts; (3) the British government should not in future approve orders for defence equipment that significantly enhances the capabilities of either side or prolongs the conflict; and (4) all applications to export equipment to Iran and Iraq should be scrutinised rigorously (Hansard, 1985a, p. 444; 1985b, p. 131; Scott, 1996, D1.59). In reality, ministers,

… had agreed that although lethal arms and ammunition would not be supplied to either side, every opportunity should be taken to exploit Iraq's potentialities as a promising market for the sale of defence equipment; and to this end ‘lethal items’ should be interpreted in the narrowest possible sense, and the obligations of neutrality as flexibly as possible (Scott, 1996, D1.10).

The DTI Minister, Alan Clark, described the guidelines as ‘high sounding, combining it seemed both moral and practical considerations and yet imprecise enough to allow real policy considerations on override in exceptional circumstances’ (Norton-Taylor et al., 1996, p. 47). Customs and Excise officers raided the offices of the company Matrix Churchill in October 1990 and arrested Paul Henderson (an MI6 asset) as well as two other company directors. These arrests led to a High Court trial in which the government sought to prevent evidence being submitted to court on the grounds of trying to protect the names of MI6 assets – which thereby removed Henderson's defence (Norton-Taylor, 1995, pp. 163–4). The trial judge ruled that this evidence could not be suppressed and would be available to the defence lawyers, which prompted Alan Clark, former Defence Minister, to tell the Court and six years later the Scott Inquiry that he had given a ‘nod and a wink’ to Matrix Churchill to export these materials to Iraq. The Scott Report demonstrated that the British government was willing to go outside the regulatory frameworks for a strategically important sale of arms and, moreover, the supportive use of the central intelligence agencies when this is the case.

According to evidence given to the Scott Inquiry by Lieutenant-Colonel Glazebrook, defence intelligence plays an important role in the licensing process itself. Glazebrook was tasked to provide assessments on the potential use of weapons and dual-use materials (Scott, 1996, D5.72–5). During the transfer that became central to the Scott Report, Glazebrook had been bypassed in the licensing process because of his prior objections to the transfer of materials to Iraq in contravention of government policy (Scott, 1996, E2.17). It had, therefore, become a matter of expediency to disregard Glazebrook's judgement. What the Scott Report also showed is that it is frequently the MoD who provides the most important objections to licence applications, with DIS assessments providing the strongest grounds for refusing a transfer of materials, and this highlights the impact that intelligence can have on the licensing of arms transfers (interview 36IS; interview 26IS).

End-use Monitoring

It is a curiosity within the transfer process that the UK parliament is entirely excluded from licensing until after the transfer of materials is complete. There is no prior role for the House of Commons Quadripartite Committee, which scrutinises strategic export licences, is constituted by the Select Committees of the MoD, FCO, DTI and DFID, and came into existence in 1999 (Quadripartite Committee, 2005). The main function of the committee is to provide judgements about the licensing process in the previous parliamentary year and to do so largely on policy grounds rather than in respect of the detail of particular transfers (interview 26IS; interview 34IS). The government has argued ‘that prior scrutiny of export licence applications raises unacceptable constitutional, legal and practical difficulties’ (HC 145, 2005, p. 12). Moving scrutiny from post hoc to prior would remove the delegated responsibility that the DTI holds within the arms transfer process, and also put pressure on the relevant government departments to respond transparently to questions that are posed to them about process – including accountability for the ‘stovepiped’ intelligence practices that support the commercial aspects of this trade, something the government has been unwilling to do even after the publication of the Scott Report. The Quadripartite Committee in the previous parliament (2001–2004) argued strongly for there to be a limited trial of prior parliamentary scrutiny, for example on government gifts of military equipment, which do not require a licence. This would remove some of the latitude the government has to make arms transfers to governments that government-insider NGOs, like Saferworld, suggest have established track records in transferring arms to problematic destinations or are problematic recipients in their own right (Saferworld, 2005a; 2005b).

There is little in the way of end-use monitoring of arms transfers by the government. The rationale for this is that once the manufacturer has cleared the F680 and full licence process the potential for misuse or misappropriation of the materials transferred has been reduced beyond a point where it is ‘economically feasible’ for the government to investigate (interview 26IS; interview 24IS). As a result, and coupled with the customer-driven intelligence cycle in the UK, there are no ‘requirements’ placed on MI6 to deliver information regarding the end-use of products, save for situations where the government places conditions on end-use as a part of the licence, like, for example, the prohibition on the Indonesian government using Hawk jets in offensive operations over East Timor (Nevins, 2002, p. 632; Zelter, 2004, p. 126). In these circumstances DAs are tasked with monitoring the end-use of these technologies, which is structurally problematic in the context of one of their key functions being to facilitate arms sales. The priority for ‘requirements’ rests on monitoring the illicit arms trade and this involves MI6 officers, who conduct ad hoc operations in conjunction with information gathered via GCHQ (interview 26IS; interview 05IS). NGOs concerned with the arms trade have campaigned strongly for a formal system of end-use monitoring. As with a sizeable part of the intelligence effort in the arms trade – as seen earlier – information being produced and provided under existing and unrelated requirements is used tangentially to provide an in-country analysis of how arms are being used, and indeed there is cross-over between this and MI6's role in monitoring the proliferation of weapons of mass destruction and the illicit arms trade (Butler, 2004, pp. 38–9). Again, this is a role for the intelligence services that is peripherally applied to the legitimate arms trade and notably appears to be more effective in assisting the sale of military equipment than monitoring its end-use. This can be explained by the disparity between the political motivation to sell military equipment and the less compelling desire to discover reasons for halting transfers or monitoring end-use. The more benevolent explanation is that stovepiped intelligence practices that assist sales are more effective than the centrally driven efforts to monitor end-use.

The final government institution engaged in end-user monitoring is HM Customs and Excise. Customs officials serve the function of monitoring and examining air, sea and land passengers and their baggage, freight and mail to ensure there is no smuggling of goods or transfer of goods in excess of established regulations (Butler, 2004, p. 37; Scott, 1996, C3.1–10). They process customs documentation and therefore have a role to play in the export of weapons and dual-use material. Customs officials played a significant role in the events leading to the Scott Inquiry, succeeding ultimately in interrupting an attempt to breach official government policy towards Iraq (Scott, 1996, C3.1–71). Outside this very high-profile example of where two government departments worked against each other, Customs officers play an important role in preventing the transfer of arms through the illegitimate trade, which is their main focus (DTI, 2005b). There is considerable liaison between the Secret Service, MI6 and Customs in providing a two-way tunnel of information about the movement of goods and where Customs can enforce existing guidelines (Butler, 2004, pp. 37–8). Precise details of this relationship are particularly difficult to trace. Their role within the legal trade is to ensure that exports are being made within the licensing paperwork agreed by the DTI. We might also note that although Customs and Excise produces a great deal of intelligence and performs a role on the Counter Proliferation Implementation Committee, this is another area of arms trade intelligence that is somewhat removed from the ‘central machinery’ and oversight of the JIC (Butler, 2004, p. 38).


Intelligence agencies act as both poacher and gamekeeper in the arms trade. In the gamekeeper role all of the UK's intelligence services and Customs and Excise are deployed to prevent illegal transfers of materials and to assist in providing the information, centrally, on which licensing decisions are made. DIS, with support from GCHQ and MI6, plays a particularly large role in providing information and analysis that feeds into the licensing process. In the poacher role these same intelligence services are used to support the UK's legitimate arms trade, often in an ad hoc and stovepiped manner. In strategically and economically significant cases secret intelligence is procured through the central intelligence services, while in routine arms transfers recycled and non-secret intelligence is used. The use of intelligence to facilitate sales strengthens Gill and Davies’ assertions that the role of the intelligence services should be seen in the context of their public administration function, rather than constructed in a way that emphasises the exceptionalism of the functions they perform.

In terms of the intelligence cycle and lines of accountability, this article has argued that for the vast majority of arms transfers, where the government is supportive, the ‘normal’ intelligence cycle is replaced by a heavily stovepiped and ad hoc process of informal networks, one that can best be characterised as a conversation between the intelligence services and the customer – the recycling of intelligence and analysis is conducted on a localised basis. The localisation of intelligence dissemination can also be seen in the way the embassies and Defence Attachés facilitate the negotiation and operation of the transfers in the host country; they take a very active approach to using British government assets in support of these commercial activities. The role of the DA is particularly important – these officials, outside formal intelligence structures, use their privileged access and information to provide manufacturers with a competitive advantage over rival manufacturers and to iron out contractual problems. This research has established that it is problematic to discuss DAs without making reference to DESO, the Department within the MoD dedicated to supporting arms sales. DESO can commission intelligence work, sits on the F680 pre-licensing committee, assists in placing DAs in countries and tasks DAs with work in support of manufacturers. The importance of DESO's role in intelligence terms is in skewing the intelligence effort in support of a set of niche business interests and as a result effectively forcing information outside the centralised intelligence frameworks.

A key theme that emerged during this research was that UK government officials have elided the government's identity and interests with those of the commercial manufacturers. There was a notable use of collective terms by government officials such as ‘we go and sell’, ‘we negotiate’ and ‘our kit’ while maintaining a notable caution in providing information on the processes that lay behind transfers. Several small units within the MoD and DTI were unwilling to cooperate with this research because it touched on issues that were ‘too sensitive’ or ‘at the moment, because of various business pressures, those in the licensing community are not in a position to discuss in detail the issues raised, and they are not something that one part of the community in isolation could discuss’ (Letter A, 2005; Letter B, 2005). Both the interview evidence and correspondence from officials suggest an assimilation of secure government information and identification with the arms manufacturers, which are private business concerns and increasingly, with the proposed flotation of QinetiQ (the UK's government's defence Research and Development arm), removed from public ownership (Boles, 2005). Thus, the governmental and private spheres have gradually elided, raising large questions about the role of intelligence in the UK's economic foreign policy and the extent to which a state–private network operates within British government.

The European Defence Agency (EDA) is a good example of how Whitehall's state–private network and self-defined shared identity with arms manufacturers has permeated, in this case, up to the European level. The EDA's role is to advance collaboration between companies and countries on the development of defence equipment. Nick Witney, the current British Chief Executive of the EDA and leader of the group tasked to design its institutional elements, has been keen to extend the role of the EDA and, critics argue, to generate further manufacturing contracts for British industry (Tigner, 2004a; interview 14IS). Arms manufacturers played a large role in the working groups that designed the EDA: high-ranking European Commission officials sat with representatives of BAE Systems and EADS as well as the President of the European Defence Industries Group to advise on how the new institution should operate, providing a key voice opportunity to manufacturers. More notable perhaps are the comments of two interviewees that the installation of Witney as the head of the EDA was a key UK government position in the intergovernmental negotiations that secured the establishment of the agency (interview 13IS; interview 14IS). Having a ‘Brit’ as the head of the agency would ensure that British interests were secured. Rathmell seeks to explain identity formation in intelligence officers in postmodernist terms: the construction of interests and identities is for him problematic in the post-Cold War world as targets and tasking agencies have diversified (Rathmell, 2002, pp. 97–8). The elision of interests raises two points for further research, firstly about the efficacy of public money being spent in the promotion and facilitation of arms sales, and secondly, whether the protection of sensitive markets for private businesses is a realisation of an age-old function of state – the further development of economic diplomacy. The privileged position of the defence industry in the decision-making processes of the government is a contested area. Explanations range from the extent to which defence manufacturers provide funding for political parties, to the level of integration and interchangeability between employees of manufacturers and civil servants, right up to meta-level critiques such as those made by Seymour Melman, for example, that feed into the classic military-industrial complex canon that views government as top-level managers of defence capitalism (Melman, 1970, p. 13). There are also foreign policy-led explanations for the pre-eminence of defence industries, namely that high-end defence technologies produce additional political credibility for the British government on the international stage, through enhanced capabilities and trade links. What remains out of these many competing explanations, some of which require further empirical testing, is that defence manufacturers hold a special position within the government's policy formulation processes.

The intelligence services cost the UK taxpayer £1 billion annually and since 2001 an extra £54 million has been invested in MI5, MI6 and GCHQ with additional year-on-year increases (Intelligence and Security Committee, 2004; Glees, 2005, p. 142;Phythian, 2005, p. 671). The Defence Intelligence Staff and Defence Attachés fall outside the formal intelligence budget and therefore their costs and evaluated performance sit outside the Parliamentary Intelligence and Security Committee's ambit, which is problematic in this instance given the prominence of both of their roles in providing information for licensing and supporting sales. The public policy questions about whether the public wants its taxes to be spent assisting this industry is for future publications or campaigners to tackle. This research has shown intelligence to be used in support of British-based private commercial businesses, and occasionally in providing intelligence on the negotiating positions of rival manufacturers. This in itself raises some important questions about the role of the state in the private sphere, and particularly with reference to using sensitive assets that imply that this industry has a core governmental function. The elite interviews conducted with government officials revealed an interesting trend of eliding the interests of the state with the commercial success of a set of industrial manufacturers. That the elision of interests has been allowed to develop is no surprise; what is more surprising is that there is little critical engagement among officials, politicians and the intelligence agencies on the issue of their very commercial role, or of how this work fits into ‘New’ Labour's foreign policy with its ‘ethical dimension’.

About the Author

Robert Dover, Defence Studies Department, King's College London, Faringdon Road, Shrivenham, SN6 8TS, UK; email: