Here is a selection of some of my recent key speeches:
The Cambridge Intelligence Seminar, Corpus Christi College, Cambridge
11th October 2019
INTELLIGENCE AND DEFENCE POLICY
“The tongue may hide the truth, the eyes never” (Bulgakov, The Master and Margarita)
I welcome this opportunity to explore the utility and importance of intelligence to the conduct and development of defence policy. I begin with some caveats. I am not an intelligence professional: my discipline was classics and ancient history. Much of what I cover remains operational, both in the Middle East and in contesting Russian aggression and Chinese ambition. I am in any case bound by the Act for life, not only not to divulge the secret material that I handled but also not to reveal details of the processes involved, including the systems of classification and compartments.
I will focus on three principal areas from my three and a half years as Defence Secretary before offering some final thoughts on the future of intelligence in the defence world. The first is the Russian intervention in Crimea and the Donbas. The MH17 airliner was shot down in eastern Ukraine on my second day in office, and from that summer onwards we were dealing with a much more aggressive Russia not only in the Donbas, in the Black Sea and Eastern Mediterranean, but also in the Baltic States and the North Atlantic.
Second is the campaign against the Daesh which spanned almost exactly my time in office. The Caliphate had been proclaimed just a few weeks before I took up post, and it was declared defeated, a little prematurely, just a month after I left. And third, the extent to which intelligence underpinned our strategic policy-making, especially in the run-up to and through the Strategic Defence and Security Review 2015, including its effect on the prioritisation of the threats we faced and the major investment decisions that flowed from that.
Of course, intelligence profoundly affected our approach in many other areas, elsewhere in the Gulf, towards the conflict in Libya, our work to constrain missile development by North Korea, and our monitoring of the Chinese build-up in the South China Sea, but it is these three areas that I concentrate on this evening.
In both the first and second, the conflict on Crimea and the Donbas and the establishment of the Caliphate in Iraq and Syria, the intelligence failures in the early 2000s and the consequent reports by Butler and Chilcott, still cast a long shadow. That shadow may well, initially at least, have inhibited our response in both theatres.
This is not the place to go back on the particular failings of intelligence and analysis that preceded the 2003 invasion of Iraq. These are well-documented. But it is worth, perhaps, adding one element that I believe is underplayed in both Butler and Chilcott, that is the extent that we wanted to believe – that the WMD was there. Parliament and commentariat, and perhaps public opinion too, weren’t concerned with the niceties of whether WMD had been present earlier but was later removed. We knew that Saddam had used chemical weapons before: we were ready to believe that he could do so again. We were, not unlike the Gauls that Caesar consulted in the Gallic War who were: “slaves of uncertain rumours, and most men respond by inventing the answers most likely to please their questioners” (Gallic War 4.5).
When I arrived as Secretary of State, the changes recommended by Butler were all in place. And it’s easy to forget now, that whatever the conclusions about the claims around WMD, significant changes were made to the processing and handling of intelligence. First, there was much better integration between the various agencies, and between the agencies and Defence Intelligence, the former DIS. Analysis and judgements were properly shared. Contrast that with the dysfunctionality in the USA in the run-up to 9/11that Christopher Andrew brings out so well in The Secret World.
Second, the National Security Council was firmly established as the principal, often only, forum for the government’s determination of foreign and defence policy. It was the place where senior Ministers could weigh intelligence assessments and question the judgements, with all the agencies and departments represented around the table. And third, we had a much improved Joint Intelligence Committee, chaired independently of the individual agencies, and a safeguard against any free-wheeling. The revised JIC machinery had put in place a system of rigorous grading that ensured that its customers were fully aware of the level of confidence that its analysts had in both their sources and the material.
These were welcome improvements but in my view there were still limitations. Defence Intelligence remained a separate entity, responsible directly to me as Secretary of State, and working to its own objectives. The National Security Council had a very large membership: by the time we had heard from the deputy national security adviser, the JIC chairman, each of the agency chiefs, the FCO Permanent Under-Secretary, the Ambassador to the country concerned, and CDS, there was relatively little time for the senior ministers present to review and test the merits of the policy or findings under discussion. It became a cumbersome process, with too many officials, and too little political input.
The JIC itself also failed, I suggest, to escape entirely the shadow of the past. It became a slave to phraseology: hours were clearly spent in debating the precise wording of each assessment, placing far more weight on the nuances of key sentences than they could bear. Busy Ministers in practice did not always have the time or patience to reflect on minute differences of wording that lay behind each judgement. The result was often over-caution, with key findings heavily caveated, to avoid the danger of the JIC being caught out again in over-confidence.
In any event, the two big events of 2014 saw the new machinery fully tested, and, if we are harsh, it failed on both. Nobody told us that Putin was going to annex Crimea or commit so heavily to the Donbas; and nobody foresaw the scale and speed of Daesh’s military advance through Syria and Iraq, from nowhere to the very gates of Baghdad. Now, of course, there were strong mitigating factors. Putin already had forces stationed in Crimea: he didn’t need to invade it in the conventional sense. The build-up in the Donbas was gradual, with the initial insertion of disguised Russian personnel. The idea of a physical caliphate was not considered a serious possibility by Western intelligence that was more concerned with Arab nationalism than religious extremism.
None the less, the failure to forecast these two seismic events earlier had consequences. In eastern Ukraine it took too long, weeks and months, to confirm the scale of the incursion into the Donbas, even when imagery showed the build-up of personnel and light armour at and around the Rostov railhead. The delays and caveating inhibited, in my view, the rapid and clear reaction in the West to what Russia was doing, and especially to the shooting down of MH17 in which 10 British citizens were killed. The morning after the shooting down I had on my desk photos of the likely missile launcher, its positioning for the firing, and its crossing back over the border. But it took another four years for the painstaking Dutch investigation to establish clearly the facts and the blame. During those four years, and especially in the first months, Russian propaganda was able to develop an alternative narrative, that the airliner had in fact been shot down by Ukrainian forces.
Contrast that with the response to the Salisbury poisonings last year. Here we saw impressive joint working between the police, the agencies and the Ministry of Defence, including the application of counter-terrorism techniques such as data washing, to the analysis of the actions of a hostile state. Within weeks we had a series of confident, well-choregraphed statements combined with the effective use of imagery, that led to the creditable identification of the suspects. That provided us with a much faster truth, enabling the UK and its allies to properly call out Russian criminality.
Nonetheless, as I suggested earlier, at least part of the responsibility in 2014 lies not with the agencies but with ourselves, the politicians who were the customers. We wanted to believe that Putin was different. We invited him into the G8 for fifteen years. Merkel, Blair, Sarkosy all travelled to Moscow in search of deeper trade and cheaper energy. We chose to put aside Russia’ repeated breaches of the international agreements that it had signed, including the Helsinki, Vienna, Bucharest and Istanbul Agreements and other treaties on nuclear and chemical weapons. We were like the prisoners in Plato’s cave, mistaking shadows for the substance.
Similar illusions permeated our view of the Middle East. As Christopher Andrew demonstrates so well, we had developed a kind of Westphalian mindset: we worried far more about the threat from Arab nationalism than religious fundamentalism. While we were hoping that the Arab Spring might lead into a more liberal summer, we ignored the deeper, darker forces at work. We did not envisage, let alone foresee, the speed and brutality with which the caliphate established itself that summer, sweeping through thousands of miles of Syria and Iraq, and taking control of some 40,000 square miles of territory (almost half the size of the UK), and enslaving around 8-10 million people on a scale that nobody, nobody predicted.
The consequence was that we may have been too slow to help organise the defence of Iraq, to mobilise the necessary coalition, and to build a consensus once again for military operations in the Middle East. And we had to move fast to deal with the challenge posed by this latest extremist threat to our way of life in western Europe. So I now turn to Operation Shader itself, and review the role of intelligence within it. How central was it? Were there weaknesses and how did any weaknesses affect the decisions that we took?
Looking back, I can recall that reading intelligence reports was a big part of my daily workload. These were reports that couldn’t normally be read out of the office or forwarded elsewhere in the world. There was always a backlog whenever I returned from abroad, and even though my staff would filter out some of the less relevant material, that still left several hours of reading each week. Alongside specific Strap material from the agencies there were voluminous reports from the military and our embassies, often combining to corroborate or complete the picture. It was important in handling all this not to be mesmerised by the classification alone: there was often no direct correlation between the classification awarded to a report and its significance, not least because the classification tended to reflect the need to protect the source rather than to signal its substance. I was also aware that I was almost always not the principal customer.
What were the actual uses of intelligence in Operation Shader, the campaign against the Daesh? Our aerial coverage was exceptional: without any air defence to worry about, our surveillance and strike aircraft enjoyed the freedom of the skies. We had a pretty good picture of the state of the campaign on the ground, and often a more accurate one than that being provided by Iraq intelligence to my defence counterpart.
There were some limitations: fairly obviously given the operating conditions, it was at the start impossible to expect HUMINT that could complement other sources. It was only in the latter stages, leading up to the liberation of Mosul and Raqqa that we had a fuller picture of Daesh dispositions on the ground. We had good information about Daesh sources of revenue – from oil, extortion, taxation and the sale of artefacts – but relatively little about its military strategy. And we lacked, again understandably, specific information about people: we never knew where Baghdadi was, nor did we find John Cantlie. Worse, during the early hostage-taking period, we did not have specific enough information about the exact location of a particular hostage to be able to mount what would otherwise have been a very risky rescue operation. What we did have, thanks to the extraordinary skills of GCHQ and the other agencies was an impressive amount of intelligence about plots being planned from inside the caliphate against targets in the UK and elsewhere in western Europe.
Overall, it was intelligence that enabled us to start planning the campaign of air strikes and to target these against Daesh infrastructure – the logistics dumps, oil wellheads, bridges and command posts – as well as in close support of Iraq troops on the ground. Pulling all the necessary intelligence together was a key part of the work of the international coalition and in ensuring that air strikes by the half dozen or so allies who were committing aircraft were properly co-ordinated.
Intelligence also underpinned the rules of engagement. I was very conscious of the criticism of the strikes in Libya and of the subsequent loss of the House of Commons vote on potential action against Assad in Syria the previous year. I set fresh rules of engagement in order to minimise the risk of civilian casualties, and I insisted from the start on retaining approval at Ministerial level. I knew that having obtained Parliamentary permission for strikes in the autumn of 2014 it was vital to keep the confidence of Parliament if we were to have any chance of returning to the Commons for authority to strike in Syria too. Hitting a wedding-party or a small mosque in error could have ended the campaign overnight.
We were helped by the precision of the weapons that we were using, much advanced from the first and second Gulf wars, and by the skill of our RAF pilots. But above all we had the intelligence from sophisticated surveillance aircraft above that enabled us to examine rigorously each potential target. We could check, over a period of several days or weeks, for patterns of life and the potential for civilian presence. We could weigh the choice of weapon and its likely blast area. And we could go back and check afterwards the precise battle damage in each case. And where we were targeting specific terrorist threats to the UK or European mainland that could not be forestalled in any other way, it was all the more important to have absolute confidence that civilians were not caught up in those strikes. Intelligence, therefore, played a huge and essential part in the defeat of Daesh in that campaign, and was also absolutely critical to the prevention of further terrorist attacks in the UK itself.
Let me turn now to my third and final area of focus, the role of intelligence in formulating defence policy at the strategic level, and in particular in the run up to the Strategic Defence and Security Review of 2015. SDSR 2015 was originally intended as a light touch refresh of SDSR 2010. In fact, it proved a more significant review that that: it led to a re0ordering our defence priorities, the reversal of some of the earlier defence cuts, and the taking of some different investment decisions. Intelligence had a crucial role in all of that.
By 2014 there had been a huge increase in intelligence around the threat of terrorism generally, the threats from groups such as Al Quaeda and Khorasan that were directed against the UK from abroad, and the threat of homegrown radicalisation. That drove the pressure to increase overall defence and security spending, and within it to allocate growing budgets to the security agencies themselves. As well, we had firmer intelligence of the growing assertiveness of Russia post-Crimea, with operations in the Balkans, in the Baltic and most particularly with air and naval operations in the North Atlantic. That led directly to one of the biggest decisions of SDSR 2015, the decision to restore, at considerable expense, the capability of maritime patrol aircraft. These were needed not only to protect our nuclear deterrent but to monitor the increasing Russian threat to our Atlantic communications cables.
Intelligence also alerted us to the growing strength of the Chines military, and especially its adoption of anti-access air denial (A2AD). Ballistic missile defence had been a gap left by the 2010 SDSR: five years on, we decided to accelerate our defensive and offensive missile capabilities. And finally, it was intelligence that foreshadowed the extraordinary increase in offensive cyber, including attacks by hostile states, by their para agencies, and by individual actors, whether terrorist or criminal. We were seeing the coming of age of information warfare, and the need to respond to the weaponising of information. That led to a much increased spend on cyber, both defensive and offensive, and to a sharper focus on the more rigorous protection of central government systems and critical national infrastructure.
That leads me to some final thoughts. Yes, intelligence may have been under a cloud post 2003. Yes, it may have been too cautious and caveated even a decade later in assessing Russian intentions and Daesh might. It may still be too cautious today. But in my view we should learn the lessons from both those two challenges. Both, the aggression of Russia and the brutality of Daesh, were fuelled and fortified by strong narratives. Those narratives, enwrapped in shameless but skilful propaganda, were developed at speed. It’s speed that Virgil in his Aeneid singles out as the great strength of propaganda: mobilitate viget, viresque eundo acquirit (“it grows by moving, it gathers strength as it goes along” 4.175).
Because we are now firmly in the age of information warfare, we cannot afford to lag behind in dealing with its weaponization. We need, more than ever, intelligence and analysis of that intelligence to the highest quality and above all at the highest possible speed. We must make our priority the faster truth.
That may mean being a little less cautious in our processes. It should certainly mean being prepared, where we can, to use intelligence less coyly. Where we have the evidence in plain sight, we must find ways of getting it out more rapidly , alongside or ahead of our opponents. If we don’t, we risk losing the propaganda battle from the start, and we will find it even more difficult to persuade our politicians and the public – always the shadows in Plato’s cave – of the reality that we are facing. That points therefore to an even more central role for intelligence – intelligence that is cleverly retrieved, properly analysed and capable of wider dissemination. After all, as Bulgakov puts it in The Master and Margarita: “The tongue may hide the truth, the eyes never”.
“The Government must look again at our infrastructure, school funding and share ownership plans”
“Salisbury must be our wake-up call”
“Digital giants must do more to tackle terrorism”
“These increased threats must mean a bigger defence budget. Our security is at stake.”
“Brexit Britain requires a bigger vision, more confident, outward-looking, self-rewarding.”
“President Trump has spoken about the need for engagement with Russia – he’s right.”
“Wider society has a lot to learn from small businesses.”