Aformer​ MI6 agent is commissioned, by the opposition, to investigate an American presidential campaign that most people regard as a joke. He uncovers an international conspiracy led by the Russian secret services to put their man in the White House. He tells the British government about it, but they ignore him, believing the ‘Russian’ candidate will never be elected. He turns to the FBI, which takes his dossier more seriously. Then, two weeks before the election, the FBI’s director announces an investigation into the other candidate, the ‘Russian’ candidate’s rival. The Russians’ man wins and the British agent goes into hiding … Oh, and there’s a great sex scene in the Presidential Suite of the Ritz-Carlton in Moscow.

The events that came to light last week – when a dossier of intelligence reports surfaced online alleging Donald Trump’s eccentric sexual exploits, a long-running conspiracy between Trump and the Russian regime, and inappropriate financial deals over sanctions against Russian companies – read like the plot of a spy novel. None of the claims made in the dossier has yet been verified, but that doesn’t mean we shouldn’t take it seriously. Intelligence is information, from a privileged source, that supports decision-making. It is seldom verifiable because that information is rarely in the public domain.

This is especially true of ‘strategic’ intelligence, as opposed to ‘tactical’ intelligence. The latter concerns specific incidents or individuals: a suspected terrorist who is known to be operating in a particular city at a specific address. Tactical intelligence is verified by operational activity, the bin Laden operation being a good example. Strategic intelligence, on the other hand, seeks to explain people’s opinions, intentions and aspirations. France’s foreign intelligence agency, the DGSE, for example, might persuade a senior member of staff at Number 10 to provide intelligence on Britain’s negotiating strategy for Brexit (let’s pretend such a strategy exists). The French can’t verify this intelligence, because until the negotiations begin there is little information available to the public of what that strategy might be. The DGSE could try to suborn another member of May’s team to provide corroboration, but this is risky and could expose both the DGSE and their contact.

At the heart of this game of betrayal is trust: the source of the intelligence must be trusted by his or her handler. The reader of the intelligence report has to trust the provider of the intelligence while remaining critical. Intelligence is about degrees of credibility, and reading it is not the same as reading reportage, or a piece of political analysis. In order to make an assessment of its reliability, a reader needs to examine how it’s been sourced, insofar as that’s possible.

A large number of news media organisations have reported that the dossier is the work of Christopher Steele of Orbis Business Intelligence, a British corporate intelligence consultancy. Steele’s background in government service, including a stint in Moscow, is not in doubt, and the format of the reports in the Trump dossier is the same as that used by most Western intelligence agencies. Surnames are fully capitalised, as they are in MI6 house style.

The dossier’s most explosive report claims that ‘the Russian authorities had been cultivating and supporting … TRUMP for at least five years,’ and that ‘the TRUMP operation was both supported and directed by Russian President Vladimir PUTIN.’ The same report makes allegations of Trump’s ‘sexual perversion’ (Trump is supposed to have paid prostitutes to urinate on a bed that had been slept in by Barack and Michelle Obama), which the Russians apparently documented in order to possess useful kompromat (compromising material). It explains that two separate sources have attested to a long-term Russian plan to support Trump. One of those sources is a senior figure in the Russian Foreign Ministry, another is a ‘former top level Russian intelligence officer still active inside the Kremlin’. They affirmed the existence of the Russian plan, the report says, while in conversation with ‘a trusted compatriot’. This is an important detail, because it tells us they weren’t speaking to the British author of the report and spinning a line for his benefit, but also because it implies that the chain of information is long, which can easily lead to misunderstandings. A third source, also a Russian official, comments on the Trump operation without demonstrating any specific knowledge about how it was conducted, thereby supplying only limited corroboration. There’s a further allegation that ‘the Kremlin had been feeding TRUMP and his team valuable intelligence on his opponents.’ This claim is made by the Foreign Ministry source and confirmed by another source, described as a ‘close associate’ of the President-elect, who organised Trump’s visits to Moscow and accompanied him on them. The Russians might have confected the allegations and fed them to Steele in order to discredit Trump; but that argument can’t account for why one of Trump’s own people repeated them, unless we suppose he had been suborned by the Russians.

The allegations of a Russian campaign to support Trump are examples of strategic intelligence. The claims about Trump’s unusual sexual activities in a Moscow hotel suite, on the other hand, are tactical: the incident either occurred or it didn’t. The report mentions four different sources referring to it. The ‘close associate’ who arranged the Moscow trips is one. It’s also claimed that the incident was ‘confirmed [by another source] … S/he and several of the staff were aware of it at the time and subsequently.’ This source appears to have had some connection with the hotel where the incident took place, and is said to have introduced one of the intelligence company’s team to ‘a female staffer at the hotel … who also confirmed the story’. All the other sources in the dossier have had their gender obscured to make it harder to identify them, so this female staffer, we can assume, was a one-off contact used to verify the hotel story rather than an established source. Finally, the Kremlin-based former intelligence officer mentioned earlier is reported to have said that ‘TRUMP’s unorthodox behaviour in Russia’ gave the authorities sufficient material for blackmail.

So, three sources with differing perspectives claim to have specific knowledge of the incident. Two of them appear to be connected with the hotel; one is Trump’s ‘close associate’. A Russian official also claims to have enough material on Trump to be able to blackmail him but doesn’t say what the material concerns. Neither of the hotel-related sources are described as having any formal relationship with the Russian authorities. It is entirely possible that the authorities pressured the hotel employees into confirming the allegations, but again, that doesn’t explain why Trump’s ‘close associate’ made the same allegations. The incident remains unverified, but from an intelligence perspective the reporting of it is at least well-sourced.

Not all of the reports in the dossier have the same degree of sourcing (and some of the reports are not about Trump at all). A report about Trump’s adviser Carter Page’s alleged meetings with senior Russians is based on the testimony of a single source. The report’s content is striking – it claims that discussions took place concerning the removal of sanctions against Russia for its policy in Ukraine – but given the lack of corroboration the reporting lacks credibility. Other reports are based solely on the word of ‘a Russian émigré figure close to’ Trump. There is no explanation of why this émigré would be party to Kremlin thinking; one report says that s/he merely ‘opined’ and ‘understood’ the Kremlin’s intentions. In another report, a source described as ‘an ethnic Russian associate of … TRUMP’, and who is possibly the same person as the émigré (is the author trying to give the impression of more sources by varying descriptors?), speaks of a ‘well-developed conspiracy of co-operation’ between Trump’s team and the Russian leadership. He/she doesn’t explain how they managed to get access to this information. An experienced reader of intelligence would have little reason to credit the claim without more evidence of co-ordination between Trump’s team and the Kremlin.

One report, possibly based on a statement by an associate of the Kremlin’s former chief of staff, Sergei Ivanov, alleges that Michael Cohen, a high-profile lawyer working for Trump, met Kremlin officials in Prague in August 2016 to discuss making ‘deniable cash payments … to hackers who had worked in Europe under Kremlin direction against the CLINTON campaign’. This allegation has been widely discredited, not least by Cohen himself, who has stated that he was in California at the time – this has been confirmed by witnesses at the University of Southern California, where Cohen was attending a baseball game with his son. Some in the media have suggested that it was a different Michael Cohen who travelled to Prague, but have said nothing about who he might be.

The reporting on Cohen is flawed at best, though the dossier’s author does seem to have received the intelligence from two separate sources, at different times, and to have produced at least three separate reports on the subject. This may be evidence of a Russian misinformation campaign – both sources are described as Kremlin insiders – but elements of the reports can be checked. A man called Oleg Solodukhin is claimed to have been one of the key organisers of the meetings. Solodukhin is a Russian official based in Prague in a quasi-diplomatic role. ‘No such meeting took place,’ he has claimed in interviews with Czech media. In one of the Cohen reports there is a reference to a company called XBT/Webzilla that allegedly engaged in hacking against the Democratic Party leadership. XBT’s CEO, Aleksej Gubarev, has denied the allegations in interviews and the report itself offers nothing in the way of corroboration, but government agencies such as the NSA or GCHQ may be able to establish whether or not there’s anything in them.

What should an experienced reader of intelligence make of the dossier overall? A small number of the reports appear to contain well-sourced, triangulated intelligence. That does not make them true, but the reader may usefully assume their likelihood while considering wider evidence. Other reports are hard to credit; this does not make them false, but the reader should treat their claims with scepticism even where there seems to be corroboration or confirmation. Many uncertainties remain. The Russians say Christopher Steele (if he really is the author) fabricated the entire dossier as part of a misinformation campaign by MI6; perhaps the Russians fabricated the dossier and fed it to Steele in order to destabilise Trump. But there are enough testimonials vouching for the reliability of Steele and his company to make those interpretations seem unlikely.

For intelligence to have value the reader must trust the author, without being clear on the identities of sources, the methods of collection or the integrity of the process. Blair on Iraq, Obama on bin Laden; both trusted the spooks with little, if any, means of further corroboration, and with varying results.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences