What is a warning for? Ideally the aim is not to warn of the future but to change the future; enabling action to be taken to produce a better outcome. Some threats cannot be prevented entirely, but a disaster can be avoided through sufficient preparation. Others, such as a terrorist attack or the Russian invasion of Ukraine, could in theory be stopped if you had luck and guile on your side and got the warning right.

There was no failure to warn of the Russian invasion of Ukraine. It was not a surprise. Whereas the shock collapse of the western-backed government in Afghanistan could in part be attributed to a lack of anticipation and insight, this was not true with Ukraine. The US and UK governments, and Baltic and Central and Eastern European countries in particular had warned in precise language publicly (and it has to be assumed in more detail privately) of Putin’s intentions. This warning was sounded very clearly and consistently throughout the autumn of 2021. It was sounded from then on through snapshots on Twitter which illustrated starkly Russian troop build-up around the perimeter of Ukraine, clearly positioned to correspond to population centres and strategic targets. It was sounded through the release by US and UK governments of intelligence regarding Russian plans to create a casus belli through sabotage and misinformation. It was sounded by Putin himself, by his actions since 2014 and with absolute clarity in his article ‘On the Historical Unity of Russians and Ukrainians’ in July 2021 in which he said that Russians and Ukrainians were one people, a single whole and that Russia was robbed. That sovereignty of Ukraine is only possible in partnership with Russia. ‘… For we are one people.’ Ukrainians would say that this warning was sounded long ago.

But for a warning to be effective, it has to be believed, and it has to be acted upon. Issuing a warning is a decision point, passing on the risk to those who decide the response. The audience for the warning should never be a passive recipient. A warning failure can be a consequence either of failure to interpret information (analysis) or a failure to respond to the interpretation (a policy choice) or both.

Why was this warning not believed, and why was it not acted upon? The seeress Cassandra foretold the fall of Troy but was fated not to be believed. Cassandra is marginalised because of her role as a seeress; but nonetheless feels that she is to blame for not having been able to stop it. I delivered a talk on analysis and warning to Ukrainian civil servants last autumn. They knew what was coming; they just didn’t know how to get others to believe them.

There is a straightforward historical parallel with the build-up of Soviet Bloc forces around Czechoslovakia in 1968, also presented as exercises. In January 1968 Alexander Dubček became First Secretary of the Communist Party of Czechoslovakia, and began a dramatic liberalisation known as the ‘Prague Spring’.  The USSR saw it as an immediate threat. The question for NATO was to what lengths Russia and Eastern Europe were prepared to go to  in order to stop it, and the analysis focused on a build-up of forces along Czechoslovakia’s border. As tensions heightened during the spring and summer of 1968, NATO collected as much intelligence as possible through satellite imagery, reconnaissance aircraft and signals intercept. This showed the USSR concentrating troops along the border between Czechoslovakia and East Germany, and Soviet troops in Poland moving south from Krakow towards the Czech border. Was this build-up intended as a show of force (aiming to persuade Dubcek to moderate his reformist tendencies) or was it a preparation for invasion? It could have been both, of course: on 17th of July the US Office of National Estimates warned the Director of the CIA that, ‘we know of no way of foretelling the precise event in Czechoslovakia which might trigger … extreme Soviet reaction, or of foreseeing the precise circumstances which might produce within the Soviet leadership an agreement to move with force.’ From June to August the USSR conducted a series of exercises across the Warsaw Pact, including recalling reservists and mobilising forces from Latvia to Ukraine. Military preparations had been put in place such that by the end of July, the Warsaw Pact was mobilised for an invasion of Czechoslovakia should such an order be given. The West had clear understanding of that fact. Still most analysts thought that in the end the USSR would show restraint – they did not believe the decision would be taken to invade. No action was taken and on the 20th of August, a Soviet special forces battalion landed at and occupied Prague airport while the next day Czechoslovakia was invaded from the north, east and south by 20 Warsaw Pact divisions totalling a quarter of a million troops.

In 1968 the analytical community remained divided right up until the end on whether it was an exercise or a prelude to invasion. It has always been an interesting intellectual exercise to ask whether a bold judgement in the spring of 1968 which warned of an imminent invasion might have set in train a series of events which would have prevented the invasion (thus of course at risk of proving itself to be incorrect). But now we have a new case to examine: with Ukraine there was a bold judgment which warned of an imminent invasion.

On 11 November 2021 Bloomberg reported that the US had briefed the EU that Russia was considering an invasion of Ukraine. This briefing was based on information which had not yet been shared with European governments, and the response from the EU was that this information would have to be shared before any decision would be taken as a collective response. This difficulty in building a baseline common understanding of the threat from Russia has been consistent in Europe and comes down of course to questions of trust. It is unrealistic to expect the US to share sensitive information, and it is understandable that precedents of the use of intelligence to make a case for military action mean that it can be treated with suspicion. But the fact that even allies could not be persuaded without doing so meant that it was impossible to form a consensus on any decisive deterrence or pre-emptive action. Security partnerships and common purpose have to be underpinned by a shared perspective and understanding of the threat. That in turn rests on situation awareness — what the threat looks like and how it is changing.

An effective deterrent response would have had to alter Putin’s risk calculus. It is hard to see where the question: ‘What would we need to do to stop this happening?’ was addressed in a co-ordinated way. As a result, the West ended up in a position where its actions were configured as a response rather than as pre-emption, despite having had the clear warnings. And so Russia achieved complete dominance over the escalation.

Part of this may well be because even those who apparently believed the warnings did not completely believe them. Debate and noise can be catastrophic to the effectiveness of a warning. There has been so much analysis of ‘what Putin might do next’, but this leads to multiple scenarios, which can confuse, rather than clarify, a warning. It is also weirdly abstract; Putin’s actions do not exist in a vacuum; analysis should also consider how our actions might affect what he might do. Knowing how we impact on a threat or risk is a critical part of understanding when to warn; disaster risk reduction techniques present likely impacts to help galvanise decision-makers to prepare responses.

There is a danger that we are instead stuck between two fashionable buzzwords: ‘Evidence-Based Policy’ and ‘Horizon Scanning’.  Neither of these really help to anticipate and act before a threat emerges: action may need to be taken on the basis of scraps of evidence and on short term projections of the future. Even acknowledging a reasonable hesitance to be seen to be the escalator, for fear of creating a self-fulfilling prophecy, it is difficult not to conclude that the West did not have its escalation steps ready.  A defensive alliance relies on picking the right moment to be proactive and pre-emptive in order to be defensive.

Britain, the Baltic States, Poland and the US started to act seriously. The UK airlifted anti-tank weaponry to Ukraine with renewed vigour in January. A row broke out in the EU about who would help and who wouldn’t, and whether or not the Germans had denied access to their airspace.  Pressure to ‘do something’ about Nordstream 2 grew. Israel apparently blocked the US provision of Iron Dome to Ukraine, which may or may not have made a difference. The repercussions of this inaction – judgements on who helped, who didn’t, who helped enough, soured co-operation and offered fissures for exploitation by the adversary. Imagine what could have been done in advance, on sanctions and money laundering, on preparing for refugees, if we really had prepared for the worst and strived to make sure it didn’t happen. Was this a question of optics, of not wanting to escalate? Or of not wanting to appear to be preparing for the inevitable?

Sadly, we can’t now know whether this invasion could have been stopped. We are left with another counterfactual: would a more robust response to the show of force have deterred the invasion? We should focus now on how to improve the entirety of our warning systems, from the alert, through belief to collective action, to give ourselves the best possible chance of getting it right next time.