How did Israel miss a surprise Arab attack on the fiftieth anniversary of the Yom Kippur War? Because knowing what the enemy is planning is very difficult, and especially so in this case, where the last fifty years have seen a constant refining of capabilities and techniques on each side. Palestinians know that they have to be surprising if they are to catch Israel off guard. Israel had developed an extraordinary set of sensors and defences to make sure that didn’t happen, to the point where they seemed impenetrable. But defences mask vulnerabilities, and may reduce the efficacy of other warning systems. Sounding an alarm about the enemy’s intent to act will always require a bold judgement about the unknown; it depends on a deep understanding and feel, which are almost impossible to gather from afar.

A strategic surprise knocks an adversary completely off balance. To be surprising, an actor needs to be able to act pre-emptively and keep the surprise secret. By definition, successful terrorist attacks will be a surprise, because those are the ones not discovered in advance. In a state of enduring conflict, the assumption has to be that the absence of indications of planning for an attack does not per se indicate the absence of planning for an attack. So the onus lies with the intelligence analysts, whose role is to look for those indications among all the noise (while also acknowledging that there might not even be a surprise among the noise). The perennial question is how best to organise that analysis to give the best possible chance of anticipating shocks and providing warning. There is, really, no such thing as a black swan. Rather, an event occurs of which the earlier stages or preparation were unseen, either because they were not looked for, or they were hidden, or both.

After the the colossal surprise of the attack by Egypt and Syria in the Yom Kippur war of 1973, an independent function was created within Israeli military intelligence, known as the Review Section. Its role was to review and to criticise the final intelligence assessment produced by the Military Intelligence Research Unit. Its purpose was expressly to ask: could the opposite also be true? Its specialness lay in the fact that it stood apart from all the other processes which formed a part of assessment and analysis, giving it greater powers of persuasion and warning. It was to challenge groupthink, a collective, assumed understanding of the intentions of the adversary. It was tasked to provide an alternative view, to ask whether there could be another important, legitimate conclusion from all available data, and whether anything had been omitted or excluded which maybe could change the conclusion.

The rationale in 1973 had been that the Arabs would not go to war because they would lose; therefore the danger of war was minimal. This became the concept of threat that underpinned all subsequent interpretations of their actions. In recent times, there had arguably again been an assumption that Hamas would not attack because it was too difficult and not in their interests. The Times of Israel reported on 7 October that the Israeli Defence Force’s assumption in recent years was that Hamas was deterred from attacking Israel because it feared the strength of the response and knew it would lead to the devastation of Gaza. Haaretz has reported that officials last week assessed that Hamas wished to avoid full/scale war, and that they would not want to jeopardise achievements which had improved the lives of Gaza’s residents. We don’t know what those assessments actually were; they will be the focus of future enquiries.

The idea behind the Review Section – a specific role of legitimate, institutionalised dissent – dates back to the office of the Devil’s Advocate (advocatus diaboli), created during the papacy of Pope Sixtus V (1585-90). The Devil’s Advocate was a canon lawyer appointed by the church to argue against the canonisation of a candidate. His purpose was to ensure rigorous challenge before an individual was appointed to sainthood, to scrutinise the miracles they had performed, and uncover any character flaws and misrepresentation. Since it was a formal appointment with a requirement to be the independent dissenting voice, the Devil’s Advocate could stand apart from rival factions. He was tasked to look for evidence which undermined the case – for example that alleged miracles performed may have been fraudulent. In a century which had seen the extreme shock of the reformation in Europe, the church needed to preserve the integrity and esteem of sainthood, and protect it from abuse. There needed to be credible answers to the critical questions of why a person should be a saint, and whether they had indeed performed miracles.

Having a formal Devil’s Advocate broadens the range of possible assessments based on partial information, but there are still limitations in how the role contributes to warning. A permanent challenge function soon undermines itself: is a contrary viewpoint automatically more valid than the orthodoxy? Introducing a phase of thoughtful debate may deepen analytical conclusions, but persistent challenge around uncertainties can be paralysing. In the case of Israel, the purpose of the Review Section was to warn of impending war when the whole system dismisses such a possibility. If it is designed always to look for war, there is a risk that it sees war in everything. And if the Devil’s Advocate always sees war, how long before it ceases to be useful? Those sounding an alarm for war have to pick their moment very carefully. When to use it and when not is very hard to call. The quality of challenge depends upon the depth of understanding of the other; the Devil’s Advocate makes the strongest case when they have the strongest evidence.

The problems of collecting evidence are compounded by barriers. In the case of Gaza, a network of sources, technological surveillance, walls, checkpoints, the Iron Dome all provided a defensive screen. But they were only a screen, and a screen works both ways. Despite its promises, technology cannot protect forever against surprise; defences will always only hold until they don’t. The other side will size up the obstacles and – one day – find a way to get around or over them. The more rigid the infrastructure, the greater the surge when it breaks.

Tsunamis in Japan are regular and catastrophic. The most recent, in 2011, was caused by a massive undersea earthquake of a magnitude of 9.0 on the Richter Scale which lasted about six minutes. The tsunami waves may have reached heights of 40 metres, travelling at a speed of 435 miles per hour. Despite the warnings sent from the Japanese Earthquake Early Warning system and the tsunami warning from the Japan Meteorological Agency, there were 20,000 fatalities. The warnings doubtless saved many lives. But a far older warning system exists, in the form of hundreds of tsunami stones – like mile markers – which stand around Japan’s coast marking the high point of a tsunami. The oldest date back about six hundred years and were built to warn future generations not to build any homes between the markers and the sea.

The demand for space for development in the twentieth century led to new communities being built between the stones and the sea. They did not ignore the warning stones. They put their faith instead in the massive sea walls built by the Japanese government. This twentieth-century innovation was taken to be better than the warning stones. The sea walls had been built to withstand lesser tsunamis, which had been scientifically proven to occur repeatedly. When such a massive one happened, the sea walls held until suddenly they didn’t, leading to an increased surge as the walls collapsed. The risks of dependence on sea walls became most evident in the case of nuclear facilities, including Fukushima, which had been built below the stones but behind sea walls. The tsunami washed over the walls which were supposed to protect the plants, disabling cooling systems.

The debate still rages about whether the construction of sea walls by the Japanese government was a mistaken, hubristic attempt to control nature, or whether they increase the odds of survival. The response has been to build a massive new sea wall and wave barrier, stretching 396 km from Iwate to Fukushima and reaching 15.5 m in height. People living right beside the sea can now no longer see it. Residents have argued that the wall will sever the human connection they had developed over generations with the sea, worrying that their understanding of the ocean’s character would be changed permanently. A more fundamental argument is that in the event of another major earthquake they simply won’t be able to see changes in the character of the sea or notice the draining effect that heralds a tsunami.

Defences need to allow for continuing proximity and connectedness to the threat to give the greatest chance of anticipating what it will do. The challenge for Israel in responding to this attack will be to consider not only how to improve the nature of their defences but also their understanding of the nature of the threat.