The launch of Australia’s new national terrorism alert system last week got me thinking about how and why the intelligence community communicates threats to the public
De-classifying sensitive intelligence to generate a specific warning about a specific threat is vital. Who wouldn’t want to know that their office building, transport route or holiday destination was a terrorist target? This is particularly important for threats against overseas locations, where the issuing Government has (at best) limited control over the identification and disruption of specific threats, or lacks confidence in the ability of domestic law enforcement and intelligence agencies to do so.
Unfortunately however, specific warnings that protect tourists in the short-term can still generate fear, disruption and publicity for the terrorist’s cause. In the worst-case scenario, they may even help achieve the desired outcomes of the attack planners. For example, if the perpetrators of the Tunisian attack aimed to degrade the tourist industry and undermine the democratic Tunisian Government, surely the subsequent UK travel ban has only accelerated that process?
National threat levels
General and ongoing domestic threat levels are a less exact science. They rely on the amalgamation of multi-source intelligence of variable reliability, regarding the activities of individuals based across the country and overseas. And they’re designed to provide a national picture, even if, as is the case here in Australia, the actual threat level varies wildly between states or cities.
Critics argue that the primary function of domestic threat levels is to provide top-cover for the issuing Government. Threat levels remain high so that if an attack occurs, the Government can argue that the public were given sufficient warning.
This cynicism is often bolstered by the lack of demonstrable change in security posture or Government behaviour when threat levels alter. And by the continued assertion that the general public should not change their own behaviour. For example, the guidance provided by the Australian Government states that regardless of threat level, “there is no need to change your daily routine. The increased threat level should not change the way you live your life”.
There seems to be a disconnect here. Governments explicitly ask citizens to change their behaviour when warning of threats overseas. But domestically, even if a terrorist attack is certain, we ask people to continue as normal.
This may make political and economic sense – disrupting tourism or business overseas is easier than doing so domestically – but does little to inspire confidence that the purpose of domestic warning systems is to provide the public with meaningful information.
Terrorism alert systems internationally
The Australian Strategic Policy Institute has collated a thorough comparison of international threat warning models. Most have tiers or levels of threat, with variance in the number of levels and the frequency with which Governments communicate about the system.
Given the Australian Government’s counter-terrorism Anglophilia, perhaps the most relevant comparison is with the UK public alert system. This has five threat levels, with comments clarifying the likelihood of an attack occurring:
- Critical – attack expected imminently
- Severe – attack highly likely
- Substantial – attack strong possibility
- Moderate – attack possible but not likely
- Low – attack unlikely
The new Australian system unsurprisingly also has five threat levels, but with each designed to explicitly indicate the likelihood of an attack occurring.
The Australian Government review that preceded last week’s announcement had stated that the previous system “did not allow for enough flexibility or precision.” The additional “flexibility and precision” has been added to the highest tier of threat.
The new Australian system – misleading terminology
The primary contrast with the UK system appears to lie in the Australian system’s use of short, clear terminology requiring no explanatory statement. Unfortunately however, the terminology is either ambiguous or too definitive.
A terrorist attack is “possible”? As Australian PM Tony Abbott himself has said, all you need to carry out an attack is a knife, an iPhone and a victim. Is such an attack ever going to be beyond the realms of possibility? Using this terminology, I’d suggest that the tier is superfluous.
And if nothing is certain but death and taxes, can a terrorist attack in Australia ever be ‘certain’? Highly likely? Yes. Expected? Probably. But certain? As an intelligence analyst, I would be reluctant to refer to an attack (or indeed any behaviour) as ‘certain’, even in circumstances where my target had revealed their attack plan, conducted reconnaissance, obtained the necessary weaponry and filmed a martyrdom video. Because regardless of how comprehensive an intelligence picture is, reliably predicting behaviour is as much art as science.
And clearly if you had the above intelligence, then what is certain is that the authorities would already be carrying out arrests and raids targeting the individual(s) (and probably anyone they’ve ever met or spoken to). Aside from in circumstances when an attack has already occurred – the definitive ‘closing the stable door after the horse has bolted’ – I can’t foresee a scenario in which ‘certain’ would be an appropriate level of threat.
So why the complexity?
From a public understanding perspective, a traffic light or high/medium/low system would be the simplest way to communicate the threat. But could a simple warning system maintain credibility if, given Government risk aversion and the current high threat environment, it is unlikely to be any lower than ‘high’ (or red) in the foreseeable future?
Beyond the complexity of a system and the terminology that it employs, the reality is that in the intelligence world, the more you discover, the more you realise how much you’ve been missing. And this uncertainty leads to what I see as the ‘rounding up’ of the threat picture, intelligence gaps filled-in with the worst-case scenario.
So even if intelligence agencies believe they have comprehensive coverage of domestic networks linked to ISIS or al-Nusra foreign fighters, can they be certain that someone hasn’t slipped through the net and has a yet to be discovered network of followers? Or that a known target is using a communications device they have no visibility of? In this scenario, you’re really looking at fifty shades of ‘imminent threat’.
Here in Australia then, we’re left with a five-tiered threat alert system, of which two levels could (or perhaps should) never be used, and only two levels – ‘probable’ or ‘expected’ – that are likely to be used for the foreseeable future. And regardless of which of these two threat levels is in place, the Government does not want the general public to change their behaviour.
This may seem like a ludicrous situation to find ourselves in. But I’m not sure how any system can balance informing the public in an understandable way and accurately reflecting the complex and changeable threat environment. And maybe that conflict makes it a problem too difficult to solve?
David Wells worked for UK and Australian intelligence agencies between 2005 and 2014, specialising in counter-terrorism.