Why Australians are not very clever when it comes to fire
Fire has been an integral part of the Australian ecosystem for tens of thousands of years. Both the landscape and it’s native inhabitants have adapted to this periodic cycle of fire and regeneration. Nor are these fires random bolts from the blue, they occur regularly and predictably. Yet modern Australians seem to have difficulty understanding that their land will burn, regularly, and sometimes catastrophically.So why do we Australians studiously avoid serious consideration of the hazards of living in a country that regularly produces firestorms? Why, in the time of fire, do we go through the same cycle of shock, recrimination, exhortations to do better, diminishing interest and finally forgetfulness? To put it bluntly, why does the clever country act just plain dumb when it comes to thinking about bushfire risk Unfortunately, if you read the Interim Report of the 2009 Victorian Bushfires Commission you are not going to find answers to the fundamental question of why we behave in such an apparently irrational way and what follows is my attempt to explain why there seems to be a fundamental disconnect between the reality and our behaviour.
Those who do not remember the past are condemned to repeat it
We’d all like to believe that individually and collectively we’re rational beings, however in practice there are limits to what we know, the time we have to make decisions and our ability to hold all the facts in our heads and process them. So while we’re rational actors our rationality is limited. To deal with this problem we humans use ‘heuristics’ or rules of thumb to improve the efficiency of our our thinking. Unfortunately these short cuts also expose us to potential biases or errors in thinking.
The affect heuristic or why we don’t fear what we love
A visitor to Australia would wonder why so many people deliberately choose to live in the bush, knowing that fires regularly reduce the landscape to ashes? The answer lies in the affect heuristic, that is humans conflate perceived benefits with their perception of risk (Finucane 2000). To put it simply we’re hard wired emotionally to discount the risks of those things about which we have a positive emotional response. Things are either good-no risk or bad-high risk. So because we love our home amongst the gum trees (1), we ignore the danger that that it poses to us. As a happy coincidence this also allows us to avoid a hard discussion of the costs and benefits of ‘tree change’, lifestyles. After all if it’s good, there can’t be a down side surely?
Affect and the decision to stay or go
When we deal with unfamiliar stressful scenarios or the situation changes rapidly we also tend to rely on affect to a greater degree. Thus it’s unreasonable of us to assume that people can and will make a decision, such as stay or go, ahead of time and stick to it in the face of a strong emotional affect.
“I am certain that events occurred very suddenly and very dramatically when the wind changed on the night in question. Daylight almost instantly turned into total blackness and burning bark and leaves showered every person and animal not undercover. People who had been hosing their homes or clearing gutters in case the fire did threaten them suddenly felt compelled to flee. Concern which had existed in the minds of many people for some time suddenly turned to extreme urgency.”
Survivor of the Ash Wednesday Fire
The quote from a survivor of the Ash Wednesday fires illustrates the extreme affect of a rapid change from a normal environment to the hellish conditions of the pre-firefront can have upon the decisions of those exposed to it. In such circumstances it’s going to be inevitable that there will be late evacuations, no matter how irrational leaving a place of shelter as the fire-front approaches may appear.
Conformity bias and panic in the herd
Decisions are also made in a social context, here a powerful effect known as conformity bias can drive people to behave similarly to the others in a group, even if doing so goes against their own judgment. Thus the decision of a few to leave in the face of an iminent fire front can also influence even those who believe leaving late is irrational and high risk.
The information deficit trap, information ≠ enlightenment
Which brings us neatly to the next problem, how can authorities provide data that will convince someone of the correct course of action. The idea is that the other person is simply uninformed and the right information will convince them to change their mind. In practice when you are communicating across a difference in belief structure providing information will not change people’s opinion, or sadly their behaviour. We also know from work on biases and heuristics in decision making that people tend to ‘cherry pick’ the information that supports their position and ignore that which does not. Therefore any public communication on fire risk is always going to be misunderstood, ignored or used in a context for which it was not intended. The more complex the message the more likely it’s failure.
The availability heuritic or she’ll be right, most of the time
Read back through the accounts of major fire events in Australia and you’ll find a recurring theme of surprise over the the lack of preparation for what is a predictable risk (2). So why if we know the risk why do we do nothing? A principal method we use the evaluate the likelihood aspect of risk is the availability heuristic that is, the easier an event is is to recall the more likely the event is judged to be (Lichtenstein 1978).
This leads for rare events, to the absurdity bias e.g. when a bushfire has not recently occurred people will judge the event as ‘absurd’, and refuse to take such actions such as insuring their property or making plans in advance to stay or go. Another availability bias is that we can be lulled into a false sense of security by the precautions we do put in place (3). Thus our country fire brigades are quite effective in dealing with smaller fires thereby reducing the likelihood of loss, most of the time. People then build their houses in the bush in the belief, justified most of the time, that the local fire authority will be able to bail them out. True for most of the time but not for extreme events such as occurred on Black Saturday. As a society it seems that we tend to treat the severity of disasters as having an upper bound defined by the level of severity for which we have designed our defences.
Thinking the unthinkable (4)
Another element of our present dilemma is the reluctance of people and organisations to plan in advance for the absolute worst case. Here we struggle against the absurdity bias, if we have not experienced it we find it difficult to encompass the consequences in our heads. At an individual level this translates to planning not just for a scenario where emergency services are functioning and information is readily available but also for the ‘broken back’ case. Similarly if defending your home you need to plan for the worst case scenario of where to go if your defence fails. For government organisations this means planning and publishing what can be done for extreme events such as the 2009 Victorian fire complex (or worse). Even a blunt, ‘we will do what we can but you will likely be on your own’, has the value of preventing illusory expectations of help driving individual decisions before the event, as well as fear and uncertainty on the day.
I was fairly sure that we would die there, as I didn’t think we could stand up to the flames and the heat. …. Throughout the time that I was there I felt I was expecting help from some outside quarter. It wasn’t until the time that I decided to get up that I realised fully that our only help lay with ourselves…..”
Survivor of the Ash Wednesday Bush-fires
So how unthinkable could it get? The likelihood of a fire versus it’s severity can be credibly modelled as a power law a particular type of heavy tailed distribution (Clauset et al. 2007). This means that extreme events in the tail of the distribution are far more likely than predicted by a gaussian (bell curve) distribution. So while a mega fire ten times the size of the Black Saturday fires is far less likely it is not completely improbable as our intuitive availability heuristic would indicate. In fact it’s much worshippers than that, in a heavy tail distribution we can apply what’s called the mean excess heuristic which really translates to the next worst event is going to be much worse…
In a market the expectation of government bailout increasing the taking of risk is termed ‘moral hazard’, the low number of properties that were insured and (Houston 2009) destroyed in the Black Saturday fire represents a version of this effect. Here the un-insured benefited from the local fire authority (the bailout), without having to carry the CFA levy placed on insurance (the cost).
Trying to do better next time
I’ll close this discussion as I opened with the point that the philosopher George Santayana made, if we don’t understand the past we’re doomed to repeat it. In the case of risk if we don’t understand how we made past decisions about fire risk and the biases that drove these decisions we will inevitably repeat them.
I hope we can do better.
1. For non-Australian readers, this is a play on a famous Australian folk song whose refrain is ‘give me a home among the gum trees…’.
2. The classical De’Moivres definition is used, i.e. the likelihood of an event times the severity of that event.
3. This effect was originally noticed in flood precautions, e.g. dams and levees reduce the likelihood of smaller floods, people then build on the flood plains guaranteeing higher losses for the rarer extreme events that overwhelm the defences (Burton 1978).
4. With apologies to Herman Kahn.
Clauset, A., Shalizi, C.R., Newman, M.E.J., Power Law Distributions in Empirical Data, arXiv:0706.1062v2 , 2007.
Burton, I., Kates, R. and White, G., Environment as Hazard, Oxford University Press, New York, 1978.
Finucane, M. L., Alhakami, A., Slovic, P., & Johnson, S. M., The affect heuristic in judgments of risks and benefits, Journal of Behavioral Decision Making, 13, 1-17, 2000.
Houston, C., Push on for forced fire insurance, The Age, 9 March 2009.
Lichtenstein, S., Slovic, P., Fischhoff, B., Layman, M. and Combs, B., Judged Frequency of Lethal Events, Journal of Experimental Psychology: Human Learning and Memory, 4(6), 551-78, 1978.