Risk is an unavoidable feature of our lives. Anything that we do, and certainly anything worthwhile — taking a plane to see family, leaving a job to follow a dream, pursuing the love of one’s life — requires us to take risks. For certain activities, such as extreme sports, it is risk that gives the activity meaning. But how well do we understand what risk is?
There is a natural way to think about risk, which is in terms of probabilities. Juggling dynamite is risky because there’s a high probability that it will kill you, but juggling oranges is not at all risky, as the chances of serious harm (an orange in the eye?) are at most minimal. So high risk entails a high probability of the relevant risk event (in this case getting killed), while low risk entails a low probability of the relevant risk event.
In a nutshell, this is the probabilistic account of risk, and it’s the standard view in the literature. Of course, the details can make things quite complicated as we move away from toy cases like this one and consider real-world scenarios where the probabilities are hard to determine, where we have multiple risk events to consider, where we are trading off between different kinds of unwanted outcomes (e.g., some very worrying, some less so), and so on. But the general motivation for the probabilistic accountis fairly clear, and it certainly looks very compelling, at least at first blush. While thinking about risk in terms of probabilities handles lots of cases, however, it also leaves out something very important. In particular, there seems to be a class of low probability scenarios that are nonetheless in a certain sense high risk.
Consider an experienced free climber like Alex Honnold, who in spectacular fashion recently climbed the sheer face of El Capitan, a unique feat captured by the award-winning documentary Free Solo. Like all free climbers, Honnold climbs without any safety net — there are no ropes, or anything to break his fall. It is hard to think of a riskier activity than what Honnold is doing. If he puts the slightest foot wrong, then he’s a dead man. As mountain climbers often put it, death is right there with you on the mountain.
And yet, is it really likely that he’s going to die on this mountain? That is, would we say that the risk event of him dying on the mountain is high probability? Remember that Honnold is the best in the world at this activity, and has been training for this feat for a lifetime. This is someone who knows what he’s doing, and in particular knows what to do to stay alive in these extreme conditions. What he’s doing is risky, but it’s not because it’s likely that he’ll die. If I were to try to free climb, with my lack of expertise, then this would practically be a death sentence, but that’s not the situation that Honnold is in. He’s taking a big risk, for sure, but this doesn’t equate to him being sanguine about being likely to die. (Indeed, if we thought it was likely that he was going to die on the mountain, then presumably we should prevent him from attempting this feat. I would certainly hope that someone would prevent me from trying it, if I ever got the insane idea in my head to give it a go)
Or take another example. Imagine an experienced jazz musician like Miles Davies playing to the very edge of his abilities, pushing his improvisational skills to the limit. This would clearly be someone taking an artistic risk. But is it likely that he would fail? That is, is there a high probability that the risk event of him messing up the performance occurs? Remember that this is Miles Davies that we are describing here. Part of his skill precisely consists in being able to take his performance to the very limit without letting it fall over the edge into catastrophe. And yet it is undeniable that he is taking a big risk; in fact, that’s what is so impressive about his performance. For Honnold on the mountain, it was death that was close. For Davies it is not death that’s close, but a disastrous performance. But the risk is the same.
The point is that probabilities only tell us part of the story about risk. The other part of the story concerns how easy it would be to fail. Often if it’s likely that we fail, then failure is easy too. But it isn’t always like that, as the cases just described illustrate. Honnold could easily fall to his death, even though this is unlikely, given his expertise. Similarly, Davies could easily mess up his performance, even though this is unlikely, given his expertise. Risk-taking isn’t just about doing something where one could likely fail; it’s also about putting oneself in a position where failure is right around the corner.
But what do we mean when we say that failure is “right around the corner” when one takes risks of this kind, if that isn’t to be understood probabilistically as the idea that failure is likely to happen? This is where the modal account of risk comes in. Modality concerns ways the world could have been — what kinds of events are possible, necessary, impossible, and so on. There is the way the world actually is, and then there are all the ways that it could have been. Some of these possibilities are easy possibilities, in that very little about the actual world would need to change in order for them to occur, while others are hard possibilities, in that a great deal would need to change. It’s an easy possibility that one’s clapped-out old car breaks down, as it is always on the brink of doing. Not much would need to change about the actual world to bring this about (e.g., if a part that’s close to breaking finally breaks, then that would suffice). But lots of other scenarios depict hard possibilities, where a lot would need to change about the way the world is for them to arise. Take the possibility that I win a 100m gold medal at the next Olympics. Given the nature of my age and physical condition (you’ll need to trust me on this one), it would take an incredible run of events to make such a thing happen.
For the most part, modal closeness — i.e., whether a possibility is easy or hard — tracks probability. It’s an easy possibility that my unreliable old car breaks down, and it’s also quite likely. It’s a hard possibility that I win the 100m gold at the next Olympics, and it’s also very unlikely. Interestingly, however, there are cases where the two things come apart, and this is important to our understanding of risk. The easiest way to bring this out is by considering a rather gruesome kind of lottery. Like a normal lottery, the number is drawn by randomly selecting from a tombola of numbered balls and the probability that a particular number (such as the number randomly allocated to you) is drawn is astronomically low. Unlike a normal lottery, however, if your number gets drawn you don’t win a prize. Instead, something terrible happens to you, such as your head exploding (please bear with me, there is a point to all this). Here is the crux of the matter. Is it likely that your head is going to explode as a result of this lottery? Clearly not, as the odds are astronomically in your favour — in all likelihood someone else’s head will be exploding as a result of the draw. But notice that it is an easy possibility that your head is going to explode as a result of this lottery draw. All that needs to happen to ensure that your head explodes is that a few coloured balls fall in a slightly different configuration. The point is that while low probability events are usually hard possibilities, this is not always the case — sometimes low probability events can be easy possibilities.
This is important to our understanding of risk, since if we understand risk purely in terms of probabilities then we miss something important, which is those cases where although the target risk event is low probability, it is nonetheless an easy possibility. This is just what’s happening in the cases described above. Honnold isn’t likely to die on that mountain, but he is taking a big risk, and this is captured by recognising how it’s an easy possibility that he dies. Even despite his prodigious skill as a climber, it wouldn’t take much for disaster to strike on that mountain. Similarly, Davies isn’t likely to screw up his performance by pushing his creative abilities to the limit, but it is an easy possibility that everything goes awry, given that he’s playing at the very edge of what he’s capable of. In both cases they are taking huge risks — Honnold with his life, Davis with his performance — but this feature of what they’re doing is lost if we just focus on the probabilities involved. The point isn’t that the probabilities don’t matter, but rather they don’t tell us the full story about risk.
We need to appreciate this point about risk for a number of reasons, not least in order to understand how to take risks in a rational fashion. Like Honnold and Davies, we are at our best when we push ourselves to the limit. But that doesn’t mean being reckless. It doesn’t mean doing things where failure is likely. If Honnold is likely to die on that mountain, then we should stop him from attempting his climb. (Just as you should stop an amateur like me if I attempted it. If I tried to do this, then death would be both likely and an easy possibility). And if Davies’s artistic risks are doomed to failure, then we should pity him rather than laud him. The reason why we acclaim these risk-takers is not because we think that they are careless, but because they expertly take the risks that they do. They skilfully sail right to the very edge of the world, but no further. In doing so they demonstrate the limits of human potential, and thereby show what it is to be alive, but they do so without being at the same time foolhardy.
The point is that some activities involve good risks, even though they are no less risky for being so. This is the sense of risk that we need to understand, and which we need to preserve in our lives, but doing so requires us to complicate our account of risk so that it can capture possibilities that are both unlikely and easy. Honnold is taking a great risk in climbing that mountain, but this is nothing like the risk that I would be taking by attempting this climb, and we make a fundamental error if we don’t recognise the important way in which Honnold’s risk-taking is different. In particular, we fail to recognise how his risk-taking is the manifestation of expertise, rather than foolhardiness.
If we want to make sense of how risk can be an ingredient in a rational and flourishing life, then we need to take this kind of risk seriously. Although the Honnold and Davies cases are extreme, it doesn’t take much imagination to see how each of us takes similar kinds of risk in our lives, such as when we push ourselves to our limits, or make a major life decision. If Honnold and Davies are simply being foolhardy, then that conclusion would carry over to us as well, and hence would mandate excessive caution. But no-one seriously believes that always opting for the safe choice is the route to a good life. By making sense of the extreme risk-takers among us we are thus able to understand how risk can be an essential ingredient of a life of flourishing. What’s needed to achieve this is a more nuanced conception of the nature of risk.