Navigating life’s hard choices – when ranking doesn’t work
Choosing between two options can be simple: compare them, rank them, and pick the best one. But sometimes ranking seems impossible. Then what?
Choosing between two options can be simple: compare them, rank them, and pick the best one. But sometimes ranking seems impossible. Then what?
Many of the decisions we made involve some form of exchange: we offer up something (say, money or time), and get something in return (goods, or money). Often this happens in some kind of market, which sets an effective exchange rate (the price of a loaf of bread or a litre of fuel, an annual salary or an hourly wage for a particular position). Even if there is no actual market to inform us, we can often work out an exchange rate that both we and the other party are happy with (imagine a flea market, but also, say, as an amateur artist, creating a mural or composing a song for a friend).
Living in the immaterial world
Things can get more difficult though, especially when one side of the exchange is immaterial. After a system error at your bank locks you out of your account for 48 hours, you may be offered compensation. If, as a result of this issue, you were unable to settle a bill in time and incurred a financial penalty, that is a material loss for which the bank most likely will be obliged to reimburse you. But the inconvenience of having your card rejected and being unable to check your balance is something else – how to value this?
An even more compelling example appeared in my Twitter timeline this week. Someone was taking a ride home in an Uber, when the driver suddenly dozed off on the highway, nearly crashing into heavy traffic while passed out. This is not so much inconvenience, as severe unpleasantness or even moral harm. Uber responded by offering a refund of 15% on the fare (about $7), evoking the following reaction from the client: “Certainly an interesting approach to valuing a customer’s life”, while a commenter remarked “the offer is abysmal, and Uber should rethink its approach to customer support and safety at once”.
Most people will indeed see this level of compensation as laughably small. Yet clearly Uber does not. In itself, this is not surprising: in a transaction where one party pays money to the other, the former will always seek to minimize the amount, while the latter will seek to gain the maximum. Through a process of negotiation (think of the flea market again), a compromise can be found that satisfies both the seller by being at least the Willingness to Accept, and the buyer by not exceeding the Willingness to Pay. It would be interesting to understand Uber’s reasoning behind their offer, but we can only speculate. Perhaps, aside from the fact that they are naturally reluctant to spend money, they might have a policy that sets compensation for any reason as a fraction of the price of the ride.
But what would be a mutually acceptable compensation for the rider? We can certainly think of several factors that might influence it. Some are quantitative, relating to the degree of immaterial harm that was caused. If, for example, the driver had actually crashed (with only material damage to the vehicle), the acceptable level of compensation might be higher than if there had only been near misses. Others are qualitative and more context-related. If the taxi driver had been operating independently, a more modest compensation might be acceptable than with the ride taking place under the flag of a technology giant turning over billions. But while these perspectives give us an idea of what might increase or decrease the demanded compensation relative to the circumstances, they do not help determine a compensation level where the rider would express indifference between experiencing the scary trip and receiving some money, and experiencing a uneventful trip.
This incident illustrates a common problem in decision-making: how do we compare, value and rank things that seem so intensely different? The rider's unsettling experience and Uber's monetary offer don't naturally align on a common scale. This dilemma is a fundamental issue in ethics, economics, and everyday life. To shed light on such perplexing situations, let's turn to the work of philosopher Ruth Chang, whose insights on 'hard choices' offer a helpful perspective on how to handle seemingly incomparable values.
Ac-cent-tchu-ate the qualitative
In her paper Hard Choices, the absence of a common scale on which to measure, for example, inconvenience, unpleasantness, or moral harm and monetary compensation is referred to as incommensurability. But, she argues, this in itself does not mean we cannot compare distress and money – clearly, both the Uber rider and the commenter have successfully compared the two, and found the compensation much too small to outweigh the negative experience.
Chang also considers the situation where two options are incomparable. As we are often reminded, we should not compare apples with pears, yet imagine that you can choose either an apple or a pear for dessert. It is not because they cannot be ranked, that you are paralysed, like Buridan’s ass which, positioned precisely between two piles of hay, cannot choose between them and dies of hunger. You will simply decide on the basis of which fruit you fancy the most in the moment, or failing that, toss a coin. Relating this back to the Uber compensation, it is obvious that $7 suits Uber, but not the shaken customer; similarly, $500 would presumably satisfy most riders in the situation, but be unacceptably high for Uber. We seem to be able to easily rank the unpleasantness and the compensation, if the latter is extremely low or extremely high.
<buridans ass.jpg>
[Don’t look so sad; it’s not because you can’t rank them that you can’t choose! (via Ideogram)]
In between, however, there is a range where we cannot say whether the compensation is too much, too little, or precisely right. This violates what Chang calls the “Trichotomy Thesis”, namely that, if two options are comparable, one must be better than the other, or worse, or they are equally good. There is a fourth condition – parity, describing situations where two options are comparable but not rankable. Imagine, for instance, two potential compensations for our Uber rider: $100 cash, or a $150 credit for future rides plus a personal apology from a company executive. These options are clearly in the same 'neighbourhood' of value, but it's not obvious which is better. The cash might be preferable for its immediacy and flexibility, while the larger credit plus apology might feel more validating of the rider's experience. They're comparable, but not easily rankable – they're on a par. This concept of parity helps us understand why some choices feel so difficult: it's not always about finding the 'best' option, but rather recognizing the distinct values each option represents. Chang offers the example of a lemon sorbet and an apple pie – one a tart, refreshing dessert, the other a soft and comforting one, quite different, but neither can be said to be “better” than the other. It is the qualitative difference that underpins both the difference, and its lack of direction.
Might these insights inform us on our quest to determining an adequate level of compensation for a scary taxi ride with a dozy driver? The harm that was done cannot be altered, but it is possible for Uber to acknowledge its qualitative nature – what specifically represents the unpleasantness of the experience? Similarly, despite the ostensible fungibility of money, the amount in compensation has a qualitative element to it as well: while $7 might be seen as adding insult to injury, the framing of the actual amount can present a very different message. 100 might still be acceptable to Uber without setting an unaffordable precedent, and framed as equivalent to a nice dinner-for-two or six bottles of fine wine, it might be accepted with grace. In addition, it may be a good idea for Uber to complement the monetary compensation with a sincere, personal apology, and a credible commitment to take steps to avoid this from happening again.
Faced with hard choices, it is a good idea to question our tendency to focus on quantifying and ranking everything. When we need to weigh up apples and pears – or lemon sorbet and apple pie – the answer likely lies in their qualitative differences.