Kwame Anthony Appiah is one of the world’s leading philosophers and the author of numerous books on ethics, cosmopolitanism and identity, among other subjects. He is currently Professor of Philosophy and Law at New York University and in 2016 delivered the Reith Lectures. In his new book, As If, he discusses the role of idealisation in science, ethics and political philosophy. He spoke to Prospect’s Sameer Rahim.
Sameer Rahim: Philosophy is often thought of as the pursuit of truth. But in your new book, As If, you argue that our perceptions are shaped by worldviews that we know to be false—and that that might not be a bad thing.
Kwame Anthony Appiah: I’m not sure I’d put it quite that way. Of course, our perceptions are shaped by our beliefs, and we know some of our beliefs are false. We just don’t know which ones… otherwise, presumably, we’d abandon them. But the point the book makes is a different one: much of our thinking, in the sciences, in philosophy, in everyday life, proceeds with assumptions we know to be not quite right. Hans Vaihinger, the German philosopher whose work inspired me, gave the example of his own religious life. He didn’t believe in the literal truth of the claim that God exists, but he thought there was much to be said for acting morally as if the Christian God were real. And he saw this as pretty much like the way scientists in his day (the turn of the twentieth century) as in ours, use models in which they make assumptions they know are wrong. In his day, for example, when physicists were trying to explain the way temperature and pressure are related in gases, they built models in which a gas was assumed to consist of a collection of tiny perfectly elastic spheres. They knew that atoms really weren’t like that: but if you assumed they were, you got pretty good predictions and a pretty good understanding of why, when you raise the pressure of a gas, the temperature goes up. When we do this sort of thing in physics we call it idealising: in fact the theory was called the ideal gas theory.
You write, “an idealisation is a useful untruth.” Is that simply because the world is too complicated to comprehend and we need simple ideas?
I think complexity is one reason why we need idealisations. We need to simplify. But it isn’t the only reason. As I say in the book, Vaihinger’s general thought was that “idealisation involves acting in some respects as if what we know is false is true, this is justifiable because it is useful for some purpose, and the purposes in question are various.” So sometimes we idealise because things are complicated. Other times, though, we do it because the assumption we’re making is close enough to the truth for some purposes that it would be pointless to take that truth into account. So it’s not that it’s too complicated to do, but it’s not necessary. Galileo did a famous experiment in which he calculated the value of the Earth’s gravitational constant, g, by letting balls roll down a surface. In doing the calculation he didn’t take account of the fact that there’s some air-resistance, because it’s so small. If you were to try to calculate the value of g by dropping a feather from a window, you’d get a very bad estimate, because in that case the wind resistance matters. So both things matter: what our purpose is and the nature of the phenomena we’re trying to manage or understand.
How can we have faith in them if we know them to be false?
Because it works!
Could you reflect on three areas in which idealisations are involved: political ideology, religious faith, and romantic love?
Well, I’ve already mentioned Vaihinger’s religion. He was at one extreme end of a spectrum. Most religious believers—people who are theists, unlike Vaihinger—will concede that they aren’t sure about the exact nature of God. So they’d concede that what they are thinking, when they think about God, is unlikely to be totally correct. Religious people often stress the limits of human intelligence, its finitude, in fact. But, like Vaihinger, that doesn’t stop them saying and doing Christian things. In democratic political thinking, we grant every citizen the right to manage her own life. We treat them as if they are rational beings, capable of making good decisions for themselves. But in other contexts we’ll acknowledge that they’re often irrational. So we make the assumption of rationality—we all pretend we’re all reasonable—for some reason other than that we believe it.
That’s something that a lot of economic modellers do as well, of course. We explain the law of supply and demand as reflecting the outcome of market decisions by rational actors aiming to maximize their wealth. But everybody knows that in many contexts that assumption would be unhelpful. Which brings us, pretty naturally, to love. One thing we do in thinking about those we love is ignore small failings. We’re not blind to them, but love makes it easier to set them aside. Someone who loves you won’t, in a sense, see you always as they know you are. They’ll think of you as better than you are, and that will often help you to be better than you might otherwise have been. This is one of the ways in which human love and divine love would have to be different.
What are the dangers of such idealisations?
I think the main danger is not recognising that we’re doing it. That can get you into trouble in lots of ways. If you’re assuming away air resistance, that’s fine if you know you’re doing it. Because you can take it into account when it becomes relevant. And similarly in economics, it seems fine to use models that assume rationality, as long as you recognise that it’s a false assumption, and that sometimes our irrationality will make a difference to the outcome. That’s why economists should look at actual data, actual decisions and actual economies.
What happens when these idealisations clash? A creationist who is also a biologist, for example.
Well, I’d say that we all do this sort of thing all the time. We all have partial pictures that work for some purposes and not for others and that make incompatible assumptions. But creationism is a problem because it’s usually held by people who don’t think of it as a picture that’s useful for some purpose of theirs, but as the complete truth. So they fit my model of people who don’t realize they are idealising: or, at least, that they should be. And if they think that current biology is the complete truth—as opposed to a partial image of it, useful for many explanatory purposes—then they’ll be faced with straight out contradictions: the universe is about 6,000 years old versus it’s more than 13 billion years old. Which is why what most religious people actually tend to do in those circumstances is use one picture some of the time and another picture at other times, which is what Vaihinger would recommend. Christians who use the story of the Creation to think about their dependence on God, say, for moral purposes, can perfectly well use the evolutionary story for the different purpose of trying to understand the fossil record or predict the development of drug-resistance in bacteria.
You’ve mentioned that you draw on the ideas of Hans Vaihinger. Could you tell us a bit more about him?
Vaihinger was born near Tübingen in the mid-nineteenth century and was raised with a father who was a Lutheran priest. He got a fine education in theology and philosophy and the classics: pretty much the best sort of humanistic education available in Germany at the time. He studied Kant, as everybody did then, and he was the founding editor of Kantstudien, one the great journals of Kant studies. As a young man he lost his faith in the literal truth of Christian theology. But, as he himself said, he could see that there was an alternative to simply abandoning Christianity for that reason. You could take up the attitude that some educated Romans had to the gods: that they were a useful fiction. And essentially what he did was to take that idea and apply it to other fields. So, for example, he was quite knowledgeable about mathematics, and in his day many people were worried about the idea of complex numbers, which involve, as you know, the idea of the square root of a negative number. Now if you multiply any two normal numbers together to get their square you’ll get a positive number: but the square root of n is just a number which, when multiplied by itself, produces n. So it looks like there can’t be square roots of negative numbers. Vaihinger proposed that we just accept that it turns out to be useful for certain mathematical purposes to behave, nevertheless, as if there could be. And he made the same sort of suggestion as I did earlier about ideal gas theory. His great book Die Philosophie des Also Ob (The Philosophy of As If), published in 1911, is a great assemblage of treatments of many subjects involving idealisation.
You write of how Adam Smith put forward two ideas of the human subject—the self-interested subject in The Wealth of Nations and the sympathetic subject in The Theory of Moral Sentiments. How did they fit together?
This is one of Vaihinger’s examples, which he borrows from an English intellectual historian. The idea is that in the Wealth of Nations Adam Smith builds a theory that assumes that human beings are rational profit maximizers. If they were, Smith’s claims about how the economy worked might be correct. But, as I said earlier, we aren’t rational profit-maximizers: it’s too hard intellectually to keep track of the facts and too hard, too, to resist our emotions, which will often lead us to do unprofitable things. But in The Theory of Moral Sentiments, Smith took account of the other side of our nature, the side that is driven by our feelings. If you’re trying to explain how individual human beings will behave, you have to take into account more than their acquisitiveness; but in many market contexts you can assume that people are trying to get the largest payoff they can.
Is that what you mean by saying “we need a multitude of pictures in the world”?
Well, if we tried to have just one picture, we’d have to avoid all the inconsistent assumptions that allow us to make each of our pictures work. And that would mean that we’d never be able to do anything at all, since we’d have to spend all our effort on conceptual hygiene. It’s not possible to avoid idealisation; and it’s not necessary, as long as we keep each picture in its place.
You write that “our most astonishing human capacity… the ability to access ways the world is not but might have been.” How important is imagination to your theory?
Every act, every theoretical reflection, involves the imagination. If I were to do this, what would happen? Would that be desirable? If this were true, what else would be true? All the time, in the simplest and the most complex of our activities, we’re thinking about how things might be, how they could be otherwise. And that’s the central work of the imagination. Without imagination, no “as if”!
Sameer Rahim: Philosophy is often thought of as the pursuit of truth. But in your new book, As If, you argue that our perceptions are shaped by worldviews that we know to be false—and that that might not be a bad thing.
Kwame Anthony Appiah: I’m not sure I’d put it quite that way. Of course, our perceptions are shaped by our beliefs, and we know some of our beliefs are false. We just don’t know which ones… otherwise, presumably, we’d abandon them. But the point the book makes is a different one: much of our thinking, in the sciences, in philosophy, in everyday life, proceeds with assumptions we know to be not quite right. Hans Vaihinger, the German philosopher whose work inspired me, gave the example of his own religious life. He didn’t believe in the literal truth of the claim that God exists, but he thought there was much to be said for acting morally as if the Christian God were real. And he saw this as pretty much like the way scientists in his day (the turn of the twentieth century) as in ours, use models in which they make assumptions they know are wrong. In his day, for example, when physicists were trying to explain the way temperature and pressure are related in gases, they built models in which a gas was assumed to consist of a collection of tiny perfectly elastic spheres. They knew that atoms really weren’t like that: but if you assumed they were, you got pretty good predictions and a pretty good understanding of why, when you raise the pressure of a gas, the temperature goes up. When we do this sort of thing in physics we call it idealising: in fact the theory was called the ideal gas theory.
You write, “an idealisation is a useful untruth.” Is that simply because the world is too complicated to comprehend and we need simple ideas?
I think complexity is one reason why we need idealisations. We need to simplify. But it isn’t the only reason. As I say in the book, Vaihinger’s general thought was that “idealisation involves acting in some respects as if what we know is false is true, this is justifiable because it is useful for some purpose, and the purposes in question are various.” So sometimes we idealise because things are complicated. Other times, though, we do it because the assumption we’re making is close enough to the truth for some purposes that it would be pointless to take that truth into account. So it’s not that it’s too complicated to do, but it’s not necessary. Galileo did a famous experiment in which he calculated the value of the Earth’s gravitational constant, g, by letting balls roll down a surface. In doing the calculation he didn’t take account of the fact that there’s some air-resistance, because it’s so small. If you were to try to calculate the value of g by dropping a feather from a window, you’d get a very bad estimate, because in that case the wind resistance matters. So both things matter: what our purpose is and the nature of the phenomena we’re trying to manage or understand.
How can we have faith in them if we know them to be false?
Because it works!
Could you reflect on three areas in which idealisations are involved: political ideology, religious faith, and romantic love?
Well, I’ve already mentioned Vaihinger’s religion. He was at one extreme end of a spectrum. Most religious believers—people who are theists, unlike Vaihinger—will concede that they aren’t sure about the exact nature of God. So they’d concede that what they are thinking, when they think about God, is unlikely to be totally correct. Religious people often stress the limits of human intelligence, its finitude, in fact. But, like Vaihinger, that doesn’t stop them saying and doing Christian things. In democratic political thinking, we grant every citizen the right to manage her own life. We treat them as if they are rational beings, capable of making good decisions for themselves. But in other contexts we’ll acknowledge that they’re often irrational. So we make the assumption of rationality—we all pretend we’re all reasonable—for some reason other than that we believe it.
That’s something that a lot of economic modellers do as well, of course. We explain the law of supply and demand as reflecting the outcome of market decisions by rational actors aiming to maximize their wealth. But everybody knows that in many contexts that assumption would be unhelpful. Which brings us, pretty naturally, to love. One thing we do in thinking about those we love is ignore small failings. We’re not blind to them, but love makes it easier to set them aside. Someone who loves you won’t, in a sense, see you always as they know you are. They’ll think of you as better than you are, and that will often help you to be better than you might otherwise have been. This is one of the ways in which human love and divine love would have to be different.
What are the dangers of such idealisations?
I think the main danger is not recognising that we’re doing it. That can get you into trouble in lots of ways. If you’re assuming away air resistance, that’s fine if you know you’re doing it. Because you can take it into account when it becomes relevant. And similarly in economics, it seems fine to use models that assume rationality, as long as you recognise that it’s a false assumption, and that sometimes our irrationality will make a difference to the outcome. That’s why economists should look at actual data, actual decisions and actual economies.
What happens when these idealisations clash? A creationist who is also a biologist, for example.
Well, I’d say that we all do this sort of thing all the time. We all have partial pictures that work for some purposes and not for others and that make incompatible assumptions. But creationism is a problem because it’s usually held by people who don’t think of it as a picture that’s useful for some purpose of theirs, but as the complete truth. So they fit my model of people who don’t realize they are idealising: or, at least, that they should be. And if they think that current biology is the complete truth—as opposed to a partial image of it, useful for many explanatory purposes—then they’ll be faced with straight out contradictions: the universe is about 6,000 years old versus it’s more than 13 billion years old. Which is why what most religious people actually tend to do in those circumstances is use one picture some of the time and another picture at other times, which is what Vaihinger would recommend. Christians who use the story of the Creation to think about their dependence on God, say, for moral purposes, can perfectly well use the evolutionary story for the different purpose of trying to understand the fossil record or predict the development of drug-resistance in bacteria.
You’ve mentioned that you draw on the ideas of Hans Vaihinger. Could you tell us a bit more about him?
Vaihinger was born near Tübingen in the mid-nineteenth century and was raised with a father who was a Lutheran priest. He got a fine education in theology and philosophy and the classics: pretty much the best sort of humanistic education available in Germany at the time. He studied Kant, as everybody did then, and he was the founding editor of Kantstudien, one the great journals of Kant studies. As a young man he lost his faith in the literal truth of Christian theology. But, as he himself said, he could see that there was an alternative to simply abandoning Christianity for that reason. You could take up the attitude that some educated Romans had to the gods: that they were a useful fiction. And essentially what he did was to take that idea and apply it to other fields. So, for example, he was quite knowledgeable about mathematics, and in his day many people were worried about the idea of complex numbers, which involve, as you know, the idea of the square root of a negative number. Now if you multiply any two normal numbers together to get their square you’ll get a positive number: but the square root of n is just a number which, when multiplied by itself, produces n. So it looks like there can’t be square roots of negative numbers. Vaihinger proposed that we just accept that it turns out to be useful for certain mathematical purposes to behave, nevertheless, as if there could be. And he made the same sort of suggestion as I did earlier about ideal gas theory. His great book Die Philosophie des Also Ob (The Philosophy of As If), published in 1911, is a great assemblage of treatments of many subjects involving idealisation.
You write of how Adam Smith put forward two ideas of the human subject—the self-interested subject in The Wealth of Nations and the sympathetic subject in The Theory of Moral Sentiments. How did they fit together?
This is one of Vaihinger’s examples, which he borrows from an English intellectual historian. The idea is that in the Wealth of Nations Adam Smith builds a theory that assumes that human beings are rational profit maximizers. If they were, Smith’s claims about how the economy worked might be correct. But, as I said earlier, we aren’t rational profit-maximizers: it’s too hard intellectually to keep track of the facts and too hard, too, to resist our emotions, which will often lead us to do unprofitable things. But in The Theory of Moral Sentiments, Smith took account of the other side of our nature, the side that is driven by our feelings. If you’re trying to explain how individual human beings will behave, you have to take into account more than their acquisitiveness; but in many market contexts you can assume that people are trying to get the largest payoff they can.
Is that what you mean by saying “we need a multitude of pictures in the world”?
Well, if we tried to have just one picture, we’d have to avoid all the inconsistent assumptions that allow us to make each of our pictures work. And that would mean that we’d never be able to do anything at all, since we’d have to spend all our effort on conceptual hygiene. It’s not possible to avoid idealisation; and it’s not necessary, as long as we keep each picture in its place.
You write that “our most astonishing human capacity… the ability to access ways the world is not but might have been.” How important is imagination to your theory?
Every act, every theoretical reflection, involves the imagination. If I were to do this, what would happen? Would that be desirable? If this were true, what else would be true? All the time, in the simplest and the most complex of our activities, we’re thinking about how things might be, how they could be otherwise. And that’s the central work of the imagination. Without imagination, no “as if”!
