Humans are curious creatures, and our curiosity drives a search for explanations. So while this search may fit squarely in the realm of science, it is hardly confined to the pursuits of scientists and intellectuals. Even preschoolers ask why, and indeed may do so to the exasperation of adults. Yet adults seek to understand things, too. They want to know why their partner responded angrily to their request, why the train was late, or why the weather changed so suddenly. By helping us understand our environment, explanations give us some control over our lives. I interviewed psychologist Tania Lombrozo at the University of California at Berkeley to find out more about the types of explanations people want, why we value these, and which types of human reasoning bring us closer to the truth.
Steve Ayan: Do people prefer some explanations of the world to others?
Dr. Lombrozo: Absolutely – there’s good evidence that people have strong and systematic preferences for some types of explanations over others. In my work, for example, I’ve found that people prefer explanations that provide a function or purpose, called “teleological explanations.” If I ask, for instance, “why does your cup have a handle?” a teleological explanations might be, “The cup has a handle so that you can lift it without burning your fingers.” In many cases, we’re more attracted to teleological explanations than to “mechanistic” alternatives, such as explaining the cup’s handle by going through the process that was used to manufacture that shape. Not everything has a function or purpose, however, so not everything has a teleological explanation. Why are there mountains? Why are there rivers? Such entities simply won’t support teleological explanations, unless you’re committed to an underlying designer, such as God, who designed then for some purpose.
Do we ever look for teleological explanations where none exist?
Yes. The urge to find teleological explanations arguably lies at the root of some religious beliefs like creationism, but it can also fuel secular notions, such as the idea of fate or that everything happens for a reason. Young children prefer teleological explanations in cases where adult would reject them. For example kids might ask, “Why are there mountains? So we can climb them!” or “Why do I have a nose? For holding glasses.” These types of explanations seem pretty silly, but adults indulge in them too, under the right conditions.
What if this kind of explanation is misleading?
Sometimes we just have to accept that not everything has a purpose. And if we over-explain we might end up with false beliefs. Conspiracy theories are a good example. There is often some truth in them, but they typically try to explain many different data points by appeal to a single group or entity, trying to make sense of everything in a unified way. But sometimes those data points are simply coincidental, or due to chance. Appealing to chance isn’t a very satisfying explanation, so if we’re looking for a satisfying explanation we might reject chance and mistakenly posit something that isn’t really there. Sometimes we should just accept that we’re uncertain, or that things are messy, or that something is pure chance.
And uncertainty is ubiquitous in the real world, isn’t it?
Sure. In the real world things rarely conform to perfect patterns; there are often exceptions. Let me give you an example from an experiment we’ve run in the lab. Let’s say you’re presented with descriptions of a bunch of people who do or do not give money to charity, and your task is to learn which ones do. You get information about each person, such as age, personality characteristics, names and photos. In 80 percent of the cases, you can predict who gives to charity based only on age, but 20 percent of the time you’d be wrong. So there is a pattern, but it is not 100 percent perfect, and you’d be able to predict more accurately if you gave up on looking for a perfect pattern and instead just memorized which individuals do or don’t give to charity. In a case like this, people who are prompted to explain why individuals give to charity actually make more mistakes than those who aren’t prompted to explain. Those who explain keep trying to find a perfect pattern that isn’t there.
What makes some explanations better than others?
Besides accuracy-that is, having explanations that fit the data-we care a lot about an explanation’s simplicity and breadth. I’ve already mentioned teleology, which we often prefer, and we often also care that explanations cohere with our prior beliefs and generate a sense of understanding.
But aren’t many things multicausal, requiring complex explanations?
Isaac Newton said, “We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances.” For Newton, this recommendation was backed by an assumption about what the world is like: he wrote, “nature affects not the pomp of superfluous causes.” Most contemporary philosophers are more skeptical about the notion that the world is actually simple. But there are more subtle reasons to prefer simpler explanations. For example, we might consider simpler explanations first because they are typically easier to falsify. So one argument is not that nature itself is simple, but that a preference for simplicity helps us arrive at the truth more efficiently. Simpler explanations also have cognitive benefits: they are easier to remember, easier to reason with, and easier to communicate. But that’s no guarantee that they’re true!
Conveying science to lay people often involves the translation of complex terms into simpler ones, the use of metaphors, and the removal of some details. In the process, information is lost. Is this the price we pay for simple explanations?
Often it is, but strangely enough, people sometimes seem to value complicated explanations. One study showed that when you add completely irrelevant math to the abstract of a scientific paper, non-experts judge the work as better. In another study, adding irrelevant neuroscientific information to psychological explanations made non-experts less effective at differentiating circular from non-circular explanations. But in general, we do need explanations to be presented in terms we can understand, and for non-experts that will often involve leaving things out. The challenge for science educators is doing so in a way that isn’t misleading.
Sometimes we create new terms to explain the world. For example, psychologists may introduce concepts such as the unconscious or self-confidence and then study them as if they were real. Does a belief in our own names for things lead us astray?
As far as I know, there isn’t research directly addressing your question, but my guess is that it’s true: we sometimes mistake the name of a phenomenon for an explanation of it. There is a famous example from a play by Moliere: “Why do some pills make you sleepy?” Answer: “Because they have a dormitive virtue.” This doesn’t actually explain anything, but introducing a concept such as dormitive virtue might make you feel you understand. This might seem like a reflection of human stupidity, but I think we may be doing it for a smart reason, because language doesn’t carve up the world arbitrarily. If something has a name, it’s usually safe to assume that it corresponds to a class of objects or events with something in common, and that what they have in common can provide a legitimate explanation for some of their properties. Even though we might not know much about it ourselves, we can assume that there are experts who do. I don’t know much about quarks or muons, for example, but I assume that they play a useful explanatory role in contemporary physics, and that there are experts who do understand them. Similary, people may assume that named concepts correspond to underlying explanations, even when they do no more than name what they purport to explain.
Are explanations soothing because they seem to reduce the chaos around us?
Sometimes, explanations give us a sense of control. This affective component can be strong and it’s also part of the reason why we prefer some explanations to others. But not all explanations are comforting. For example, the cognitive anthropologist Pascal Boyer points out that some religious explanations and beliefs offer comfort (such as the idea of heaven), but others don’t (such as the idea of hell and eternal damnation). And some researchers have found that changing people’s sense of control can influence the kinds of scientific explanations they prefer: if you feel that you don’t have control, you’ll be more drawn to explanations that promise order and predictability.
How can we create better explanations?
Explaining the present can help us prepare for the future. In doing so, we should, however, balance the need to find explanations that strike us as satisfying-that is, those that are simple and broad-with those that best fit all the facts. We should try to explain the world around us, but we shouldn’t expect all explanations to be satisfying on an intuitive level.