A better way to ask “why” (from behavioral science)

Short of time? Click here to read the key takeaways.

🧠 The word ‘why’ hides multiple different questions. An animal behavior scientist from the 1960s noticed that, when we ask ‘why’, we could be asking multiple different questions. About mechanisms, individual development, fitness effects, evolutionary history, or intentions.

🔍 Different ‘why’ questions explain the same behavior in different ways. A single action, such as procrastination, can be explained using each of the different approaches to ‘why’.

🗣️ Clarifying which ‘why’ you mean can help you, by allowing you to investigate things more deeply and by reducing the chances of talking past people.

My partner recently asked me whether I knew why The Financial Times is printed on distinctive pink pages. I remembered reading that it was a way to distinguish themselves from competitors, so I said as much, but my partner (a brilliant geologist) replied that the pages are pink because the China clay they originally used in making the pages came from a clay pit in Cornwall (England) which was unusually discolored pink.

I realized, in that moment, that I had assumed I was being asked a question about the intentions of the people involved, but my partner had in mind a question about the mechanics of producing paper. The single ‘why’ question could be interpreted in multiple ways. That immediately made me think of an animal behavior scientist from the 1960s called Nikolaas Tinbergen, who distinguished between several different meanings of ‘why’, and I was inspired to write this article about his insights.

Let’s start with a question: Who is the person you love most in the world? 

Some lucky people will find that a very difficult question to answer. One reason it can be so difficult is because the love we feel for parents, partners, friends, and fellows (to name just a few) seems to be different in kind rather than just different in degree. That is to say, they seem like different feelings described by the same word. That means a question like “Who do you love most?” can feel like it’s forcing you to rank feelings that don’t belong on the same scale.

Perhaps you’ve heard the popular idea that the ancient Greeks recognized many different ‘types’ of love and had words for each of them. Popular sources like to oversimplify matters by presenting neat lists of “the n types of love in ancient Greek” (though they often disagree about how many there were - numbers typically range from three to eight). If you’re looking for a sure-fire way to annoy classicists, just show them one of those lists. In reality, there was no single canonical ancient Greek taxonomy of love. Instead, although it was common to treat it as a concept with multiple senses, different writers theorized about it in different ways, and several of the words commonly attributed to the ancient Greeks’ conception of love were really developed by Christian theologians much later.

What you probably haven’t heard is that the word ‘why’ is similar to the word ‘love’ because it also has many meanings or kinds, but we (in English) cover them all with only one word. This article will help you to clarify your thinking about ‘why’ questions (which are of great value to any curious mind!), specifically in the context of understanding the behavior of human and non-human animals. We’re going to differentiate between five meanings of ‘why’ that you might not realize are available to you any time you want to understand behavior. Knowing these different meanings can help you to:

1. avoid talking past people (who might be using a different meaning than you are), and 

2. investigate things more deeply (by asking several different ‘why’ questions, instead of one).

We’ll follow the ancient Greeks’ approach to love by giving you different words for these different meanings of ‘why’.

Most of the framework for this article comes from a classic paper in the field of animal behavior studies (known as ‘ethology’). In 1963, Nikolaas Tinbergen was attempting to lay the theoretical foundations of research for this field of study, which was then in its infancy. In doing so, he pointed out four different kinds of ‘why’ questions that people studying animal behavior should be aware of. But his insights can be applied far beyond the behavior of non-human animals, and any careful, critical thinker could do with knowing about them. In this article, we’ll cover his four kinds of ‘why’ and we’ll add one more that covers things ethologists in the 1960s weren’t interested in.

The First Meaning of ‘Why’: Mechanism

This kind of ‘why’ is asking what the mechanism is that produces the behavior being explained. In other words, it asks what is happening inside the system such that the behavior is the output. We can make this clearer by focusing on an example:

Why did Sam procrastinate on the task she wanted to get done?

If we interpret this as a mechanism-why question, the answer may involve brain activity and its causal links to procrastination. It might suggest that Sam’s procrastination occurs due to activity in regions linked to memory and thinking about the future (such as the parahippocampal cortex), increasing how aversive the task feels and making its future rewards feel distant and uncertain. To oversimplify tremendously, when this influence outweighs activity in brain systems supporting planning and action initiation, the task may fail to trigger action.

The Second Meaning of ‘Why’: Development

Sometimes, when we ask ‘why’ a behavior occurred, we’re wondering how the behavior developed in this particular individual. These kinds of questions focus on past experiences, training, and things learned over time. If we interpret the question about Sam’s procrastination as a development-why question, then our answer will involve saying things like:

• Sam’s experiences with similar tasks in the past have taught her that she can get away with delaying this task and get more instant gratification in the meantime.

• Sam’s personality traits, such as conscientiousness, which developed (at least in part) as a result of her individual history and genetics and which studies indicate are linked to procrastination, make her prone to acting this way.

Unlike mechanistic explanations (which describe what is happening in the moment), developmental explanations describe how the relevant tendencies came to be present in the agent at all.

The Third Meaning of ‘Why’: Fitness Function

When it comes to human and non-human animals, it often makes sense to ask, “What is that behavior for?” In other words, to ask about its fitness function. The word ‘function’ can mean lots of things but, in this context, we’re not talking about intention (more on that below). Here, the fitness function of a behavior (what it is for) is the reliable effects of that behavior that tend to support survival and reproduction (a.k.a fitness).

If we interpret the question about Sam’s procrastination as a fitness-function-why question, then we might find it hard to answer because there is evidence that procrastination is generally maladaptive (i.e., has negative effects on fitness). It’s certainly easier to answer this kind of why-question about behaviors that clearly have mostly positive effects. For example, it’s easy to see the fitness-function of pain-avoidance behaviors like pulling your hand away from a hot surface: the effect is to put distance between you and danger, which makes it more likely that you survive.

But that doesn’t mean we can’t say anything about the fitness function of procrastination. For one thing, not all researchers think procrastination is entirely maladaptive; it has been argued that procrastination can serve “the adaptive function of avoiding the cost of a current effort when there may not be a future in which the payoffs can be realized.”

For another, when discussing maladaptive behaviors, it’s common to consider the possibility that the behavior is a consequence of some other behavior or system that does contribute to fitness (e.g., superstitious behaviors as a by-product of pattern-learning mechanisms).

This means that, if we interpret the question about Sam’s procrastination as a fitness-function-why, we might say things along the lines of: 

Sam’s procrastination reduced immediate discomfort and delayed an unpleasant task, which made her feel better in the short term. Perhaps this has net-positive effects on fitness by tending to avoid costly efforts when payoffs are uncertain. Or, if Sam’s procrastination is maladaptive, it’s possible that it is an unfortunate side-effect of other behavioral dispositions like impulse-control and avoidance that tend to contribute to fitness by reducing unnecessary effort or exposure to risk.

Of course, not all things that ‘behave’ have functions in this sense. The gathering of the storm clouds isn’t for getting you wet on your way home from work. So, even though it can be tempting to apply fitness-function-why explanations to things that haven’t evolved, it’s best to avoid doing so. As the philosophers John Hawthorne and Daniel Nolan have pointed out, the orthodox view in modern science is that applying function talk to things like storm clouds is “part of a superseded, pre-scientific muddle about how the world works”. 

It’s very easy to confuse this concept of fitness function for two things that it’s not:

1. A behavior’s fitness function isn’t anything to do with what goal the agent is consciously or intentionally aimed at achieving (if anything). That’s our fifth meaning of ‘why’, discussed below. Tinbergen himself warned against confusing function with goals, when he praised a colleague for being “too clear a thinker to confuse teleology [i.e., being directed towards a goal] with the study of survival value.”

2. A behavior’s fitness function isn’t the same as its evolutionary explanation. That’s our fourth meaning of ‘why’. We can make the difference between function and evolution clearer by turning to evolution now. 

The Fourth Meaning of ‘Why’: Evolutionary History

Sometimes, when we ask why a behavior occurs, we’re asking how this kind of behavior came to exist in this species (or kind of system) at all. That is, how did it evolve? While fitness-function-why questions (that we addressed in the last section) are asking about what kind of effect the behavior tends to have, evolutionary-history-why questions are asking about the history that led to the behavior. It’s easy to confuse the two because functions are ultimately grounded in evolutionary history: a behavior has a function because traits with those effects were favored in the past. But the questions are distinct. Fitness-function-whys are about effects, and evolutionary-history-whys are about origins.

If we interpret the question about Sam’s procrastination as an evolutionary-history-why question, then our answer will focus on the evolutionary history of human behavior. It might hypothesize that human brains evolved under conditions of uncertainty, in environments where delaying or abandoning costly efforts was often safer than carrying on with them. In such environments, a tendency to avoid immediate costs (rather than balancing them against long-term benefits) might have been favored by natural selection. It might then hypothesize that this tendency sometimes misfires in modern contexts, producing maladaptive procrastination.

When engaging in explanations like these, it’s always worth being careful that we don’t slip into telling ‘just-so stories’. Just-so stories are plausible-sounding explanations for why certain traits may have evolved, but they go well beyond what empirical evidence supports. Stephen Jay Gould used the phrase to deride what he considered “facile speculation”, and we’ve written about this in another article (which you can read here.)

The Fifth Meaning of ‘Why’: Intention

If someone asks you why you opened the curtains in the office, you’ll probably talk about your intentions. You might say, “I wanted it to be lighter in here.” Or if someone accuses you of wrongdoing, you might jump to saying that the negative consequences weren’t your intention (“I didn’t mean for that to happen!”). Intentions are ubiquitous in our everyday explanations of human behavior. So, why didn’t Tinbergen (mentioned at the beginning of this piece) include this as one of his questions?

Well, Tinbergen worried that explanations of non-human animal behavior slipped too easily into subjective anthropomorphizing and psychologizing. Even when researchers avoided what he called “crude” claims like “the animal attacks because it feels angry”, he argued that they relied on concepts like play and learning that “have not yet been satisfactorily defined objectively, and this might well prove impossible.” Thus, he excluded intention-like explanations on the grounds that they could not (at the time) be studied scientifically.

But, in this article, we’re looking beyond the scientific study of non-human animal behavior, so Tinbergen’s restriction no longer makes sense. Humans routinely act for reasons they can articulate, reflect on, revise, and even regret. And even though we famously cannot know what it’s like to be a bat, the possibility remains that non-human animals also routinely act with intention (many people believe that at least some animals do). Thus, leaving intention out of the picture would make the framework incomplete for wider application.

If we take our question about Sam’s procrastination as an intention-why question, an answer might be something like: Sam wanted to avoid the task that seemed unpleasant and wanted to do something more enjoyable. 

Of course, this kind of question won’t always be relevant. If you want to know why a volcano erupted, it won’t make sense to think about intention. However, Philosopher Dan Dennett argued that there are also contexts in which it can make sense to attribute intentions to things that plausibly don’t have them. For instance, if you’re playing chess against a computer and you want to know why the computer made a certain move. If you’re not a computer scientist, you probably don’t have the technical knowledge of computers in general (or chess computers in particular) to describe the mechanism, development, function, or (figurative) evolution of the chess computer that explains why it made the move it did. However, you can ascribe beliefs and intentions to it by saying things like “It intends to provoke me to weaken my king’s position, and it believes that move will cause me to do so.” Even if this is literally false, it’s at least a helpful way of talking because it enables us to make predictions (e.g., which move will come next) in the absence of the technical knowledge needed for the other ‘why’ questions. This is not an explanation in the same sense as it is when talking about human (or perhaps non-human animal) intentions; instead, it’s a pragmatic shortcut for when other explanations are unavailable.

Putting This Into Practice

If you want to avoid talking past people in discussions of ‘why’ (whether that’s about procrastination, the Financial Times, or something else), you can remember this framework for making ‘why’ questions more precise:

Five meanings of ‘why’

1. Mechanism: How does it work?

2. Development: How did it get this way for this agent?

3. Fitness Function: What is it good for?

4. Evolutionary History: Why does this kind of thing exist at all?

5. Intention: What was the agent trying to do?

And, if you want to understand a behavior more deeply, don’t just ask “Why?” once; ask it five times! 

Finally, I’ll end with a bad joke:

Question: Why did the chicken cross the road?

Answer: Because its leg muscles moved in a way that produced walking; because it learned to cross roads; because similar behaviors historically aided survival; because evolution built creatures that explore their environments; and to get to the other side.

Okay, I’m no comedian.