When we are in conflict with others, we sometimes start to believe they are irrational.

This article makes the case why we should always assume that our counterpart in a collaborative relation is acting in a perfectly rational manner. The argument has two parts: First, starting from a definition of rationality, we will see that we can never actually know whether our counterpart is rational. Second, as the argument goes, we should always assume that our counterpart is perfectly rational for pragmatic reasons in our own self-interest.

The article is a reflection on psychological phenomena occurring in discussions and conflict. It unfolds a model of human beings and interactions which aims at improving collaboration.

1. We Can Never Know

Let’s start with a definition of rationality. Stanford Encycopledia of Philosophy has a good definition:

Someone displays instrumental rationality insofar as she adopts suitable means to her ends.

Wikipedia has a similar one:

In its most common sense, rationality is the quality of being guided by reasons or being reasonable. For example, a person who acts rationally has good reasons for what they do. This usually implies that they reflected on the possible consequences of their action and the goal it is supposed to realize.

The point of these definitions is: rationality is instrumental. Rationality means to apply the best means to achieve a certain goal. A certain behavior cannot be judged as absolutely rational (or irrational), but this question is relative. Rationality is not an attribute of actions, but a quality of the relation between actions and goals.

While we can directly observe the behavior of someone, we have no direct access to their preferences. If we are not aware of the fact that the preferences of them are not accessible to us, we tend to assume the goals of the person are similar to our own goals. Lacking actual information, we insert an assumption. This assumption seems to be so natural that we often don’t even notice it, and we are not aware that this is but a hypothesis. Forgetting this can lead to flawed results – Whenever a person has preferences differing a lot from our own we might attribute irrationality to them when in fact we just fail to understand the difference in preferences.

Whenever we conclude that someone is irrational, because they act in a way we do not understand, this conclusion is a fallacy.¹ The behavior of a person does not give us access to their preferences. Since the rationality of a certain behavior depends on these unknown preferences, we cannot tell whether a person is acting rationally from their behavior alone.

Asking “Is this behavior rational?” without knowing the preferences is similar to asking “Is a hammer a good tool?” without knowing what we want to achieve. Whether the hammer is a good tool, certainly depends on the goal we want to achieve with it. When we need to get a nail into a wall, a hammer may be a good instrument. When we want to cure our headache, or clean our windows, we better look for something else.

The above does not only apply to what a person is doing, but also to what someone is uttering about themselves and their preferences. If someone says they prefer strawberry ice cream, although they always order raspberry ice cream, they might be just irrational. Or they might have a motive to say (or even to believe) that strawberry is their favorite, while it is actually raspberry. Psychologists, economists and UX-designers can tell: people are not always aware of their motives. Them telling us about their preferences is not a direct access to their actual preferences either. Even if they claim some preferences and act against them, we still cannot be sure whether they are behaving irrationally or whether they are unaware of their preferences (or not telling the truth about them).

The hypothesis that someone is rational or irrational can actually not be tested. In that regard, it is like the hypothesis that the universe just got created a second ago (including all the video records of our past birthdays and the photos in our smartphone of our last trip to Paris, and also including our own state of mind, including all our memories). There is no argument per se against this hypothesis. But it is a remarkable feature of that theory that it is self-immunizing against empirical falsification.

Notice that this does not mean that it was false to assume someone’s irrationality – it may or may not be true. We just can never know.

2. We Should Always Assume the Other One is Perfectly Rational

The whole point of this rather philosophical encounter is to demonstrate that we have a choice in believing whether someone is rational or not. That choice is not bound by our observations. As any observation cannot prove or disprove someone’s rationality, our choice is independent from that.

That being established, the second step of the argument contains a pragmatic appeal to why, in general, we should choose to believe that someone is rational.

That first step above is quite a shift in the understanding of the claim that someone is irrational. The following second part of the argument – that we should always assume someone is rational – is fully based on that shift of perspective. We are not longer talking about whether someone is rational or not (because we have seen above that it does not make sense to ask this question, as we cannot answer it). Instead, we take a step back and have a look at what the better assumption is.

When we are confronted with a person whose actions we do not understand (in descriptive terms) and do not accept (in normative terms), we have a tendency to perceive them as irrational. It makes sense to take a short break and ask why we do so. We might do so because we feel threatened: If their opinion is different, it might be the case we are wrong. Labelling the other person as irrational can be understood as an immunization strategy, which helps us avoid the painful question whether it is them or actually it is us who is wrong. The latter outcome would be irritating and involve a big mental effort. So it makes sense, that we try to avoid the question in the first place.

It is psychologically understandable why we do this. And sometimes it may actually be a good strategy: For instance, if we do not want to collaborate with someone, it is easier to not invest so many resources into establishing complex explanations for their behavior.

However, this article is based on the premise that we want to improve our collaboration. Hence, let’s focus on an argument why, under that premise, we should always assume the other person is perfectly rational.

For that, it makes sense to shift focus and look at what happens to ourselves whenever we claim someone to be irrational: As soon as we assume that someone is “just” irrational, we stop making attempts to understand them. We basically say: There is no consistent way to describe the decision making of this person. Our explanation puts a big question mark at the center of the other person’s decisions. And then, that’s the end of it.² That way, we will never be able to understand the person.

From this perspective, assuming that the other person is irrational is actually not so much a statement about them, but rather a statement about our willingness to create a more complex model of their preferences. It means as much as the economic statement: “It is not worth the effort investigating how the decision making process of this person works.³

If instead, we assume that the other person is rational, we have to search for a difference in their preferences and motives in order to explain their behavior. That way, we will still maintain some testable model of their mind. From the perspective of an empirical researcher, we have to assume the person is rational, as doing otherwise would simply break the empirical paradigm – it would be like saying: “Thunder and lightning are magical forces which we must not seek to explain”. Sure, it is possible to see the world that way. But we pay a very big price in terms of lack of understanding the world around us for the psychological feeling of safety that we gain with this attitude.

In work contexts, there is often a gain in collaboration. It is always our choice whether we want to put extra effort into improving collaboration with people we have problems with. The article just provides a mindset which helps broaden the repertoire of possible behaviors. We can still choose to consider others as irrational. But in case we do see value in working better with them, we now have a strategy how to resolve this unproductive hypothesis that blocks the relation with that person from growing into better understanding, deeper trust and more productive collaboration.