How we define “reciprocity”: the good, the broad, and the ugly

I hope this is the last blogpost I ever write about semantics. I always want to point people to a good reference on what the words that I use mean (and there isn’t a short quick guide), and Wikipedia does not work here.

People use the terms “reciprocal altruism” and “reciprocity”in very different ways in the scientific literature. So it’s very confusing for a new student in this field. I don’t use the term “reciprocal altruism” at all (because it is not altruism). Instead, I use the term “reciprocity”. Authors have very strong and contrasting opinions on what this word means, and I’m frequently told by overconfident researchers (and reviewers) that I don’t understand the original concept, or perhaps I haven’t read it carefully enough. But I wrote a review of its meaning and the history of its meaning, and the Prof. Trivers (who coined the term and invented the concept) wrote me to tell me he agreed and loved the paper. So there.

For the deeply confused, there are 3 basic ways that people will often use the term reciprocity. I call them the good, the broad, and the ugly:

The good

Direct reciprocity occurs when an organism makes cooperative investments that are in some way contingent on the experience (or memory) of cooperative returns from the recipient. This behavior requires repeated interactions, cognition, and in most cases, individual recognition. I call this definition “good” because  I believe it is most in line with what Trivers meant by reciprocal altruism.

An example: Monkey A grooms Monkey B on day 1, which causes monkey B to share its food with Monkey A on day 3. In contrast, Monkey C has never groomed Monkey B, which causes Monkey B to not groom or share food with Monkey C. Over many generations, monkeys that do not take past experience into account like this, when deciding who to groom or give food to, have fewer offspring.

This is similar to what Frans de Waal called “attitudinal reciprocity”. I like this definition, but I like the next one too.

The broad

A broader usage of reciprocity says that it occurs when an organism makes cooperative investments in a recipient that are in some way contingent on cooperative returns from the recipient, regardless of the mechanism. Note that this does not require cognition. So even plants and fungi can perform this version of reciprocity. This has also been called reciprocal rewards, sanctions, partner choice, or reciprocation. I like this definition too, although it means that all enforced forms of mutual benefits are a form of reciprocity.

An example: A plant provides carbon to Fungi A and Fungi B, and both Fungi A and B provide phosphorus to the plant. One day, Fungi A stops providing phosophorus to the plant. The plant responds by only providing carbon to Fungi B. As a result, Fungi A dies (or reproduces less). Plants that alter their carbon outputs to fungi based on the returns have higher reproductive success than those that do not.

What makes this different than “cooperation” more generally is that the plant’s actions reward and punish the cooperative partner based on the partner’s returns. In other forms of cooperation, the investments in the partner might be fixed. For example, in a byproduct mutualism, the plant might even be leaking carbon as a waste product, in which case it might not benefit from contingently altering its output one way or another.

The ugly

Some people define reciprocity as when organisms make cognitively calculated cooperative investments based on the cognitive expectation of the cooperative returns from the recipient. I call this the ugly definition because, although common, it’s a distortion of the original concept, it’s not what we should expect animals to do, and it’s not really even what humans do most of the time in their social lives, unless they are nasty Machiavellian psychopaths.

An example: Monkey A wants to get food from Monkey B. Monkey A doesn’t enjoy grooming Monkey B, but it believes that if it grooms Monkey B today, then B might give it food later. So Monkey A grooms Monkey B with the expectation of getting a reward later. Monkey B doesn’t want to give food to Monkey A but now it thinks that if it gives A some food now, than later when B has no food maybe A will feed it. So Monkey B suppresses it’s desire to eat all its food, and feeds A based on the expectation of a reward.

This has been called “calculated reciprocity” and I don’t think it’s very important except when 1) animals are trained to do it and or 2) humans are not relying on their normal intuitive emotion-based guidance systems, because the situation is relatively novel from an evolutionary perspective (like a business transaction between strangers), or maybe because they lack normal prosocial emotions (like a person that has to take a moment to calculate the costs and benefits of trying to help a friend who begins choking).

To make strictly “rational” economic decisions, a person must use calculated reciprocity. And people often use calculated reciprocity when playing a social dilemma game. However, the vast majority of reciprocity that occurs in humans is subconscious and is built into our intuitive emotions. When we invest time and energy in friendships, we do it intuitively and we find it inherently rewarding. We don’t think “I will give my friend a ride home, but I better get something out of this next week” (well, at least not most of us, most of the time).

For this reason, many social scientists think that friendship has nothing to do with reciprocity. But this is confusing reciprocity the good with reciprocity the ugly. People believe, and they will report, that their devotion to their friends is unconditional and unbreakable. But is that true? When someone invests in a cooperative relationship like a friendship, and they experience something that is not an adequate “return”, they will experience negative emotions like sadness or anger or disappointment. These emotions are not an accident. This “social pain” prevents us from being socially exploited in the same way that the sensation of pain can prevent us from getting burnt by touching a flame. Social pain can reduce social investments in others and in effect it can punish social partners (if only in a subtly way). This is not the same as consciously or strategically punishing someone for their behavior.

In other words, people don’t need to use calculated reciprocity, because their emotions are already strategically designed to do the job. And that’s still reciprocity.

If we want to understand friendships, we might not need a new theory that is an alternative to reciprocity. We might just need a better and deeper understanding of how reciprocity actually works. This means taking into account how social bonding is regulated by emotions (Do more emotional people have more social bonds? Are people who are most generous, also most spiteful when cheated?  Do more variable emotional experiences make social bonds stronger or weaker?). It also means testing exactly how cooperative investments are influenced by various social factors (such as short-term experience, long-term experience, partner options, supply and demand of partners, one’s assessment of one’s own value as a social partner, etc). These are tests that can be done with both human and non-humans (including  <wink-wink> vampire bats!).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s