Originally published on Medium on March 19, 2018.
In my last essay, I noted that utility is a useful tool to have in our mental toolboxes when formalizing ethical rules. One assumption implicit in that essay what that we do, in fact, need to systematize rules. It is worth taking some time to explore that such a need is, in fact, non-obvious.
Rules do not normally motivate us to act. Humans do not look at the rules, weigh the cost and benefits of compliance or defiance, and decide what to do. Despite the popularity of homo economicus, humans are motivated by more than maximizing self-interest. We are influenced by our social context. As Mark Granovetter writes in his book Society and Economy: Framework and Principles:
Actors do not behave or decide as atoms outside a social context, nor do they adhere slavishly to a script written for them by the particular intersection of sociocultural categories they happen to occupy. Their attempts at purposive actions are instead embedded in concrete, ongoing systems of social relations. … In ongoing relations, human beings do not start fresh each day but carry the baggage of previous interactions into each new one.
We are influenced by other people, culture, politics, religion, and all the other institutions we interact with. These influences add up to a set of norms and values where, most of the time, we do what feels right, not what some analysis tells us is right.
The norms and values of groups are intimately intertwined with those of individuals. Individuals interact with each other. Repeated interactions in sufficiently dense networks give rise to norms about behavior. Morals, those norms that separate right from wrong, give rise to broader values about what defines the good in life and in society. Yet because people are embedded in social context, societal ethics is not merely the sum of the individual morals. Individual morals are shaped by the broader societal ethics and are enforced (usually informally) by the members of the networks we are embedded in.
Most of the time, this is good enough. Nearly all of the actions we take each day are influenced by informal norms, not by consciously deciding a course of action based on rules. When we do make decisions consciously, they feel wrong unless we align them with our internalized norms. When necessary, people will find ways to do this…often by finding a way to justify our own behavior as a special exception.
The organic emergence of norm and values does not work as well when we are designing systems which will influence the behavior of others. These systems of influence can be command oriented, such as the law. Such explicit rules are not the only, or best, way to influence behavior. In many cases, it works better to structure the choices people make. In their book, Nudge, Richard Thaler and Cass Sunstein call ways of influencing behavior not based on explicit rules choice architectures. The important thing to realize about influencing the behavior of others is that there is no neutral choice. It may not seem like it should, yet the way items are shelved at the grocery store influences behavior. It is also worth noting that we cannot get away from systems which influence our behavior. They include more than government programs and policies. Anytime someone makes a decision that influences the choices of others, they are a choice architect, although not necessarily an effective one.
Designed systems have an interesting relationship with norms and values. Sometimes they merely formalize existing values. Laws against murder formalize commonly held values against murder and set expectations about how violations of those values will be punished. Other designed systems create new norms, although not always intentionally. It use to be normal to only text someone after asking their permission. As more people switched to plans where they did not pay per text, the norm changed and now texting is something we assume we can do. Designed systems are least successful when they try to impose behaviors that conflict with existing norms. Homeless shelter policies often break up families because facilities often take women and children or men. As a result, homeless families may refuse services because the system design conflicts with the value they put on their family.
What these brief examples tell us is that we need to be aware of how behavioral systems interact with norms. Awareness is what is required, not honoring the norms. Sometimes the goal is to change norms, such as programs to reduce drug use and violence in urban youth. Even when change is the goal, programs and policies will only be effective if existing norms are understood. To change norms we must understand norms.
Another interesting connection between organic values and designed systems is that organically arising values do not aim to improve an explicit goal. Systems designed to influence behavior do (at least, they should). This includes the direct goal for the system (e.g., helping homeless families), but it also needs to include our broader sense of how the system influences the greater good. This good, however, is not some externally imposed system of values. It is derived from the organic norms and values that inform the target audience.
Utility still matters. An effective choice architect will explicitly think about how their system increases or decreases it. The relationship is more complicated than a straight forward application of utility to decide what rules to create or what actions to take. Without taking our norms and values into account, utility directed decisions will ring false, leading in the worse case to situations where making the “right” choice pits us against what we hold most dear. Instead, utility is better thought of as a way of comparing options when creating behavioral systems.
At this point, I have spent two essays not actually saying what I think utility is. Don’t panic, that‘s next! (Or, at least, next in this informal series. I reserve the right to get distracted.)
(Today’s title is inspired by an excellent example of social context influencing individual behavior, Bonnie Raitt’s Something To Talk About.)