About the SA Blog Network

The Moral Universe

The Moral Universe

Dialogues on the psychology of right and wrong
The Moral Universe HomeAboutContact

A Moral Paradox

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint


As much as I want to offer a thoughtful response to your point on science and communication in the last post, I really have nothing else to add because I think you nailed it.  Let’s get into content.

I really appreciate that you brought up the Hobbes-Rousseau debate.  A related tension in the literature is the following question:  Does behaving morally at some initial point make people behave more morally or less morally later?  Decades of self-perception and self-consistency research find that people have a surprising lack of insights into their own attitudes.  Instead, we often infer what we like, who we are, etc. based on our behaviors, almost as though we’re drawing inferences about another person.  Therefore, if we behave morally and, say, donate to the Hurricane Katrina relief fund, we’ll observe this behavior and think, “Huh, I must be the type of person who behaves morally.”  This observation then should lead us to behave morally when we have a subsequent opportunity—when we are asked to donate to the Japanese tsunami relief fund.  We like to view ourselves as consistent beings, so doing good should lead us to see ourselves as good people in general–behaving selfishly.

Of course, the opposite view is nicely captured by the literature on moral licensing.  The moral licensing literature says, sure we’ll make the same observation for our initial moral behavior (“I see that I am a moral person”) but then this observation leads us to say, “Well, I guess I have fulfilled my moral quota.”  So now, when the second email comes around, asking us to donate to the Tsunami fund, we’ll say, “No need.  I’ve already proved myself to be a good person, so this time I can keep my money to myself.”

A man gives food to a boy in need


There is considerable evidence for both processes.  In terms of licensing, the classic work by Benoit Monin and Dale Miller shows that after people have shown themselves to be egalitarian in a hiring decision (giving clear preference to a female versus male candidate) then are more likely to discriminate against women in a subsequent decision.  This all makes me wonder about the NFL’s Rooney Rule, by the way.  Does the moral boost NFL teams get from shortlisting minority candidates then license them to pass over these same candidates at the final stage?  (we can get back to this).

My favorite example of “moral consistency” comes from a recent paper by Adam Grant and Jane Dutton.  They asked a slightly different question: What will make people behave more prosocially—acting as a benefactor or as a beneficiary?  Although reciprocity suggests people will behave more prosocially when they consider the ways that others have given to them, the consistency hypothesis suggests that people will behave more prosocially when they consider themselves to be benefactors, which is exactly what the studies found.  People who kept a four-day journal or wrote briefly about ways that they had been benefactors in the recent past (given to others, assisted others) volunteered more and gave more to charity than people who kept journals or wrote about ways that others had assisted them and given to them.

At any rate, there is a puzzle here about when licensing or consistency occurs, and people have started to suggest some clues.  Ayelet Gneezy and her colleagues recently showed that minimal, costless prosocial acts (attaching your name to a charitable donation that somebody else makes) produce licensing, because those acts don’t become an important part of one’s identity.  But costly act (actually digging into your pockets to make the charitable donation) instead cause people to take on the identity of “benefactor,” inclining them towards more prosocial behavior later on.  I also recently saw Paul Conway present some great work of his that mentions two other factors that could tip people between licensing and consistency: characteristics of the target and whether you’re thinking abstractly or concretely.  His work shows that thinking of yourself as moral makes you subsequently more prosocial toward positive targets (like schoolchildren) and less prosocial toward negative targets (like criminals).  Further, your concrete moral behaviors produces licensing whereas thinking about yourself as moral in the abstract produces consistency.  Also, some new work by Gert Cornelissen and his colleagues shows that ethical mindset is the critical moderator of these effects.  When people base moral judgments on strict rules and principles, they tend towards moral consistency, but individuals who judge the morality of an action based on its consequences tend towards licensing.

Clearly this topic is in the air.  Lots to discuss here.


Image courtesy of Geo C. Needham via Wikimedia Commons

Adam Waytz About the Author: Adam Waytz is a psychologist who studies the attribution and denial of mental states to other agents, and the moral and ethical implications of these processes. Follow on twitter: @awaytz

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 3 Comments

Add Comment
  1. 1. davenussbaum 10:54 am 03/21/2013

    Great question — your Kellogg colleague Daniel Effron was recently giving a talk on moral licensing and I came away with exactly that question.

    I think the suggestions you offer are intriguing — the identity question makes me think of the opposite of moral licensing, like what happens in Cialdini’s door-in-the-face studies. After refusing a very large (identity consistent) request, people become more likely to comply with a smaller one. The identity explanation is that they feel the need to repair the moral identity that’s been undermined.

    The other take on it, from my perspective, focuses on people’s ability to rationalize the subsequent (post-licensing) behavior. We know that people are very good at rationalizing things, but we also know from old school dissonance theory that the cognition that will change is the one that has more flexibility — so, as with the Gneezy research — when the initial behavior is staring you right in the face, it makes it harder to ignore. It also suggests that the more costly/difficult it is to do the right thing, the more likely we should be to rationalize not doing it.

    Link to this
  2. 2. Mike North 6:03 pm 03/21/2013

    Congrats to both JZ and AW on the new blog. I’d add that there is a temporal element worth throwing into the equation: that is, it might matter how far into the future we are talking about when considering people’s likelihood of enacting subsequent behaviors.

    The reason this comes to mind is that temporal distance matters when it comes to morality in general. For instance, all things equal, people are more likely to act immorally concerning the future than the present–at least that’s what Milkman, Akinola & Chugh (2012) showed in what they call a “temporal discrimination” effect, using emails sent to professors(!). When the email proposed a meeting that same day, the profs were just as likely to agree to meet with an apparently White student as they were an apparently Black one. But when the meeting was proposed for a week later, they were 25% more likely to meet with the White students than the Black ones. In other words, the immoral behavior emerged for the future but not the present.

    Assuming that the amount of “futureness” has a direct relation to people’s morality, this seems like an important factor to consider–including resolving the licensing vs. consistency question. People might be motivated to maintain consistency (or infer licensing) from prior behaviors, but perhaps only to a point; the more separated those initial behaviors are from the subsequent behaviors, the less relevance they might have to one another. Or if we take Conway’s work as a general guide, then perhaps the more distance between initial action and subsequent action there is, the more likely people will be to maintain consistency over licensing (since the future is more abstract). Regardless, the magnitude of temporal distance seems like it warrants future investigation in order to resolve the consistency-licensing debate. (Admittedly this is not my area of expertise, so perhaps there is a study out there that indeed addresses this by systematically manipulating multiple points of temporal distance.)

    Link to this
  3. 3. gesimsek 6:44 pm 03/22/2013

    If morality was connected to identity it would be a contingent act, however, contingent acts cannot be the basis of moral acts. If a moral principle cannot be shown as true in all cases (timing and frequency of the application might change according to capacity of the agent), it cannot be a moral principle.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article