Did morality really evolve?

I say no. Watch the video (from my talk at UCLA in the Brain Evolution Culture series) and the Questions and Answers

Comments and suggestions welcome as usual.

5 Comments

  1. Bryce Huebner

    Edouard,

    I really liked the talk and I think that you are right that the existing data don’t warrant the debunking argument. But I have one worry.

    You use the data from Haidt et al and Kelly et al to argue against the moral/conventional distinction. While these data clearly show that there are some cases that don’t fit easily into either category, they don’t show that the distinction isn’t viable for a huge range of cases. As far as these data are concerned, there may still be clearly moral transgressions and clearly conventional transgressions, and they may be line up on the issues of punishment, authority independence, and generality exactly as Turiel et al suggest. The data to which you appeal only show that there are some cases that fit so easily into either category. I’m inclined to think that the distinction ends up being pretty robust, but I think that you can get it off the ground with something like your 1 + 2. So, I’m sympathetic to your conclusion in the talk.

    (By the way, I am writing up some data on the M/C distinction as we speak, and I’ll send the paper along when we finish writing)

  2. edouard machery

    Bryce,

    Thanks! About the M/C question:
    1. I heartily agree that the question is still open. Paolo Sousa has a response to Kelly et al. coming out in M&L and another one coming out in Cognition (with a response by Kelly et al.). 
    2. Whether or not the M/C distinction can be defended partly depends on what you think Turiel and other are claiming:
    – I agree with Kelly et al. that Turiel and others claim that  authority-independence, universality, severity of transgression, and justification by appeal to rights, harm or justice cooccur. This predicts that there should be *few* cases where these aspects or properties of norms are dissociated. I think this claim is wrong: My empirical bet is that there are many such dissociations.
    – At the same time, it is true that some norms are judged (at least by some people) to be authority-independent and universal and that they are justified on the basis of harm or rights or justice while their violations are judged to be severe. But I doubt that this is Turiel’s claim.
    – Now, a more controversial claim would be that we think that only those norms that are authority-independent and universal and that tare justified on the basis of harm or rights or justice while their violations are severe are moral. This claim strikes as clearly false. 
    Thoughts?
    e
  3. Bryce

    Edouard,

    So, I was thinking in terms of the data that we have just collected. We ran a huge number of moral/conventional transgressions across a variety of different domains and on a huge number of participants (1500 or so, i think). What we found is that there were almost perfect correlations between authority dependence, universality, badness, and punishability for every one of the transgressions that we looked at. There was variation in where things landed across the spectrum, but it didn’t matter where things landed, judgments in all of these categories lined up.

    There is a bunch more to the data, but that will have to wait until we get it all written up.

    All of that said, my guess is that there are a lot of ways of pulling apart those various categories for particular scenarios. I think that it is an interesting empirical question where and why we get overlap in the cases that we do (ditto for where and why we fail to get overlap in the cases where we don’t)

    Finally, I think that I share your hunch that there will be cases where these things come apart but we still see a transgression as a moral transgression–but that, of course, is still an open empirical question too!

    b

  4. Bryce

    Hey Edouard,

    I was actually thinking in terms of the data that we’ve collected recently. I don’t have a sense for the shape of the whole data set yet, but the one thing that I can tell you is that we ran a huge number of transgressions in a wide variety of contexts on a very large sample. We found that *for every transgression* that we looked at there was an incredibly high degree of correlation between responses to questions about punishability, badness, universality, and authority independence.

    We also found that there was a spectrum of different judgments, some of the cases were at the extreme for being seen as moral transgressions, some sat at the extreme for being seen as conventional transgressions, and there were some that sat somewhere in the middle. But in every case, each of the measures were highly correlated.

    As for the stronger claim, I’m inclined to think that there may be ways of modulating where things fall along this spectrum; but even if this turns out to be a highly stable trend, my guess is that the data is consistent with an explanation that appeals to an evolved capacity for norm sensitivity and cultural pressures that determine where along the m/c spectrum a case will end up.

    At best, those who want to defend a strong claim about the evolution of morality have a long ways to go in defending their claims.

    b

  5. edouard machery

    Bryce,

    You write: “We ran a huge number of moral/conventional transgressions across a variety of different domains and on a huge number of participants (1500 or so, i think). What we found is that there were almost perfect correlations between authority dependence, universality, badness, and punishability for every one of the transgressions that we looked at.”
    This is very interesting and, I admit, it seems to speak against the claim made by Kelly et al. I am now curious to see the report.
    I should also note that theoretical reasons seem to weigh against the idea that we have evolved to distinguish between norms as a function of their universality or their locality. The notion of a universal human species, to which norms would apply, is a very recent one, and, according to the anthropologists I talk to, it is (or at least was) unknown in small-scale societies. 
    E

Comments are closed.