Testing Moral Progress

Mike Huemer just published his version of the familiar argument that changing moral views is evidence for moral realism. Here is the progress datum he seeks to explain:

Mainstream illiberal views of earlier centuries are shocking and absurd to modern readers. The trend is consistent across many issues: war, murder, slavery, democracy, women’s suffrage, racial segregation, torture, execution, colonization. It is difficult to think of any issue on which attitudes have moved in the other direction. This trend has been ongoing for millennia, accelerating in the last two centuries, and even the last 50 years, and it affects virtually every country on Earth. … All the changes are consistent with a certain coherent ethical standpoint. Furthermore, the change has been proceeding in the same direction for centuries, and the changes have affected nearly all societies across the globe. This is not a random walk.

Huemer’s favored explanation:

If there are objective ethical truths to which human beings have some epistemic access, then we should expect moral beliefs across societies to converge over time, if only very slowly.

But note three other implications of this moral-learning process, at least if we assume the usual (e.g., Bayesian) rational belief framework:

  1. The rate at which moral beliefs have been changing should track the rate at which we get relevant info, such as via life experience or careful thought. If we’ve seen a lot more change recently than thousands of years ago, we need a reason to think we’ve had a lot more thinking or experience lately.
  2. If people are at least crudely aware of the moral beliefs of others in the world, then they should be learning from each other much more than from their personal thoughts and experience. Thus moral learning should be a worldwide phenomena; it might explain average world moral beliefs, but it can’t explain much of belief differences at a time.
  3. Rational learning of any expected value via a stream of info should produce a random walk in those expectations, not a steady trend. But as Huemer notes, what we mostly see lately are steady trends.

For Age of Em, I read a lot about cultural value variation, and related factor analyses. One of the two main factors by which national values vary correlates strongly with average national wealth. At each point in time, richer nations have more of this factor, over time nations get more of it as they get richer, and when a nation has an unusual jump in wealth it gets an unusual jump in this factor. And this favor explains an awful lot of the value choices Huemer seeks to explain. All this even though people within a nation that have these values more are not richer on average.

The usual view in this field is that the direction of causation here is mostly from wealth to this value factor. This makes sense because this is the usual situation for variables that correlate with wealth. For example, if length of roads or number of TVs correlate with wealth, that is much more because wealth causes roads and TVs, and much less because roads and TV cause wealth. Since wealth is the main “power” factor of a society, this main factor tends to cause other small things more than they cause it.

This is as close as Hummer gets to addressing this usual view:

Perhaps there is a gene that inclines one toward illiberal beliefs if one’s society as a whole is primitive and poor, but inclines one toward liberal beliefs if one’s society is advanced and prosperous. Again, it is unclear why such a gene would be especially advantageous, as compared with a gene that causes one to be liberal in all conditions, or illiberal in all conditions. Even if such a gene would be advantageous, there has not been sufficient opportunity for it to be selected, since for almost all of the history of the species, human beings have lived in poor, primitive societies.

Well if you insist on explaining things in terms of genes, everything is “unclear”; we just don’t have good full explanations to take us all the way from genes to how values vary with cultural context. I’ve suggested that we industry folks are reverting to forager values in many ways with increasing wealth, because wealth cuts the fear that made foragers into farmers. But you don’t have to buy my story to find it plausible that humans are just built so that their values vary as their society gets rich. (This change need not at all be adaptive in today’s environment.)

Note that we already see many variables that change between rich vs. poor societies, but which don’t change the same way between rich and poor people within a society. For example rich people in a society save more, but rich societies don’t save more. Richer societies spend a larger fraction of income on medicine, but richer people spend a smaller fraction. And rich societies have much lower fertility even when rich people have about the same fertility.

Also not that “convergence” is about variance of opinion; it isn’t obvious to me that variance is lower now than it was thousands of years. What we’ve seen is change, not convergence.

Bottom line: the usual social science story that increasing wealth causes certain predictable value changes fits the value variation data a lot better than the theory that the world is slowly learning moral truth. Even if we accepted moral learning as explaining some of the variation, we’ll need wealth causes values to explain a lot of the rest of the variation. So why not let it explain all? Maybe someone can come up with variations on the moral learning theory that fit the data better. But at the moment, the choice isn’t even close.

GD Star Rating
Tagged as: , ,
Trackback URL:
  • Rafal Smigrodzki


    And yes, I completely agree with your opinion. Some additional mediators of the change in moral attitudes aside from wealth could be increasing IQ, and larger influence of global mass-media but the argument stands – a secular change in moral opinions is not evidence of moral realism.

    • Michael Huemer

      If moral realism is true, that would explain why rising IQ and larger influence of global mass-media would contribute to moral progress. Again, it just doesn’t seem to me that you’re engaging with what I’m saying.

      • Rafal Smigrodzki

        I do not have access to the full text of your article, thus I address only the brief quotation (“If there are objective ethical truths…) in Robin’s post. Correct me if I am wrong but the word “objective” seems to point towards a robust form of moral realism, including the thesis that moral propositions are not metaphysically different from other, non-moral propositions.

        The three features of modern societies we mentioned (wealth, intelligence and homogenization through mass media) may influence common ethical attitudes through diverse mechanisms. Robin points out that wealth allows some common, primitive, hunter-gatherer attitudes to be expressed with less danger to survival and in a larger number of people. Higher IQ allows better predictions of consequences, thus reducing the number of erroneous choices – the ultimate mind would reliably make only one choice, the best one. Global mass media reduce variance of opinions by reducing the variance of data on which such opinions are built. These mechanisms can explain secular trends in moral beliefs without referring to the metaphysical status of moral propositions, and fit well with other philosophical frameworks, such as moral subjectivism, or consequentialism.

        Thus, 21st century moral attitudes, while consistent with robust moral realism, are also consistent other meta-ethical views, and therefore do not provide strong evidence for either.

      • Let’s be frank: moral realism is not part of serious and intelligent 21st century ontology. ( http://juridicalcoherence.blogspot.com/2011/12/14-why-do-what-oughta-habit-theory-of.html )

        In proclaiming that liberalism is a truth, Huemer advances a position as vile and idiotic as creationism: it implies “liberal values” should be taught as truths in the schools.

    • Thursday1

      IQ is rising in highly specific ways, as Steve Sailer notes here:

      Catholic writer James Kalb has some similar thoughts, and applies them to changes in morality:

      Short version: we’re not really getting that much smarter overall, but we are thinking in more technologically oriented ways. Our style of thinking has changed.

      • Thursday1

        And interesting question is whether wealth is causing us to think in more technological style.

        Or is thinking in a more technological style causing us to be more wealthy.

        The precise causal link between all this and our changes in morality adds yet another layer of complexity.

      • No, Steve notwithstanding, we’re getting smarter – due to getting wealthier. Most of the Flynn Effect seems to result from nutritional advance, and it proceeds at a rate comparable to the increase in height.

        We’re not just learning to think differently. The proof is that the Flynn Effect applies to tests of infant IQ. Infants are walking and talking earlier: this doesn’t reflect learning a way of thinking.

      • Thursday1

        No, not all, or even most, of the Flynn Effect can be attributed to nutrition.

      • Could you point me to your main argument for so thinking? I’ve pointed to height and developmental changes that are equal in magnitude to the Flynn Effect. Note also that life expectancy has risen in this period by the same degree. (Perhaps you challenge these data.)

        My tone was too strong in my original comment (perhaps because I wanted to draw refuttal). I’m not an expert on population genetics. But Flynn seems to have gone off on a wrong tangent and never recovered: he once reasoned that perhaps the ancient Greeks and Romans were extreme retardates if you extend the Flynn Effect back. This is ridiculous, and the reason seems to be that the Flynn Effect depends on nutrition, which has improved in recent centuries, not a very long-term trend based on the development of culture.

      • Thursday1

        One would expect nutrition to affect fluid equally with crystalized IQ gains. That’s not what we see.

      • The actual findings on nutrition, including the effects of prenatal nutrition, are (predominantly) that fluid abilities are affected much more by nutrition than crystallized abilities. This is as you might expect, given the general principle of the sparing of language abilities from early insult to the brain.

        The exception to a generalization that the Flynn Effect pertains more to fluid ability is the Similarities subtest of the WAIS. But it’s an outlier, and the scoring for Similarities is particularly subjective.

      • chaosmosis

        Can I get a citation? This is really neat. I had no clue there was a potential viable explanation of the Flynn effect.

      • wikipedia has a pretty thorough treatment of nutrition and the Flynn Effect – https://en.wikipedia.org/wiki/Flynn_effect#Nutrition

        However, an authoritative article dismisses nutrition: http://www.apa.org/pubs/journals/releases/amp-67-2-130.pdf

        [Notice that Flynn is co-author.]

        This literature review maintains nutrition can’t explain the Flynn Effect because the patterns of gains don’t match any nutritional hypothesis: sometimes occurring in the lower end of the distribution and sometimes in the higher (in different countries). That this is their main reason for rejecting nutrition doesn’t seem to me a strong position, since various micro-nutrients can be involved. (Iodine has been found unexpectedly important to intellectual development.)

        The early proponent of nutrition was Richard Lynn, a somewhat notorious racialist – although the integrity of his research hasn’t been impugned. But this might impede acceptance. (It did with me – not that I’m by any means certain.)

  • Michael Huemer

    Thanks for your thoughts, Robin. However, this looks to me like an episode of talking past each other.

    By my lights, for instance, you didn’t offer any explanation of moral progress. By an “explanation of x”, I don’t mean “a variable with a statistical correlation with x”; I mean “a theory that enables me to understand why x is the case.”

    I meant to offer an explanation in the latter sense, not in the former sense; I think you intended to offer the former and not the latter. But it seems that you thought I was trying to do the former, i.e., you thought I was trying to cite a variable with a statistical correlation with moral progress. But I’m not sure what you think I’m saying, i.e., what variable you think I’m saying is statistically correlated with what other variable.

    In response to the putative predictions you mentioned, note that I suggest that moral progress results from overcoming biases (a particularly appropriate topic for this blog!), more so than gathering new data. If you get too preoccupied with Bayesianism and robots, you might forget what actual human beings are like.

    • Overcoming biases should look the same as experience or thinking in terms of the predictions I claim follow.

      Whatever else you were doing in your paper you were *definitely* pointing to data and talking about the relative abilities of different accounts to explain that data, in part by predicting that data. That is the part of your paper I’m addressing in my post. I point to predictions of your favored account that do not predict the data we see, and predictions of another account that do fit the data we see. I can’t see how you don’t think that is relevant to your paper.

      • chaosmosis

        > Overcoming biases should look the same as experience or thinking in terms of the predictions I claim follow.

        I think this assumes that the biases are just as likely to support liberalism as oppose it. If we’re systematically biased towards our in-group, as we eliminate our biases we’ll approach a more and more universal form of ethics.

      • Michael Huemer

        Yeah, I don’t get why Robin said that. I think Robin Hanson might be a Bayesian robot, and that’s why he believes the “random walk” theory. He doesn’t get the idea of bias. If you gradually overcome biases, you don’t make a random walk. You move toward the truth.

        That’s also why Robin writes books trying to prepare the world for the day when his kind will take over.

      • One can be unbiased and yet ignorant (that is our prior, unless we have reason to believe one has knowledge). Once one is unbiased, one can random-walk. The biased don’t walk randomly.

        This might come off as disrespectful, but Robin’s writings rely to a significant extent on the findings of behavioral psychology on actual human beings, not “Bayesian robots”. This empiricism is more common in the social sciences than philosophy, so right now I don’t have any particular confidence that Huemer has a more accurate model of human beings regardless of how many times he throws around the term “Bayesian robot”.

      • I think this assumes that the biases are just as likely to support liberalism as oppose it.

        No more than it assumes that new moral information of a positive nature supports liberalism.

        Of Robin’s three reasons, #2 seems neither here nor there with respect to moral progress; I don’t understand #3. But #1 is critical. There seems no reason why moral progress due to greater moral knowledge should accelerate the way moral progress seems to have accelerated (or at least is claimed to have).

        [There is reason to think, however, that scientific acceleration has removed biases faster in recent centuries, and when those biases are eliminated from moral thought, you will get change that seems like progress – simply because the change has avoided what we now know is bias.]

    • note that I suggest that moral progress results from overcoming biases

      But doesn’t this provide an alternative explanation for what you call moral progress? If there’s an historical tendency to overcome cognitive biases (which is quite plausible), then this tendency alone could explain “moral progress.” As we rip away more and more biases, the remaining biases can shape the new morality in a consistent manner. What happens when and if we eliminate all bias? Then (I would contend) there would be no morality as such. ( http://juridicalcoherence.blogspot.com/2012/04/143-unraveling-mystery-of-morality.html )

      An example. Morality has become less punitive with the rise of our
      increasingly scientific determinist worldview, which (at least in my
      book) is progress. I think the illusion of free will (
      is a bias of near-mode thought that folks increasingly are overcoming when they approach far-mode subject matter. It is intellectual progress, not
      moral progress per se, but it produces different moral sensibilities.

      This account requires no convergence on “moral truth.” (In fact, the belief in moral truth is one of the biases needing to be overcome.)

      • Michael Huemer

        Well, my claim is that liberalism represents a coherent, initially plausible ethical perspective (it’s not a mere default or absence of values), and the observed moral progress is all in line with liberalism.

      • Jonathan Haidt finds that liberalism represents a hypertrophy of certain values and atrophy of others. This fits my model where certain biases are eliminated and others strengthened.

    • chaosmosis

      One of Robin’s arguments is that a moral trend is not good evidence of moral realism because other factors than learning can cause moral trends. Even if there were no real morals, history still might see a growing consensus of liberal values over time for purely amoral reasons. Your response to points 2 and 3 seems on point, though.

      • Michael Huemer

        In the paper, I do not say that the existence of a trend entails moral realism. That doesn’t mean that it isn’t *evidence for* moral realism. “Even if ~P, E still *might* occur” does not imply “E doesn’t support P.”

        In the paper, I explain how I think moral progress works, and I discuss why other views that people have advanced are not good.

      • chaosmosis

        I don’t agree with the arguments you make in the paper against alternate explanations, but I want to set that issue aside. For the sake of this comment, I’m granting you that the alternate explanations you mentioned don’t work.

        This is still not a good reason to believe in moral realism. You’re making an argument from ignorance, saying that because we don’t know what caused increasing liberalism that should increase the odds we assign to the explanation that morality did it. The problem with arguments from ignorance is that they unjustifiably focus on one specific possible explanation. The moral realism explanation is infinitesimally strengthened by the failure of the 3 specific explanations you addressed, but so too strengthened are all other imaginable explanations. So your argument seems weak.

        If you want good evidence for moral realism, your arguments against alternate explanations need to be sufficiently general that they make all non-moral-realist explanations less palatable, regardless of their specifics. But instead you attacked the alternate explanations in very specific ways.

        The only sort of arguments I can imagine that might promote the relative plausibility of moral realism as an explanation over other possibilities are arguments which specifically look at the association between morality and learning, claiming that non-learning oriented accounts cannot explain the increased rise of liberalism. But for the 3 reasons given by Robin above, it seems like the opposite is more likely.

    • Swami Cat

      Michael, great paper. I agree with you that prosperity does not explain the shift to liberal values. That said, I disagree that there is a natural tendency for values to progress toward a priori liberal values.

      I would explain it as such:

      1). Morality is about cooperation, it is an evolved sense which allows us to prosper cooperatively within a range of social conventions.

      2). Evolution is broader than morality, and includes switching off moral senses when exploitation or defection is the dominant strategy (and it often is).

      3). Liberal morality matches quite closely (as far as I can tell exactly) to what anthropologist Boehm calls forager egalitarianism. It is the dominant ethos of nomadic foragers toward band members of their own gender. Thus it is the evolutionary adaptive ethos of the majority of our evolution.

      4). Defection and exploitation have always been the dominant strategy if you can get away with it. Hence the alpha males of Chimps. However, among well armed, nomadic foragers with communication skills, the dominant strategy is for non alphas to unite against the alpha wanna be, or even easier, just leave him and wander off with fellow “equals”. Violence was still endemic between bands or tribes because there was no way to escape the cooperative dilemma, indeed the threat of external violence was one reason cooperation among egalitarians dominated — they needed to cooperate to flourish among competing bands.

      5). Agriculture was a short term genetic cornucopia leading to a long term Malthusian trap which killed off our nomadic freedom and which fostered a class of alpha elites who could specialize in violence and thus neutralize the egalitarian cooperation of foragers. Farmers became for thousands of years the livestock of elites. Note there was no gradual evolution for thousands of years toward liberal values. No indication that Qing Dynasty was more ethical according to liberal values than the Song, or that the Romans were more ethical than the Greeks (indeed they were less liberal).

      6). Competition between hundreds of interconnected states for hundreds of years in classical Greece led to the partial rise of egalitarian values among citizen males. This lasted for centuries and spread to over half of all Hellenic city states. Ober explains why and documents how the Greeks were more prone to cooperation, but the net results were more a egalitarian ethos and more egalitarian institutions (democracy, impartial laws, markets, etc). After the collapse of Greece, we became less liberal again for several thousand years.

      7). Competition between hundreds of interconnected states for hundred of years led to similar selective pressures for enhanced within-state cooperation in Europe. An ethos of enlightenment philosophy which stresses liberal egalitarian values again comes to the forefront along with science, open access political institutions and markets. Those societies with liberal values fostering astronomically effective cooperation rose to overwhelming dominance, leading to emulation and arms races of constructive competition — competition to become more cooperative and thus more dependent upon liberal morality which is the ethos of cooperation. Successful states have markets, open access institutions and science, and these depend upon and further stimulate the liberal ethos in a virtuous cycle.

      Nothing is inevitable about it. Indeed if there is not competition between cooperative units, I would assume alphas and exploiters/defectors will again assume control and we would see a demise in liberal morality.

      In summary, no a priori morality. No intrinsic teleological pull toward morality, indeed no consistent trend at all, but rather one of phase transitions first away from then back toward cooperation. And in the end morality is our evolved sense to flourish in cooperative environments. Cooperation and egalitarian/liberal ethos are self reinforcing, but this is a matter of logic, not moral realism.

      At least that is my interpretation…

  • Ben Kennedy

    The wealth/morality link? Perhaps cosmopolitanism – when people live together, we would expect both rising liberalism due to simple proximity, and rising wealth due to increased specialization and efficiency. Isn’t it common in most countries to have the (relatively) rich/expensive liberal cities and the poor backwoods bumpkins?

  • I would argue that the supposed trend is also not as consistent as supposed, since the rights of unborn children, God, and dead people have been tending to decrease recently. I suspect that this supports Robin’s account.

  • Michael Huemer

    Robin, I think we’re talking past one another again. I thought you were trying to offer a better explanation for moral progress. I was then suggesting that I don’t think you offered a better explanation, because I don’t think you offered an explanation. That’s because I don’t think you gave a theory that would enable me to understand why there’s been moral progress. From what I can tell, you just said that wealth is correlated with morality.

    I wasn’t saying that your message was completely irrelevant to my paper, or that correct predictions don’t matter.

    Now, I don’t see that you cited any genuine mistaken predictions of my view. Consider your three numbered points:

    (1) It wasn’t clear to me whether this was supposed to be a mistaken prediction or not, because I couldn’t tell whether you think that there has been a lot more thinking and experience lately, or not. But now forget that, because this is what matters: cognitive progress in many areas has accelerated in modern times. Given that, if morality is cognitive and moral progress is cognitive progress, I would also predict that moral progress would have accelerated in modern times … which it has.

    (2) Not sure I followed the import of this. Moral learning is indeed a worldwide phenomenon, and I didn’t claim to be explaining belief variation at a given time. So I’m not seeing how this was supposed to be a criticism (if it was?).
    (3) The best I can figure is that this one depended on the assumption that we’re Bayesian calculators updating on new data by conditionalization, rather than humans gradually overcoming biases.

    • On #3, agents trying to overcome biases, if they do that in a rational manner, should follow a random walk in their sequence of estimates.

      On #2, the point is that there is a lot of variation at a time that isn’t explained by your account, but is explained by wealth causing liberal moral views. And if we are going to invoke that explanation there, then why not just apply it to the variation across time and be done?

    • On whether I offer an explanation for moral progress, we are trying to compare explanations of moral CHANGE. One explanation is moral progress, as you suggest. But another explanation is that humans are built to vary their moral views with their wealth. That can also explain many of the variations in moral views across time and space. I don’t see why that can’t count as valid explanation as well.

    • To elaborate on #3, I could show you formal models in which an agent rationally deals with the possibility of their being biased in certain ways. In that model, beliefs follow a random walk. Can you show me a formal model of your concept of beliefs changing in light of finding a bias?

      • chaosmosis

        Would such a model depend on the agent anticipating and attempting to compensate for the possibility of future biases as yet unknown? Without this mechanism, I’m unable to imagine why a random walk might occur.

  • Michael Huemer

    Another possible empirical test: If moral progress is a matter of figuring out the correct moral views, and liberalism is the correct moral stance, then this might lead us to expect that intelligence, education, and rationality will be correlated with liberalism. Furthermore, I suppose that this correlation should be stronger than the correlation between wealth and liberalism. Is that true?

    • Intelligence and rationality seem reasonable predictors to consider, but education seems too close to enculturation. Some subcultures that can afford more education may happen to have certain view tendencies.

      • Thursday1

        Intelligence and rationality aren’t the only things though that lead to correct conclusions. Things like perception matter too.

      • Thursday1

        We have to worry about certain kinds of rationality making us less perceptive, dumber in other ways.

    • If progress is about figuring out correct views, would you expect that process to take place over an individual’s lifetime, and thus be greater in the elderly?

      • That seems just as fair as a data point as IQ or rationality correlations.

    • Rafal Smigrodzki

      They say that if you are not a socialist as a teenager, you have no heart but if you are still a socialist in middle age, you have no brain. Indeed, as we age, there is an empirically observed progression in moral views, away from attitudes commonly termed “liberal”, and towards “conservative” ones. Empirically, intelligence is anti-correlated with liberalism in the US, in the common 21st political group meaning. We have thus two data points indicating that liberalism is associated with lack of moral reflection or lack of intelligence.

      I am not quite sure what the term of art “liberal” means in this discussion (again, regrettably, I do not have access to the full text of your article) but I have the impression it may have many odious connotations to commenters here, as it nowadays commonly refers to radically intolerant, reactionary and socially corrosive attitudes. You may want to clarify this term for our benefit.

      • IMASBA

        “They say that if you are not a socialist as a teenager, you have no heart but if you are still a socialist in middle age, you have no brain. Indeed, as we age, there is an empirically observed progression in moral views, away from attitudes commonly termed “liberal”, and towards “conservative” ones.”

        But this is commonly explained as moving from a state where you mostly have things to gain to a state where you mostly have things to lose, not as an increase in intelligence. The transition occurs when people get children, large financial obligations and lots of material possessions, regardless of their age. Interestingly the very old can (I’ve seen this a couple of times) swing back to more liberal views after they’ve paid off their debts, know their kids are doing alright and in general know they don’t have to take anyone’s BS anymore when they’re secure and long since retired.

  • I’m tempted to call ‘moral progress, US-EU-Facebook-mainstream 2015’ a self-fulfilling social dynamic without any ‘random walk about the truth’ to it. Unfortunately your (very informative) ‘here’s what we’d see [the opposite of a trend; oscillations around the truth] if people were really truth-seeking’ doesn’t tell me whether the moral truth is (mostly) ahead of the trend or behind it. So really we have some issues on which the public signaling race is out of control and even reversing the direction would be an improvement (though reversed stupidity is not optimal), and some where the old-guard still need to die off. The point that people around the world would be converging on moral-truth-agreement is decent – maybe in the long run, but that need not be true today; there really are backward nations out there (in an objective material sense, not just “we haven’t forced them to fall in line with our norms yet”). Overall, you lay a couple theoretical supports for the skepticism about modern ‘moral progress’ and ‘single objective moral truth’ (which at least seems mostly hypocritical/empty). Anyway today it’s easy to imagine we’re far from Pareto-optimal; why not just negotiate consensually better agreements, instead of fighting ‘moral truth’ holy wars?

  • Recall Feynman’s complaint about empirical physics: published measurements of fundamental constants had errors more centered around the original (randomly wrong) high-prestige groundbreaking measurements than the truth. It took a generation.

  • Michael Huemer

    I want to add a clarification to my view of moral knowledge (since I suspect that not everyone in the comments thread has read the paper!). Individuals don’t all figure out morality by themselves. Most moral beliefs of most people come from their culture. (Compare the fact that most of my non-moral knowledge also comes from my culture. I didn’t figure out myself that the Earth is 4.5 billion years old.) As long as your culture is generally reliable, it counts as knowledge.

    And our culture develops over time to become more reliable. The mechanism causing this development involves smart, rational, morally sensitive individuals trying to improve their society’s practices — but even those individuals get most of their moral beliefs from their culture. That’s okay. (Compare the fact that even scientists who are advancing scientific knowledge have inherited most of their own scientific beliefs from their society.)

    • Ben Kennedy

      In your conclusion, you state “Purely cultural accounts of the source of morals leave us at a loss to explain why the culture itself has moved in a given direction over time.”

      Two things that move liberalism forward are economics and empathy. On the first point, Libertarian economists are always very pleased to note how racisim is expensive and punished in the market. When liberal behavior leads to greater profits, we see more of it. On the second point, empathy is a base human emotion. The concept of justive is tied to victims, and victims are recipients of empathy.

      This really great Joe Posnanski piece I think illustratates these concepts (http://joeposnanski.com/a-baseball-story). Bobby Bragan was a backup catcher on the Brooklyn Dodgers when Jackie Robinson joined the team. Bobby had to decide between indulging his racism and getting a paycheck. He chose to stay (+1 economics). Then, over time, he directly observed that Jackie Robinson was a perfectly nice human being and Bobby began to react defensively when his friends would disparage him (+1 empathy). His liberal transformation is perfectly explainable by ordinary forces.

      On a different note, I’d like to point out on page 5 you seem to speak derisively of the pro-life position, using the term “schliberalism” and lumping it in with some kind hedonism and promise-breaking. Perhaps you could elaborate here. If anything has convinced me of non-cognitivsm, it is this issue. My empathic response to the millions of abortions in this country is deep and profound, yet according to my facebook feed I am enslaver of women or otherwise horrible person. As a society, we can’t even agree on the moral permissibility of killing the unborn. If moral realism is true, then a whole lot of really intelligent people are in massive cosmic moral error. The alternative non-cognitive approach is that broad segments of the population are having different and non-erroneous responses to the unborn and mothers. This is the one that gets me through the day

      • Michael Huemer

        Ben, that passage was not meant to attack pro-lifers. “Schliberalism” was simply introduced as an example of a “theory” that is just a random collection of unrelated moral views.

        The ethics literature on abortion is complex and subtle. Issues about the source of rights, the criteria of personhood, the moral import of potentiality, and criteria of personal identity all enter in. On a theoretical level, it’s a very interesting and difficult issue.

        On economics and empathy, I’m not sure why economics and empathy would not have operated effectively much earlier in human history.

        I also think economic theories are just ignoring what we know about the actual history. If what happened in the 1960’s was that non-discriminatory businesses started expanding due to being able to hire better/cheaper workers, while racist businesses started shrinking due to having to charge higher prices, and the racist businesses started going out of business or being bought out by the non-racist competition — then the economic explanation would be correct. But we know that that’s not what happened.

        Come on, people. It wasn’t that long ago — we know what the motivations for the civil rights movement were. We know that it was a moral argument that won the day. The most obvious thing to say is that the civil rights protesters had the better moral case. This just seems like a “well, duh” moment. Why is everyone bizarrely struggling to avoid admitting that?

      • The most obvious thing to say is that the civil rights protesters had the better moral case. This just seems like a “well, duh” moment. Why is everyone bizarrely struggling to avoid admitting that?

        It’s you who is failing to “engage” with critics and misrepresenting their arguments. Nobody need deny that civil rights won the moral argument. What’s medieval is your tacit premise that the moral argument implies moral truth.

      • Ben Kennedy

        Thanks for the clarification on “schliberalism”, that makes sense.

        “On economics and empathy, I’m not sure why economics and empathy would not have operated effectively much earlier in human history.”

        People can only be empathic to what they are aware of. In order to have muckrakers to successfully agitate for social change, they need a platform for mass distribution of ideas, and a literate audience – both relatively recent developments. The quick change on gay marriage could be a consequence of faster and cheaper communication

        “The most obvious thing to say is that the civil rights protesters had the better moral case. This just seems like a ‘well, duh’ moment. Why is everyone bizarrely struggling to avoid admitting that?”

        That’s not your argument – one data point proves nothing. Sometimes society converges on the “wrong” answer, like eugenics. If I understand you correctly, you are looking at trajectory and convergence across societies as evidence of moral truth. But when liberalism and wealth are highly correlated, maybe there is another answer. I suggested cosmopolitanism, which has predictable consequences of both increased wealth and increased social tolerance. Why go to cosmic moral truth to explain things?

  • zarzuelazen

    Wealth is certainly correlated with values, but where did all this wealth come from in the first place? Is wealth produced out of thin air? It seems that changes in values have to be *prior* to leaps in wealth creation, because the values of a society have to be compatible with conditions in which wealth can be created.

    I’m reasonably convinced there are universal (‘platonic’) values that all minds capable of moral reflection have to agree upon.

    But these take the form of abstract virtues or ideals, rather than concrete prescriptions. For example, I now have overwhelming evidence (I have greater than 90% confidence) that ‘Beauty’ is the most fundamental platonic value.

    [Note: How does the above relate to Bostrom/Yudkowsky’s orthogonality thesis: I’m still not sure about that, I must admit , because it certainly still seems possible to build unfriendly super-intelligences’ that are not capable of moral reflection. So moral realism still doesn’t seem sufficient to save us from UFAI]

    By the way Robin, as I’ve been saying on your blog for over ten years, the Bayesian framework is definitely *not* the correct ideal for reasoning (again, I have greater than 90% confidence about this).

    Even respected ‘Less Wrong’ experts are now starting to suspect I was right all along. Listen to Wei Dai:

    “lately my faith in Bayesian probability theory as an ideal for reasoning (under logical omniscience) has been dropping a bit, due to lack of progress on the problems of understanding what one’s ideal ultimate prior represents and how it ought to be constructed or derived. It seems like one way that Bayesian probability theory could ultimately fail to be a suitable ideal for reasoning is if those problems turn out to be unsolvable.”

    For moral reasoning, I think categorization/analogical inference is more fundamental than probability updates. Again, I’ve stated as much in the comments on this blog multiple times over the last ten years.

    The question is where does wealth come from in the first place. And the answer is from the mind. So the changes in values have to be prior to the wealth creation.

    • Michael Huemer

      On Bayesianism: I think the relevant question here is not so much whether
      it is an adequate account of rationality, but whether it is an adequate
      account of _actual human psychology_. Even Robin should agree that the answer to the latter is “no”.

      Years ago, Robin wrote a paper about how Bayesian agents should be able to agree with each other on everything, as a result of exchanging information about their degrees of belief. This doesn’t seem to happen with real people. I thought Robin would draw the inference that actual humans are not Bayesians (never mind whether they should be!).

      Regarding economic progress and shifts in values: both have been going on for so long that I don’t think you can say one of them is temporally prior to the other.

      Maybe you (zarz) just mean that value shifts are causally sustaining economic growth. I suspect that there’s some two-way causation going on. Prosperity might lead to more people going to college, for example, where they are more likely to absorb enlightened values . . . which in turn might make them more productive.

      • Bayesian models at least make concrete predictions, including about bias. It is fair to offer another concrete model of bias to consider. Not fair to say “non-Bayesian” and then just make up any predictions you want.

  • Sam Dangremond

    I can think of one moral issue where evolutionary pressure has resulted in a “less liberal” perspective over time…


    • IMASBA

      It’s an interesting thought, but probably not true in this case. Abortion mostly just delays a woman having children. Also, just like there aren’t many atheists in foxholes, historically there haven’t been many faithful among those who had a major problem that could be solved by an abortion. It could be that the modern welfare state will subsidize the ultra-religious’ uncontrolled breeding (this is happening in Israel, although there is increasing popular resistance against it as the effect on the state’s budget increases) enough for such effects to become real in the future but time will have to tell.

      • Sam Dangremond

        “Abortion mostly just delays a woman having children.”

        I don’t think this is the case.

        Example: ” the absolute effect of abortion on fertility … is often substantial. In most of the countries examined, the TFR would have been from about 20 percent to nearly 90 percent higher than it actually was (other things being equal) had no induced abortions been performed. ”


    • Why hasn’t evolutionary pressure produced a “less liberal” perspective on contraception?

      • Sam Dangremond

        It already has.

  • Rational learning of any expected value via a stream of info should produce a random walk in those expectations, not a steady trend. But as Huemer notes, what we mostly see lately are steady trends.

    Hasn’t humanity experienced increased moral randomness (expressed in deep moral differences within societies)?

    • A random walk over time is a very different as a prediction that variance at the same time.


    Robin, I too think it’s mostly a case of genetics (of course “discovering” moral truths would also require some genetic basis deep down, but that’s not what you and I mean): modern liberal values are the forager values that were actually the norm for most of humanity’s existence. The farmer era is the aberration, not the norm! It’s not that we are exceptionally wealthy and safe now, it’s that we were exceptionally poor and unsafe in the farmer era! We don’t need flatscreen TVs and nuclear deterrence to be liberal: we need a basic feeling of mental freedom, physical safety and economic security to be liberal, we happen to have these ingredients in some dominant, wealthy parts of the world. If a dystopian, cutthroat em-nightmare is to be the future then we will lose at least some of those requirements and we will become more brutal again.

    I do wonder though if modern liberal society is really that strongly linked to just our wealth. Perhaps it was simply the most liberal societies of the past centuries who used that liberty to develop technologically (science requires elementary freedom of thought and speech and an openness to new ways and ideas) and become the major/superpowers of the world, in doing so spreading liberalism through soft power? If the Nazis or Soviets had conquered the world, or if China, Russia, etc… become dominant would their regimes inevitably become more liberal over time as a result of wealth, or would it have been perfectly possible for their autocratic nature to remain for an extended period (proving that wealth does not create liberalism single-handedly)?

    • As I said in the post, we see DETAILED correlation between liberalism and wealth, across nations at the same time, and within each nation over time. Your selection theory doesn’t predict that.

      • IMASBA

        What your text seemed to imply to me is that a nation that is wealthy compared to its contemporaries tends to be more liberal, this does not conflict with the selection process I theorized (and which I don’t necessarily believe in, I just thought it was an interesting perspective, based on the last 5 centuries, since it’s very difficult for a non-expert to know how liberal nations really were before that: Romans are often characterized as decadent and liberal while their moral philosophies were actually quite conservative, even for their time, it’s difficult what the actual balance was in ordinary people’s lives).

        If you also meant to imply that absolute wealth matters then that would conflict (and an empire under Nazi-rule would inevitably become liberal over time), did you mean to imply that?

      • Thursday1

        Liberal values are like forager values in some ways. Foragers are often quite hostile and suspicious of outgroups, for example. That doesn’t sound very liberal to me.

      • Agreed, we are quite non-forager-like at work, for example.

      • IMASBA

        We STILL are quite non-forager-like at work. Naturally there will be some limit to how forager-like we can become while still maintaining an advanced technological base, but I really doubt we’ve reached that limit already and I know for a fact American society hasn’t reached it yet because Western Europe is more forager-like and is still standing.

        “Foragers are often quite hostile and suspicious of outgroups, for example. That doesn’t sound very liberal to me.”

        Foragers have limited knowledge about outgroups, but do know those outgroups are likely to have farmer values and have more advanced weapons. Their suspicion is hardly a sign of some irrational xenophopbia. Likewise, you don’t cease to be a liberal when you fear invasion by an outgroup with hostile values and superior military strength (case in point, the liberal Hong Kong protester’s fear of mainland China).

    • The farmer era is the aberration, not the norm!

      There’s evidence (I haven’t seriously evaluated it, but it seems to make sense) that the process of human biological evolution was greatly accelerated during the farmer and industrial eras. This would imply (although I haven’t seen it so applied) that farmer values might have been reinforced genetically.

      It’s also curious (and to me surprising) that twin studies find left-right ideology to be almost entirely inherited. (The results are so extreme that I suspect they’re flawed, but the importance of heredity seems clear). This suggests to me that any truth does not align with the modern distinction between conservative and liberal – those may be mere hereditary biases.

      • IMASBA

        “It’s also curious (and to me surprising) that twin studies find left-right ideology to be almost entirely inherited.”

        And still societies as a whole become more liberal. Almost no one today wishes for the return of slavery, even though if you had done a twin study in the 1850s you would have found support for slavery to seem like an inherited trait. What is genetically inherited is a penchant for fearing change and for fear in general, this leads people to choose the conservative side of an argument when given the choice, I don’t know whether this genetic predisposition has become more common since the stone age, it’s not impossible, I suppose. What exactly constitutes the conservative side changes over time, may even reverse as things that were once hallmarks of progressivism become entrenched traditions. By definition, the liberal (or more accurately: “progressive”) side has to change over time as well and at some point in the future farmer values may actually be considered “liberal” because they would be a change with respect to the established values of the day.

        So yes, forager values will not exactly map to what is considered “liberalism” in every time period and in every place, but forager values do map pretty well to the broad strokes of modern liberalism that Huemer mentioned.

        A great deal of confusion is added by the term liberal having become a placeholder for “progressive” in the United States. Essentially, forager values map well onto the European definition of liberalism. This reminds me of those experiments where chimpanzees would not accept food if another friendly chimpanzee didn’t get food as well.

      • gda

        “And still societies as a whole become more liberal”

        Thats what entropy is – a gradual decline into disorder.

      • jhertzli

        Societies become more liberal because victorious movements are reclassified as liberal.

        Jackson’s racism was a left-wing movement.

        Lincoln’s anti-racism was a left-wing movement.

        Wilson’s racism was a left-wing movement.

        Truman’s anti-racism was a left-wing movement.

        … and we have always been at war with Eurasia.

  • Trimegistus

    Plenty of things our ancestors would consider as proof that we’ve degenerated morally: Abortion. Contraception, even. Promiscuity. Basically all the Seven Deadly Sins (except for Anger) have become modern virtues, and all of the Cardinal Virtues are derided as irrelevant. Patriotism is derided as “nationalism.”

    The worst moral inversion of all: most people can’t even imagine that things might be otherwise.

    • gda

      The trend towards leftism is best explained by the fact that the universe tends towards entropy. In the big picture, as in life, entropy is a bad thing.

      Interestingly, conservatism is a force against entropy.

      • Thiago

        So is formaldehyde. Lenin’s body is well-conservated.

    • Thiago

      “The worst moral inversion of all: most people can’t even imagine that things might be otherwise.”
      Oh, we know they can be otherwise: serfdom, racial segregation, slavery, book fires, colonial empires, wars of conquest, torture, inquisitions, religious wars (Catholics vs Protestants), blood libel, the Index Librorum Prohibitorum. Maybe we are just being luck for now.

  • sleepmon


    The trend towards leftism is explained by the fact that leftism tends to increase the influence of an individual and their personal utility, and so more influential people spread more leftist ideas in turn etc… See Soviet Russia

    I don’t know where you get the idea that this is moral progress. Telling people what they want to hear tends to increase the influence of an individual but that is in fact often less moral than telling the truth about the matter.

    Morality that creates bubbles of cooperation that are eventually exploited by defectors is an inferior morality compared to the morality that protects itself from parasites to preserve cooperation.

    Scientific ideas can be tested and falsehoods eliminated. Clearly modern moral beliefs are being put up to a selection process but its not clear that what they are selecting for is conducive to long term welfare.

  • redlobster

    It could just as well be that wealthier societies can afford more immorality without failing. Every one choosing a luxury doesn’t make it a virtue.

    • IMASBA

      In many cases “immorality” (as in “it’s immoral because a book written by a bunch of violent tribal bronze age sheepherders says it’s immoral”) actually makes economic and strategic sense.

    • Thiago

      “Immorality” such as democracy, universal suffrage, abolishing colonial empires, banning slavery, drastically curtailing torture and avoiding wars of conquest.

  • Pingback: Why They Don’t Return | Entirely Useless()