Friday, March 08, 2013

Science cannot answer moral questions II

I wrote about the same topic three years ago and probably at many other places, too. However, Sean Carroll revived the topic again,
Science, Morality, Possible Worlds, Scientism, and Ways of Knowing
and because I still agree with him, let me review and add some relevant arguments.

Carroll says that scientifically meaningful propositions are questions about the past, the present, the future, or the eternal laws that
  • might in principle be both false and true
  • admit a method, at least in principle, to evaluate their truth values.
He mentions three examples. Oxygen is distributed throughout the body by circulating blood. The Universe is expanding. The Sun will run out of the thermonuclear fuel in several (7.5 or so) billion years. They happen to be true.

He also gives you examples of propositions that don't belong to science because one of the disqualifying conditions below holds:
  • they're purely mathematical in character so they require no empirical input at all
  • they're statements about fictional objects such as Hamlet that can't be decided from the only available data, in this case the text of Hamlet (there's no "real Hamlet" offering "additional data")
  • they depend on subjective opinions and preferences
The last group happens to contain not only the people's favorite ice creams but also most moral questions, whether or not someone likes the fact that morality is in the same category with ice creams.

I completely agree with all these propositions. It may be useful to discuss some additional opposition. Many people who tend to say that morality may be derived scientifically are inclined to define morality as something positively correlated with the principle
We should work to maximize the well-being of conscious creatures.
Well, not everyone agrees with this thesis – count me as an agnostic here – but even those who do agree can't transform this assumption into science because pretty much by construction, this thesis (about the rightness of improvements of the well-being) is as vague and subjective as the original one (about morality in general).

How do you quantify the well-being of conscious creatures? The first problem is that even the "conscious creature" isn't really well-defined. This point is ultimately nearly equivalent to the deep moral disagreement of people of different religions or ideologies. I am referring to women's rights, animal rights, savages' rights, infidels' rights, poor people's special rights, homosexual rights, humanrightism in general, and so on. Different ideologies inevitably differ in the weight they attribute to different subjects' right to contribute to the "body of conscious creatures" and in the evaluation of their right to enjoy "well-being in their characteristic way".

Second, even if you knew who is a conscious creature and how important this creature should be in the definition of morality, it's still true that their well-being is not measured in meters, seconds, kilograms, joules, kelvins, electronvolts, candelas, and not even in dollars. Some people may decide that well-being is very close to some of those – the more sensible people could express it in dollars but the really unhinged ones may use an arbitrarily crazy unit, e.g. one negative gigaton of carbon dioxide, to measure the well-being of conscious creatures.

In science, we often lack the "best definition" of a concept – for example, we still feel that there might be some universal equations of string/M-theory – and it's therefore sensible to expect that certain things will become not only scientific but rigorously defined, high-precision concepts in the future.

However, pretty much by its basic character, this is not the case of morality. Imagine that you find some quantity M encoded in the equations of M(orality)-theory in the future and you will claim that it measures morality. It may be given by a pretty formula. I don't know whether it would be a local or global quantity – none of the options makes too much sense – but there will be a nice formula.

The problem is that even with this nice and well-defined formula, one may always legitimately refuse such a measure of morality and choose a completely different one. We may consider M as the measure of morality to be misguided for particular reasons. We may even consider the very idea of converting morality into a mathematical expression to be deeply immoral. At most, M would be another competitor of the Bible and Quran and Marx's books.

We may look what's wrong with various attempts to quantify the well-being. For example, Hugo Chavez died and some people – including some politicians who are supposed to represent the people of Europe but they behave shamefully instead – have argued that Chavez has been a great politician who has improved the well-being of the people of Venezuela. Now, I must say: give me a break with this communist garbage!

Venezuela has lots of fossil fuels and a great potential but what we see is an oppressed country with B-class ratings, unexplainable debt that is already at 50% of the GDP. They're 177th in the "easy of doing business rank" which is next to catastrophic. Lots of people remain unemployed, lots of people are below the poverty line. Relatively to other excessively oil-rich nations, Venezuela's GDP per capita is low.

The White Town's 1997 song, Your Woman [I could never be your woman], has made it to Lumo's all-time 100 hits. Because it's a song by a straight girl in love with a lying, two-timing, fake-ass Marxist who has manipulated her, I dedicate it to Hugo Chavez – the girl is a metaphor for the Venezuelan nation – and wish him a smooth and uninterrupted trip to Hell in whose existence I temporarily believe just for this special occasion.

Chavez's posthumous children may refer to some special indices created by the United Nations. That's great but these indices are artificial pieces of propaganda composed by Marxists and by the terrorist friends of Chavez, Ahmadinejad, and their Western politically correct apologists – all these bastards de facto "own" the United Nations. There's nothing credible about those things. They're deeply immoral.

Many of these indices – attempts to quantify the well-being of conscious creatures – incorporate egalitarianism as a positive condition. The smaller the wealth inequality is, the better, these commies want to argue. Give me a break with this junk. There is nothing nice or ethical about egalitarianism and there is nothing wrong about wealth inequality. Wealth inequality – the concentration of capital – has been both an inevitable and by definition fair and just consequence of people's varying contributions to others (plus some fluctuations that are explained by less noble causes but there's no reason for the total inequality to equal zero); as well as a necessary precondition for the progress of the human society.

People who back egalitarianism are downgrading the humans into cattle of a sort, cattle that is measured either in kilograms or pieces. Mass brainwashing and indoctrination of the citizenry becomes pretty much another "must" for everyone who wants to create an egalitarian society. When people are treated in this way, they obviously become less conscious, less thinking. People actively supporting egalitarianism are inhuman hyenas who should be tried for crimes against humanity and perhaps executed if the well-being of the conscious creatures is our goal. Now, I can't give you a scientific proof – neither a rigorous mathematical proof nor a rock-solid piece of scientific evidence – that the previous sentence is true. What I say is still true although it's true in a different sense than scientific propositions.

Some people who suggest that moral questions may be settled by science are usually narrow-minded folks who agree with a couple of their friends – or a mass movement of a sort – and because these friends or movements also tend to pay a lip service to science, they identify science with the morality. But even if their support for science went beyond the lip service, it would still be incorrect to identify science and a particular kind of morality just because both are supported by the same group(s) of people!

At the end, it seems to me that the people promoting a "scientifically established morality for everyone" haven't even tried to think about the immense diversity of viewpoints, values, and priorities that make the moral opinions of the mankind across the space and time so non-uniform. Let me give you an example.

Last night, our Pilsner soccer team, Victoria, was beaten 1-to-0 by the visiting Fenerbahce Instanbul, a very rich team supported by 1/3 of the Turks worldwide. It was a sort of surprise for our celebrated team but the Turks were playing better than us – and therefore much better (and in a more organized way) than teams such as Napoli. But that's not what I want to talk about. ;-) Twenty minutes ago, I bought and ate a kebab box in my favorite nearby kebab shop. They raised the prices today, bastards! But the prices are also not the topic here.

What I want to say is that these Muslim nations are slaughtering animals in the halal way – it's their version of the Jewish kosher way. They find blood to be unacceptable in food and the execution of the animals corresponds to these principles. The Egyptian guy has warned me about that many times. I checked – and yes, the halal slaughtering involves cutting the animal near the throat and discharging pretty much all the blood from the body.

It looks cruel, I want to cry, vomit, or whatever, but I was nearly killed 16 years ago in almost the same way (and it was my birthday) so it's no real science-fiction (some seven stitches or so did the job). Moreover, the European ways to slaughter animals seem terribly cruel to me, too. And at the end, even if you make an animal (or human) sleep so that it doesn't realize that it dies when it does, I still find it horrible. Differently horrible, creepy, but still horrible.

If you believe that some of these methods are particularly moral or immoral, you may feel extremely strongly about it but as long as you stay rational, you must realize that your feelings (and even the "right" suggestive analogies between things) are emotional, formed by your education, background, experience, perhaps genes of your family and nation, and it's completely possible that someone feels differently or very differently. These preferences are like the preferences for the best ice cream. I am surely not saying that all viewpoints are equally good; of course that only mine are any good. I am just saying that the validity of a moral system can't be proved in the same way as statements about hard sciences.

I want to mention one more alternative albeit related definition of immorality to compete with the well-being:
A habit, tradition, behavior, or even opinion that threatens the very fabric of the society is immoral.
Again, one could get many people subscribe to this thesis. It's perhaps more conservative than the morality based on the well-being (the latter smells of a progressive bias – it is not shocking that it was picked by Sean Carroll) but it's not too different. The problem is that the "fabric of the society" is inevitably so much ill-defined that there can't ever be a rigorous "scientific verdict" about morality based on this criterion, either. Moreover, whether or not the society will be undermined by a particular behavior is not only "unpredictable deterministically" but it is really unknown in advance. Even if the "fabric of the society" and its preservation were rigorously defined, which is too much to ask in the real world, we could at most calculate the probability that this fabric will be destroyed because of a particular behavior.

This probability would be a number between 0 and 1. Which threshold would be tolerable? There's clearly no objective, unique, canonical answer. Moreover, the calculated probability depends on many other assumptions.

An example. During the life of Giordano Bruno, many powerful Catholics believed – unfortunately for Bruno – that heliocentrism was undermining the very fabric of the society. If people started to believe heliocentrism, something terrible would happen. What would the Armageddon be? Perhaps, the inertia from the floating Earth would throw us to outer space. Well, that's not really possible because the Sun's gravitational force acts equally on the Earth and the objects on it which is why the centrifugal and centripetal (gravitational from the Sun) forces pretty much cancel, up to tides, but yes, the understandable misunderstanding of this point (those people lived before Isaac Newton and his laws) was a reason why many people couldn't internalize the physics of heliocentrism.

But of course, the actual worries were different. It was believed (although some people had to promote this belief first: this anti-heliocentric hysteria wasn't dominant during Copernicus' life) that the Bible stated that the Earth had to be the center of everything and it was a key proposition. So any violation of this wisdom would undermine God, the body of His or Her believers, and their faith. They would quickly begin to sleep with their siblings, eat each other, Mr or Mrs God could get pis*ed off very quickly and very severely, and roughly speaking, the mankind would go extinct very rapidly.

This was the ideology behind the belief that heliocentrism was immoral and Giordano Bruno had to be burned at stake. Well, this worry hasn't materialized. The mankind no longer believes geocentrism much. Even the Catholic Church has largely abandoned it. But the mankind still exists. Even the Catholic Church is still around. It still has a Pope – well, not right now, but maybe it will have one next week.

So some belief about the "collapse of the human society" because of something turned out to be immaterial. There are lots of wrong beliefs – wrong because not sufficiently scientifically backed – underlying many or all ideologies that claim to determine what is moral and what is immoral. This is the easy part of the problem. I think that everyone understands that the worries turned out to be misplaced which means that this particular reason to promote execution of Giordano Bruno for heresy was bogus. The society and even the Church may work independently of beliefs in geocentrism or heliocentrism.

But I also want to emphasize a different point, one that is easily overlooked with the hindsight. And the point is that the moral arguments of the Catholic prosecutors had a logic and they could be right. The coupling between the fate of geocentrism and the fate of the mostly cultural Western society was weak but it was nonzero. The probability that the loss of the Church's authority would turn the Western nations into hordes of savages was arguably nonzero. More importantly, there are other situations in which the coupling is perhaps stronger and similar worries turn out to be justified.

In some cases, the society adjusts to some "potentially crippling" changes. In other cases, the changes may be so crippling and unadjustable that the society gets crippled, indeed. When the society gets adjusted, it's a slightly different society – we could say it is a mutated society – afterwords. For such a mutated society, someone who claims to quantify or "scientize" morality has to adjust all the rules, too. This quickly becomes an unsolvable problem where everything is coupled to nearly everything, all the coupling constants are unknown, their effects' morality is undefined, and it's just a mess.

One method to appreciate how unscientific all propositions about morality are is to try to apply your moral rules to some very distant societies – ideally societies that are separated by thousands of years (or more) from ours. If you're the best morality expert in the world of 2013, how will you teach the Neanderthals to be ethical? What about the future people who live in the Idiocracy of the 26th century? Clearly, the folks in both distant worlds live completely different lives and have different interests, concerns, and priorities. Even the notion of murder may get a totally different flavor due to different needs, different technologies to kill, and so on.

Your moral rules will largely mean nothing to them. Even if you strongly believe in your moral rules, you shouldn't overlook that they're provincial rules that are largely inapplicable to general enough situations – situations in which the people are too far spatially, temporally, or culturally. And that's an attribute of something that is decidedly unscientific because the scientific laws should be pretty much constant in time. Otherwise they're not really laws. The laws of moralities are inevitably descriptions of a subset of people's subjective opinions that are relevant and believed to be valid in a limited region of spacetime and the society and that may only be demonstrated to be "clearly better" than competing systems of morality if the two systems sort of fairly compete and show where their beliefs go.

However, even if you could make the competition fair, you will still have trouble with this criterion because a society may thrive due to good luck or some (genetic and other) prerequisites that are independent of the moral system so no victorious outcome is a clear proof of moral propositions, either. Moreover, you cannot step into the same river twice. Important enough "moral experiments" are unrepeatable (because the society or at least its knowledge is changing) and fail to obey the scientific requirements of reproducibility.

So I surely think that science will never become adequate to settle moral questions in practice – except for totalitarian societies where all the people are forced to believe certain things and nothing else – but I also think that moral questions are outside science even in principle.


  1. Unfortunately, a bias can be held by the very life one has life then can be perceived as "black and white." In a religious sense, a heaven and hell?:)


  2. The "Trolley Problems" that Sean Caroll mentions is a bit lame. Nobody would kill his son or daughter to avoid killing a crowd of people !

  3. How is "chocolate is your favourite ice cream" not a scientificly meaningful proposition?
    It might in principle be true or false.
    It admits methods to evaluate its truth value. (Say "Lobos, what is your favourite ice cream?" and listen to the answer, observe which kind you chose, watch the expression on your face when offered different kinds, etc.)

    Knowing the answer has more practical applications than for "Are there particles with mass less than 10TeV not predicted by the Standard Model?" You might not think the question is as important (I don't), but some people (who want to sell you ice cream) might .

    Of course, "chocolate is the best ice cream", and "One should follow the Klingon code of honor", are another matter.

    I can't conceive of any way to test those, I don't even know what it would *mean* to test them.

    People once thought that the heavens were unknowable too. There is nothing *logically* inconsistent about the existence of a fundamentally objective morality, that no one could conceivably disagree with, but none has been found (and not for lack of trying), and there is good reason to think it impossible. I'd put the probability in the same range as for "I am a Botzman's brain" (not zero but too low to be worth considering).

  4. The issue of morality is fundamentally about how humans interact with one another; as such, some would say is falls in the realm of "social science". Of course, the phrase "social science" is itself an oxymoron, as there is little or no science involved in such studies. So, morality has little to do with science.

    Those who maintain that there there is anything scientific about morality remind me of the old communists who would claim that communism was "scientific" and other such rubbish. Such claims sound remarkably similar to the ideas of those who claim that atheism is scientific, in contrast to religion. Atheism seems to me to be just as much an article of faith as any religion, and as such as no connection with science, and certainly no scientific justification.

  5. This is back to word games and interpretation of -----.
    No one I know of other than Sam Harris believes that morals arise or are justified scientifically. If I say, for example, like John Stuart Mill, that morality consists of the greatest good (or like Harris, greatest happiness) for the greatest number, I can use scientific techniques to study the results of implementing this on society or individuals, but the statement of what is moral (or what is not) is a statement of belief. Science can study the results of that belief, but the belief itself is like an axiom.
    It is an a priori statement not subject to proof. Its effects are often subject to examination using the scientific method.
    Beyond this, people just get into language games by defining morality to suit their prejudices and interpretations, and just look at the turmoil when TRF introduces "interpretation of QM":)

  6. Difficult topic. In part because so much depends on one's definition of "moral". And the more you try to pin any word down, the less substantive it appears to be.
    Ethics and morals. I see ethical behaviour as that which promotes the general social "good", meaning continued health, organizational strength, integrity, optimism etc. Heliocentrism was viewed as dangerous for undermining the absolutism of Biblical interpretation as per the Catholic Church, in the same way that Catharism was a threat: the Church absolutism removed disquiet with the Church's controls, and since the power of Kings lay with the Church's (God's) selection of the Royals, a Church without abosolute correctness on its side allowed unhappy citizens to question the continuation of displeasing monarchist rule.
    So questioning heliocentrism (or shifting your beliefs to Catharism) could be seen as "unethical" if your primary concern was social stability (whatever that stability meant).
    The thing about ethics is that they are pragmatic, concerned about ends. It is ethical to eat the bodies of plane crash survivors in order to survive - even the Pope has said so. But "morality" might say it is not.
    Morals are transient and ideological. Smoking dope is immoral by some, but unethical only if there is signficant harm relative to other recreational pursuits.
    What science does is provide data from which one might envision a future before being committed to that future. It can help develop ethical behaviour, as you might better understand the "if" scenarios, and your intent is long-term health/weath/happiness/general pleasantness. But none of that touches on morals.
    Morals are in our heads. You can be immoral just looking and thinking. You can't be unethical just looking and thinking, you have to act. Science gives us a pre-view of our actions and their results. But none of it addresses morals.
    That's up for us.
    Sometimes they come together, which is why the subject is so difficult. Like this:
    Al Gore sold Current TV and got oil-sourced money in his pocket. That was unethical as it encourages the behaviour he claims is going to kill the planet. It is immoral because he betrayed the trust of his acolytes by personal benefiting in ways he has worked hard to deny (by persuasion) in other people.

  7. Dear Ralph, Carroll actually wrote

    Chunky Monkey is the best possible ice cream.

    without any "your" in front of the ice cream. So the proposition wants to suggest that there is an objectively best ice cream.

    I wrote something else but I am also willing to say that "XY is my best ice cream" isn't a scientific proposition, either. At different moments, I surely have different tastes, but even if I focus on a moment, I just can't say what is the best one. They're different. Trying to get some answer from the shape of my face is pure superstition, it is idiocy. Even if my face expresses some feelings of sourness or bitterness or whatever, I may still love the food.

    I've always considered people who would be picking the best color or best *** anything in the lifestyle to be stupid, while I had to politely pretend that I thought these interesting questions, and I am sorry that I must count you in that group. Your comparison of ice creams to the Beyond the Standard Model physics - which moreover ends up unfavorably for particle physics - shows that we have absolute nothing in common. I can hardly imagine a more tangible proof of yours that you are a superficial consumerist animal with no education.

    The existence of below-10-TeV particles, whether the answer is Yes or No, has been essential for the existence of the Universe in the last 13.7 billion years, especially during the first formative second. It determined the formation of everything, including stupid animals who think that there is a best possible ice cream.

    Boltzmann is spelled with a double N.

  8. When Jefferson penned his first draft of the Declaration of Independence he wrote, "We hold these truths to be sacred and eternal . . ." Franklin, who knew something about math and science, crossed out "sacred and eternal" and wrote in "self-evident." In effect he was making them axiomatic, and thus beyond discussion.

  9. Luke, to me the "empirical methods of science" used to draw moral conclusions are borrowed from statistical mathematics. However math is not science (according to Carroll).

  10. I think you are too intellectual about these matters. Morality is a matter of having the instinctual feeling for what is wrong and what is right. To a moral person, you don´t have to rationalize why hurting others is wrong. A psychopath doesn´t know why it is wrong to hurt others. He doesn´t feel it. Unfortunately research shows that not all people in a society are moral. There is something like 1-5% of clinical psychopaths and maybe 10-20% subclinical psychopats (called deprivants by Koukolík). All this can be shown not only by psychological tests but also by direct objective images of human brain investigated by fMRI (you can observe the brain while presenting the proband with some kind of moral task or empathy task). So you can study morality objectively and scientifically and it is being done. You can watch a lecture if you like

    Off-topic: I recently watched an amazing movie from Akira Kurosawa - Redbeard (Akahige in Japanese). It has become one of my all-time favourite movies. It is a movie about morality in life

  11. This is bullshit, Mephisto. fMRI may be showing many things but it's surely not "morality". For a related issue, see this Penn and Teller's Bullshit episode

    on why lie detectors are bullshit.

  12. Of course it is not directly morality, but strictly speaking some kind of difference in the brain of people with psychopathic traits. fMRI is a subtraction method. You investigate the activations in the brain of healthy controls, then of psychopaths (matched for age, sex, other counfounding variables) and you compare the two. You get some statistically significant difference is certain brain regions which shows you that there is a difference in the brain activation of the two groups. So morality and psychopathy are "wired in the brain" somehow. Something is dysfunctional in the way the brain of psychopats reacts to the suffering of others, for example the mirror neurons

  13. You may abuse the likely fact that the two of us might agree about some particular psychopaths' being immoral.

    But the broader point is that even if you can see various differences between humans - and I mean more ordinary ones - via fMRI or any other method, you can't establish which of the results is more moral and which of them is less moral.

    Morality is probably wired in the brain but the point is that there is no objective universal way to distinguish morality wired in the brain from immorality wired in the brain.

  14. Shannon, the law of diminishing returns is a scientific fact which shows up all over the place in economic theory: in the marginal theory of productivity (law of variable proportions), the decreasing marginal utlity of income, increasing marginal disutility of labor, etc. These phenomena can be modeled as concave and convex curve, the curvature of which cannot be measured, granted, but the convexity or concavity of which allows certain geometrical conclusions. The theoretical existence of general equilibrium for example, which is what Adam Smith is all about. Math is important of course -- in this case geometry (or whatever branch is concerned with shapes only as opposed to quantitative measurement. The really controversial parts however -- such as issue of how much economic redistribution is ideal in terms of maximizing general welfare (pitting growth against against equality) can only be settled by the political process, which means voting in a democracy. There is no scientific solution to these questions.

    Let me add that I don't disagree with anything Lubos says here because what I am I am saying is not inconsistent with it.

  15. Hi Mephisto, my favorite movie is one of Kurozawa too "Ran" ;-) His movies seem to be "Shakespearian"...

  16. Interesting video Lubos. Cruel, so cruel.

  17. "There is nothing right or wrong but thinking makes it so." Shakespeare

    Even science assumes the principle of induction as an article of faith. Hume

  18. Lubos: Hmmm, first agreement with Bee, and now with Sean Carroll....are you sure that the real Lubos hasn't been kidnapped and that you are not some sort of changeling? :)

  19. Unfortunately it's exactly the same agreement with Carroll as 3 years ago

    so nothing essential of this sort has changed about the relationship or him or me.

  20. Of the four un-scientific propositions below, which is the only non-trivial one worth thinking about:

    a) 1+1=2.
    b) Hamlet was really crazy, he wasn’t just faking it.
    c) Chunky Monkey is the best possible ice cream.
    d) All else equal, a dollar is worth more to a poor man than a rich one.

    Bonus question:

    Why do philosophers choose such silly examples?

  21. Some nice quotes from Hume:

    There has been an opinion very industriously propagated by certain philosophers, that morality is susceptible of demonstration; and tho’ no one has ever been able to advance a single step in those demonstrations; yet ’tis taken for granted, that this science may be brought to an equal certainty with geometry or algebra.

    If morality had naturally no influence on human passions and actions, ’twere in vain to take such pains to inculcate it; and nothing wou’d be more fruitless than that multitude of rules and precepts, with which all moralities abound.

    Since morals, therefore, have an influence on the actions and affections, it follows, that they cannot be deriv’d from reason; and that because reason alone, as we have already prov’d, can never have any such influence. Morals excite passions, and produce or prevent actions. Reason of itself is utterly impotent in the particular. The rules of morality, therefore, are not conclusions of reason…

    …reason can never immediately prevent or produce any action by contradicting or approving of it, it cannot be the source of the distinction betwixt moral good and evil, which are found to have that influence. Actions may be laudable or blameable; but they cannot be reasonable or unreasonable.

    Reason is wholly inactive, and can never be the source of so active a principle as conscience, or a sense of morals.