1

Recently, I advocated utilitarianism based on consciousness on Change my View subreddit and I was quite suprised by the opposition against the bare idea of a universal moral system. From my limited observation, those who think in such way argue by the role of "emotions" in our decisions as opposed to "cold reason".

In my opinion, this view is determined to fail, as it introduces our animal perception, evolutionary built to generate a constant flow of naturalistic fallacy arguments, as a universal moral judge. I think those moral dilemmas where utilitariansim is taken to the "extremes" look as dilemmas just because they are emotionally strong enough for us to feel bad about logically the best option, even though it can be calculated it'll bring more goodness whatever that is (5 random people's happinesses are more than 1).

So is it philosophically/logicaly tenable to refuse a universal moral system in general?

  • 3
    Yes, and it is the prevailing secular view today. Even modern supporters of moral realism, like Nozick or McDowell, tie morality to our "second nature", which means that it is not "universal", on common view utilitarianism isn't "universal" either. As for evolutionary ethics, it purports not to be normative ethics but rather to explain why people have an illusion of morality and what the underlying mechanisms are, so there is no naturalistic fallacy. Others, like Putnam, do not consider "naturalistic fallacy" to even be a fallacy because all "facts" are value-laden. – Conifold Sep 29 '17 at 20:55
  • No, because you know better. If you base your reasoning on evolution, you might come to the conclusions that Conifold outlines. Logically, the best evolution can do is to explain an "illusion of morality." However, our sense of morality tells us that these conclusion are false, and philosophy is about seeking truth. – user3017 Sep 29 '17 at 21:20
  • 1
    People who oppose multiculturalism believe that multiple cultures cannot live in close proximity for precisely this reason: living harmoniously in close proximity requires laws and rules, and, if there no universal morality, there is conflict over what these laws and rules should be. Although I personally disagree with this argument, it could be a valid point. – barrycarter Sep 30 '17 at 13:57
  • 1
    most people aren't philosophers. the question seems a bit vague, if warmly motivated. maybe a better question is whether ethics without "universality" is robust enough to think ethically. i think the answer to that is likely yes, as galling as that is. that's philosophers though! – user28660 Oct 5 '17 at 7:36
1

It is definitely philosophically and logically defensible to claim that morality is not universal. There are some good comments made here which make reference to how this is the accepted view in meta-ethics. Meta-ethics is the study of what ethical statements mean, as opposed to what ethical acts are and how to be ethical. The latter is discussed in applied ethics and ethics more broadly when we think about different ethical systems such as utilitarianism, consequentialism, etc.

It's great to see more-and-more acceptance of the evolutionary origins of our morality in philosophy and society. Many still disagree with this point of view: that the origins of our morality are wholly evolved (or a result of evolution). The counter-view is that morality is universal and we either: 1) don't know the origin of this universal morality or 2) we do know the origin of this universal morality — such as divine, rational inquiry, etc. I will not talk about the possibility of a universal morality, only arguments that could contradict this view.

Yes, objectively we can't say anything is absolutely right or wrong as all claims about moral acts are value judgments. However, most humans don't seem to have a choice in the matter of making moral judgments and acting on them. We make moral judgments because we can't not make them. Hence why there is a sharp focus in some areas of moral philosophy on the evolutionary origins of morality. This will hopefully shed more light on why it is we act morally and don't have a choice in the matter for the most part even though we think we absolutely do.

Some humans, on the other hand, don't act how most of us would think is moral. Somebody whose brain does not function neuro-typically (due to developmental differences in the brain) may lack the capacity to be empathetic. They understand that their actions are wrong on a rational level, but do not feel their actions are good or bad. They may not feel that killing or stealing for their own gain is good or bad even if they know perfectly well that this is not acceptable behavior.

There's a difference then between people who know what's morally wrong and still act in this way but feel guilty about it afterward; and those who know what's considered morally wrong to others, still act in this way but feel zero guilt about it afterward. Neuroscientists have studied how the brain functions, for example, in psychopaths in prison who lack empathy and compared this data with those who don't have this type of disorder (see: Probing psychopathic brains). It seems that psychopaths can switch on and off their empathy when they choose to. Whereas, most of us, don't have a choice in that. If we see something we consider morally wrong, we immediately feel a negative emotional response to it.

There are many reasons why this would be advantageous to us in an evolutionary context. Some of the reasons include group bonding and fitness, evading confrontation with enemies and rivals, rearing children so that they have the best chance of survival and so on (see: Morality and Evolutionary Biology). We are not the only animals to display what could be considered moral behavior and have a sense of fairness. Other animals cannot articulate they know this of course, so we can't say they "know" they are acting morally.

As a society, if we want to think about what makes something moral or not we should consider how the brain evolved to feel emotions which direct our moral behavior. This knowledge can also be used when we think about ethical dilemmas such as the trolly problem. We may feel bad about sacrificing 1 person to save 5 if we are emotionally isolated from this, but if we have to feel the discomfort of physically pushing somebody to their death, this is much more of an emotional experience and a much harder decision to make. And when asked to sacrifice your loved one to save the 5 people, not many would save the strangers — all things being equal. A utilitarian may argue that which places human happiness as the highest good ought to be done. Therefore, these are all morally equivalent scenarios and you should sacrafice your own family to save the strangers. There are specific logical and rational defenses as to why your own family has more utility to you and not to others etc. But it is clear emotions (which are determined wholly by the brain) play a big role in making moral decisions and we should, therefore, heavily consider how emotions work to understand morality.

If someone's brain is not capable of feeling the same emotions as others, such as empathy, should we place the same amount of blame on them as somebody who does? There is an example in neuroscience of a man who was convicted of being a pedophile who also had a brain tumor in the orbitofrontal cortex. This region of the brain is responsible for judgment, impulse control, and social behavior. The man knew his behavior was morally indefensible but continued to act in this way regardless. The brain tumor he had was eventually cut-out and then he stopped feeling these pedophilic impulses. But eventually the tumor returned, and so did the pedophilic impulses (see: Brain tumour causes uncontrollable paedophilia). This is quite a shock and is highly controversial. Nearly all of us think and know why this act is so morally repugnant and would never and couldn't even imagine actually doing so, let alone exacting this crime. But to some, it appears that even in such an extreme scenario, perhaps they have no real "choice" in the matter.

If emotions can overpower moral, logical, and rational decision making, what does that say about the absolutism or universalism of our morals? Not only does this go beyond morality being relative to cultures, it also questions whether people that know the same things are morally indefensible all act in the same ways morally. How should we then think about blaming or praising moral actions? Therefore, these questions are highly linked with notions of free will. Do we have it? What is it? Is it real? If we do or do not have free will, what does this mean for moral decision making?

More on this by the philosopher Derk Pereboom:

http://joelvelasco.net/teaching/hum9/pereboom-nofreewill.pdf

https://www.youtube.com/watch?v=bObzpWrhH-Q

And I recommend checking out the neurobiologist Robert Sapolsky. He talks about how morality, evolutionary biology, free will, and morality intersect from a scientific point of view:

https://www.amazon.com/Behave-Biology-Humans-Best-Worst/dp/1594205078

https://www.youtube.com/watch?v=Cx8xEUYrb74

  • 1
    I think those arguments include the exact bias I wrote about. Yes, we're driven by evolutionary morality through our emotions but although I don't understand anything but theoretical evolution, I think it pushes against our sense of morality most of the times. It's our instincts that allow us to steal or cheat on others without contemplating the good for the society. – Probably Oct 8 '17 at 7:41
  • 1
    Yes, but we do have degrees of control. Even if the universe is either completely or mostly deterministic. Others may think that we are in full control of our actions — libertarian free will. I think when you discuss morality you have to consider how much control somebody has over their actions on a neurobiological level. Otherwise, you could be holding onto outlived paradigms. Many think that if you don't have free will you can't have morality because immoral acts were going to happen anyway. We can't know with absolute certainty if our actions are predetermined and how much control we have. – user28485 Oct 8 '17 at 8:08
  • I think we all'd agree retributional theory of punishment doesn't make any sense, thus justice is irrelevant for this debate. – Probably Oct 8 '17 at 10:34
  • What is relevant tho, is the evolutionary morality. Race segregation has been observed in some Amazonian tribes and chimpanzees which were observed going to a neighbouring tribe and massacre. Minority segregation can be observed on any animal with a social hierarchy. – Probably Oct 8 '17 at 10:34
  • Of course, we can build societies because of our common (cultural) social contract but overwhelming majority of our morality isn't inherent. It doesn't matter what happens inside our brains, for morality debate it's important what comes out of them. – Probably Oct 8 '17 at 10:37
0

Two moral systems, one Absolute, one relational (casuist, if you must):

A Killing = 100% wrong.

B No killing > less killing > more killing.

You wake up in a room tied to a seat and are given two choices: press the button and you kill 1 person, don't press the button and 100 people are killed by someone.

An Absolute, objective moral system does not provide the efficient solution (you can't argue with 100%). Someone who believes A would choose to not kill 1 person, and let the 100 die. This way they are not the efficient cause.

The relational moral system chooses to kill 1 person accurately.

Why? Because creating Absolute, arbitrary, 100% rules (no hypothesis test can get you to 100%, so one would wonder how such certainty is arrived at, if not by observation/measurement) creates a disconnect between the rule and our relational experience. They also have no power to choose between 'killing' and 'killing'.. only a relational addendum would do that, which would make one question the need for the objective moral rule in the first place.

Relational rules are not universal. The above example can be changed - kill 1 newborn or 100 terminally ill people who die tomorrow will be killed. There you go - now A gets the right(er) answer, and we need to adjust B to:

B1: no loss of life years > less loss of life years > more loss of life years.

And ooon, and ooon, and ooon..

An objective moral system is complete but inconsistent. A subjective moral system is consistent but incomplete. Choose your poison - I like consistency over completeness - horses for courses as they say!

  • If I get you right, you're saying that universal moral systems are incomplete because they have to be formulated as "this is bad, more of this is worse"? – Probably Oct 8 '17 at 10:41
  • 1
    Because the only moral rule I have is "the most good for the most consciousness". This seems to me to be infinitely times simpler than our current judeo-christian moral system with many exclusions and ambiguities. We would do anything for love (if isn't homosexual love which we find wrong) but we would never kill (only if the jury finds it reasonable). We shall not judge others and yet we shall take "eye for an eye". Also, it's wrong to repay evil for evil but it's ok to flood the humanity for their sins – Probably Oct 8 '17 at 10:48
  • The world is relational. The evidence is Copernican, Galilean, special, general relativity and relational quantum mechanics (arxiv.org/abs/quant-ph/9609002) .. effectively properties and objects cannot be said to exists independent of observation/measurement. Observation though, by definition is frame dependent - you and I watch the world through different screens. Universality cannot be declared for any concept but the relational nature of Existence (at least QM based existence, and there are strong logical arguments for why any other theory of motion cannot be complete)... – Ilya Grushevskiy Oct 10 '17 at 7:57
  • Universal rules are complete, because they present simple, unarguable rules (whether right or wrong). but because experience does not possess this 'universal' frame of reference (such as the pilot wave in De-Broglie Bohm), one can only examine relational frames for rules. Hence the relational rule B - it is consistent because the rule will always lead to numerically less killed, if followed. But it would not by definitino lead to less suffering or 'wrongness', because that is not defined in the rule.. it's a bit like X=X is a 0 information tautology. – Ilya Grushevskiy Oct 10 '17 at 7:59
  • I use the information imperfect (negational) Platinum rule as a primary: 'do not do unto others, as you think/feel/believe they would not wish to be done unto themselves.. if you don't know, ask.' .. if you can't ask, drop to the secondary Silver rule: 'do not do unto others, as you would not have them do unto you'. – Ilya Grushevskiy Oct 10 '17 at 8:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.