What you describe about in-groups and out-groups happens all the time with religious communities. Even people sharing the same religion suddenly become "unworthy".

> And below all that: What values do we use when evaluating what ethical systems are better, and where do THOSE values come from?

Precisely. If you don't believe in any gods, where do you suppose religious value systems come from? :)

However, if we look at different societies, some work better than others. I think most people on HN live in societies that actually work quite well.

To think that you can somehow reason your way to anything better, from first principles (which have to be values, not rules), is very optimistic, in my opinion. First of all, you have to be very certain that you have the correct first principles. (I think this is a much harder problem in itself than most people realize), but even if you do, creating a full ethical system that maximizes those values, is computationally impossible.

To my knowledge, most attempts to introduce radical changes in this manner, has lead to societies that function worse, not better.

I think this problem is very closely related to defining a utility function to a superhuman AI, that you can trust to lead to good outcomes.

How atheists behave can also be observed now, regardless of how young the "society": some commit crimes, some don't. Some follow strict value systems, some don't. Some are empathic, some aren't. Some are monsters, some are heroes. Some are flawed, common people.

Just like religious people ;)

There are some exceptions, though, such as Stalin's USSR, Hitler's Germany and Mao's China. They each killed many, many millions. All three thought they could create a utopia by turning traditional morality upside down.

It is unlikely that the murder rate will ever reach absoulte 0 (unless we have an extremely totalitarian state, such as a chip operated into every individual to monitor their actions), but it is much better to have murder rates of below 10 per 100k than to have them in the 100s or 1000s per 100k, like we had through much of history.

I happen to think that in many ways, the western world has actually moved in the right direction under christianity. You present the existence of theft and murder as a boolean yes/no situation.

I think such argument are misleading. For these statistics, it only makes sense to present the RATES of theft and murder for each system.

As an athetist myself, I'm very aware of the urge to try to construct an utopia from first principles by revolutionary means. But I happen to think that this is very, very dangerous. It can easily take murder rates to less than 10 per 100k to killing a significant percentage of the population, as seen many times during last century.

It is not that I think that religion is very different from other cultural sources of value. I happen to consider relgion to actually BE part of culture. But just because the epistemology of religion is likely wrong, from a scientific point of view, I think the cultures that we live in, including much of a christian baggage has proved that it can sustain a fairly good society.

As an analogy, linux is not perfect, but if random people start modifying code in the kernel more at less at random and without deep understanding, the stability is likely to suffer.

Society is like the social operating system, and if it breaks down, everything breaks down.

EDIT: If you didn't notice, the argument above is pretty much the argument for atheistic liberal-conservativism.

So in practice, I think we are left to make incremental attempts at improvements, while trying to make changes both small and reversible.

In computer science terms, this is gradient decent. It may not allow us to reach all global minima, but it is much more likely to lead to a good place than large semi-random changes.

Sorry, I simply don't see where anyone is talking about revolutionary utopias here, or first principles, or totalitarian anything. If you want to argue that non-religious value systems inevitably lead to genocide, feel free to do that -- and good luck with that, I'm uninterested in that argument.

All I'm saying, is that the western culture we have inherited, has proven to be pretty good at providing a good life for most people.

As we discover that the foundation of the religous belief system is wrong, we need be careful about how we dispense with the rules that have been justified in religous dogma up until now.

I think it can go wrong in two ways:

1. The far right atheists can argue that without Jesus, there is no objective reason for universal empathy, which we can use to justify extreme selfishness, eugenics, racism, etc.

2. Left wingers, on the other hand, can argue that ONLY empathy and happiness matters when we dispense with the supernatural soul. Because the religous reasons for free will are wrong, we should not treat individuals as moral agents, but instead treat all unwanted behaviour or unwanted outcomes as health problems or something caused by unfair treatment by society.

In small doses, I don't see either view as a big problem. But there have been cases where they have been taken to the extreme, and that went blody on both sides.

Scientifically, I think both views are simplifications to the point of being just wrong. In order to function as a society and as individuals, we need a balance between individual responsibility and empathy where it is due.

I think it is possible to reach this conclusion through science and philosophy, but that this is extremely difficult, and even more difficult to communicate to the electorate.

Which is why I argue that we should not change our moral rules too quickly, even if the traditional justification for them fall away.

Btw, it may very well be that we more or less agree on this. The discussion started further up, in response to the claim that moral behaviour (not killing) was a consequence of "being an empathic person", and nothing else.

Indeed, it may be that we agree. The problem is that the discussion got so sidetracked it's hard to see what we're talking about anymore.

As an example of how we're talking past each other, I wasn't responding that empathy is or should be the only restraint. I was specifically arguing against this assertion further up:

> I think that's a good point that lots of things can override empathy. It would require something much stronger [...] Religion with a belief in immortality and eternal justice with eternal possibility of happiness gives a very strong reason [...]

This is demonstrably, patently false: religion gets overridden and fails ALL the time. It should be noted that, traditionally, the argument goes "atheists forget religion and therefore have no value systems; what prevents murder when there is no god?", yadda yadda yadda.

To clarify, things I am arguing:

- Religion is not enough or better than empathy or other secular/atheist moral value systems. It works (sometimes), but it's not better or the only way. Arguments that start with "but without religion...!" are therefore wrong.

Things I'm NOT arguing:

- Religion always fails.

- Empathy is enough.

- Atheists never commit crimes.

Things I'm NOT interested in debating, because they're wrongheaded, red herrings, appeals to ignorance and/or "not even wrong":

- "But atheist societies aren't old enough!"

- Software development as an analogy to moral value systems.

- Hitler, Stalin, Mao, Patton, Torquemada, the French Revolution, etc.

The difference seems to be on emphasis: - You seem more concerned about religiously inspired fallacies. (I agree that they are fallacies, but I don't take them very seriously any more). - I am more concerned with preventing gulags. (Which you are not interested in dicussing).

I suspect the difference in emphasis may come from us living in different countries (I live in Norway, which is more or less atheist socialist in its value system)

Btw, the original article was: "Things I learned in the Gulag"