Could ego be a[n evolutionary] basis for morality?

[Peter Grant]

I posted this in the philosophy forum but they seem, for the moment at least, to have run out of arguments there:

viewtopic.php?f=18&t=63834

Here is the basic outline of the argument. I really want to be sure that I'm right about this, so please point out any flaws in the theory:

- In a human society certain types of behaviour will be more effective than others.
- Evolution has equipped us with the ability to create models of ourselves in order to predict which types of behaviour will be more effective.
- These models are based on how we observe and interpret reality.
- The more realistic this model, the more accurate predictions based on it will be.
- We each have our own personal concept of good based on our own needs and desires.
- We use this concept to judge the behaviour of both ourselves and others.
- Behaviour which is less frequently judged as bad will thus be more effective.
- Individuals with more realistic self models will behave more effectively than those with less realistic ones.
- Societies comprised of individuals with more realistic self models will behave more effectively than those comprised of individuals with less realistic ones.
- These individuals and the societies they are part of will tend to behave more morally and more generally be considered good.

[Marios]

[b]- In a human society certain types of behaviour will be more effective than others.
- Evolution has equipped us with the ability to create models of ourselves in order to predict which types of behaviour will be more effective.

How is "effective" defined?

- These models are based on how we observe and interpret reality.
- The more realistic this model, the more accurate predictions based on it will be.

Not really, no - unless you have an unusual definition of realism. In AI, the idea that you can create an intelligent machine by just chucking more and more data into a bucket is pretty much dead - 'intelligent' behaviour is all about paring down the data into the simplest, most robust model which does the job - that often has very little if anything to do with realism.

If you want a very simple published example, you could do worse than this:
http://jasss.soc.surrey.ac.uk/1/1/3.html

- We each have our own personal concept of good based on our own needs and desires.

Don't notions of good correlate more closely with geography and social context? What evidence do you have that notions of good correlate with "needs and desires"?

- Individuals with more realistic self models will behave more effectively than those with less realistic ones.

Still not sure what you mean by "effective" or "realistic", but it's hard to see what would necessarily benefit from "realism" except "realism".

- These individuals and the societies they are part of will tend to behave more morally and more generally be considered good.

By whom will they be considered good? If "- We each have our own personal concept of good based on our own needs and desires" then how can you say "moral morally"/"considered good" without specifying who is the person doing the judging?
Marios

[Peter Grant]

[b]- In a human society certain types of behaviour will be more effective than others.
- Evolution has equipped us with the ability to create models of ourselves in order to predict which types of behaviour will be more effective.

How is "effective" defined?

In the evolutionary sense. Effective behaviour is behaviour that increases the likelihood of of the genes that build the model building brain being passed on to the next generation.

- These models are based on how we observe and interpret reality.
- The more realistic this model, the more accurate predictions based on it will be.

Not really, no - unless you have an unusual definition of realism. In AI, the idea that you can create an intelligent machine by just chucking more and more data into a bucket is pretty much dead - 'intelligent' behaviour is all about paring down the data into the simplest, most robust model which does the job - that often has very little if anything to do with realism.

If you want a very simple published example, you could do worse than this:
http://jasss.soc.surrey.ac.uk/1/1/3.html

In this study the behaviours were being programmed directly into the bots. I'm suggesting that social norms change too quickly to act directly on the genes. That better model building brains are what would be selected in the long run.

- We each have our own personal concept of good based on our own needs and desires.

Don't notions of good correlate more closely with geography and social context? What evidence do you have that notions of good correlate with "needs and desires"?

If we're going to argue about the definition of "the good", please lets do it back in the philosophy forum with the other relativists. The reason I did not define "the good" is that I think it can be seen as an environmental factor which evolution would have taken into account when selecting model building brains.

- Individuals with more realistic self models will behave more effectively than those with less realistic ones.

Still not sure what you mean by "effective" or "realistic", but it's hard to see what would necessarily benefit from "realism" except "realism".

Well if I need to guess what a potential rival or mate is thinking then I'm sure it would help. An understanding of my own motivations coupled with good observational evidence of their previous behaviour should do the trick.

- These individuals and the societies they are part of will tend to behave more morally and more generally be considered good.

By whom will they be considered good? If "- We each have our own personal concept of good based on our own needs and desires" then how can you say "moral morally"/"considered good" without specifying who is the person doing the judging?
Marios

Humans, who have been selecting ever more effective model building brains based on their own experiences of good for generations.

[Marios]

In the evolutionary sense. Effective behaviour is behaviour that increases the likelihood of of the genes that build the model building brain being passed on to the next generation.

Okay, you mean biological fitness? Remember that the "effective behaviour" is a rapidly moving target and will tend to involve the behaviours of other conspecifics.

In this study the behaviours were being programmed directly into the bots. I'm suggesting that social norms change too quickly to act directly on the genes. That better model building brains are what would be selected in the long run.

An interesting verbal argument, but do you have anything more than?

If we're going to argue about the definition of "the good", please lets do it back in the philosophy forum with the other relativists.

Actually, the philosophy forum is where I expect to find the absolutists - evolution and natural selection is where I expect to find the methodological naturalists.

The reason I did not define "the good" is that I think it can be seen as an environmental factor which evolution would have taken into account when selecting model building brains.

Sorry, if I'm reading you right, that's really the sort of thing that's only allowed in philosophy. In evolutionary biology, that would get you hauled up as a social darwinist. You can't just project your beliefs about what "the good" entails onto your model and call it science.

Given that your whole argument depends on whether what you're claiming evolution selects for is actually selected for, I think this is where you need to make your case very clearly without skipping any steps.

Well if I need to guess what a potential rival or mate is thinking then I'm sure it would help. An understanding of my own motivations coupled with good observational evidence of their previous behaviour should do the trick.

It might do and it might not - it's certainly not automatically the case.

Humans, who have been selecting ever more effective model building brains based on their own experiences of good for generations.

Which humans are doing the judging? How do you measure the level of perceived morality/good?

You seem to be conflating conformity with morality - are you really sure you want to do that?
Marios

[Peter Grant]

In the evolutionary sense. Effective behaviour is behaviour that increases the likelihood of of the genes that build the model building brain being passed on to the next generation.

Okay, you mean biological fitness? Remember that the "effective behaviour" is a rapidly moving target and will tend to involve the behaviours of other conspecifics.

Exactly, I'm not saying that the behaviour is directly governed by genes, but the ability to deduce which behaviour would be more effective must be. By conspecifics do you mean other humans? If so I agree. Other humans would probably have been the most difficult to predict aspects of our early environment.

In this study the behaviours were being programmed directly into the bots. I'm suggesting that social norms change too quickly to act directly on the genes. That better model building brains are what would be selected in the long run.

An interesting verbal argument, but do you have anything more than?

If you mean evidence, I only have that which I can readily deduce from our modern society. It seems that all of us have a concept of good, although we define it somewhat differently. Children seem to absorb the same basic moral norms of the society in which they grow up, regardless of genetic background. We are good at making models of the universe in which we live. We do make moral judgements all the time, usually based at least partly on intuition.

If we're going to argue about the definition of "the good", please lets do it back in the philosophy forum with the other relativists.

Actually, the philosophy forum is where I expect to find the absolutists - evolution and natural selection is where I expect to find the methodological naturalists.

Yeah well I'm getting a bit sick of the whole beauty is truth and truth is beauty thing:

viewtopic.php?f=18&t=63834&sid=0b38ae8f05298876697c80e56f2139b2#p1523463

The reason I did not define "the good" is that I think it can be seen as an environmental factor which evolution would have taken into account when selecting model building brains.

Sorry, if I'm reading you right, that's really the sort of thing that's only allowed in philosophy. In evolutionary biology, that would get you hauled up as a social darwinist. You can't just project your beliefs about what "the good" entails onto your model and call it science.

Given that your whole argument depends on whether what you're claiming evolution selects for is actually selected for, I think this is where you need to make your case very clearly without skipping any steps.

Please don't apologise. I wish to express this idea as clearly as possible to both myself and others. Your arguments are helping.

I'm not trying to project any beliefs about "the good". I agree it is a subjective, indefinable concept. It is, however, something that we all have and use to judge other's behaviours all the time. I'm looking at morality and trying to see what parts of the problem we can solve. Where are the objective facts that we can test? Saying that there is a subjective element to morality is all well and good, but there the work seems to stop. I began by trying to figure out which elements aren't subjective.

However subjective "the good" may be, the ability to judge other's actions as either good or bad must be genetically based. Even my cat seems quite capable of getting angry with me. My cat's behaviour is relatively simple. By comparison, all my ex-girlfriends were ridiculously complex. To stand any chance at all, I had to try imagine things from their point of view in order to predict their responses. For the girl-down-the-road this is not excessively difficult, but for someone on the other side of the world it gets a bit harder. By a strange coincidence the girl-down-the-road's parents came from the other side of the world, but she behaves as predictably as anyone else in my neighbourhood. This tells me that her subjective element, "the good", is probably not all that different from mine. It's the objective positioning of where you are on the globe that determines the differences in your morals from mine. If morals really were as subjective as is made out then it would be impossible for any multi-cultural society to function. In seems that there would be a strong selective pressure in favour of uniformity in all subjective elements that influence moral decisions so that the individuals would make good moral judgements whatever culture they grew up in, and this seems to be the case.

As to the question of why different cultures have different morals. One's model need only be as accurate as those around one. While I don't think their morals are really that different to begin with, where they do differ it is based on their objective, testable beliefs. Not on some subjective, indefinable concept, which I propose we all share to a certain extent anyway thanks to natural selection.

Well if I need to guess what a potential rival or mate is thinking then I'm sure it would help. An understanding of my own motivations coupled with good observational evidence of their previous behaviour should do the trick.

It might do and it might not - it's certainly not automatically the case.

Well say I believed that when people are frowning at me, they approve of my behaviour. I don't think I'd do very well. How could the ability to predict behaviour be a disadvantage?

Humans, who have been selecting ever more effective model building brains based on their own experiences of good for generations.

Which humans are doing the judging? How do you measure the level of perceived morality/good?

You seem to be conflating conformity with morality - are you really sure you want to do that?
Marios

I don't think any model that is not at least as complex as the thing it is modelling can ever be perfect. Conformity with the most realistically based morality available would be the next best thing.

[Marios]

Exactly, I'm not saying that the behaviour is directly governed by genes, but the ability to deduce which behaviour would be more effective must be.

Why must it be _directly_ governed by the genes if "behaviour" is not?

By conspecifics do you mean other humans?

Yes, sorry, it's a useful little word that means "members of the same species"

If you mean evidence, I only have that which I can readily deduce from our modern society.

I'm afraid arguments from intuition fall down on the fact that different people "readily deduce" different things using their common sense intuition.

If you're just saying "Some components of social behaviour are correlate more with social context than genetic context" then I think that's fairly uncontroversial.

Yeah well I'm getting a bit sick of the whole beauty is truth and truth is beauty thing:

http://www.richarddawkins.net/forum/vie ... 2#p1523463

Christians get sick of atheists saying "I don't find that convincing" - I don't think that really comments on whether one or the other is wrong, but it might imply that there are key assumptions that they two parties don't share.

Please don't apologise. I wish to express this idea as clearly as possible to both myself and others. Your arguments are helping.

"Good".

I'm not trying to project any beliefs about "the good". I agree it is a subjective, indefinable concept. It is, however, something that we all have and use to judge other's behaviours all the time. I'm looking at morality and trying to see what parts of the problem we can solve.

I'm not sure if we're in total agreement or total disagreement. When you say "problem we can solve" do you mean the four scientific questions you'd ask about any evolving system -
http://en.wikipedia.org/wiki/Tinbergen%27s_four_questions
- or do you mean "which of these moralities is the right/righter one?"?

Where are the objective facts that we can test? Saying that there is a subjective element to morality is all well and good, but there the work seems to stop. I began by trying to figure out which elements aren't subjective.

Hmm - I think you might be confusing two senses of the word 'subjective'. Morality is subjective in the sense that there's no external reference to compare it to. If I say I think female circumcision is right and proper, you can say you think it's wrong - what you can't do is say I think it's wrong, but I'm just confused. If I say a metre is 83 centimetres long, then I am in just in error (insofar as we accept that the metric system/dictionary should define how words should be used).

I don't morality is usually viewed subjective in the sense of a hidden experience which cannot be observed - depending how you use the word you could be talking about observed expressed beliefs or observed actions.

However subjective "the good" may be, the ability to judge other's actions as either good or bad must be genetically based.

I think you'd probably be interested in "Theory of Mind" literature:
http://en.wikipedia.org/wiki/Theory_of_mind

If morals really were as subjective as is made out then it would be impossible for any multi-cultural society to function.

I think you have to clarify how the word "multicultural" is used. What it _does not_ mean is two distinct cultures living entwined among another - there is no culture in the world that I know of where one culture isn't dominant. There is nowhere in the world where half the people go to Western Christian Churches and the other half worship Huitzlipoctli through biennial human sacrifice - either Huitzipoctli is in charge and a few small Churches are tolerated/run underground or Christianity is in charge and Huitzlipoctli human sacrifices have to go on underground. In Britain, multicultural means that we tolerate foreign cultures so long as they don't do anything we really don't like. Funny hats, weird architecture, yes - wearing bits of cloth over your face, body alteration of the kind we don't like, espousing alien political systems, no. Even then, multiculturalism as a term is becoming increasingly unpopular - no one wants to say "assimilation" - but most everyone kind of wants it to happen, just as long as we don't have to _see_ it happening. Nor is it just the UK - Holland, most liberal and progressive of countries, has recently undergone the same shift. People liked living in a 'multicultural' society, just so long as the culture showed no signs of changing.

In seems that there would be a strong selective pressure in favour of uniformity in all subjective elements that influence moral decisions so that the individuals would make good moral judgements whatever culture they grew up in, and this seems to be the case.

Again, you probably don't want to use the word "good" when you're talking about morality, particularly when the most obvious heuristic for 'getting along' in society is 'keep your head down'.

As to the question of why different cultures have different morals. One's model need only be as accurate as those around one. While I don't think their morals are really that different to begin with, where they do differ it is based on their objective, testable beliefs. Not on some subjective, indefinable concept, which I propose we all share to a certain extent anyway thanks to natural selection.

I can't see what you're saying here - do you have any anthropology references to back this up?

How could the ability to predict behaviour be a disadvantage?

For one, because the interests of the individual do not align with the interests of the genes. For instance, in humans, unlike almost any other animal, the fertility of the female is deliberately obscured from both the female and everyone else. This isn't just an example of a failure to evolve a capacity to make certain vital predictions (and there are few things more vital to biological fitness than when you are/are not fertile), but a clear example where ignorance has been selected for.

More generally, as the little toy sim by Doran showed, in any situation where you have (i) less than total knowledge
and (ii) less than infinite computing power it is not necessarily the case that more/more accurate information will increase biological fitness.

I don't think any model that is not at least as complex as the thing it is modelling can ever be perfect. Conformity with the most realistically based morality available would be the next best thing.

No, I meant conformity in the sense that you were saying "It is good because other people judge it to be good". Aztec human sacrifice is good if most Aztecs agree that it's good? A definition of "good" that I'm happy with, but which doesn't seem to suit the rest of your model.

MArios

[Peter Grant]

Exactly, I'm not saying that the behaviour is directly governed by genes, but the ability to deduce which behaviour would be more effective must be.

Why must it be _directly_ governed by the genes if "behaviour" is not?

Behaviour effects survival and thus genes. If behaviour is not governed directly by genes, and this is obviously the case, then they must ultimately effect behaviour through some other means of indirect control.

By conspecifics do you mean other humans?

Yes, sorry, it's a useful little word that means "members of the same species"

If you mean evidence, I only have that which I can readily deduce from our modern society.

I'm afraid arguments from intuition fall down on the fact that different people "readily deduce" different things using their common sense intuition.

If you're just saying "Some components of social behaviour are correlate more with social context than genetic context" then I think that's fairly uncontroversial.

That's exactly what I'm saying, thanks for stating it so clearly.

Yeah well I'm getting a bit sick of the whole beauty is truth and truth is beauty thing:

viewtopic.php?f=18&t=63834&sid=0b38ae8f05298876697c80e56f2139b2#p1523463

Christians get sick of atheists saying "I don't find that convincing" - I don't think that really comments on whether one or the other is wrong, but it might imply that there are key assumptions that they two parties don't share.

Fair enough, but I don't want to argue about the subjective good here in the evolution forum. For the purposes of this discussion can we simply define "the good" as that unknown individual factor which decides, within the context of any social system, what behaviour will more likely be judged as good and thus be more effective.

Please don't apologise. I wish to express this idea as clearly as possible to both myself and others. Your arguments are helping.

"Good".

I'm not trying to project any beliefs about "the good". I agree it is a subjective, indefinable concept. It is, however, something that we all have and use to judge other's behaviours all the time. I'm looking at morality and trying to see what parts of the problem we can solve.

I'm not sure if we're in total agreement or total disagreement. When you say "problem we can solve" do you mean the four scientific questions you'd ask about any evolving system -
http://en.wikipedia.org/wiki/Tinbergen%27s_four_questions
- or do you mean "which of these moralities is the right/righter one?"?

The four questions definitely, though I'm not too sure which one. What question would one ask about a peacock's tail? I'm thinking in therms of an evolutionary arms race happening within the same species. I'm talking about a theory which would predict which behaviour would be judged right more often. A theory that would predict the least subjective harm on an individual basis.

Where are the objective facts that we can test? Saying that there is a subjective element to morality is all well and good, but there the work seems to stop. I began by trying to figure out which elements aren't subjective.

Hmm - I think you might be confusing two senses of the word 'subjective'. Morality is subjective in the sense that there's no external reference to compare it to. If I say I think female circumcision is right and proper, you can say you think it's wrong - what you can't do is say I think it's wrong, but I'm just confused. If I say a metre is 83 centimetres long, then I am in just in error (insofar as we accept that the metric system/dictionary should define how words should be used).

I don't morality is usually viewed subjective in the sense of a hidden experience which cannot be observed - depending how you use the word you could be talking about observed expressed beliefs or observed actions.

A hidden experience which cannot be observed is exactly how I would define a subjective experience. We can only observe the results of that experience, but so far we cannot observe it directly. I think that, in principal at least, it is observable, but either way we can theorise about it. Advertisers and fashion designers certainly do, and rather successfully too.

However subjective "the good" may be, the ability to judge other's actions as either good or bad must be genetically based.

I think you'd probably be interested in "Theory of Mind" literature:
http://en.wikipedia.org/wiki/Theory_of_mind

Interesting article, I particularly like the simulation theory and the false-belief task. Amazing that 3-4 year old children can already lie.

If morals really were as subjective as is made out then it would be impossible for any multi-cultural society to function.

I think you have to clarify how the word "multicultural" is used. What it _does not_ mean is two distinct cultures living entwined among another - there is no culture in the world that I know of where one culture isn't dominant. There is nowhere in the world where half the people go to Western Christian Churches and the other half worship Huitzlipoctli through biennial human sacrifice - either Huitzipoctli is in charge and a few small Churches are tolerated/run underground or Christianity is in charge and Huitzlipoctli human sacrifices have to go on underground. In Britain, multicultural means that we tolerate foreign cultures so long as they don't do anything we really don't like. Funny hats, weird architecture, yes - wearing bits of cloth over your face, body alteration of the kind we don't like, espousing alien political systems, no. Even then, multiculturalism as a term is becoming increasingly unpopular - no one wants to say "assimilation" - but most everyone kind of wants it to happen, just as long as we don't have to _see_ it happening. Nor is it just the UK - Holland, most liberal and progressive of countries, has recently undergone the same shift. People liked living in a 'multicultural' society, just so long as the culture showed no signs of changing.

I'm just illustrating that differences in morality are not genetically based, but cultural.

In seems that there would be a strong selective pressure in favour of uniformity in all subjective elements that influence moral decisions so that the individuals would make good moral judgements whatever culture they grew up in, and this seems to be the case.

Again, you probably don't want to use the word "good" when you're talking about morality, particularly when the most obvious heuristic for 'getting along' in society is 'keep your head down'.

Only "good" in the sense I defined above.

As to the question of why different cultures have different morals. One's model need only be as accurate as those around one. While I don't think their morals are really that different to begin with, where they do differ it is based on their objective, testable beliefs. Not on some subjective, indefinable concept, which I propose we all share to a certain extent anyway thanks to natural selection.

I can't see what you're saying here - do you have any anthropology references to back this up?

No, this is just where I would like to get with the theory. I don't know if any research has been done yet.

How could the ability to predict behaviour be a disadvantage?

For one, because the interests of the individual do not align with the interests of the genes. For instance, in humans, unlike almost any other animal, the fertility of the female is deliberately obscured from both the female and everyone else. This isn't just an example of a failure to evolve a capacity to make certain vital predictions (and there are few things more vital to biological fitness than when you are/are not fertile), but a clear example where ignorance has been selected for.

In this example there is no way to know if the female is fertile. It is not a failure to predict behaviour that has been selected but the impossibility there of. If females showed visible signs of fertility and males failed to recognise this then it would be a clear example of ignorance being selected for.

More generally, as the little toy sim by Doran showed, in any situation where you have (i) less than total knowledge
and (ii) less than infinite computing power it is not necessarily the case that more/more accurate information will increase biological fitness.

In this example the behaviour was programmed directly. The bots did not construct models of their own environment and base their decisions on them.

I don't think any model that is not at least as complex as the thing it is modelling can ever be perfect. Conformity with the most realistically based morality available would be the next best thing.

No, I meant conformity in the sense that you were saying "It is good because other people judge it to be good". Aztec human sacrifice is good if most Aztecs agree that it's good? A definition of "good" that I'm happy with, but which doesn't seem to suit the rest of your model.

MArios

It's good enough for the Aztecs and only for the Aztecs. The main difference between us and the Aztecs is that they thought the the sun would not come up without human sacrifice. Based on their model of the universe their behaviour was completely justified. Our model is better.

[Marios]

Behaviour effects survival and thus genes. If behaviour is not governed directly by genes, and this is obviously the case, then they must ultimately effect behaviour through some other means of indirect control.

Yes, I don't have a problem with behaviour being 'indirectly' governed by genes and environment - in the same way that all phenotypic traits depend on the interactions of genes with their environment - I'm just not clear if you're saying anything more than "there are genes which affect how behaviours are socially transmitted and received" - which I don't think anyone would disagree with (short of removing themselves from any scientific discussion).

Fair enough, but I don't want to argue about the subjective good here in the evolution forum. For the purposes of this discussion can we simply define "the good" as that unknown individual factor which decides, within the context of any social system, what behaviour will more likely be judged as good and thus be more effective.

No, I'm sorry, the validity of any scientific discussion touching on culture really does hang on whether you're willing to adopt methodological cultural relativism. It doesn't reflect on what you believe in your personal life anymore than it matters whether chemists go to Mass once they leave the lab - so long as they are methodological naturalists in the lab.

I'm not really clear on what you're saying here 'unknown individual factor' within 'the context of any social system'? Sounds like you're assuming a supervening moral system - like CS Lewis did when he said:

The answer [to the claim that cultures radically differ on their moral codes] is that this is a lie—a good, solid, resounding lie.

He then went on to argue that witchhunters weren't evil because they were acting according to universal morality in the context of the mental models they had (i.e. witches are effectively mass-murdering terrorists - worse, there is no safe place you can lock them up).

Granted, that's a step up from someone says "Witch hunters are evil because, according to contemporary moral principle and contemporary models of the world, they did evil things", but it still assumes that there's such a thing as Universal Morality - which is a faith position (CS Lewis wasn't inconsistent here - yes, he dismissed anthropology as a pack of lies - but he was clear that all his arguments were founded a basic commitment to faith - he wasn't trying to present a scientific model).

Maybe that's not what you mean, but any discussion on applying scientific principles to human cultures requires that you first state whether you're adopting methodological cultural relativism (I'm not sure there's another kind!) to present a scientific model or whether you're presenting a model partisan to a specific moral conviction. The worst thing you can do is not be clear about which it is.

I'm talking about a theory which would predict which behaviour would be judged right more often. A theory that would predict the least subjective harm on an individual basis.

What does "least subjective harm on an individual basis" mean?

A hidden experience which cannot be observed is exactly how I would define a subjective experience.

Yes, that's what I mean - there are two senses of the word "subjective"
(1) "Subjective experience" is an experience no third party one can observe (qualia)
(2) "Subjective opinion" is an opinion that we can observe easily ("Do you think abortion is wrong?" "Yes"), but that we can't objectively determine to be true or false (it's true that they said it, but is it true that it's wrong?).

I think you're morality is subject in sense (2), but you seem to be describing it in terms of sense (1).

I'm just illustrating that differences in morality are not genetically based, but cultural.

In that paragraph, yes, but not in the sentence I quoted:

If morals really were as subjective as is made out then it would be impossible for any multi-cultural society to function.

The implication here is that you seem to be following CS Lewis in assuming that cultural variation is vastly overemphasised.

In this example there is no way to know if the female is fertile. It is not a failure to predict behaviour that has been selected but the impossibility there of.

No it isn't - chimps and bonobos both have visible signs of female fertility - subsequently selection has acted to make women and observers unaware of the fertility of that woman. If more and more accurate information was always advantageous - as you seem to be assuming - then how come human females have deliberately been selected for ignorance as to the state of their fertility, while female chimps and bonobos have not?

In this example the behaviour was programmed directly. The bots did not construct models of their own environment and base their decisions on them.

No so - have another read through:

In general, sources of limited or mistaken belief in a multiple agent system may easily be identified. They include:

* limited and faulty perception. An agent may be unable to perceive what is the case (e.g. because it is too far away) or may misperceive.
* limited and faulty communication. One agent may fail to pass information to another, or may do so in such a way that errors are introduced.
* beliefs becoming inaccurate with the passage of time. An agent may persist in a belief long after it has ceased to be accurate.
* faulty generalization.

Agents construct models of their world over time and communicate them to eachother.

It's good enough for the Aztecs and only for the Aztecs. The main difference between us and the Aztecs is that they thought the the sun would not come up without human sacrifice. Based on their model of the universe their behaviour was completely justified. Our model is better.

Are you saying that we think our model is better or that the Aztecs would (without having first been decimated by disease and conquest) have somehow naturally viewed our model as better? Or are you assuming some independent viewpoint from which cultures can be objectively defined as superior and inferior?

Marios

[Peter Grant]

Behaviour effects survival and thus genes. If behaviour is not governed directly by genes, and this is obviously the case, then they must ultimately effect behaviour through some other means of indirect control.

Yes, I don't have a problem with behaviour being 'indirectly' governed by genes and environment - in the same way that all phenotypic traits depend on the interactions of genes with their environment - I'm just not clear if you're saying anything more than "there are genes which affect how behaviours are socially transmitted and received" - which I don't think anyone would disagree with (short of removing themselves from any scientific discussion).

"there are genes which affect how behaviours are socially transmitted and received" I like that, it's a good way of putting it.

Fair enough, but I don't want to argue about the subjective good here in the evolution forum. For the purposes of this discussion can we simply define "the good" as that unknown individual factor which decides, within the context of any social system, what behaviour will more likely be judged as good and thus be more effective.

No, I'm sorry, the validity of any scientific discussion touching on culture really does hang on whether you're willing to adopt methodological cultural relativism. It doesn't reflect on what you believe in your personal life anymore than it matters whether chemists go to Mass once they leave the lab - so long as they are methodological naturalists in the lab.

I have no problem accepting cultural relativism, it's sort of essential to my argument really. What I do disagree with is the assumption that cultural relativism necessarily implies moral relativism:

Virtually all anthropologists today subscribe to the methodological and heuristic principles of Boas and his students in their research. But, according to Marcus and Fischer, when the principle of cultural relativism was popularized after World War II, it came to be understood "more as a doctrine, or position, than as a method." As a consequence, people misinterpreted cultural relativism to mean that all cultures are both separate and equal, and that all value systems, however different, are equally valid. Thus, people came to use the phrase "cultural relativism" erroneously to signify "moral relativism."

The evolutionary perspective on moral relativism makes sense:

Evolutionary biologists believe that morality is a natural phenomenon that evolved by natural selection acting at the individual level, and through group selection. Consequently they view morality as being relative, constituting any set of social behaviors that promoted the survival and successful reproduction of humans

I am quite willing to accept that if dolphins evolved a type of morality it might be completely different to ours. However, it seems that some sets of social behaviours are more likely to promote the survival and successful reproduction of humans than others.

I'm not really clear on what you're saying here 'unknown individual factor' within 'the context of any social system'? Sounds like you're assuming a supervening moral system - like CS Lewis did when he said:

The answer [to the claim that cultures radically differ on their moral codes] is that this is a lie—a good, solid, resounding lie.

He then went on to argue that witchhunters weren't evil because they were acting according to universal morality in the context of the mental models they had (i.e. witches are effectively mass-murdering terrorists - worse, there is no safe place you can lock them up).

Granted, that's a step up from someone says "Witch hunters are evil because, according to contemporary moral principle and contemporary models of the world, they did evil things", but it still assumes that there's such a thing as Universal Morality - which is a faith position (CS Lewis wasn't inconsistent here - yes, he dismissed anthropology as a pack of lies - but he was clear that all his arguments were founded a basic commitment to faith - he wasn't trying to present a scientific model).

Maybe that's not what you mean, but any discussion on applying scientific principles to human cultures requires that you first state whether you're adopting methodological cultural relativism (I'm not sure there's another kind!) to present a scientific model or whether you're presenting a model partisan to a specific moral conviction. The worst thing you can do is not be clear about which it is.

I agree that cultures radically differ on their moral codes, but I think they use roughly the same methods of devising them based on the available information.

I'm talking about a theory which would predict which behaviour would be judged right more often. A theory that would predict the least subjective harm on an individual basis.

What does "least subjective harm on an individual basis" mean?

Behaviour which is less likely to be judged as harmful by any given individual in the population, particularly when that behaviour affects the individual in question.

A hidden experience which cannot be observed is exactly how I would define a subjective experience.

Yes, that's what I mean - there are two senses of the word "subjective"
(1) "Subjective experience" is an experience no third party one can observe (qualia)
(2) "Subjective opinion" is an opinion that we can observe easily ("Do you think abortion is wrong?" "Yes"), but that we can't objectively determine to be true or false (it's true that they said it, but is it true that it's wrong?).

I think you're morality is subject in sense (2), but you seem to be describing it in terms of sense (1).

Agreed we cannot objectively determine if the opinion is false. This is because the opinion is formed using both objective facts and subjective experience of those facts. We can, however, determine that the subjective opinion is more likely to be wrong if one of the objective facts it is (in part) based on can be demonstrated to be false.

I'm just illustrating that differences in morality are not genetically based, but cultural.

In that paragraph, yes, but not in the sentence I quoted:

If morals really were as subjective as is made out then it would be impossible for any multi-cultural society to function.

The implication here is that you seem to be following CS Lewis in assuming that cultural variation is vastly overemphasised.

I think that cultural variation is assumed to imply moral relativism within the same species by incorrectly assuming that morality is purely a cultural phenomenon.

In this example there is no way to know if the female is fertile. It is not a failure to predict behaviour that has been selected but the impossibility there of.

No it isn't - chimps and bonobos both have visible signs of female fertility - subsequently selection has acted to make women and observers unaware of the fertility of that woman. If more and more accurate information was always advantageous - as you seem to be assuming - then how come human females have deliberately been selected for ignorance as to the state of their fertility, while female chimps and bonobos have not?

The ability to predict another individual's behaviour is an advantage to the predictor, but having another individual predict yours might not be. The obvious solution to this is to lie, but an individual who is very good at predicting behaviour might be able to see through this deception. When there would be more advantage in others not knowing than you yourself knowing there might be evolutionary pressure for this knowledge to become impossible for both parties to determine. The ability to predict behaviour generally would be far to useful to forfeit entirely.

In this example the behaviour was programmed directly. The bots did not construct models of their own environment and base their decisions on them.

No so - have another read through:

In general, sources of limited or mistaken belief in a multiple agent system may easily be identified. They include:

* limited and faulty perception. An agent may be unable to perceive what is the case (e.g. because it is too far away) or may misperceive.
* limited and faulty communication. One agent may fail to pass information to another, or may do so in such a way that errors are introduced.
* beliefs becoming inaccurate with the passage of time. An agent may persist in a belief long after it has ceased to be accurate.
* faulty generalization.

Agents construct models of their world over time and communicate them to eachother.

OK I missed that bit. After re-reading it though I still think that the model is a little too simplistic. It does predict that holding false beliefs is only effective when the majority of agents in a society hold that belief, which I agree with. It also says that those false beliefs are only effective when they correlate with reality in some way:

It is clear that belief and collective misbelief are inevitably at the heart of the behaviour of natural and artificial societies. Further, collective misbelief is not necessarily something to be avoided -- it may be functional if matched to the agents' environment.

His future work also looks interesting:

Is it the case that in more realistic scenarios the possible benefits of collective misbelief tend to be outweighed by disadvantages?

It's good enough for the Aztecs and only for the Aztecs. The main difference between us and the Aztecs is that they thought the the sun would not come up without human sacrifice. Based on their model of the universe their behaviour was completely justified. Our model is better.

Are you saying that we think our model is better or that the Aztecs would (without having first been decimated by disease and conquest) have somehow naturally viewed our model as better? Or are you assuming some independent viewpoint from which cultures can be objectively defined as superior and inferior?

Marios

I am making no comments about Aztec culture. I'm simply claiming that our model of the solar system is better and pointing out that one is less likely to draw conclusions from it linking human sacrifice to sunrise events.

[Peter Grant]

Well since there seem to be no further arguments, so far, I would like to direct attention to two articles which seem to support this theory:

Sense of justice discovered in the brain

Monkeys Show Sense Of Fairness, Study Says

[Marios]

I have no problem accepting cultural relativism, it's sort of essential to my argument really. What I do disagree with is the assumption that cultural relativism necessarily implies moral relativism:

Yes, that's what I mean by 'methodological cultural relativism' (like methodological naturalism). I think that to work in anthropology (science) you have to be embrace these as methodological requirements. Most scientists aren't actually naturalists outside of the lab - they are perfectly comfortable going to Church or using dualist constructs like "I" and "you" - equally, most anthropologists do actually seem to have their own personal preferences for when they are in their own societies. It's important to grasp that it isn't an all-or-nothing "you must adopt this for your entire life" - it's just for anyone who want to engage in anthropological/scientific discourse.

If you start discussing anthropology and - half way through - it suddenly becomes clear that you work from the premise that some societies, some behaviours are 'just naturally' superior/inferior - I think you can expect that most of the people interested in anthropology will leave suddenly - just as if it suddenly turned out that you were rejecting naturalism in a natural science discourse. All you'll be left with is the people who view science as a moral/political tool - the people who don't care about the subject, but do care about trying to use science/anthropology to lend authority to their political opinions.

The evolutionary perspective on moral relativism makes sense:

Evolutionary biologists believe that morality is a natural phenomenon that evolved by natural selection acting at the individual level, and through group selection. Consequently they view morality as being relative, constituting any set of social behaviors that promoted the survival and successful reproduction of humans

Worth noting that there's only one reference there and it's for Michael Shermer, a science writer, not an evolutionary biologist (I've had a brief glance at The Science of Good and Evil, not sure I'd agree with all of it, but at least the reviewers seem to be complaining that he's engaging in too much cultural relativism
http://en.wikipedia.org/wiki/Michael_Shermer
http://en.wikipedia.org/wiki/The_Science_of_Good_and_Evil:_Why_People_Cheat,_Gossip,_Care,_Share,_and_Follow_the_Golden_Rule

I am quite willing to accept that if dolphins evolved a type of morality it might be completely different to ours. However, it seems that some sets of social behaviours are more likely to promote the survival and successful reproduction of humans than others.

Yes, and some sets of genes seem to have been more likely to replicate themselves in the past - the result is not one kind of SuperAnimal but a vast diversity of organisms in different niches. Again, remember that group selection is no longer valid in any sense that distinguishes it from individual selection (I think the terms are _really_ unhelpful here, but some people do prefer to view the maths that way around) - social behaviours promote the survival and successful reproduction of those social behaviours insofar as they are socially transmitted. If you want to consider the effect of gene-culture coevolution you end up with some quite gnarly maths which doesn't really fold down into neat little answers.

I agree that cultures radically differ on their moral codes, but I think they use roughly the same methods of devising them based on the available information.

They do? This sounds very much like trying to ignore culture variation, which is problematic if you're proposing to study culture.

Behaviour which is less likely to be judged as harmful by any given individual in the population, particularly when that behaviour affects the individual in question.

Right - so not only are you edging towards ignoring variation between cultures, but you're also ignoring variation within a culture - trying to get a simple answer by dumping everyone into a blender and hoping something meaningful will come out? I'm sorry, but this approach seems considerably cruder than the sorts of agent-based models used to study less socially complex primates.

Agreed we cannot objectively determine if the opinion is false. This is because the opinion is formed using both objective facts and subjective experience of those facts.

No, I think you're missing my point - or assuming some induction model for _all beliefs_. If someone says "I think abortion/female circumcision is wrong" that is what I would view as a subjective opinion. I'm not sure what you mean by 'objective facts' and 'subjective experience' there

We can, however, determine that the subjective opinion is more likely to be wrong if one of the objective facts it is (in part) based on can be demonstrated to be false.

But I've shown you (Doran link) that this isn't necessarily true - do you have some sort of weak convergence proof that it is so, or is it just an intuition that it 'must be so'? Would you be willing to consider that it might be a false intuition?

I think that cultural variation is assumed to imply moral relativism within the same species by incorrectly assuming that morality is purely a cultural phenomenon.

Incorrectly assuming? I think that rather depends on what you mean by "purely a cultural phenonemenon". Certainly C.S. Lewis felt that it wasn't "purely cultural" - he felt that morality came from God - when someone makes a claim that they think there's some 'privileged morality' that underlies the 'superficial' variation I think it's perfectly reasonable to be skeptical, given that almost every culture seems to view their specific cultural morality as in some sense universal.

The ability to predict another individual's behaviour is an advantage to the predictor, but having another individual predict yours might not be. The obvious solution to this is to lie, but an individual who is very good at predicting behaviour might be able to see through this deception. When there would be more advantage in others not knowing than you yourself knowing there might be evolutionary pressure for this knowledge to become impossible for both parties to determine. The ability to predict behaviour generally would be far to useful to forfeit entirely.

It's an interesting line of conjecture - although perhaps a little implicitly Chauvinist. Women evolve the inability to be aware of their own fertility because they wouldn't be able to keep it from men who would then exploit them? Surely _first_ we should consider the simpler conjecture that women evolve the inability to be aware of their own fertility _because of how it influences their own behaviour_ (i.e. sex without pregnancy)?

Either way, you seem to have accepted that actually it isn't necessarily always advantageous to have more/more accurate information (even if we assumed this was costless). Once that's conceded we can no longer assume that there's just one optima and the beliefs and desires of individual agents needs to be considered separate from biological fitness.

OK I missed that bit. After re-reading it though I still think that the model is a little too simplistic.

Scientific models should never be any more complicated than they need to be - the point of the model was to communicate the point that it isn't necessarily the case that "more accurate information implies higher biological fitness" - if anything, he's maybe put more than he needs to in one paper (the second simulation).

It does predict that holding false beliefs is only effective when the majority of agents in a society hold that belief, which I agree with. It also says that those false beliefs are only effective when they correlate with reality in some way:

Yes, but bear in mind that for social organisms one of the principle elements of the "agents' environment" are the beliefs of the other agents.

Is it the case that in more realistic scenarios the possible benefits of collective misbelief tend to be outweighed by disadvantages?

This is the sort of question you need to look to answering - but it's important to grasp that there's a limit to what you can do just by referring to your own intuition (otherwise scientists wouldn't waste time with maths and experiments - we'd still be Natural Philosophers chewing the end of a pencil and trying to decide which essay seems 'more plausible' than the other).

I am making no comments about Aztec culture. I'm simply claiming that our model of the solar system is better and pointing out that one is less likely to draw conclusions from it linking human sacrifice to sunrise events.

Sorry, I don't agree - I think you're falling down here:

The main difference between us and the Aztecs is that they thought the the sun would not come up without human sacrifice.

By ignoring the fact that the Aztecs had a totally different culture you're implicitly assuming that they share all our values about what defines a model of the solar system as 'better'. I think the main difference between 21st century Westerners and the Aztecs is that we live in totally different cultures. I don't think it's informative to reduce an entire culture to "they had this weird idea about the sun not coming up if they didn't do blah" and then infer that all differences herald from that - you're putting the rationale for a behaviour _before_ the behaviour, which on a historical timescale is usually a big mistake (as far as I recall, we're talking about 500 years of human sacrifice prior to the rise of the Aztecs and their subsequent development of theology, so I think you're confusing cause with effect).

Based on their model of the universe their behaviour was completely justified. Our model is better.

This is exactly the same as C.S.Lewis's argument about witchfinders not being bad people. I'm certainly not disagreeing with the idea that you've got to look at belief systems in context, but the suggestion that you can assume that there's two contexts - one of which is changing and fallible human knowledge - the other is ultimate, underlying eternal truth - is implicitly religious. Obviously, that's not a problem for C.S. Lewis because he's arguing that religion is necessary, but it strikes me as a problem for you if you're trying to present this as a scientific naturalist approach.

If you don't do that, all you can conclude is that our model is better from our point of view (which is fairly trite). If you want to argue that Aztecs would have automatically viewed our approach as better, then I'm going to ask for evidence (which is hard since when the two cultures came into contact, Aztec culture was rapidly destabilised with disease and conquest).

Marios

[Marios]

Sense of justice discovered in the brain

This article is rather careless - have you noticed how it contradicts itself?

Humans are the only animals to act spitefully or to mete out "justice", dishing out punishment to people seen to be behaving unfairly - even if it is not in the punisher's own best interests.

"This form of spite is a bit of an evolutionary puzzle. There are few examples in the animal kingdom."

Few is not 0! There are plenty of examples of spite/reciprocal altruism in the animal kingdom - especially considering how hard it is to observe creatures like bonobos relative to humans.

I don't see anything in there that supports your theory.

Monkeys Show Sense Of Fairness, Study Says

You need to be careful about interpreting these - differential reporting (and the disciplinary biases in neighbouring subjects like economics) tends to sift out primatological results which can be interpreted as supporting current social norms. So, papers which show that some particular kind of primate seems to enjoy being hugged get the headlines as "Proving Hugging To Be Natural", whereas another piece of research that shows that chimps engage in genocide-like behaviour will be dismissed as Unnatural Behaviour due to Too Much or Too Few Bananas.

If you really want to look at primate social manouevring, I suggest de Waal's "Chimpanzee Politics" - real societies - even 10 chimp societies - have structure. The experiments that are quoted tend to do their best to remove structure (since experimenters very rightly want to minimise confounding factors), but when people report the experiments they often don't stress that the results are based on social interactions _in the absence of a pre-existing social structure_.

That's effectively like trying to learn about biochemistry by studying inorganic chemistry - certainly the two overlap, but if you adopt the assumptions of inorganic chemistry (no catalysts, no structure, well-mixed reactants) then you end up with results which are totally at odds with observation. In a sense, chemists do use simple organic chemistry to define "natural chemical interactions" (which makes most of biochemistry distinctly unnatural - if the "natural interactions" were to occur, we should explode or dissolve) - but I think that's a fairly safe use of the term "natural" to mean "simplistic". I don't think anyone _actually believes_ that we ought to stop breathing because the maldistribution of entropy is morally unacceptable.

However, some people do hold beliefs which reject certain social structures, so they tend to leap on the 'structure-free' research to claim that structurelessness is somehow natural (not merely an artefact of those specific kinds of experiment).

Again, I don't see how this supports your theory (accept in the irrelevant sense that it doesn't directly contradict it).

Marios

[Peter Grant]

I have no problem accepting cultural relativism, it's sort of essential to my argument really. What I do disagree with is the assumption that cultural relativism necessarily implies moral relativism:

Yes, that's what I mean by 'methodological cultural relativism' (like methodological naturalism). I think that to work in anthropology (science) you have to be embrace these as methodological requirements. Most scientists aren't actually naturalists outside of the lab - they are perfectly comfortable going to Church or using dualist constructs like "I" and "you" - equally, most anthropologists do actually seem to have their own personal preferences for when they are in their own societies. It's important to grasp that it isn't an all-or-nothing "you must adopt this for your entire life" - it's just for anyone who want to engage in anthropological/scientific discourse.

I do not think it is valid to separate the two. I saw no scientific reason to believe in God, though I had many personal reasons to believe. If my personal preferences differ from the scientific evaluation then it is my personal preferences which will have to change.

If you start discussing anthropology and - half way through - it suddenly becomes clear that you work from the premise that some societies, some behaviours are 'just naturally' superior/inferior - I think you can expect that most of the people interested in anthropology will leave suddenly - just as if it suddenly turned out that you were rejecting naturalism in a natural science discourse. All you'll be left with is the people who view science as a moral/political tool - the people who don't care about the subject, but do care about trying to use science/anthropology to lend authority to their political opinions.

I'm not suggesting that societies as a whole are superior or inferior. I am simply suggesting that certain subset of behaviours are less likely to be judged as bad in any cultural setting. These behaviours would ultimately be the most effective. I am trying to find a way to determine, as objectively as possible, what these behaviours would be. I have no political agenda, I am genuinely searching for a way to determine which behaviour will promote the most good, as it is independently judged by each individual.

The evolutionary perspective on moral relativism makes sense:

Evolutionary biologists believe that morality is a natural phenomenon that evolved by natural selection acting at the individual level, and through group selection. Consequently they view morality as being relative, constituting any set of social behaviors that promoted the survival and successful reproduction of humans

Worth noting that there's only one reference there and it's for Michael Shermer, a science writer, not an evolutionary biologist (I've had a brief glance at The Science of Good and Evil, not sure I'd agree with all of it, but at least the reviewers seem to be complaining that he's engaging in too much cultural relativism
http://en.wikipedia.org/wiki/Michael_Shermer
http://en.wikipedia.org/wiki/The_Science_of_Good_and_Evil:_Why_People_Cheat,_Gossip,_Care,_Share,_and_Follow_the_Golden_Rule

Too much cultural relativism? Now I'm confused, I thought "the validity of any scientific discussion touching on culture really does hang on whether you're willing to adopt methodological cultural relativism".

I am quite willing to accept that if dolphins evolved a type of morality it might be completely different to ours. However, it seems that some sets of social behaviours are more likely to promote the survival and successful reproduction of humans than others.

Yes, and some sets of genes seem to have been more likely to replicate themselves in the past - the result is not one kind of SuperAnimal but a vast diversity of organisms in different niches. Again, remember that group selection is no longer valid in any sense that distinguishes it from individual selection (I think the terms are _really_ unhelpful here, but some people do prefer to view the maths that way around) - social behaviours promote the survival and successful reproduction of those social behaviours insofar as they are socially transmitted. If you want to consider the effect of gene-culture coevolution you end up with some quite gnarly maths which doesn't really fold down into neat little answers.

What if we turn it around and say that the ability to reproduce social behaviours, whatever they may be, promotes the individual's survival?

I agree that cultures radically differ on their moral codes, but I think they use roughly the same methods of devising them based on the available information.

They do? This sounds very much like trying to ignore culture variation, which is problematic if you're proposing to study culture.

Cultural variation is the key to determining the underlying method. Without it, we would have no comparisons to make.

Behaviour which is less likely to be judged as harmful by any given individual in the population, particularly when that behaviour affects the individual in question.

Right - so not only are you edging towards ignoring variation between cultures, but you're also ignoring variation within a culture - trying to get a simple answer by dumping everyone into a blender and hoping something meaningful will come out? I'm sorry, but this approach seems considerably cruder than the sorts of agent-based models used to study less socially complex primates.

It is conceivable that moral ability would vary as much as physical fitness and intellectual ability in the general population. Empathy, for instance, varies greatly among individuals but most of us are capable of it to some extent.

Agreed we cannot objectively determine if the opinion is false. This is because the opinion is formed using both objective facts and subjective experience of those facts.

No, I think you're missing my point - or assuming some induction model for _all beliefs_. If someone says "I think abortion/female circumcision is wrong" that is what I would view as a subjective opinion. I'm not sure what you mean by 'objective facts' and 'subjective experience' there

If someone says they don't like ice-cream they are simply referring to their own subjective experience of eating it. If someone says they think abortion is wrong they are referring to a model of reality they have in their minds. If this model includes souls and spirits, how can one be sure they are making a valid judgement? One cannot be sure that one doesn't like ice-cream if one isn't certain that what one is experiencing eating is actually ice-cream.

We can, however, determine that the subjective opinion is more likely to be wrong if one of the objective facts it is (in part) based on can be demonstrated to be false.

But I've shown you (Doran link) that this isn't necessarily true - do you have some sort of weak convergence proof that it is so, or is it just an intuition that it 'must be so'? Would you be willing to consider that it might be a false intuition?

The Doran link showed that if the majority of a population holds a false belief then there may be some evolutionary advantage in holding that belief. There are infinitely more potentially false beliefs than true beliefs though, as there is only one reality. A false belief might result in the right behaviour purely by chance, but it is just as likely to result in the wrong behaviour. A true belief will result in the right behaviour every time.

I think that cultural variation is assumed to imply moral relativism within the same species by incorrectly assuming that morality is purely a cultural phenomenon.

Incorrectly assuming? I think that rather depends on what you mean by "purely a cultural phenonemenon". Certainly C.S. Lewis felt that it wasn't "purely cultural" - he felt that morality came from God - when someone makes a claim that they think there's some 'privileged morality' that underlies the 'superficial' variation I think it's perfectly reasonable to be skeptical, given that almost every culture seems to view their specific cultural morality as in some sense universal.

I doubt C.S. Lewis would have suggested an evolutionary basis for morality, he believed in original sin after all.

The ability to predict another individual's behaviour is an advantage to the predictor, but having another individual predict yours might not be. The obvious solution to this is to lie, but an individual who is very good at predicting behaviour might be able to see through this deception. When there would be more advantage in others not knowing than you yourself knowing there might be evolutionary pressure for this knowledge to become impossible for both parties to determine. The ability to predict behaviour generally would be far to useful to forfeit entirely.

It's an interesting line of conjecture - although perhaps a little implicitly Chauvinist. Women evolve the inability to be aware of their own fertility because they wouldn't be able to keep it from men who would then exploit them? Surely _first_ we should consider the simpler conjecture that women evolve the inability to be aware of their own fertility _because of how it influences their own behaviour_ (i.e. sex without pregnancy)?

Either way, you seem to have accepted that actually it isn't necessarily always advantageous to have more/more accurate information (even if we assumed this was costless). Once that's conceded we can no longer assume that there's just one optima and the beliefs and desires of individual agents needs to be considered separate from biological fitness.

I am not saying that it was always advantageous to have more accurate information, simply that it was more advantageous to have a brain capable of building the most accurate model possible based on the information available. How else do we explain our current ability to create such models today?

OK I missed that bit. After re-reading it though I still think that the model is a little too simplistic.

Scientific models should never be any more complicated than they need to be - the point of the model was to communicate the point that it isn't necessarily the case that "more accurate information implies higher biological fitness" - if anything, he's maybe put more than he needs to in one paper (the second simulation).

It is not the information that is being selected for but the ability to create models based on that information.

It does predict that holding false beliefs is only effective when the majority of agents in a society hold that belief, which I agree with. It also says that those false beliefs are only effective when they correlate with reality in some way:

Yes, but bear in mind that for social organisms one of the principle elements of the "agents' environment" are the beliefs of the other agents.

Exactly my point. The whole reason the agent creates a model is to predict the behaviour of other agents.

Is it the case that in more realistic scenarios the possible benefits of collective misbelief tend to be outweighed by disadvantages?

This is the sort of question you need to look to answering - but it's important to grasp that there's a limit to what you can do just by referring to your own intuition (otherwise scientists wouldn't waste time with maths and experiments - we'd still be Natural Philosophers chewing the end of a pencil and trying to decide which essay seems 'more plausible' than the other).

Agreed, but intuition is often a basis for further research. I'm hoping our discussion might lead to it.

I am making no comments about Aztec culture. I'm simply claiming that our model of the solar system is better and pointing out that one is less likely to draw conclusions from it linking human sacrifice to sunrise events.

Sorry, I don't agree - I think you're falling down here:

The main difference between us and the Aztecs is that they thought the the sun would not come up without human sacrifice.

By ignoring the fact that the Aztecs had a totally different culture you're implicitly assuming that they share all our values about what defines a model of the solar system as 'better'. I think the main difference between 21st century Westerners and the Aztecs is that we live in totally different cultures. I don't think it's informative to reduce an entire culture to "they had this weird idea about the sun not coming up if they didn't do blah" and then infer that all differences herald from that - you're putting the rationale for a behaviour _before_ the behaviour, which on a historical timescale is usually a big mistake (as far as I recall, we're talking about 500 years of human sacrifice prior to the rise of the Aztecs and their subsequent development of theology, so I think you're confusing cause with effect).

Aztec values concerning the model of the solar system are irrelevant. Our model makes better predictions. Any moral values they held based on the apparent movement of the sun was therefore probably wrong in any context other than Aztec culture. This is not to say that all of their values were wrong, only those that do not conform with reality. I'm sure that they loved their children etc. Their morals were most probably superior to ours in other ways.

Based on their model of the universe their behaviour was completely justified. Our model is better.

This is exactly the same as C.S.Lewis's argument about witchfinders not being bad people. I'm certainly not disagreeing with the idea that you've got to look at belief systems in context, but the suggestion that you can assume that there's two contexts - one of which is changing and fallible human knowledge - the other is ultimate, underlying eternal truth - is implicitly religious. Obviously, that's not a problem for C.S. Lewis because he's arguing that religion is necessary, but it strikes me as a problem for you if you're trying to present this as a scientific naturalist approach.

I don't think the witchfinders, in their cultural context, were necessarily bad either. The two contexts I propose is ever changing, and improving, human knowledge with an underlying evolved predisposition toward moral behaviour.

If you don't do that, all you can conclude is that our model is better from our point of view (which is fairly trite). If you want to argue that Aztecs would have automatically viewed our approach as better, then I'm going to ask for evidence (which is hard since when the two cultures came into contact, Aztec culture was rapidly destabilised with disease and conquest).

Marios

Our model of the solar system is demonstrably better. As to the rest of their culture I have not done enough research.

[Peter Grant]

Regarding Human sacrifice in Aztec culture I found the following on Wikipedia:

The cycle of fifty-two years was central to Mesoamerican cultures. The Nahua's religious beliefs were based on a great fear that the universe would collapse after each cycle if the gods were not strong enough. Every fifty-two years a special New Fire ceremony was performed.[18] All fires were extinguished and at midnight a human sacrifice was made. The Aztecs waited for the dawn. If the Sun appeared it meant that the sacrifices for this cycle had been enough. A fire was ignited on the body of a victim, and this new fire was taken to every house, city and town. Rejoicing was general: a new cycle of fifty-two years was beginning, and the end of the world had been postponed, at least for another 52-year century. (A similar ceremony is still practiced by small indigenous groups, but without human sacrifice.) The ceremony was older than the Aztecs. While originally it was believed it was a matter of luck to survive, the Aztecs thought that constant sacrifice through the fifty-two year cycle could postpone the end.

The above has been demonstrated to be false as there have been numerous fifty-two year cycles since the practice has ceased. Presumably the god's strength is not dependant on human sacrifice after all.

[Marios]

The above has been demonstrated to be false as there have been numerous fifty-two year cycles since the practice has ceased. Presumably the god's strength is not dependant on human sacrifice after all.

As I said, that would be true if:
(i) that's an accurate and complete reconstruction of a prevalent Aztec worldview and
(ii) mental models dictate cultural practices strictly (rather than cultural practices dictating mental models or some more realistic coevolution model)

For instance, a lot of people believed (and believe) that "Science will create a better world" before WWI/WWII. Many of their predictions turned out to be entirely false (have a read over late 19th century/very early 20th century utopian literature) - but not many people said "Hmm, well that's clearly incorrect then, I had best resign my comfortable little professorship and give the Mormons a go". What they did was that they discarded inconvenient predictions and redefined terms - even the old favourite ("Hmm, well it would have, but Bad People stopped it").

That's not an unrepresentative sample, either - I would say that's generally the rule for any cultural system (think of any of the well documented cultural models of the last century) - the only system where that doesn't occur is science (and even then it's an uphill struggle to persaud people to abadon theories just because the predictions don't seem to have panned out).

I think the mistake is in assuming that the observable predictions of a system of thought are what's actually important about it - outside of the natural sciences, I can't think of a system where that's true. Observable predictions are more like are like origin myths in religion - just because they logically or historically seem to be predate the religion doesn't mean they actually do (people who have no experience of religion beyond Christianity often seem to get confused by the idea that other religions - including early Christianity - often have multiple origin myths).

Will come back to this later.
Marios

[ScholasticSpastic]

This is certainly a philosophical discussion- or has become so.

The specific area of philosophy is meta-ethics.

In meta-ethics, it is equally valid to state that ethical ends and ethical means can be objectively assessed or that they cannot be objectively assessed.

If the participants in this discussion cannot agree about meta-ethical premises, I do not see how it can result in anything but frustration for everyone concerned. I can well imagine a moral realist arguing with a subjectivist indefinitely and to little positive effect for either party.

[Peter Grant]

Hey thanks for the post ScholasticSpastic, I thought this thread was dead for sure.

Would that make me a moral realist? I kind of like that, it has a nice ring to it.

If I may ask, what do you think of the original argument at the top of this page?

[ScholasticSpastic]

If I may ask, what do you think of the original argument at the top of this page?

I tried to frame a very similar argument in another thread. Got into it with Marios there, too. As the frustration mounted, I started fishing around for some way to understand how I could so disagree with Marios's conclusions while failing to find fault with his reasoning. Then I found meta-ethics and it all made sense. Unfortunately, that thread fell off the first page as soon as I suggested that our disagreement might have a meta-ethical basis and I don't expect it to come back.

I do not know enough about meta-ethics to discern which type we might subscibe to, but I'm sure there is one for us. I also suspect that the applicable meta-ethical premises will be identical or similar for us as we've both attempted to argue the same point from those premises.

It does seem rather parsimonious that there would be selection for an ethical system which tends to better ensure attainment of biological imperatives over time within a population of social organisms subject to natural selection.

A great many old myths and superstitions have now been linked to real benefits. For example, there is an aboriginal American superstition that pregnant women and their partners are not allowed to come into contact with birds or reptiles (and I think amphibians figure in there somewhere). Now we know that salmonella would have been a problem- especially in the absence of antibiotics and reliable cooking and cleaning methods. This does not, of course, validate their superstition. But it is strongly indicative that we cannot consider culture as independant from our embodiment. And we've certainly demonstrated that variation within populations (which can include cultures) is greater than variation between populations where humans are concerned.

[Marios]

Unfortunately, that thread fell off the first page as soon as I suggested that our disagreement might have a meta-ethical basis and I don't expect it to come back.

Will come back to that too in a bit.
Marios

[ScholasticSpastic]

Unfortunately, that thread fell off the first page as soon as I suggested that our disagreement might have a meta-ethical basis and I don't expect it to come back.

Will come back to that too in a bit.
Marios

[hotshoe]

And we've certainly demonstrated that variation within populations (which can include cultures) is greater than variation between populations where humans are concerned.

I think you're missing a clause that would be necessary for this sentence to mean what you wanted it to mean.

[ScholasticSpastic]

And we've certainly demonstrated that variation within populations (which can include cultures) is greater than variation between populations where humans are concerned.

I think you're missing a clause that would be necessary for this sentence to mean what you wanted it to mean.

Yes. They declaused me because I was scratching up the sofa.

Edit: Are you sure? I may be being obtuse, but I'm not seeing it. Perhaps when I said I liked full frontal they took me to the wrong clinic.

[hotshoe]

I'm sorry, I guess it was me who was missing something today.

I guess you mean that the variation within populations would be sufficient to give selection something to work on, variation in how much an individual's ethical system tends to better work with biological imperatives ?

[ScholasticSpastic]

I'm sorry, I guess it was me who was missing something today.

I guess you mean that the variation within populations would be sufficient to give selection something to work on, variation in how much an individual's ethical system tends to better work with biological imperatives ?

Nah, it was just a statement of general homogeneity- pointing out that we are a very uniform species. As such, we have very uniform physiological requirements and pathogenic vulnerabilities which would serve as a universal basis for a minimal degree of commonality regarding possible "good lives".

[Peter Grant]

Hi Marios. Sorry I missed your post.

The above has been demonstrated to be false as there have been numerous fifty-two year cycles since the practice has ceased. Presumably the god's strength is not dependant on human sacrifice after all.

As I said, that would be true if:
(i) that's an accurate and complete reconstruction of a prevalent Aztec worldview and
(ii) mental models dictate cultural practices strictly (rather than cultural practices dictating mental models or some more realistic coevolution model)

Well I'm not judging the entire Aztec culture, just the human sacrifice part. Human sacrifice would be fine in Aztec society but the moment they interact with another culture who doesn't share their belief in the same gods difficulties begin to arise. Since their gods obviously don't exist, the chances of another culture independently coming up with the same idea are infinitely small. Ideas about morality that geographically separate culture's share are therefore more likely to be based on genetic or environmental similarities.

For instance, a lot of people believed (and believe) that "Science will create a better world" before WWI/WWII. Many of their predictions turned out to be entirely false (have a read over late 19th century/very early 20th century utopian literature) - but not many people said "Hmm, well that's clearly incorrect then, I had best resign my comfortable little professorship and give the Mormons a go". What they did was that they discarded inconvenient predictions and redefined terms - even the old favourite ("Hmm, well it would have, but Bad People stopped it").

I think it could be argued that until recently science has had more of a technological effect that an effect on world-views. Improved communication was as effective at spreading propaganda and disinformation as it was at spreading knowledge. Perhaps it is only now with the internet, where we all have a more equal voice, that true selection of ideas can occur.

That's not an unrepresentative sample, either - I would say that's generally the rule for any cultural system (think of any of the well documented cultural models of the last century) - the only system where that doesn't occur is science (and even then it's an uphill struggle to persaud people to abadon theories just because the predictions don't seem to have panned out).

That's it, I'm hoping we can one day apply scientific principals to moral theory.

I think the mistake is in assuming that the observable predictions of a system of thought are what's actually important about it - outside of the natural sciences, I can't think of a system where that's true. Observable predictions are more like are like origin myths in religion - just because they logically or historically seem to be predate the religion doesn't mean they actually do (people who have no experience of religion beyond Christianity often seem to get confused by the idea that other religions - including early Christianity - often have multiple origin myths).

If that's true, then it seems there is little point in thinking. One has to ask the question why we are so good at it compared to other animals. Since the scientific method is a relatively new invention, it couldn't have had much effect on the evolution of our brains.