right and wrong

[B.H.]

I have a question.

Are ideas of right and wrong in a moral sense really just facades?

I guess we can go one step farther and ask the question whether laws are facades too.

[Sidhe747]

Moral pluralism is the only Ethics, we can't place morals into a moral=0 framework so essentially morals=infinity or something else by definition.

If not then what are ethics any more than what are morals? Worthless?:dunno:

[B.H.]

Moral pluralism is the only Ethics, we can't place morals into a moral=0 framework so essentially morals=infinity or something else by definition.

If not then what are ethics any more than what are morals? Worthless?:dunno:

Can you explain some more?

[Sidhe747]

Moral pluralism is the only Ethics, we can't place morals into a moral=0 framework so essentially morals=infinity or something else by definition.

If not then what are ethics any more than what are morals? Worthless?:dunno:

Can you explain some more?

Ok there are two extremes moral nihilism and moral certainty.

Between the two are morals outside them are ethics, and how we justify morals.

Nihilism in morals is all very well but what do you replace it with when you have battered down your moral icons?

In the spectrum of moral fortune what is The - value Divine command, and what is the + value moral relativism?

And how can we have ethics in the first place? Without moral questions? And without a God who is certain?

Are ideas right or wrong?

Are they moral?

And how would we know?

Right and wrong is a value judgement so who decides?

Good and evil?

Right/wrong?

Bad/Good?

[B.H.]

What if you think that good is what is best for you regardless of others, even to the point you think it is okay for rules to apply to others but not you as long as you have the power to do so and get away with it?

[Preno]

I have a question.

Are ideas of right and wrong in a moral sense really just facades?

I guess we can go one step farther and ask the question whether laws are facades too.Meaning? :dunno:

[Jobar]

"Facades"- that seems a strange term to apply, here.

I would say that the judgment of right or wrong has to be relative to the judge; it's not an absolute. Nevertheless, if we recognize the relative nature of right/wrong, they can still be useful and relevant terms when discussing ethics.

[Yahzi]

Are ideas of right and wrong in a moral sense really just facades?

No, they are biologically evolved coping strategies for the pressures of social living.

Because they are products of evolution (complete with specialized mental hardware) they are objective. Because they are unique to each species and the situations that specie evolved to, they are not absolute.

[His Noodly Appendage]

Morality is the value placed on an action by the person making the judgement. I don't see what's so nihilistic about that.

[B.H.]

Are ideas of right and wrong in a moral sense really just facades?

No, they are biologically evolved coping strategies for the pressures of social living.

Because they are products of evolution (complete with specialized mental hardware) they are objective. Because they are unique to each species and the situations that specie evolved to, they are not absolute.

Can you give me an example?

[Alex]

If morality is hard-wired into the human psyche, is immoral conduct merely the result of a "short circuit" or a "blown fuse"? I ask because it's unclear where personal responsibility fits into the picture.

From what we know of the Aztecs, a least some of them appear to have believed that human sacrifice was a benefit to society. If their religious observances were among their coping strategies for social living, on what grounds can we say their behaviour was wrong?

[Sidhe747]

If morality is hard-wired into the human psyche, is immoral conduct merely the result of a "short circuit" or a "blown fuse"? I ask because it's unclear where personal responsibility fits into the picture.

From what we know of the Aztecs, a least some of them appear to have believed that human sacrifice was a benefit to society. If their religious observances were among their coping strategies for social living, on what grounds can we say their behaviour was wrong?

The Aztecs were crushed by Spain.

[Alex]

The Aztecs were crushed by Spain.

Your point being....?

[Sidhe747]

The Aztecs were crushed by Spain.

Your point being....?

Godless heathens 0 Catholic Spain 1.

[Valheru]

Godless heathens 0 Catholic Spain 1.

Your point being...?

[Sidhe747]

Godless heathens 0 Catholic Spain 1.

Your point being...?

God beats Qutaelotacpol 10-0

Is the point obviously. :bang:

[muidiri]

No, they are biologically evolved coping strategies for the pressures of social living.

Because they are products of evolution (complete with specialized mental hardware) they are objective. Because they are unique to each species and the situations that specie evolved to, they are not absolute.

:thumbup: Very clearly stated.

[muidiri]

Godless heathens 0 Catholic Spain 1.

Your point being...?

God beats Qutaelotacpol 10-0

Is the point obviously. :bang:

It's NOT clear, Sidhe. Are you trying to say that because the Spanish crushed the Aztecs, the Spanish morals were better? And if that is what you're saying, how does that address the OP? Do you see that as evidence that morality is a facade? Or what? None of us know what you're trying to say here.

[Sidhe747]

Godless heathens 0 Catholic Spain 1.

Your point being...?

God beats Qutaelotacpol 10-0

Is the point obviously. :bang:

It's NOT clear, Sidhe. Are you trying to say that because the Spanish crushed the Aztecs, the Spanish morals were better? And if that is what you're saying, how does that address the OP? Do you see that as evidence that morality is a facade? Or what? None of us know what you're trying to say here.

Yes.

[muidiri]

Yes.

:mad: Can you elaborate with more than one word, please?

[Rilx]

I have a question.

Are ideas of right and wrong in a moral sense really just facades?

I guess we can go one step farther and ask the question whether laws are facades too.
I think you are speaking of double morality. Do you mean that you don't see true morality anywhere? Laws are nothing but winners dictations?

[B.H.]

I have a question.

Are ideas of right and wrong in a moral sense really just facades?

I guess we can go one step farther and ask the question whether laws are facades too.
I think you are speaking of double morality. Do you mean that you don't see true morality anywhere? Laws are nothing but winners dictations?

Pretty much.

[Eudaimonist]

Morality is the value placed on an action by the person making the judgement. I don't see what's so nihilistic about that.

If the judgment isn't based in anything pertaining to the world -- if all such judgments are at root arbitrary and could just as correctly be something else -- then that is nihilistic.


eudaimonia,

Mak

[Alex]

Morality is the value placed on an action by the person making the judgement. I don't see what's so nihilistic about that.

If the judgment isn't based in anything pertaining to the world -- if all such judgments are at root arbitrary and could just as correctly be something else -- then that is nihilistic.

eudaimonia,

Mak

I think he was talking about moral relativism.

[Rie]

:sadyes:..Is it evil , not right or just a follow on from the old old stories that a man would lie and involve a woman's affections and then just drop off her radar and then ,this morning, come back and say he was 'sorry' and didn't know why he did it, just that he 'was married' and and and. You here at the cafe haver been good for me and really helped get me through this. I am having a computer rebuild.. a whizz bang one.. my daughter is doing it for me and even A BIG SCREEN!!! The point is that I'll be away from the computer for a little while but back in time for Xmas wishes for you all. Rie

[Yahzi]

Can you give me an example?
Monkeys demanding equal pay for equal work.

Male tigers allowing cubs (even when it's not their own) to eat first.

People making non-(strictly)-rational decisions to punish cheaters even when it costs them more than the cheater stole.

And in humans, Theory of Mind.

[Febble]

Are ideas of right and wrong in a moral sense really just facades?

No, they are biologically evolved coping strategies for the pressures of social living.

Because they are products of evolution (complete with specialized mental hardware) they are objective.

What do you mean by "objective" here? Why would a product of evolution be necessarily "objective"?

Because they are unique to each species and the situations that specie evolved to, they are not absolute.

What's the deal with "evolved"? I don't understand your logic here.

[Yahzi]

What do you mean by "objective" here? Why would a product of evolution be necessarily "objective"?
(peers suspiciously)
You're not Preno in disguise, are you?

Wel'l.. OK.

By objective I just mean non-subjective; that is, the reality that is independent of any particular perception of it. Evolution is of course objective because it (and its products) are rooted in the non-subjective world.

What's the deal with "evolved"? I don't understand your logic here.
What I am trying to say is that creatures with different biologies, and thus different problems to face, will have different moralities. Our morality is objective, in that it is a real-world response to real-world problems; but it is not absolute, in that other kinds of creatures might have other problems. For example, we place a high value on individual babies; intelligent turtles, who produce hundreds of thousands of young, probably wouldn't. Infanticide is a terrible crime for us; for them, it would be at worst a misdeamoner.

We evolved to cope with our environment. Part of our environment consists of self-aware creatures with needs virtually the same as ours - needs that are most easily met through cooperation. Ergo, we evolved to cooperate. We evolved mechanisms for encouraging, coordinating, regulating, and enforcing cooperation, as well as mechanisms for protecting ourselves when those other methods fail.

[Febble]

What do you mean by "objective" here? Why would a product of evolution be necessarily "objective"?
(peers suspiciously)
You're not Preno in disguise, are you?

Wel'l.. OK.

By objective I just mean non-subjective; that is, the reality that is independent of any particular perception of it. Evolution is of course objective because it (and its products) are rooted in the non-subjective world.

Hmm. I'm still not getting your point. Evolutionary processes are a model of reality. They are a fairly objective model because different people can observe different data that supports what can be objectively observed to be the model. And we can devise an evolutionary model that accounts for what we call "moral" behaviour - altruism, for instance, or cheater-punishing. But we can do exactly the same for "religious" behaviour, or "irrational" behaviour.

It seems to me that all you are doing is saying that it is possible to account moral behaviour objectively, not that moral behaviour is objective.

What's the deal with "evolved"? I don't understand your logic here.
What I am trying to say is that creatures with different biologies, and thus different problems to face, will have different moralities. Our morality is objective, in that it is a real-world response to real-world problems; but it is not absolute, in that other kinds of creatures might have other problems. For example, we place a high value on individual babies; intelligent turtles, who produce hundreds of thousands of young, probably wouldn't. Infanticide is a terrible crime for us; for them, it would be at worst a misdeamoner.

Well, I don't disagree with any of that.

We evolved to cope with our environment. Part of our environment consists of self-aware creatures with needs virtually the same as ours - needs that are most easily met through cooperation. Ergo, we evolved to cooperate. We evolved mechanisms for encouraging, coordinating, regulating, and enforcing cooperation, as well as mechanisms for protecting ourselves when those other methods fail.

Sure. But, as I said, that's true of other behavours as well, including religious behaviour.

I guess I am still missing your point.

[DMB]

I like this article (http://http://www.theage.com.au/opinion/contributors/it-seems-biology-not-religion-equals-morality-20091202-k5x8.html). I don't suppose that what this chap is saying is new or unfamilar to any of us, but IMO he expresses it remarkably well.

Sorry. I've been told that the link doesn't work. Here it is again: http://www.theage.com.au/opinion/contributors/it-seems-biology-not-religion-equals-morality-20091202-k5x8.html

I don'tknow what to do if that doesn't work.

[Yahzi]

Evolutionary processes are a model of reality.
The Theory of Evolution is a model humans have constructed to explain certain observations. Evolutionary processes - that is, the source of the observations - are not a model. They are part of objective reality.

Now whether or not those processes mirror or even resemble our model is not really our concern here - the fact is that those processes occur in the objective world and produce objective results (namely, people with moral intuitions).

And we can devise an evolutionary model that accounts for what we call "moral" behaviour - altruism, for instance, or cheater-punishing. But we can do exactly the same for "religious" behaviour, or "irrational" behaviour.
Yes, we can derive evolutionary origins for religion and irrationality. We sort of have to, since evolution is the only possible explanation for human nature.

It seems to me that all you are doing is saying that it is possible to account moral behaviour objectively, not that moral behaviour is objective.
I am saying it is possible to state objectively what morality is. That is, for a human being, unfairness is always immoral. This is an objective statement about human biology, just as "gaseous HCN is always poisonous" is an objective truth about human biology.

I guess I am still missing your point.
My point is that morality is an innate function of the human brain, just like language. It is flexible to some extent but also limited its expression by hard-wired heuristics. (Language is more flexible than morality). By understanding that the source of morality is evolved biology, we can dispense with worrying about gods or relativism and get down to the business of uncovering what is actually best for us. And as usual, such a journey of discovery is most rewarding when limited to empirical reality rather than cultural constructions or sheer fantasy.

[Yahzi]

When it comes to judging unfamiliar moral scenarios, your cultural background is virtually irrelevant.

http://www.theage.com.au/opinion/contributors/it-seems-biology-not-religion-equals-morality-20091202-k5x8.html
Exactly what I have been saying - culture does not construct morality; rather it constructs exemptions to morality.

Everybody intuitively knows that being a slave is bad; to make being a slaveholder good requires culture.

They are like recognising the identity relationship of 1=1, a rule that is abstract and content-free.
Is this guy channeling me or what?

I love this guy!

Lest there be any confusion about the claims I am making, I am not saying that our evolved capacity to intuitively judge what is right or wrong is sufficient to live a moral life.
Right, in exactly the same way our evolved capacity for language is not sufficient to communicate. We have to learn language to get the most out of it, and we have to learn morality for the same reason.

But morality is, at its root, a biological function. This makes it objective, not relative.

Great article (although I could have done without the gratuitous fist-pump for the Dali Lama).

[Febble]

Evolutionary processes are a model of reality.
The Theory of Evolution is a model humans have constructed to explain certain observations. Evolutionary processes - that is, the source of the observations - are not a model. They are part of objective reality.

In what sense? I think you are making a heuristic (and reasonable) distinction here, not a real one. All our observations are models. We don't have access to "objective reality". All we have are models.

Now whether or not those processes mirror or even resemble our model is not really our concern here - the fact is that those processes occur in the objective world and produce objective results (namely, people with moral intuitions).

The "fact is" that we model what we believe to be "the objective world". The "objective world" is not a fact. It is another model. And no, I won't fail your baseball test, because I think the "objective world" model is a decent model. It works. That, in my view, is the best you can say for a model - that it is useful.

And we can devise an evolutionary model that accounts for what we call "moral" behaviour - altruism, for instance, or cheater-punishing. But we can do exactly the same for "religious" behaviour, or "irrational" behaviour.
Yes, we can derive evolutionary origins for religion and irrationality. We sort of have to, since evolution is the only possible explanation for human nature.

Well, no. There are other possible explanations, and, indeed, the "evolutionary model" like all good scientific models, is subject to constant revision. However, to rule out a fundamental change to it, in advance, would invite hubris, IMO. We may have, one day, a more useful model.

It seems to me that all you are doing is saying that it is possible to account moral behaviour objectively, not that moral behaviour is objective.
I am saying it is possible to state objectively what morality is. That is, for a human being, unfairness is always immoral. This is an objective statement about human biology, just as "gaseous HCN is always poisonous" is an objective truth about human biology.

I'm afraid I think this is simply wrong. Firstly, I don't think it's possible to state "objectively what morality is". Or rather, I'm not sure what stating it non-objectively would be. Secondly, I don't think your statement is correct. In most societies, "unfairness" has been seen as highly moral - that it is "immoral" to disrupt society with claims to equality.

The rich man in his castle/the poor man at his gate/ GOD made them, high or lowly/ And ordered their estate

I don't believe that's moral, and I don't believe in a god who made them that way.

But it falsifies your claim that "for a human being, unfairness is always immoral". For the human beings who sang that, unfairness was moral.

I guess I am still missing your point.
My point is that morality is an innate function of the human brain, just like language. It is flexible to some extent but also limited its expression by hard-wired heuristics. (Language is more flexible than morality). By understanding that the source of morality is evolved biology, we can dispense with worrying about gods or relativism and get down to the business of uncovering what is actually best for us. And as usual, such a journey of discovery is most rewarding when limited to empirical reality rather than cultural constructions or sheer fantasy.

Well, I don't disagree with that. But I disagree with a lot of what you have said on the way ;)

[JamesBannon]

subjectivity != relativity

[muidiri]

subjectivity != relativity

That's true, and too often forgotten. Good call.

[Febble]

subjectivity != relativity

indeed.

[JamesBannon]

subjectivity != relativity

indeed.

I agree with much of what you said earlier, Febble. All we have are "models". We have some damn good ones, of course, but they are, nevertheless, models. Claiming that they represent the objective state of reality, whatever that claim actually means, is unsupportable.

[Yahzi]

We don't have access to "objective reality". All we have are models.
We do have access to objective reality. The observations of objective reality are what our models are based on.

If all we had was the model, then how could we ever have an observation that did not fit the model? The fact that we can revise our model based on unexpected input indicates that there is something shaping the model other than the model itself.

You are confusing what we know with what actually exists.

The "objective world" is not a fact. It is another model.
A model of what?

Are we even using model in the same sense? A model is a small, crude, incomplete replica of a thing. What is "objective reality" a small, crude, incomplete replica of?

That, in my view, is the best you can say for a model - that it is useful.
Why is it useful?

I think it is useful because it is true - or very nearly so. I think models that are truer are more useful.

If the only value a model had was usefulness, then my model would be "I am the King of Siam and all women are clones of Alyson Hannigan." That strikes me as an extremely useful model, insomuch as it would easily allow me to achieve extreme pleasure. However, I think that model is rendered useless by not being even remotely true. Why do you think that model is useless?

There are other possible explanations,
There aren't. There is only metaphysical naturalism and super-naturalism (infinite flavors, all of which are essentially the same).

Things are either caused by natural forces or they aren't. If we are restricted to the realm of naturalism, I wasn't aware there was any competition for evolution.

However, to rule out a fundamental change to it, in advance, would invite hubris, IMO. We may have, one day, a more useful model.
Are we just waiting for some creative genius to create a more useful model? What an interesting world to live in - a place where eventually any inconvenient fact (like say death or entropy) can be banished once some artist of sufficient brilliance comes up with a magic formulation.

Sadly I am stuck here in the real world, where the only way Evolutionary theory can become more useful is by becoming closer to the truth. Which means we already know the basic outlines; we already know certain truths that will never change.

At this point it would be highly productive to stop and read Isaac Asimov's "The Relativity of Wrong." It's short and on-line for free.

In most societies, "unfairness" has been seen as highly moral - that it is "immoral" to disrupt society with claims to equality.
There is an excellent article linked to in one of these threads about biological morality. In it the author makes a similar point to what I have always maintained: that the role of culture is to create exemptions from morality. Everybody is born knowing that being a slave is bad; but to learn that slaveholders are good requires a cultural education.

Well, I don't disagree with that. But I disagree with a lot of what you have said on the way ;)
That does not compute.

How can you simultaneously assert that everything - including morality and objective reality - is a model, and that morality is an innate function of a biological construction? Isn't the human brain just a model?

The "objective world" is not a fact. It is another model.
I want to repeat this question so it is not missed: A model of what?

[Yahzi]

I agree with much of what you said earlier, Febble. All we have are "models". We have some damn good ones, of course, but they are, nevertheless, models. Claiming that they represent the objective state of reality, whatever that claim actually means, is unsupportable.
This is precisely what the Yahzi Baseball Bat Test is intended to refute. I invite you to hop over to that thread and put the test into investigation post haste. :evil:

The entire enterprise of science depends on two presumptions: 1) it is possible to know reality at least in approximation, and 2) it is possible to distinguish between approximations. To argue that either of those propositions is false is to repudiate the success of the scientific endeavor. To do so through a medium that exists solely because of the staggering accuracy of science is to repudiate common sense.

[JamesBannon]

I agree with much of what you said earlier, Febble. All we have are "models". We have some damn good ones, of course, but they are, nevertheless, models. Claiming that they represent the objective state of reality, whatever that claim actually means, is unsupportable.
This is precisely what the Yahzi Baseball Bat Test is intended to refute. I invite you to hop over to that thread and put the test into investigation post haste. :evil:

The entire enterprise of science depends on two presumptions: 1) it is possible to know reality at least in approximation, and 2) it is possible to distinguish between approximations. To argue that either of those propositions is false is to repudiate the success of the scientific endeavor. To do so through a medium that exists solely because of the staggering accuracy of science is to repudiate common sense.

Not if one adopts a pragmatic stance it doesn't; i.e., we use science because it works. Whether or not scientific theories are "true" is, in some sense, irrelevant, and next to impossible to determine in any case.

To illustrate, consider performing a brain scan on a live subject whilst performing some task or thinking about some problem. Various parts of the brain light up. Further, let us suppose that we can resolve the process down to a single neurone and even determine exactly which chemicals are crossing the synaptic gaps between neurones. Does this mean we know what the subject was thinking? No, because we cannot observe the content of a thought. We can observe the process of thinking, perhaps, but we have to rely on the subject telling us what the content is. That part of reality is forever closed to us. There is no way of directly observing a thought.

Similarly, consider observing a distant star / galaxy. At any point in time, we observe what the star / galaxy was, never what it actually is, even allowing for improvements in technology. In fact, at the point we make our observations, the star / galaxy might not even exist; i.e., we are observing a non-existent entity.

We can conclude, therefore, that there are at least some parts of reality that are forever closed to any means of observation we devise. The absolute best we can do is to infer what is going on; otherwise known as guessing. True, it may be a well-informed guess backed by theory and past observations, but it is, nevertheless, a guess.

[Febble]

We don't have access to "objective reality". All we have are models.
We do have access to objective reality. The observations of objective reality are what our models are based on.

Except that our observations are also models. As I said, we don't have access to reality - all we have are models.

If all we had was the model, then how could we ever have an observation that did not fit the model? The fact that we can revise our model based on unexpected input indicates that there is something shaping the model other than the model itself.

No, we have consilient models and non-consilient models. Our best grand models are consistent with the sub-models they subsume, right down to the tiny models we call "observations". But they are all models of reality. We do not have direct access to reality, we can only infer it from the consilience of our models.

You are confusing what we know with what actually exists.

No, I am distinguishing what we can know from what actually exists.

The "objective world" is not a fact. It is another model.
A model of what?

A model of reality.

Are we even using model in the same sense? A model is a small, crude, incomplete replica of a thing. What is "objective reality" a small, crude, incomplete replica of?

I don't know what "objective" reality is. I'm not sure what the adjective is for, there. I "believe in" reality, by which I mean that our best evidence that reality exists is that we can make consilient models. And yes, those models are "small crude replicas", which we continue to enlarge and elaborate. But they will always be replicas, because, as you imply, reality isn't a replica of anything, and the only model of reality that was as accurate as the thing it modelled would be reality itself. And not much use to us, therefore.

That, in my view, is the best you can say for a model - that it is [I]useful.
Why is it useful?

All kinds of things - depends what you made it for. Scientific models are useful, though, for being predictive. And we choose the simplest model (usually) that gives us accurate-enough predictions, which will depend on the scale of data we are interested. Newtonian physics works pretty well for most terrestrial engineering, but you need Einsteinian physics for space work. Models are a way of representing the world ("reality" if you like) to ourselves. So they are as good as they are useful.

I think it is useful because it is true - or very nearly so. I think models that are truer are more useful.

Well, I think that models that fulfil their purposes are more useful than those that don't. And if your purpose is to make reliable predictions about the world, then you might describe the models that do so as "truer" than those that do so unreliably. And we do, commonly, describe predictive models as "truer" if they make better predictions. But not always. Is General Relativity truer than Quantum Mechanics? Or does it merely make better predictions over a given range of data (and worse over others)?

(Obviously the latter, I'd say.)

If the only value a model had was usefulness, then my model would be "I am the King of Siam and all women are clones of Alyson Hannigan." That strikes me as an extremely useful model, insomuch as it would easily allow me to achieve extreme pleasure. However, I think that model is rendered useless by not being even remotely true. Why do you think that model is useless?

I think it's perfectly useful if you want to use it to gain pleasure, and it works. But if you want it to be predictive, then it will be crap. And if you want it to be part of a consilient nest of predictive models then it will be even crapper.

There are other [I]possible explanations,
There aren't. There is only metaphysical naturalism and super-naturalism (infinite flavors, all of which are essentially the same).

Well, we are probably using words differently here. You said that "evolution is the only possible explanation for human nature." I assumed you meant by "evolution" something like "current evolutionary theory", which is by no means complete, and may well be substantially wrong. If you meant "anything other than naturalism" then I'd agree, but that is not, for me, synonymous with "evolution" in any commonly used sense of the word.

Things are either caused by natural forces or they aren't. If we are restricted to the realm of naturalism, I wasn't aware there was any competition for evolution.

Well, depends on what you mean by "evolution". I think evolutionary theory is "less wrong" than creationism, for instance, but it almost leaves a great deal out. It is, to quote Asimov, incomplete. And incomplete is a kind of wrong. I would be surprised if, a hundred years from now, evolutionary theory as we now conceive it, is not considered a mere detail in a much more thorough over-arching theory.

However, to rule out a fundamental change to it, in advance, would invite hubris, IMO. We may have, one day, a more useful model.
Are we just waiting for some creative genius to create a more useful model? What an interesting world to live in - a place where eventually any inconvenient fact (like say death or entropy) can be banished once some artist of sufficient brilliance comes up with a magic formulation.

WTF? I think you have read into my words something that simply wasn't there. It is very rare for any one "creative genius" to come up with some paradigm-shifting insight, and even Darwins and Einsteins "stand on the shoulders of giants". But there are some very interesting developments in non-Darwinian mechanism in the evolution (in its broadest sense - meaning change over time) of living things, including, of course, horizontal gene transfer and "neo-Lamarckian" effects like epistasis, not to mention the feedback loops between culturally-transmitted behavour and the fitness landscape.

Sadly I am stuck here in the real world, where the only way Evolutionary theory can become more useful is by becoming closer to the truth. Which means we already know the basic outlines; we already know certain truths that will never change.

I disagree with that last formulation. Possibly because I don't know what you mean by it.

At this point it would be highly productive to stop and read Isaac Asimov's "The Relativity of Wrong." It's short and on-line for free.

lol.

In most societies, "unfairness" has been seen as highly moral - that it is "immoral" to disrupt society with claims to equality.
There is an excellent article linked to in one of these threads about biological morality. In it the author makes a similar point to what I have always maintained: that the role of culture is to create exemptions from morality. Everybody is born knowing that being a slave is bad; but to learn that slaveholders are good requires a cultural education.

OK, I'll read it. But what is "non-biological" about culture? We are a culture-transmitting species. We evolved to be a culture-transmitting species. Again, I don't see the point you are making.

Well, I don't disagree with that. But I disagree with a lot of what you have said on the way ;)
That does not compute.

um. To disagree with you, is to "not compute"? Is that an objective or a subjective assertion? :evil:

How can you simultaneously assert that everything - including morality and objective reality - is a model, and that morality is an innate function of a biological construction? Isn't the human brain just a model?

Well, as I said, I don't know what "objective reality" is, as opposed to plain-jane "reality". I have said that reality is not a model. I have also said that we do not have direct access to it - that our only access to reality is via models. And our morality is a way of modelling our world, just as predictive models are. And our ability to model is granted by our human brains, which evolved (as far as I am aware - it's the best model I have right now). Yes, it's circular, but then thinking about thinking is circular - you might even call it a Strange Loop. Nonetheless, the only access we have to reality is via the models we construct in order to present it ourselves, and that includes our models of the modelling process (and, as you may recall, that is my field, and a very fascinating one it is).

The "objective world" is not a fact. It is another model.
I want to repeat this question so it is not missed: A model of what?

Reality.

ETA: Or if, by "objective world" you mean what I mean by "reality" then, the "objective world" is not a model, but also not something we have direct access to. IMO.

[Febble]

I agree with much of what you said earlier, Febble. All we have are "models". We have some damn good ones, of course, but they are, nevertheless, models. Claiming that they represent the objective state of reality, whatever that claim actually means, is unsupportable.
This is precisely what the Yahzi Baseball Bat Test is intended to refute. I invite you to hop over to that thread and put the test into investigation post haste. :evil:

The entire enterprise of science depends on two presumptions: 1) it is possible to know reality at least in approximation, and 2) it is possible to distinguish between approximations. To argue that either of those propositions is false is to repudiate the success of the scientific endeavor. To do so through a medium that exists solely because of the staggering accuracy of science is to repudiate common sense.

Not if one adopts a pragmatic stance it doesn't; i.e., we use science because it works. Whether or not scientific theories are "true" is, in some sense, irrelevant, and next to impossible to determine in any case.

To illustrate, consider performing a brain scan on a live subject whilst performing some task or thinking about some problem. Various parts of the brain light up. Further, let us suppose that we can resolve the process down to a single neurone and even determine exactly which chemicals are crossing the synaptic gaps between neurones. Does this mean we know what the subject was thinking? No, because we cannot observe the content of a thought. We can observe the process of thinking, perhaps, but we have to rely on the subject telling us what the content is. That part of reality is forever closed to us. There is no way of directly observing a thought.

Well, yes and no. Funnily enough, I'd say that your first example is too optimistic - parts of the brain don't "light up" when we do, say, an fMRI scan. What lights up, lol, is our "model" - literally, t values of beta coefficients derived using the General Linear Model, in a lot of cases. However, I think you are too pessimistic with the second part. Actually, I think it is becoming possibly to identify thoughts. Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house. The door is not, IMO "forever closed to us"/


Similarly, consider observing a distant star / galaxy. At any point in time, we observe what the star / galaxy was, never what it actually is, even allowing for improvements in technology. In fact, at the point we make our observations, the star / galaxy might not even exist; i.e., we are observing a non-existent entity.

We can conclude, therefore, that there are at least some parts of reality that are forever closed to any means of observation we devise. The absolute best we can do is to infer what is going on; otherwise known as guessing. True, it may be a well-informed guess backed by theory and past observations, but it is, nevertheless, a guess.

OK, I take your main point (indeed it was what I was trying to say to Yahzi, so thanks) - our only door to reality is via our models. I think we agree :)

Even our data are models.

[Alex]

I can understand why let's say Adam Smith's Theory of Moral Sentiments could be described as an "ethical model" - since it attempts to explain the origins of moral principles. He starts his investigation by analysing our sense of propriety - which seems to be intuitive. Whether Smith was on the right track or not, presumably intuition itself can be modelled. But for that model you'd need another model of the subconscious mind - if that's where intuition "resides".

Pursuing this line of inquiry, are we on the brink of an indefinite series of models........?

[JamesBannon]

Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house.

Firstly, there is no such thing as an objective observer. All observations are theory laden to some degree. Science recognises this by insisting on peer review of experimental data and models.

Secondly, inference != directly observing the content of. We may have correlates between what a subject reports and the experimenter's observations, suitably reviewed, but we must still rely on the subject's reports about content.

[JamesBannon]

I can understand why let's say Adam Smith's Theory of Moral Sentiments could be described as an "ethical model" - since it attempts to explain the origins of moral principles. He starts his investigation by analysing our sense of propriety - which seems to be intuitive. Whether Smith was on the right track or not, presumably intuition itself can be modelled. But for that model you'd need another model of the subconscious mind - if that's where intuition "resides".

Pursuing this line of inquiry, are we on the brink of an indefinite series of models........?

Without data, the answer would be yes. The best we can probably do, with respect to morals, is to intersubjectively agree on some set of basic principles which we will use to make value judgements about given behaviours. However, untangling principles from mere cultural mores is very difficult.

[muidiri]

Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house.

Firstly, there is no such thing as an objective observer. All observations are theory laden to some degree. Science recognises this by insisting on peer review of experimental data and models.

I can see what you're saying here, but I'm not sure I agree with this being an absolute statement. Measuring the weight, length, volume, etc. isn't subjective. And measuring things like that isn't going to be theory laden either, is it? I mean, if you're asking people what color something is, you might get a variety of subjective answers from vermilion to tomato, and every shade of red and orange in between... but the wavelength of the color is objective, isn't it?

Or am I not understanding something here? :D

[JamesBannon]

Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house.

Firstly, there is no such thing as an objective observer. All observations are theory laden to some degree. Science recognises this by insisting on peer review of experimental data and models.

I can see what you're saying here, but I'm not sure I agree with this being an absolute statement. Measuring the weight, length, volume, etc. isn't subjective. And measuring things like that isn't going to be theory laden either, is it? I mean, if you're asking people what color something is, you might get a variety of subjective answers from vermilion to tomato, and every shade of red and orange in between... but the wavelength of the color is objective, isn't it?

Or am I not understanding something here? :D

Depends on how one measures it. There was a conundrum shown on last year's Christmas Lectures on numbers. It asked the following question: How long is the British coastline? It gave four numerical answers (one of which was "infinitely long") and the fifth option was "all of the above". I chose the fifth option, and was correct. But how can that be? How can the British coastline be all those sizes at once? The answer is "it all depends on how the coastline is represented and measured".

[Yahzi]

ETA: Or if, by "objective world" you mean what I mean by "reality" then, the "objective world" is not a model, but also not something we have direct access to. IMO.
What the feck is the deal here?

I wasn't interested in arguing with Preno about the difference between "objective reality" and "reality," and I'm not interested in arguing with you about it.

When I say "objective reality," I mean "reality" in your terms. When I say "subjective reality" I mean "model" in your terms. If you are going to insist on using your own idiosyncratic terms for common concepts, the least you can do is translate them yourself.

I didn't bother to read the rest of your post. Is there anything in there that is not merely terminological pedantry?

[Yahzi]

our only door to reality is via our models...
Even our data are models.
Models of what?

What is it these "data models" are modeling?

Are they modeling observations of that reality you declared inaccessible? What does it mean for something to be "inaccessible" if you can observe it well enough to construct models of it?

[Yahzi]

Or am I not understanding something here? :D
When you figure it out, tell me. Right now I am harboring the suspicion that the entire affair is merely the trite, banal claim that we cannot know objective reality directly and completely. I don't even know what that phrase is supposed to mean, let alone comprehend why anyone would think it matters.

[Febble]

our only door to reality is via our models...
Even our data are models.
Models of what?

What is it these "data models" are modeling?

Are they modeling observations of that reality you declared inaccessible? What does it mean for something to be "inaccessible" if you can observe it well enough to construct models of it?

I already answered this. They are models of reality. Look, Yahzi, I'm no solipsist. I believe in reality because I think that the fact that we can build grand consilient models is evidence for there being a reality "out there" (even if we don't yet have a grand consilient model that includes both Relativity and Quantum Mechanics).

But I think it is misleading to think that we have Models on one hand and Data on the other. Our data are still models, just on a tighter scale. We call them "data" because they are what are "given". We say "given these observations, we can predict those observations with this model".

But the observations themselves are constructs. i.e. models. Our brains work by modelling. There is no other way they can work. I would argue.

Please present your rebuttal :)

[Febble]

Sorry, I scrolled past this post:

ETA: Or if, by "objective world" you mean what I mean by "reality" then, the "objective world" is not a model, but also not something we have direct access to. IMO.
What the feck is the deal here?

I wasn't interested in arguing with Preno about the difference between "objective reality" and "reality," and I'm not interested in arguing with you about it.

When I say "objective reality," I mean "reality" in your terms. When I say "subjective reality" I mean "model" in your terms. If you are going to insist on using your own idiosyncratic terms for common concepts, the least you can do is translate them yourself.

I'm not convinced my use of terms is any more "idiosyncratic" than yours. Maybe it's a Europe:Americas thing, I dunno. They are terms I've used all my life, and mostly been understood. And I have other good uses for "subjective" and "objective", being a cognitive scientist and all.

I didn't bother to read the rest of your post. Is there anything in there that is not merely terminological pedantry?

Well, obviously I would say so, yes. YMMV.

[Alex]

A provisional representation of a physical system is formulated from a series of observations and experiments. You can call this either a model of some part of "reality" or a model of "objective reality" - it makes no difference because they amount to the same thing. But how is "modelling" a useful concept when accounting for the existence of an abstraction like a moral code or the nature of an aesthetic experience?

A model of moral conventions, for instance, is simply a collection of conjectures or a synthesis of opinions that seem to acquire scientific authority by calling them a model.

[Febble]

Or am I not understanding something here? :D
When you figure it out, tell me. Right now I am harboring the suspicion that the entire affair is merely the trite, banal claim that we cannot know objective reality directly and completely. I don't even know what that phrase is supposed to mean, let alone comprehend why anyone would think it matters.

Well, I think it matters because to think otherwise can in some circumstances be dangerously misleading - to think we have incontrovertible facts on one hand and mere models on the other.

What appears to be a fact to one generation turns out to be a misleading model to the next (e.g. "the sun rises in the west each morning). Or : "DNA contains all the information needed to make an organism". And, specifically, I think it has misled us in this thread. You appear to be making an odd distinction between morality derived from "evolution" and morality derived from "culture", one being "objective" and one "subjective". I find this incoherent, and I think the reason is that you have confused models with "objective reality".

[Febble]

A provisional representation of a physical system is formulated from a series of observations and experiments. You can call this either a model of some part of "reality" or a model of "objective reality" - it makes no difference because they amount to the same thing. But how is "modelling" a useful concept when accounting for the existence of an abstraction like a moral code or the nature of an aesthetic experience?

A model of moral conventions, for instance, is simply a collection of conjectures or a synthesis of opinions that seem to acquire scientific authority by calling them a model.

Well, we have predictive models and normative models. We should not confuse them. The Map is not the Territory.

[Alex]

Well, we have predictive models and normative models. We should not confuse them. The Map is not the Territory.

If the difference between a predictive and a normative model is analogous to the difference between how we represent what is and what ought to be the case, where does the "data" for a normative model come from?

We are supposed to be discussing ideas of right and wrong in this thread, and I fail to see how "modelling" can help us get a grip on the slippery sources of moral principles.

[Preno]

But I think it is misleading to think that we have Models on one hand and Data on the other. Our data are still models, just on a tighter scale. We call them "data" because they are what are "given". We say "given these observations, we can predict those observations with this model".If you haven't read it already, you might be interested in Sellars's take on the "myth of the given" in Empiricism and the Philosophy of Mind (http://www.ditext.com/sellars/epm.html).

[muidiri]

Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house.

Firstly, there is no such thing as an objective observer. All observations are theory laden to some degree. Science recognises this by insisting on peer review of experimental data and models.

I can see what you're saying here, but I'm not sure I agree with this being an absolute statement. Measuring the weight, length, volume, etc. isn't subjective. And measuring things like that isn't going to be theory laden either, is it? I mean, if you're asking people what color something is, you might get a variety of subjective answers from vermilion to tomato, and every shade of red and orange in between... but the wavelength of the color is objective, isn't it?

Or am I not understanding something here? :D

Depends on how one measures it. There was a conundrum shown on last year's Christmas Lectures on numbers. It asked the following question: How long is the British coastline? It gave four numerical answers (one of which was "infinitely long") and the fifth option was "all of the above". I chose the fifth option, and was correct. But how can that be? How can the British coastline be all those sizes at once? The answer is "it all depends on how the coastline is represented and measured".

Hmm. Good point, I guess. But within any given system of measurement, the length would be objectively measurable, right? I mean - that's sort of like saying it's not an objective length, because it depends on whether you're using inches of centimeters... but the length is objectively fixed, only your units change. And as long as the units are disclosed, then any other party using the same measuring conditions should get the same answer as you (within measurement error).

[muidiri]

Or am I not understanding something here? :D
When you figure it out, tell me. Right now I am harboring the suspicion that the entire affair is merely the trite, banal claim that we cannot know objective reality directly and completely. I don't even know what that phrase is supposed to mean, let alone comprehend why anyone would think it matters.

I don't know - it confuses me. Reality is - and it is what it is. Our measurements and observations of it may be imperfect... but reality itself is rather unconcerned with how accurate our measurements are. So I don't really follow the arguments. :dunno:

[Febble]

But I think it is misleading to think that we have Models on one hand and Data on the other. Our data are still models, just on a tighter scale. We call them "data" because they are what are "given". We say "given these observations, we can predict those observations with this model".If you haven't read it already, you might be interested in Sellars's take on the "myth of the given" in Empiricism and the Philosophy of Mind (http://www.ditext.com/sellars/epm.html).

Thanks!

I have the feeling (the sense lol) that a similar fallacy lies at the bottom of all those qualia arguments too.

[Febble]

Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house.

Firstly, there is no such thing as an objective observer. All observations are theory laden to some degree. Science recognises this by insisting on peer review of experimental data and models.

I can see what you're saying here, but I'm not sure I agree with this being an absolute statement. Measuring the weight, length, volume, etc. isn't subjective. And measuring things like that isn't going to be theory laden either, is it? I mean, if you're asking people what color something is, you might get a variety of subjective answers from vermilion to tomato, and every shade of red and orange in between... but the wavelength of the color is objective, isn't it?

Or am I not understanding something here? :D

Depends on how one measures it. There was a conundrum shown on last year's Christmas Lectures on numbers. It asked the following question: How long is the British coastline? It gave four numerical answers (one of which was "infinitely long") and the fifth option was "all of the above". I chose the fifth option, and was correct. But how can that be? How can the British coastline be all those sizes at once? The answer is "it all depends on how the coastline is represented and measured".

heh. You were probably raised on the same interesting factoid that I was - that even though the area of Scotland is smaller than that of England, the coastline is longer.

Even in Primary 5 that fascinated me, and I remember asking "how is the coastline measured? Tide in or tide out? Do you count each half-submerged pebble?" And was told not to be a smartass or the fifties Scottish equivalent, which I forget. But I remember thinking that it was interesting that it was probably true, however you measured it, it's just that you'd get a different ratio for every measuring system.

So Scottish pride was vindicated, and an interest in fractals was born....

[Febble]

Well, we have predictive models and normative models. We should not confuse them. The Map is not the Territory.

If the difference between a predictive and a normative model is analogous to the difference between how we represent what is and what ought to be the case, where does the "data" for a normative model come from?

We are supposed to be discussing ideas of right and wrong in this thread, and I fail to see how "modelling" can help us get a grip on the slippery sources of moral principles.

Same place as any other kind of data. We observe causes and effects (data) and form a model of what will best produce desired effects.

I'm not sure that there's a clean line between the two model types, actually - but normative models are recursive. The models themselves become data: moral systems are themselves datapoints.

[JamesBannon]

Sure, we are still at the stage of correlating subjective reports (i.e. reports from the subject) with objective data (i.e. the experimenter's observations) but already crude content-inferences are possible. For example, we can tell, with a fair degree of confidence, whether someone attending to (i.e. thinking about) a face or a house.

Firstly, there is no such thing as an objective observer. All observations are theory laden to some degree. Science recognises this by insisting on peer review of experimental data and models.

I can see what you're saying here, but I'm not sure I agree with this being an absolute statement. Measuring the weight, length, volume, etc. isn't subjective. And measuring things like that isn't going to be theory laden either, is it? I mean, if you're asking people what color something is, you might get a variety of subjective answers from vermilion to tomato, and every shade of red and orange in between... but the wavelength of the color is objective, isn't it?

Or am I not understanding something here? :D

Depends on how one measures it. There was a conundrum shown on last year's Christmas Lectures on numbers. It asked the following question: How long is the British coastline? It gave four numerical answers (one of which was "infinitely long") and the fifth option was "all of the above". I chose the fifth option, and was correct. But how can that be? How can the British coastline be all those sizes at once? The answer is "it all depends on how the coastline is represented and measured".

Hmm. Good point, I guess. But within any given system of measurement, the length would be objectively measurable, right? I mean - that's sort of like saying it's not an objective length, because it depends on whether you're using inches of centimeters... but the length is objectively fixed, only your units change. And as long as the units are disclosed, then any other party using the same measuring conditions should get the same answer as you (within measurement error).

If we are all using the same system of measurement, then yes, it is objective within that system. The same is true when we consider "truth", in mathematics, say. Are all proven mathematical theorems forever true? Certainly, within mathematics that is.

[Alex]

Same place as any other kind of data. We observe causes and effects (data) and form a model of what will best produce desired effects.

Is this true? Scientific models for physical systems get their quantifiable raw data from the "objective" world. As we know, such models can be scrutinized and evaluated by competent scientists in whatever field of inquiry is being studied.

But we must rely on subjective information - obtained as a rule by introspection - in order to give an account of something like a moral code. A social scientist may get recursive answers to questions put to a sample of people when their moral attitude toward something or other is being investigated, but this seems to be no better than building a model based on the results from an opinion poll.


I'm not sure that there's a clean line between the two model types, actually - but normative models are recursive. The models themselves become data: moral systems are themselves datapoints.

I have no difficulty in understanding what you say about models of the material world: it makes sense to me. As I expect you've guessed, the reason for my scepticism about "normative models" is because I don't believe scientific materialism can account for the moral sense.

[JamesBannon]

Same place as any other kind of data. We observe causes and effects (data) and form a model of what will best produce desired effects.

Is this true? Scientific models for physical systems get their quantifiable raw data from the "objective" world. As we know, such models can be scrutinized and evaluated by competent scientists in whatever field of inquiry is being studied.

But we must rely on subjective information - obtained as a rule by introspection - in order to give an account of something like a moral code. A social scientist may get recursive answers to questions put to a sample of people when their moral attitude toward something or other is being investigated, but this seems to be no better than building a model based on the results from an opinion poll.


I'm not sure that there's a clean line between the two model types, actually - but normative models are recursive. The models themselves become data: moral systems are themselves datapoints.

I have no difficulty in understanding what you say about models of the material world: it makes sense to me. As I expect you've guessed, the reason for my scepticism about "normative models" is because I don't believe scientific materialism can account for the moral sense.

Then what does? (And, please, don't say "god" because that is silly). Human beings have a range, perhaps infinite, of behaviours to which we, as humans, apply value judgements. Looked at on its own, a piece of behaviour cannot be said to be moral or immoral without some system of value judgements applied by a human agent. However, this, in itself, is no different from saying "theory X is better than theory Y because it does not introduce more explanatory variables than necessary". In principle, there is no difference between a scientific claim and a moral one: both are truth claims.

[Febble]

Same place as any other kind of data. We observe causes and effects (data) and form a model of what will best produce desired effects.

Is this true? Scientific models for physical systems get their quantifiable raw data from the "objective" world. As we know, such models can be scrutinized and evaluated by competent scientists in whatever field of inquiry is being studied.

And my point is there is no such thing as "the objective world". There is reality (as I've said, I'm not solipsist), which we can infer from the fact that we are able to build consilient predictive models. But "objective world" is not an item, or, if it is, it is simply "reality" to which, as I said, we do not have direct access. And yes, of course, "competent scientists" can evaluate and scrutinize our models - but for consilience. And we assume (with justification) that the more consilient the model, the "truer" (i.e. the closer it is to unobserved "reality") it is. So yes, I think what I said is true :)

But we must rely on subjective information - obtained as a rule by introspection - in order to give an account of something like a moral code. A social scientist may get recursive answers to questions put to a sample of people when their moral attitude toward something or other is being investigated, but this seems to be no better than building a model based on the results from an opinion poll.

Well, I think there is an is/ought category confusion here. All information is "subjective" in the sense that all datapoints are observation by a subject. That many observers can agree on an observation means that there is consilience. And we tend to call such observations "objective". But objectivity itself has to be evalulated.

As for the rest, I'm not sure what you are saying - are you saying that moral codes should not be based on opinion polls or equivalent? In which case I would agree. Or are you saying that they are not based on opinion polls or equivalent? In which case I'd disagree. It seems to me pretty unarguable that moral codes are cultural artefacts (where "artefact" is not a pejorative). But culture itself is a biological artefact - we evolved to be a culture-building species, no less than termites evolved to be a mound-building species.

I'm not sure that there's a clean line between the two model types, actually - but normative models are recursive. The models themselves become data: moral systems are themselves datapoints.

I have no difficulty in understanding what you say about models of the material world: it makes sense to me. As I expect you've guessed, the reason for my scepticism about "normative models" is because I don't believe scientific materialism can account for the moral sense.

Ah. Well, that helps (should have read your whole post more thoroughly before I embarked on my response - oh well, I'll leave it now...)

OK, well, I simply disagree :) I think it is pretty easy to account for our moral sense within a naturalistic evolutionary framework. There, I'd agree (probably) with Yahzi. What do you see as the problem? What's the bit you don't think science can account for?

[Sidhe747]

Morals must be dynamic particularly when dealing with situations that are abstract or original.

There's no use saying this is a rule for when this happens because things seldom go to plan, keep your own shit together and hope everyone else does the same is my motto, I try not to badger people with my code but shit happens and sometimes you have to make exceptions.

[Alex]

Then what does? (And, please, don't say "god" because that is silly). Human beings have a range, perhaps infinite, of behaviours to which we, as humans, apply value judgements. Looked at on its own, a piece of behaviour cannot be said to be moral or immoral without some system of value judgements applied by a human agent. However, this, in itself, is no different from saying "theory X is better than theory Y because it does not introduce more explanatory variables than necessary". In principle, there is no difference between a scientific claim and a moral one: both are truth claims.
The claims of science are not based on subjective evidence. If you and I agree that something's wrong and a third party thinks it's right, we can't appeal to the arbitration of science to settle the matter. We can consult our shared "moral code" and argue about the difference (between ourselves and the third party).

But a "moral code" itself can't be modelled by scientific materialism. Its (subjective) data isn't quantifiable. In my opinion, it's a cultural artefact that's been developed in different ways by different societies over millennia. But in many vital respects these artefacts resemble each other. They have far more correspondences than differences. And it's obvious why this should be so.

One of the lures of theism is that it accounts for the existence and authority of moral standards. I don't think it's "silly" to believe this, just mistaken. In a cynical humour, I would even argue that a belief in the supernatural origin of moral principles has social utility.

[JamesBannon]

Then what does? (And, please, don't say "god" because that is silly). Human beings have a range, perhaps infinite, of behaviours to which we, as humans, apply value judgements. Looked at on its own, a piece of behaviour cannot be said to be moral or immoral without some system of value judgements applied by a human agent. However, this, in itself, is no different from saying "theory X is better than theory Y because it does not introduce more explanatory variables than necessary". In principle, there is no difference between a scientific claim and a moral one: both are truth claims.
The claims of science are not based on subjective evidence. If you and I agree that something's wrong and a third party thinks it's right, we can't appeal to the arbitration of science to settle the matter. We can consult our shared "moral code" and argue about the difference (between ourselves and the third party).

But a "moral code" itself can't be modelled by scientific materialism. Its (subjective) data isn't quantifiable. In my opinion, it's a cultural artefact that's been developed in different ways by different societies over millennia. But in many vital respects these artefacts resemble each other. They have far more correspondences than differences. And it's obvious why this should be so.

One of the lures of theism is that it accounts for the existence and authority of moral standards. I don't think it's "silly" to believe this, just mistaken. In a cynical humour, I would even argue that a belief in the supernatural origin of moral principles has social utility.

Authority has little, if anything, to do with morality. That's where theists have it all wrong. Morality, if it's to be of any use, has to be internalised, and there is no more effective way of making sure that does not happen than saying "do it, or else".

[Alex]

Authority has little, if anything, to do with morality. That's where theists have it all wrong. Morality, if it's to be of any use, has to be internalised, and there is no more effective way of making sure that does not happen than saying "do it, or else".

Practical morality can't exist without a "court of appeal". In other words, we can't settle moral questions without resorting to some form of authority. For example: the common law is itself largely a collection of moral judgments made over a long time. It's authoritative in the sense that it specifies what is and what is not permissible.

Your point about internalization of moral principles is only true for those who are willing to be guided by them. Others have to be coerced - and they are coerced by reference to an authority of some kind.

[Febble]

Morals must be dynamic particularly when dealing with situations that are abstract or original.

There's no use saying this is a rule for when this happens because things seldom go to plan, keep your own shit together and hope everyone else does the same is my motto, I try not to badger people with my code but shit happens and sometimes you have to make exceptions.

Yup.

[Preno]

Same place as any other kind of data. We observe causes and effects (data) and form a model of what will best produce desired effects.

Is this true? Scientific models for physical systems get their quantifiable raw data from the "objective" world. As we know, such models can be scrutinized and evaluated by competent scientists in whatever field of inquiry is being studied.

But we must rely on subjective information - obtained as a rule by introspection - in order to give an account of something like a moral code. A social scientist may get recursive answers to questions put to a sample of people when their moral attitude toward something or other is being investigated, but this seems to be no better than building a model based on the results from an opinion poll.I think you're making two mistakes here. First, connecting subjectivity and introspection too tightly. Ryle has shown already in the fifties that introspection isn't a method that leads to some sort of ineffably private knowledge. It may, of course, be easier for one to gain a particular piece of knowledge through introspection, but that doesn't make it knowledge gain through introspection a fundamentally different kind than knowledge of other people.

Second, we don't rely on, or certainly not primarily on, introspection to gain moral knowledge. Sure, I may consult my moral intuition to gain a first impression of sorts, but that impression may often turn out to be mistaken. One certainly has a moral intuition and it is often helpful, but that doesn't mean morality is based on this. We also have mathematical intuition, philosophical intuition, I suppose social intuition, too, and all of them are helpful in various forms of reasoning, but that doesn't mean they constitute the basis of that reasoning. To make that claim in connection with morality is to presuppose that morality is special, rather than to demonstrate it.
Practical morality can't exist without a "court of appeal". In other words, we can't settle moral questions without resorting to some form of authority. For example: the common law is itself largely a collection of moral judgments made over a long time. It's authoritative in the sense that it specifies what is and what is not permissible.In a way, yes, no reasoning can exist without a court of appeal. But the idea that this takes the form of an ultimate authority is a mistaken one. In light of the fact that even our most fundamental assumptions about space and time embedded in the very fabric of our language turned out to be subject to revision, this sort of foundationalism seems to be a very unrealistic way of looking at our reasoning, whether moral, everyday or scientific.

It's not very surprising that if one takes a foundationalist stance and assumes that morality is based on introspection, one is led to subjectivism. The question is, why would we want to grant those assumptions? Again, granting them means assuming that morality is in some sense "queer", rather than demonstrating it.

Imo, the problem is that most of the time, people overlook the essentially normative character of our reasoning and the essentially social character of our norms, but due to long-standing philosophical prejudice (and the influence of the "label theory" of meaning), they notice it when it comes to morality (because they simply run out of labels there). And perhaps in the case of morality, it is more salient than in cases where things like norms of reference are also involved, but the fact is that even other kinds of discourse are normative through and through. Representation is a special kind of normativity (one in which the norms of reference and criteria of identity play a key role), rather than being a category on the same level as Febble would have it.

eta: to expand on this slightly, conceiving of representation as not involving norms of reference and, typically, criteria of identity is just incoherent, as that would leave us without any connection between our use of language and the things represented, and vice versa, in learning those norms, we simply establish the representation, there is no residue to be added above and beyond acquiring the relevant norms.

[Yahzi]

I already answered this. They are models of reality.
Why do you think that the word "reality" refers tos an obvious, objective entity, but the phrase "objective reality" doesn't?

If I had said "reality" the first time around, you would have just said, "no, we can't access reality, we only have models of it." And then when I asked what they were models of, you would have said, "Why, objective reality of course."

Our data are still models,
Models of what?

You keep saying the only thing we have are models. Every time I ask you what they are models of, you say, "Why reality of course, Isn't that obvious?" And then when I parrot your words back to you, you tell me they only refer to models.

But the observations themselves are constructs. i.e. models. Our brains work by modelling. There is no other way they can work. I would argue.
Yep, I was right. Your entire point is that we can't have "direct access" to reality, whatever the hell that means. Well, nobody said we could. It is not necessary to have direct access to reality to be able to discern objective truths.

At this point I would remark on how we know the world will never turn out to be square, but I'm pretty sure you never actually read "The Relativity of Wrong."

Please present your rebuttal :)
My rebuttal is that, absent any concrete, objective input, there is no way to distinguish between models. You keep telling me there's nothing but models; and then you say some models are more consilent with reality; and then you say but of course consilence and reality are only models.

You want everything to be relative, even while you hold onto the absoluteness of reality. The only distinguishing characteristic seems to be that you can talk about reality and nobody else can.

[Yahzi]

Practical morality can't exist without a "court of appeal". In other words, we can't settle moral questions without resorting to some form of authority.
That's not quite true.

Even sociopaths know what is moral; they just don't act on it. We can't always enforce morality - that is to say, we can't create justice - without using force. But we can still know what the answer is.

In another sense you are right, though, because the concept of morality is also bound up to accountability. If you are not accountable to anyone, then you cannot be moral: morality is necessarily a transaction between at least two moral agents. But you still don't need to be able to enforce it.

If the only way you could settle moral answers was by appealing to a moral authority with the power to inflict its decisions, then there would never be any injustice (by definition).

[Yahzi]

But a "moral code" itself can't be modelled by scientific materialism.
Yes it can. Morality is a biologically relevant behavior, just like sex and language. Some moralities are more effective at producing cooperation than others, just like some sexual mores are more effective at producing (or reducing) procreation.

In my opinion, it's a cultural artefact that's been developed in different ways by different societies over millennia.
If that were true, communism would have worked.

In a cynical humour, I would even argue that a belief in the supernatural origin of moral principles has social utility.
It's funny 'cause it's true. If your choice is between morally degenerate anarchy and theocracy, then theocracy is better. The problem with Christianity is not that it is immoral; it is that is not moral enough. We expect "universal rights" these days, and on the scale of moral development, Christian theology stops at "social contract" at best.

[Yahzi]

And I have other good uses for "subjective" and "objective", being a cognitive scientist and all.
If you pull that snarking superiority bullshit one more time, I am going to leave.

I left IIDB because you couldn't stop touting your fucking academic qualifications as if it meant no one was allowed to disagree with you. If I recall, when I challenged you on it, you got me reprimanded.

I am not going to go through that shit again.

Get off your fucking high horse if you want to have a conversation.

[Preno]

:ironicat:

[Febble]

I already answered this. They are models of reality.
Why do you think that the word "reality" refers tos an obvious, objective entity, but the phrase "objective reality" doesn't?

If I had said "reality" the first time around, you would have just said, "no, we can't access reality, we only have models of it."

I just think the adjective is superfluous. What would non-objective reality be? Doesn't matter what you call it, my point is that we don't have direct access to it. What we have are our models of it, and by the consilience and predictive power of our models we can infer it exists.

Our data are still models,
Models of what?

You keep saying the only thing we have are models. Every time I ask you what they are models of, you say, "Why reality of course, Isn't that obvious?"

Well, that's my answer (though I note that you haven't dropped your habit of sticking words in quotes, even though they aren't quotations....) My answer to your question is that our data, and all our other models, are models of reality. I don't know if it's obvious or not, but that is my view, and I can support it. And have.

But the observations themselves are constructs. i.e. models. Our brains work by modelling. There is no other way they can work. I would argue.
Yep, I was right. Your entire point is that we can't have "direct access" to reality, whatever the hell that means. Well, nobody said we could. It is not necessary to have direct access to reality to be able to discern objective truths.

Right, and that was what I said my point was (well, it isn't my entire point, but it's certainly one of them). Sounds like you agree. That's fine. And yes, I agree that we do not have to have direct access to reality to be able to infer its existence. Which I think is a better way of putting it than "discern objective truths", for many reasons. I think the idea that we can "discern objective truths" is a dangerous one. What we can do, and should do, is continually check the consilience of our models, because what seems obviously an objective truth in one era, seems to be a rank falsehood in another. As Asimov points out :)

At this point I would remark on how we know the world will never turn out to be square, but I'm pretty sure you never actually read "The Relativity of Wrong."

I have not only read it, I have probably read it twenty of thirty times, and it's probably the link I most often insert into my posts.

Please present your rebuttal :)
My rebuttal is that, absent any concrete, objective input, there is no way to distinguish between models. You keep telling me there's nothing but models; and then you say some models are more consilent with reality; and then you say but of course consilence and reality are only models.

No, I didn't say that some models are more consilient with reality. I think they are, but that's an inference not a direct observation. I meant (and I admit, I didn't say it, I assumed my meaning would be clear, mea culpa) that our some models are more consilient with each other. In fact my meaning should have been clear because I talked about Grand models being consilient with submodels, and also pointed out that we currently have two Grand and highly useful models, that are also highly predictive, but are not actually consilient with each other.

You want everything to be relative, even while you hold onto the absoluteness of reality. The only distinguishing characteristic seems to be that you can talk about reality and nobody else can.

Not at all. I certainly don't want "everything to be relative", although I do think it's useful (lol) to regard wrongness as relative, but then so, I take it, do you :) And nowhere have I said or implied that no-one else can talk about reality except for me!!!

All I've said, and I think it's important, is that reality is something we can robustly infer, but cannot observe. What we can do is construct models of it, and check the consilience of those models with each other - that includes making predictions, and checking that our new observations are consilient with the prior model (or not). But again, those new observations are themselves models.

In the most basic sense - when we observe, with our ears and eyes, and other senses - we are making models. It's how our cognition works. I think you know this, so we probably aren't actually disagreeing.

In which case, why are we disagreeing?

[Febble]

And I have other good uses for "subjective" and "objective", being a cognitive scientist and all.
If you pull that snarking superiority bullshit one more time, I am going to leave.

I left IIDB because you couldn't stop touting your fucking academic qualifications as if it meant no one was allowed to disagree with you. If I recall, when I challenged you on it, you got me reprimanded.

I am not going to go through that shit again.

Get off your fucking high horse if you want to have a conversation.

Golly.

No, that wasn't what I meant at all. I simply meant what I said - that in my trade, those words have useful connotations, and I don't see the necessity to use them outside that context.

If you thought it meant I was pulling some kind of rank - well you couldn't be more wrong.

Oh well. I apologise, I guess, but no offence was intended.

[Jobar]

The tao that can be described is not the eternal Tao.
The name that can be spoken is not the eternal Name.

When we try to reduce reality to words, we can't, because the words or concepts we use are only parts or aspects of reality.

I agree with Yahzi that there exists an objective reality, which is the ground of all our observations. But that ground is indescribable in any ultimate sense; we can say "objective reality" as an indicator- a model- of that absolute which transcends all our mathematical or verbal models, yes; but we still can't state that "objective reality" is anything but a model.

The subject-object duality resists all efforts to break it down into 'either-or'. The Baseball Bat Test indicates that there's an objective reality outside our consciousness, the subject; but without a subject to perform the test, how can we say that objective reality exists? The two are inextricable.

[Febble]

The tao that can be described is not the eternal Tao.
The name that can be spoken is not the eternal Name.

When we try to reduce reality to words, we can't, because the words or concepts we use are only parts or aspects of reality.

I agree with Yahzi that there exists an objective reality, which is the ground of all our observations. But that ground is indescribable in any ultimate sense; we can say "objective reality" as an indicator- a model- of that absolute which transcends all our mathematical or verbal models, yes; but we still can't state that "objective reality" is anything but a model.

The subject-object duality resists all efforts to break it down into 'either-or'. The Baseball Bat Test indicates that there's an objective reality outside our consciousness, the subject; but without a subject to perform the test, how can we say that objective reality exists? The two are inextricable.

Well, I agree with Yahzi too, as I've said. I've quibbled about attaching the adjective "objective" because the word seems adequate on its own - to talk about "objective reality" implies there is also "subjective reality" which seems to me to be an oxymoron.

Yes, lots of things (not just TBBT) indicates that reality exists. But where I take issue with Yahzi's approach is the idea that we have access to two things - Objective Reality on the one hand and Models on the other. My claim is that all we have are models, and that it is from the interconsilience of those models that we infer that there is, as you say, a reality that is "the ground of all our observations". The observations themselves, however, are always models.

Is my point.

And I'd also like to say that at no point have I intended to imply that my trade gives me any edge over anyone else's argument. The great thing about the internet is that our arguments have to make sense - we cannot argue from authority because we cannot verify each other's credentials. I simply give my background as information as to where I am coming from. As a cognitive scientist, I deal with mental models. I'm not a philosopher. I don't understand a lot of philosophy. I struggled with Preno's very interesting link. Perhaps I should have said that I am "merely" a cognitive scientist, not a philosopher.

But that would be false too, because I don't privilege any particular viewpoint, I think. I was merely trying to explain where I am coming from.

FWIW.

[JamesBannon]

I don't see what on earth is so controversial here. Humans are subjects. We simply cannot perceive anything independently of mind, even a newborn doesn't do that (i.e., there is no "blank slate"). The closest we will get to that state is some kind of inter-subjective agreement on what constitutes reality (whether it be moral, physical or whatever).

[Alex]

... we don't rely on, or certainly not primarily on, introspection to gain moral knowledge. Sure, I may consult my moral intuition to gain a first impression of sorts, but that impression may often turn out to be mistaken. One certainly has a moral intuition and it is often helpful, but that doesn't mean morality is based on this. We also have mathematical intuition, philosophical intuition, I suppose social intuition, too, and all of them are helpful in various forms of reasoning, but that doesn't mean they constitute the basis of that reasoning. To make that claim in connection with morality is to presuppose that morality is special, rather than to demonstrate it.

..... no reasoning can exist without a court of appeal. But the idea that this takes the form of an ultimate authority is a mistaken one. In light of the fact that even our most fundamental assumptions about space and time embedded in the very fabric of our language turned out to be subject to revision, this sort of foundationalism seems to be a very unrealistic way of looking at our reasoning, whether moral, everyday or scientific.

It's not very surprising that if one takes a foundationalist stance and assumes that morality is based on introspection, one is led to subjectivism. The question is, why would we want to grant those assumptions? Again, granting them means assuming that morality is in some sense "queer", rather than demonstrating it.

.... the problem is that most of the time, people overlook the essentially normative character of our reasoning and the essentially social character of our norms, but due to long-standing philosophical prejudice (and the influence of the "label theory" of meaning), they notice it when it comes to morality (because they simply run out of labels there). And perhaps in the case of morality, it is more salient than in cases where things like norms of reference are also involved, but the fact is that even other kinds of discourse are normative through and through.....
When you say I'm assuming that moral statements are in some sense "queer" instead of being able to demonstrate it, I believe you're implying that I could never demonstrate that they're not "queer" because all statements - including what I think of as scientific statements of fact - are normative. In other words, there is no essential dichotomy between normative reasoning about moral problems and scientific reasoning about the physical world? Maybe I don't understand what you mean by "normative" here.

Here's two statements: in my view, the first is a statement of fact and the second a (normative) statement which expresses an opinion:

1). The earth is one of a number of planets that orbit the sun.

2). A man who runs off with another man's wife is a scoundrel.

If someone denied the truth of the first statement, I think I could give grounds for knowing it as a matter of fact - a tiny fragment of knowledge about reality that's been verified by scientific inquiry.

If someone asked me why I believed the second statement was true, I'd have to explain it as a moral judgment - perhaps something that I'd consulted my conscience about. Scoundrelism, I would claim, is a moral concept about which science has nothing to say.

What fallacy am I committing by thinking like this? I suspect it's failing to appreciate that "our most fundamental assumptions about space and time embedded in the very fabric of our language turned out to be subject to revision", and this makes distinctions between scientific and moral discourse problematic.

Being neither a moral philosopher nor a scientist, I feel like I've arrived in a dialectical limbo.

[Preno]

When you say I'm assuming that moral statements are in some sense "queer" instead of being able to demonstrate it, I believe you're implying that I could never demonstrate that they're not "queer" because all statements - including what I think of as scientific statements of fact - are normative. In other words, there is no essential dichotomy between normative reasoning about moral problems and scientific reasoning about the physical world? Maybe I don't understand what you mean by "normative" here.Yes, to be more specific, I don't think there's some sort of vast difference in their epistemological status. Obviously, moral reasoning differs from scientific reasoning in its topic, just like, say, economy differs from physics.
Here's two statements: in my view, the first is a statement of fact and the second a (normative) statement which expresses an opinion:

1). The earth is one of a number of planets that orbit the sun.

2). A man who runs off with another man's wife is a scoundrel.

If someone denied the truth of the first statement, I think I could give grounds for knowing it as a matter of fact - a tiny fragment of knowledge about reality that's been verified by scientific inquiry.

If someone asked me why I believed the second statement was true, I'd have to explain it as a moral judgment - perhaps something that I'd consulted my conscience about. Scoundrelism, I would claim, is a moral concept about which science has nothing to say.Well, that's entirely true. The first is a scientific statement and the second is a moral statement, or if you wish, a moral judgment. It's an exaggeration to say that science has nothing to say about whether someone is a scoundrel or not - psychology and sociology can have plenty to say about that, for example, but I suppose you wouldn't really dispute that (or would you). I also suppose in the social sciences the barriers aren't as distinct as one might think (see for example Sen's approach to economics, which is certainly not some marginal phenomenon to be disregarded).

The only problem here is with the idea of conscience as some sort of special capacity (fundamentally different from our other epistemological capacities) in which moral reasoning is grounded (but it's entirely possible that I'm just misinterpreting how you imagine "conscience" works). The problem is that in order to be meaningful - and moral language clearly is meaningful - moral discourse must be grounded in something publicly accessible, not in introspection. And the common public ground is constituted, just like in every other part of human language, by the norms which govern when it's appropriate to use the term and which inferences you have the right to make. For example, one such norm says you have the right to question whether an action is right when you can show that it is unfair, i.e. treats agents differing in only irrelevant respects differently. Other such norm might say that an action is wrong if it hurts people for no good reason. Such norms are constitutive of the meaning of terms such as "right". It's not like we first have a concept of rightness and then we ask about this concept whether it's right to hurt people or not. That seems to be an evidently mistaken way of looking at the matter to me.

Of course, they aren't absolute norms, in the sense that they apply to all circumstances, but that really isn't anything special, as virtually all of our non-mathematical reasoning exhibits this property (an inference is sanctioned unless there are specific grounds to believe that it fails). They are also often in conflict and this conflict can be resolved by revising or restricting some of them (which involves what one might call "moral judgment", but this doesn't seem to be different in any substantial way from the way we handle conflicts between norms in other part of our language).

A simple example of norms in the more empirical parts of language are cases like induction or inference to the best explanation. (Norms which are common to all areas of language would be logical norms.)

More fundamentally, it is a fact constitutive of the meaning of a word like "dog" that it is appropriate to use it in certain situations and inappropriate to use it in others. Obviously, one might try to reduce this normative aspect to representation and say something like "sure, in particular, it's appropriate to use it when there is a dog around (for example)". But the point is that although it describes when it is appropriate to use the word, it doesn't explain what constitutes our knowledge of the meaning of that word. That knowledge is the knowledge of a framework of (implicit) rules, just like our knowledge of moral concepts. Sure, the phrase "that dog" represents that dog, but what this means is that we've acquired a framework of rules relating to that phrase, not that there is some sort of irreducible relationship between dogs and pieces of language above and beyond such rules. So, considering all this, I don't think there's any good reason to hold that moral reasoning is fundamentally distinct epistemologically from everyday reasoning or scientific reasoning, or at any rate that this supposed distinction lies in the fact that one is normative while the others aren't.

ETA: I suppose what obscures the normativity of our everyday language is that its norms are only rarely in conflict (one such example could perhaps be situations like hallucination), so one doesn't have to judge them against one another very often, one merely follows them. But such conflicts can arise in science, for example when the evidence for quantum physics caused us to abandon some of our fundamental modes of reasoning (and reference), at least in some contexts.

[Alex]

Of course, they aren't absolute norms, in the sense that they apply to all circumstances, but that really isn't anything special, as virtually all of our non-mathematical reasoning exhibits this property (an inference is sanctioned unless there are specific grounds to believe that it fails). They are also often in conflict and this conflict can be resolved by revising or restricting some of them (which involves what one might call "moral judgment", but this doesn't seem to be different in any substantial way from the way we handle conflicts between norms in other part of our language).

A simple example of norms in the more empirical parts of language are cases like induction or inference to the best explanation. (Norms which are common to all areas of language would be logical norms.)

....... So, considering all this, I don't think there's any good reason to hold that moral reasoning is fundamentally distinct epistemologically from everyday reasoning or scientific reasoning, or at any rate that this supposed distinction lies in the fact that one is normative while the others aren't.


I've been in the habit of distinguishing between reasoning based on the observation and measurement of physical phenomena, and inferences made from information about subjective states. Very crudely, I'd say that physics, chemistry, and biology (the so-called "hard sciences") operate with quantifiable material data that is more or less accessible to each expert in whatever field. I do not know for sure whether this is true in the social sciences, but suspect that it is not.

To give a specific example: it seems to me that predicting how much radiation would result from the detonation of a nuclear device of x size etc., depends on understanding and measuring the large number of physical variables involved. Repeatable experiments would be essential to achieving that understanding. But an industrial psychologist trying to measure a change in attitude after, let's say, undermining resistance to innovation in the workplace, would not be able to count on other psychologists being able to repeat his experiment. The data is too "fuzzy" for properly controlled experimental purposes.

Sorry if my example was too long-winded or far-fetched, but I think you can see what I'm driving at.

Now if what you say about the normativity of all scientific discourse is true, and I'm far from understanding it perfectly, then I'm making so many categorical mistakes in my assumptions about different ways of reasoning that I probably need to junk all my common sense notions about the rationality of science (and the status of moral rules ) and start from scratch.

[JamesBannon]

I'm with Preno on this one. Both moral and scientific claims are "truth apt". IMO to argue otherwise, as many non-cognitivists would do for example, is a blatant case of special pleading. There appears to me to be no difference in principle between the two types of claim.

[Yahzi]

I just think the adjective is superfluous. What would non-objective reality be?
I already said on page 2 of this thread that I was not interested in quibbling about the phrase "objective reality." I provided an explanation of what I meant by the term. The post was in direct response to one of your posts.

Did you a) not read my post, or b) not understand my post?

If your entire argument boils down to your inability to equate the term "objective reality" with the term "reality" even after being provided with a definition, please advise me so I can put you on /ignore.

Here is a link to the post in case you would like to review it:

http://www.secularcafe.org/showthread.php?p=90023&#post90023

I note that you haven't dropped your habit of sticking words in quotes, even though they aren't quotations.
I guess it's a literary thing. I have other good uses for artistically expressive punctuation, being a published author and all.

See what I mean about snark? Don't apologize. Just stop doing it. Your academic credentials are irrelevant to any possible conversation here. Even if there were somehow relevant, they cannot be proved, so they are merely empty claims. There is no reason to ever mention your (alleged) academic profession.

Your arguments stand or fall on their own. Stop resorting to authority to support them. Also, refraining from being demeaning or dismissive of your fellow conversationalists would be helpful too.

And yes, I agree that we do not have to have direct access to reality to be able to infer its existence. Which I think is a better way of putting it than "discern objective truths", for many reasons. I think the idea that we can "discern objective truths" is a dangerous one. What we can do, and should do, is continually check the consilience of our models, because what seems obviously an objective truth in one era, seems to be a rank falsehood in another.
To discern an objective truth is to prove that a specific component (such as a theoretical mechanism or prediction) is consilent with reality. That is what "objective truth" means. It means true about objective reality.

When you say we should continually check the consilence of your models, you seem to imply that at any given point a specific (previously reliable) model could suddenly stop being consilent. Is this what you mean to imply?

because what seems obviously an objective truth in one era, seems to be a rank falsehood in another. As Asimov points out :)
That's what you got out of his article? Are you serious? And you've read it many times? :eek:

Are you serious?

In which case, why are we disagreeing?
The actual disagreement is this:

I meant that our some models are more consilient with each other...

What we can do is construct models of it, and check the consilience of those models with each other - that includes making predictions, and checking that our new observations are consilient with the prior model (or not).


You think the way we prove a model to be useful is by showing how well it fits with our other models.

I think that the way we prove a model is true is by showing how well it fits with reality.

Now you keep telling me that you agree there is a reality but it is inaccessible. I don't know what you mean by this.

If you mean it is impossible to judge models by how well they fit with reality, then you are implying that reality is irrelevant, and all we need to do is make up models that fit really well with each other, and therefore our models are not constrained by reality. They are merely constrained by other models. In this view the only reason we can't all have ice cream for dinner and not get fat is because of those mean-spirited people who hold onto the "fat makes you fat" models. If only they were purged from society - by unspecified means - why we could make up models that would fulfill our wildest dreams. And as long as they all matched with each other, we would be perfectly happy. And skinny.

On the other hand, if you are implying that all of these models must eventually conform to observations of the real world, then what do you mean by "inaccessible?" Other than the obvious and trite claim that we cannot "access" reality "directly," which has already been dismissed as self-evident and uninteresting.

In fact my meaning should have been clear because I talked about Grand models being consilient with submodels, and also pointed out that we currently have two Grand and highly useful models, that are also highly predictive, but are not actually consilient with each other.

I'm just gonna guess here, but I think your two Grand models are "metaphysical naturalism/science" and "supernaturalism/religion." While it is true they are not consilent, it is not true that they are equally useful or predictive.

However, this does illuminate the point of your defending "internal coherency" as a standard of truth. The point is to preserve viability for a theory which has nothing but internal coherency to recommend it (although honestly, it doesn't even that).


.
.
.
.
.
.
.
Right,
Hey, muidiri... I was right. This really is all about the sophomore philosophy observation that reality is not directly accessible.

Next will probably be the idea, "what if our entire universe were just an atom in another, bigger universe?" :dunno:

[Yahzi]

I agree with Yahzi that there exists an objective reality, which is the ground of all our observations.
As far as I can tell, Febble does not: she asserts that the ground of our observations are other models.

But that ground is indescribable in any ultimate sense; we can say "objective reality" as an indicator- a model- of that absolute which transcends all our mathematical or verbal models, yes; but we still can't state that "objective reality" is anything but a model.
Language is capable of pointing to the thing behind the symbol. It does it all the time. When I say objective reality, I mean that thing we're all making models of.

The Baseball Bat Test indicates that there's an objective reality outside our consciousness, the subject; but without a subject to perform the test, how can we say that objective reality exists? The two are inextricable.
Without a subject (i.e. a sentient mind) we can't say anything; saying is an action of sentience. However, that in no way affects the existence of objective reality. It's there where anybody is around to say it or not.

It turns out the old canard is actually quite simple to resolve:

"If a tree falls in the forest and no one hears it, does it make a sound?"

The answer depends on what you mean by sound. If you mean "turbulence in the atmosphere," then the answer is obviously yes. If you mean "cognitive linkage in a sentient brain," then the answer is obviously no.

[Yahzi]

[But an industrial psychologist trying to measure a change in attitude after, let's say, undermining resistance to innovation in the workplace, would not be able to count on other psychologists being able to repeat his experiment. The data is too "fuzzy" for properly controlled experimental purposes.
That's the same as saying psychology isn't a science.

That might be true, but it's not what psychologists claim. They expect their experiments to be repeatable.

[Febble]

I agree with Yahzi that there exists an objective reality, which is the ground of all our observations.
As far as I can tell, Febble does not: she asserts that the ground of our observations are other models.

No, I don't assert that. I agree that the ground of our observation is reality. It is also what we model, and we can infer its existence from the consilience of our models. I have said that quite a few times.

So I agree with Jobar, and with you, that reality is the ground of our observations in the sense that it is the subject matter of our models.

But that ground is indescribable in any ultimate sense; we can say "objective reality" as an indicator- a model- of that absolute which transcends all our mathematical or verbal models, yes; but we still can't state that "objective reality" is anything but a model.
Language is capable of pointing to the thing behind the symbol. It does it all the time. When I say objective reality, I mean that thing we're all making models of.

So do I. I actually said that. I just expressed the view that the adjective "objective" was unnecessary, as in wet rain.

[Febble]

[But an industrial psychologist trying to measure a change in attitude after, let's say, undermining resistance to innovation in the workplace, would not be able to count on other psychologists being able to repeat his experiment. The data is too "fuzzy" for properly controlled experimental purposes.
That's the same as saying psychology isn't a science.

That might be true, but it's not what psychologists claim. They expect their experiments to be repeatable.

Absolutely. But noisy data is an issue.

[Preno]

I've been in the habit of distinguishing between reasoning based on the observation and measurement of physical phenomena, and inferences made from information about subjective states. Very crudely, I'd say that physics, chemistry, and biology (the so-called "hard sciences") operate with quantifiable material data that is more or less accessible to each expert in whatever field. I do not know for sure whether this is true in the social sciences, but suspect that it is not.

To give a specific example: it seems to me that predicting how much radiation would result from the detonation of a nuclear device of x size etc., depends on understanding and measuring the large number of physical variables involved. Repeatable experiments would be essential to achieving that understanding. But an industrial psychologist trying to measure a change in attitude after, let's say, undermining resistance to innovation in the workplace, would not be able to count on other psychologists being able to repeat his experiment. The data is too "fuzzy" for properly controlled experimental purposes.

Sorry if my example was too long-winded or far-fetched, but I think you can see what I'm driving at.Well, yes and no. I agree that experiments in natural sciences often have a greater repeatability than, say, social scientific ones. I'm not sure where you're going with this in the context of this discussion, however.

[Rilx]

You [Febble] think the way we prove a model to be useful is by showing how well it fits with our other models.

I think that the way we prove a model is true is by showing how well it fits with reality.

Now you keep telling me that you agree there is a reality but it is inaccessible. I don't know what you mean by this.

To show how well a model fits with reality, i.e. to compare a model with reality, we must know [the absolute] reality, mustn't we? How else we could do the comparison?

But if we knew the absolute reality, we'd know everything. We wouldn't need any models at all because we could see everything as it is, "by God's eyes". We'd live in the world of absolute thruth.

[Febble]

I just think the adjective is superfluous. What would non-objective reality be?
I already said on page 2 of this thread that I was not interested in quibbling about the phrase "objective reality." I provided an explanation of what I meant by the term. The post was in direct response to one of your posts.

Did you a) not read my post, or b) not understand my post?

If your entire argument boils down to your inability to equate the term "objective reality" with the term "reality" even after being provided with a definition, please advise me so I can put you on /ignore.

Here is a link to the post in case you would like to review it:

http://www.secularcafe.org/showthread.php?p=90023&#post90023

That's fine. I just think it's a potentially misleading way of labelling things, that's all. But as you have explained, I'm not misled.

I note that you haven't dropped your habit of sticking words in quotes, even though they aren't quotations.
I guess it's a literary thing. I have other good uses for artistically expressive punctuation, being a published author and all.

See what I mean about snark? Don't apologize. Just stop doing it. Your academic credentials are irrelevant to any possible conversation here. Even if there were somehow relevant, they cannot be proved, so they are merely empty claims. There is no reason to ever mention your (alleged) academic profession.

Well, I understand now that it offends you. As I said, I was just saying where I was coming from. And I know you are a published author. Nonetheless, I don't like having people attribute words to me in quotes when they are words I didn't say. However, I accept that you did not mean to attribute them to me. I've had a lot of trouble, though, in the past, from people assuming that I've said something that turned out to be a paraphrase of what they thought I meant, rather than what I actually said (and meant). So I guess it's a sore spot for me.

Your arguments stand or fall on their own. Stop resorting to authority to support them. Also, refraining from being demeaning or dismissive of your fellow conversationalists would be helpful too.

I'm not "resorting to authority". You've complete misread my intention. I'm not blaming you for that, clearly I was misleading. However, I have now put the record straight. And I'm certainly not trying to be "demeaning or dismissive of [my] fellow conversationalists". Actually I'm not even "not trying" - I'm simply not doing it. It's not the way I think (except occasionally my patience runs out with creationists).

And yes, I agree that we do not have to have direct access to reality to be able to infer its existence. Which I think is a better way of putting it than "discern objective truths", for many reasons. I think the idea that we can "discern objective truths" is a dangerous one. What we can do, and should do, is continually check the consilience of our models, because what seems obviously an objective truth in one era, seems to be a rank falsehood in another.
To discern an objective truth is to prove that a specific component (such as a theoretical mechanism or prediction) is consilent with reality. That is what "objective truth" means. It means true about objective reality.

Well, I am putting to you a different view: that we do not "discern an objective truth" - what we do is find that our models are consilient with each other. That's where we differ. That's the point I'm trying to make.

When you say we should continually check the consilence of your models, you seem to imply that at any given point a specific (previously reliable) model could suddenly stop being consilent. Is this what you mean to imply?

Well, it could certainly cease to be consilient with other good models. As Asimov points out in his essay. So yes, old models that seem to account for all the available data can, at a later stage, prove not to account for new available data. And, occasionally, it happens that the earlier model DOES account for new available data because the new available data is the model that is out of step. That's why I'm unhappy with this division between models on one hand and data on the other. Data are models too, and sometimes those are the models that need adjusting.

because what seems obviously an objective truth in one era, seems to be a rank falsehood in another. As Asimov points out :)
That's what you got out of his article? Are you serious? And you've read it many times? :eek:

Are you serious?

Yes, indeed. But again, I don't think we are disagreeing - I think you are reading something into my words that I do not intend. It may be that you think my point is trivial, and, in a way it is. Most of the time the division between models and data is a convenient one to make (a convenient model, lol) But it ican be dangerous, IMO, to forget that even our data are models. And the reason I mentioned my field (and for no other reason) is that my field is a very good example of that. I work in brain imaging (and no, that is not an attempt to argue from authority, it is simply an argument from an example with which I am familiar) and we collect "data" and we model it. however, the data itself is fairly highly processed before I even attempt to model it, and in any case, the data is not neural, it's blood flow, and not even that, it's the Blood Oxygen Level Dependency of the magnetic resonance image. All of which are models. And it's only too easy to think "this is what the data show" and forget that often what we regard as data are artefacts of the data processing - of the modelling.

In which case, why are we disagreeing?
The actual disagreement is this:

I meant that our some models are more consilient with each other...

What we can do is construct models of it, and check the consilience of those models with each other - that includes making predictions, and checking that our new observations are consilient with the prior model (or not).


You think the way we prove a model to be useful is by showing how well it fits with our other models.

Well, if the use we have for our model is to predict new data, yes. There are other kinds of models.

I think that the way we prove a model is true is by showing how well it fits with reality.

Now you keep telling me that you agree there is a reality but it is inaccessible. I don't know what you mean by this.

Well, there's a reason that people say "proof is for math and alcohol, not science". I'd argue that we don't prove models are true in science. We demonstrate that they are consilient with each other and with the data, and as I regard the data themselves as models, then that can be simplified to "demonstrate that they are consilient with each other". But of course it is particularly convincing if your model makes a prediction about new data, and the new data is consilient with it. As long as you (generic you) don't forget that the new data are also models.

If you mean it is impossible to judge models by how well they fit with reality, then you are implying that reality is irrelevant, and all we need to do is make up models that fit really well with each other, and therefore our models are not constrained by reality.

Well, I'm turning that inside out. I'm saying our models (predictive models) are highly constrained by the requirement (to be useful) that they are consilient, and that from the fact that we can build complex, nested, consilient models we can infer that there really is reality (objective reality if you like) "out there". So rather than saying our models are constrained by reality, I'd say that predictive models are constrained to be consilient, and from their consilience we can infer reality.

They are merely constrained by other models. In this view the only reason we can't all have ice cream for dinner and not get fat is because of those mean-spirited people who hold onto the "fat makes you fat" models. If only they were purged from society - by unspecified means - why we could make up models that would fulfill our wildest dreams. And as long as they all matched with each other, we would be perfectly happy. And skinny.

Well, no. Remember that I am regarding data as models. Sometimes they are such accurately predictive models that we forget they are models at all. We just call them "facts". But a nested set of consilient predictive models isn't "merely" constrained. It's highly constrained. I'd argue.

On the other hand, if you are implying that all of these models must eventually conform to observations of the real world, then what do you mean by "inaccessible?" Other than the obvious and trite claim that we cannot "access" reality "directly," which has already been dismissed as self-evident and uninteresting.

Well, you may find it so. That's fair enough. Obviously it's not uninteresting to me, because it's what my field is.

In fact my meaning should have been clear because I talked about Grand models being consilient with submodels, and also pointed out that we currently have two Grand and highly useful models, that are also highly predictive, but are not actually consilient with each other.

I'm just gonna guess here, but I think your two Grand models are "metaphysical naturalism/science" and "supernaturalism/religion." While it is true they are not consilent, it is not true that they are equally useful or predictive.

Oh, no, I wasn't trying to be coy. I said what they were - Relativity and Quantum Mechanics. Sorry, I should have repeated that. I don't think "metaphysical naturalism/science" and "supernaturalism/religion" are comparable models at all. They are neither consilient nor non-consilient, in my view. Or rather, some forms of "supernaturalism/religion" are simply non-consilient with the whole of science, and completely unpredictive. They predict, but their predictions utterly fail. Other forms simply have different uses, and don't attempt to be predictive models at all. But Relativity and Quantum Mechanics are both comparable models in that they are useful predictive models that are consilient with a great number of observations. But they aren't consilient with each other.

However, this does illuminate the point of your defending "internal coherency" as a standard of truth. The point is to preserve viability for a theory which has nothing but internal coherency to recommend it (although honestly, it doesn't even that).

Well, we seem to have got on to a stray track here. No. I meant nothing of the kind.


.
.
.
.
.
.
Right,
Hey, muidiri... I was right. This really is all about the sophomore philosophy observation that reality is not directly accessible.

Next will probably be the idea, "what if our entire universe were just an atom in another, bigger universe?" :dunno:

Well, sophomore it may be. What was that about not demeaning your fellow conversationalists? :evil:

[Alex]

Well, yes and no. I agree that experiments in natural sciences often have a greater repeatability than, say, social scientific ones. I'm not sure where you're going with this in the context of this discussion, however.

What I'm trying to insinuate is my disposition - which I believe is widely shared - to regard the discourse of scientists when talking about physical science as being authoritative. If I heard what a nuclear physicist had to say about the construction of an atom bomb, I'd consider it as trustworthy unless I had reason not to. I would place less trust in what an economist had to say about the causes of and cures for a boom and bust cycle. My trust in what a psycho-analyst might say about the operations of the unconscious mind is close to vanishing point.

The reason I "privilege" the discourse of physical scientists above all others is because the nature of the data they examine ranges from "hard" to "fuzzy", in my opinion, and because the science involved ranges from empirical to speculative.

However, if a scientist speaks about moral problems and offers advice about what we ought to do, I take no more notice of him than the next person.

If I've understood what you say about the normative character of all scientific discourse, it's a categorical mistake to continue thinking in the way I do (about science and morality).

[Febble]

Well, yes and no. I agree that experiments in natural sciences often have a greater repeatability than, say, social scientific ones. I'm not sure where you're going with this in the context of this discussion, however.

What I'm trying to insinuate is my disposition - which I believe is widely shared - to regard the discourse of scientists when talking about physical science as being authoritative. If I heard what a nuclear physicist had to say about the construction of an atom bomb, I'd consider it as trustworthy unless I had reason not to. I would place less trust in what an economist had to say about the causes of and cures for a boom and bust cycle. My trust in what a psycho-analyst might say about the operations of the unconscious mind is close to vanishing point.

The reason I "privilege" the discourse of physical scientists above all others is because the nature of the data they examine ranges from "hard" to "fuzzy", in my opinion, and because the science involved ranges from empirical to speculative.

I think what you are calling "fuzzy" is what I would call something like "noisy", which just means that there are many more factors influencing the data than we can possibly model. I don't see that having noisy data makes inferences more "speculative" - it's just harder to get your confidence intervals down. Psychology is just as empirical as the other sciences, IMO (although I wouldn't include psychoanalysis in that).

However, if a scientist speaks about moral problems and offers advice about what we ought to do, I take no more notice of him than the next person.

I'm not sure about that, although I agree in principle. But I do think that science is shedding light on some moral issues. Abortion, for example. And the nature of moral responsibility itself.

If I've understood what you say about the normative character of all scientific discourse, it's a categorical mistake to continue thinking in the way I do (about science and morality).

I think that "normative" is the wrong word for scientific models (but I may be misunderstanding you). My categorical distinction is between normative (what should happen) and predictive (what will happen) models.

[Febble]

I think it's also worth distinguishing two categorically different questions here:


Why do human beings have moral principles?
What should those moral principles be?


The first certainly has a biological answer (in my view).
The second is completely different, but I think what scientists now know has a bearing on it.

[Alex]

I think it's also worth distinguishing two categorically different questions here:


Why do human beings have moral principles?
What should those moral principles be?


The first certainly has a biological answer (in my view).
The second is completely different, but I think what scientists now know has a bearing on it.

Thanks for making this important distinction which I failed to do in my previous post.

I remain to be convinced that evolutionary biologists can explain the origin of moral principles in the way they can explain something like genetic variation in flowers.

Even if they can, I still do not believe that when scientists jump outside their fields of expertise and speak as "mere intellectuals", they are any more fitted to give moral advice to humanity than Joe Bloggs.

[Febble]

I think it's also worth distinguishing two categorically different questions here:


Why do human beings have moral principles?
What should those moral principles be?


The first certainly has a biological answer (in my view).
The second is completely different, but I think what scientists now know has a bearing on it.

Thanks for making this important distinction which I failed to do in my previous post.

I remain to be convinced that evolutionary biologists can explain the origin of moral principles in the way they can explain something like genetic variation in flowers.

Why not? (That's not a sarcastic question by the way, I really want to know where your boggle point is).


Even if they can, I still do not believe that when scientists jump outside their fields of expertise and speak as "mere intellectuals", they are any more fitted to give moral advice to humanity than Joe Bloggs.

In one sense, I agree.

However, in other senses I profoundly disagree. I think that some fields of science can provide highly relevant information - indeed raise new ethical issues. For instance, it seems pretty clear to me that the idea that a human being exists from the moment of conception is simply not supported by the scientific evidence. So that has a bearing on the ethics of abortion. Ditto with the other end of life. Before cardiac resuscitation was possible, it was easy to mark the end of life by the cessation of heart-beat. Now cardiac is just another potentially reversible organ failure. We now know that life has a fuzzy beginning and a fuzzy end, and we need to think of "the sanctity of life" in terms of potential and reversibility, rather than existence and non-existence.

So many traditional ethical positions turn out to be, essentially, about hygiene. We know now that disease is caused by pathogens, not sin, and so behaviours that increase risk of disease aren't necessarily wrong, just risky. And the risks can be minimised.

The best ethical principle I know of is The Golden Rule. Yahzi thinks it can be derived from logic (as I understood him in the past). I'm not so sure. But wherever we get it from, it seems to me that science can help us understand What we should best do for our Neighbour. It can't tell us Whether we should do it.

[Yahzi]

It may be that you think my point is trivial
What are the sources of our disagreements?

1) Two pages on the unnecessary-ness of putting "objective" in front of "reality."

2) Four pages on the well-understood, stipulated, and never questioned observation that cognition involves modeling reality instead of directly accessing it (whatever that would mean).

Here are a sample of your quotes:

Well, I am putting to you a different view: that we do not "discern an objective truth" - what we do is find that our models are consilient with each other. That's where we differ. That's the point I'm trying to make....

I'd argue that we don't prove models are true in science. We demonstrate that they are consilient with each other and with the data, and as I regard the data themselves as models, then that can be simplified to "demonstrate that they are consilient with each other"...

So rather than saying our models are constrained by reality, I'd say that predictive models are constrained to be consilient, and from their consilience we can infer reality...

Remember that I am regarding data as models. Sometimes they are such accurately predictive models that we forget they are models at all. We just call them "facts"...

I'm saying our models (predictive models) are highly constrained by the requirement (to be useful) that they are consilient, and that from the fact that we can build complex, nested, consilient models we can infer that there really is reality (objective reality if you like) "out there"...
What we see here is a constant theme that your "usefulness" (what ordinary people call true) is predicated on consilence with other theories.

Then I point out to you that all of these observations must originate in some kind of objective, external reality, and you say:

I agree that the ground of our observation is reality. It is also what we model, and we can infer its existence from the consilience of our models.
And the minute I say anything about reality, you immediately go off on your "data are models" routine again.

We know that the human brain does not contain direct access to pumpkins. We know that thinking about pumpkins involves modeling pumpkins. No one in this thread has ever even hinted at disputing that.

The dispute is over whether or not it is possible to make truth claims about objective reality - that is, the reality that underlies all of these models. I keep saying that a model which is more consilent with real-world observations is truer than a model that isn't; that is, it is a more accurate model of the underlying reality. Ala Asimov, we can state "the world is not cubical" as an objective truth; no amount of refinement of our models will ever invalidate that statement. This qualifies it as an objective truth.

When I say this, you go to great lengths to assure me that our data are only models, too, so my position is unsound.

When I point out that the models must depend on observations from the real world, you agree.

And we go round and round and round again.

Here is my claim: There is an objective reality. We can make models of it. Given any two models, we can judge which one is more accurate. We call this "discerning truth." Some truths are so large and well-discerned that although they may be refined they can never be refuted (aka the earth may be round, egg-shaped, or pear-shaped, but will never turn out to be square).

What part of this do you disagree with? If the answer is "nothing," then this entire thread has been about nothing more than quibbling over terms. Perhaps we can put that behind us and move on.

Data are models too, and sometimes those are the models that need adjusting.
I do believe that in effect stating, "It is the data that needs adjusting to match my new theory" is grounds for automatic disqualification from the scientific enterprise.

Now if what you are trying to say is, "sometimes our observations turn out to be wrong - i.e. sometimes our data is just crap," well of course. But that doesn't have anything to do with modeling and it doesn't involve adjusting. When you have crap data, you throw it out and replace it with new data. Even when a scientist asserts that his theory will be proven right once you obtain hitherto non-existing data, he is not suggesting you "adjust" the data set. He is suggesting you correct, increase, or refine the data set. In common usage these words imply different things.

Once again you are unable to phrase your ideas in simple terms that make sense to ordinary people. You are so focused on working the modeling angle into every paragraph that you can't just say what you mean.

I've had a lot of trouble, though, in the past, from people assuming that I've said something that turned out to be a paraphrase of what they thought I meant, rather than what I actually said (and meant). So I guess it's a sore spot for me.
In normal usage (such as what has occurred consistently throughout this thread) direct quotes are marked by /quote boxes, while quotation marks are often used to mark text as important, ambiguous, or paraphrased.

[Febble]

It may be that you think my point is trivial
What are the sources of our disagreements?

1) Two pages on the unnecessary-ness of putting "objective" in front of "reality."

2) Four pages on the well-understood, stipulated, and never questioned observation that cognition involves modeling reality instead of directly accessing it (whatever that would mean).

Here are a sample of your quotes:

Well, I am putting to you a different view: that we do not "discern an objective truth" - what we do is find that our models are consilient with each other. That's where we differ. That's the point I'm trying to make....

I'd argue that we don't prove models are true in science. We demonstrate that they are consilient with each other and with the data, and as I regard the data themselves as models, then that can be simplified to "demonstrate that they are consilient with each other"...

So rather than saying our models are constrained by reality, I'd say that predictive models are constrained to be consilient, and from their consilience we can infer reality...

Remember that I am regarding data as models. Sometimes they are such accurately predictive models that we forget they are models at all. We just call them "facts"...

I'm saying our models (predictive models) are highly constrained by the requirement (to be useful) that they are consilient, and that from the fact that we can build complex, nested, consilient models we can infer that there really is reality (objective reality if you like) "out there"...
What we see here is a constant theme that your "usefulness" (what ordinary people call true) is predicated on consilence with other theories.

Then I point out to you that all of these observations must originate in some kind of objective, external reality, and you say:

I agree that the ground of our observation is reality. It is also what we model, and we can infer its existence from the consilience of our models.
And the minute I say anything about reality, you immediately go off on your "data are models" routine again.

We know that the human brain does not contain direct access to pumpkins. We know that thinking about pumpkins involves modeling pumpkins. No one in this thread has ever even hinted at disputing that.

I'm not suggesting they have. However, I am suggesting that language that implies a distinction between sense data on the one hand and models on the other can lead us down fallacious paths, although I accept that you may have resisted the lure. Although I'm not entirely convinced.

So I'm suggesting that a more fruitful way of looking at reality is to consider all our data as models, and that our grounds for assuming that there is an external reality (to which we do not have direct access, as we agree) is the fact that we can build predictive models with such a high degree of inter-consilience.

The reason I think it is (or can be) misleading to think of sense-data as difference in some essential attribute from models is that it can lead us into qualia holes. Indeed, it led me into a qualia hole for quite a while, and you were partly instrumental in extricating me. But I am intrigued that qualia holes are still so prevalent in the non-woo community.

The dispute is over whether or not it is possible to make truth claims about objective reality - that is, the reality that underlies all of these models. I keep saying that a model which is more consilent with real-world observations is truer than a model that isn't; that is, it is a more accurate model of the underlying reality. Ala Asimov, we can state "the world is not cubical" as an objective truth; no amount of refinement of our models will ever invalidate that statement. This qualifies it as an objective truth.

What qualifies it? I mean, I agree that it is an excellent model, and consilient with a great many other models (including the models we make of our measurements) and that working with a cuboid model would be hugely complicated by comparison with working with a spherical model, but I don't know that I agree that "the world is not cubical" is an "objective truth". I think it's just an extremely poor model for most purposes. For some purposes, though, it isn't actually bad, which is of course why Mercator's projection is still used.

And that isn't meant to be snark. As I said, I think that pitting "objective truth" on one side against "models" on another is misleading. And I know of plenty of instances when we have been misled, with the best intentions in the world.

When I say this, you go to great lengths to assure me that our data are only models, too, so my position is unsound.

When I point out that the models must depend on observations from the real world, you agree.

And we go round and round and round again.

Well, I guess we do, because I don't seem to have made my point very clearly. Those observations are models. And we agree on that. So what do those models "depend on"?

Here is my claim: There is an objective reality. We can make models of it. Given any two models, we can judge which one is more accurate. We call this "discerning truth." Some truths are so large and well-discerned that although they may be refined they can never be refuted (aka the earth may be round, egg-shaped, or pear-shaped, but will never turn out to be square).

What part of this do you disagree with? If the answer is "nothing," then this entire thread has been about nothing more than quibbling over terms.

The part I disagree with, is probably the part you regard as "quibbling over terms". As a friend of mine recently conjugated: "I explain; you assert; he/she quibbles". I think it's important. And that part is "Given any two models, we can judge which one is more accurate." My quibble is, that I would say: "Given any two models, we can judge which one is more consilient with our data" with the proviso that the data themselves are models. And so the whole system is infinitely recursive, and so we can never say: "we have discerned truth". And sometimes, it turns out that two models can explain exactly the same data, and be equally useful, but for different purposes. Those circumstances, I would argue, call into question the idea that what we do when we compare models is "discern truth". I have seen a pair of astrolabes, one ptolomaic, one copernican, and both work beautifully (they are in beautiful condition) They explain the data equally well. And for many purposes (e.g. a person standing on the earth with a telescope) the ptolomaic astrolabe will be more convenient to use when it comes to predicting where a given heavenly body will be in the sky at any one time.

Perhaps we can put that behind us and move on.

Well, we could talk about ethics, but I'm going to hold on to that quibble of mine. As I said, I think it's important.

Data are models too, and sometimes those are the models that need adjusting.
I do believe that in effect stating, "It is the data that needs adjusting to match my new theory" is grounds for automatic disqualification from the scientific enterprise.

Well, that is a false paraphrase.

Let me give you an example. I do a lot of work with EEG, and we use Fast Fourier Transforms to decompose our EEG samples into frequency bands. And we do a lot of work, first of all, to clean up the data, so that we eliminate gross artefacts, like eye blinks. That clean-up involves modelling. Then we do analyses on things like the phase or amplitude of each frequency over time, relative to some stimulus. And we see very intersesting things. But we have to be constantly aware that our data are models (even the raw EEG samples are models), and that just because we observe, say, a delta band component, or a gamma band component, that doesn't mean that the brain has delta oscillators and gamma oscillators. It could be that our fundamental data model is wrong.

Same with virtually any data set. There is always measurement error, and the extraction of data from noise involves modelling. Even, as you agree, at the most basic level, the sense-data level, we are modelling. And sometimes those are the models that are wrong.

Now if what you are trying to say is, "sometimes our observations turn out to be wrong - i.e. sometimes our data is just crap," well of course. But that doesn't have anything to do with modeling and it doesn't involve adjusting. When you have crap data, you throw it out and replace it with new data.

Well, not always. That's my basic point, actually. The very criteria we use to decide whether data is "crap" are models. There aren't (again) two bins, marked "good" and "crap". Sorting the signal from the noise is exactly what modelling is, whether it's figuring out what is in front of our eyes, or trying to interpret a brain image, or even counting lake varves.

Even when a scientist asserts that his theory will be proven right once you obtain hitherto non-existing data, he is not suggesting you "adjust" the data set. He is suggesting you correct, increase, or refine the data set. In common usage these words imply different things.

Which words? I'm saying you adjust the model. That's what "correcting" data, or "refining" it involves. Increasing it doesn't, of course. It's always good to have new data.

Once again you are unable to phrase your ideas in simple terms that make sense to ordinary people. You are so focused on working the modeling angle into every paragraph that you can't just say what you mean.

I'm saying exactly what I mean.

I've had a lot of trouble, though, in the past, from people assuming that I've said something that turned out to be a paraphrase of what they thought I meant, rather than what I actually said (and meant). So I guess it's a sore spot for me.
In normal usage (such as what has occurred consistently throughout this thread) direct quotes are marked by /quote boxes, while quotation marks are often used to mark text as important, ambiguous, or paraphrased.

Well, in the normal usage I'm used to, "direct speech" is indicated by quotation marks. It's true that on bulletin boards we often use quote boxes instead, but I still find it disconcerting to have words in quote marks (as well as boxes) attributed to me when they aren't mine. Couldn't you use italics or something? That's what most people around here seem to do.

[Alex]

I remain to be convinced that evolutionary biologists can explain the origin of moral principles in the way they can explain something like genetic variation in flowers.

Why not? (That's not a sarcastic question by the way, I really want to know where your boggle point is).
My boggle point is to do with the nature of the data. I've said this before, but it seems to me that the physical data on which scientists make inferences about what is the case are (and must be) qualitatively different from the information on which they might "model" the origin of moral principles or rules that guide human conduct.
..... I think that some fields of science can provide highly relevant information - indeed raise new ethical issues. For instance, it seems pretty clear to me that the idea that a human being exists from the moment of conception is simply not supported by the scientific evidence. So that has a bearing on the ethics of abortion. Ditto with the other end of life. Before cardiac resuscitation was possible, it was easy to mark the end of life by the cessation of heart-beat. Now cardiac is just another potentially reversible organ failure. We now know that life has a fuzzy beginning and a fuzzy end, and we need to think of "the sanctity of life" in terms of potential and reversibility, rather than existence and non-existence.
I agree that biologists can give a scientific account of conception, but the question of whether a human being is conceived at that critical instant depends on what you mean by "human being", and that is a moral problem. Scientists can offer an opinion on this point of course - but I don't see why their opinions should carry more weight than the views of any other educated person
So many traditional ethical positions turn out to be, essentially, about hygiene. We know now that disease is caused by pathogens, not sin, and so behaviours that increase risk of disease aren't necessarily wrong, just risky. And the risks can be minimised.
I also question your claim that behaviours which increase the risk of disease aren't necessarily wrong. Well, how could they be "right"? If we can agree that a preventable disease is at the very least a misfortune and possibly a calamity - then any behaviour which certainly increases the likelihood of contracting and/or spreading diseases must be wrong. Should I knowingly take the risk of infecting another person with a disease which I had acquired, surely that would be wrong?

Of course simply being alive is itself a risk factor, and there are diseases which can't be prevented: it would be absurd to include some of the cancers and genetic disorders, for example, in a generalization about how we can choose to avoid disease.

[Febble]

I remain to be convinced that evolutionary biologists can explain the origin of moral principles in the way they can explain something like genetic variation in flowers.

Why not? (That's not a sarcastic question by the way, I really want to know where your boggle point is).
My boggle point is to do with the nature of the data. I've said this before, but it seems to me that the physical data on which scientists make inferences about what is the case are (and must be) qualitatively different from the information on which they might "model" the origin of moral principles or rules that guide human conduct.

Well, again, I'd make a categorical distinction between the question: why do human beings have ethical principles? And "what should those principles be?"

The first is, IMO, a perfectly scientific question, and the second isn't. And the data relevant to the first (the data that any scientific model would have to account for) are, IMO, not fundamentally different from any other "physical data", and would range from behavioural data, from both individuals and populations, right down to the neurons and ion channels, taking in genetics on the way.

..... I think that some fields of science can provide highly relevant information - indeed raise new ethical issues. For instance, it seems pretty clear to me that the idea that a human being exists from the moment of conception is simply not supported by the scientific evidence. So that has a bearing on the ethics of abortion. Ditto with the other end of life. Before cardiac resuscitation was possible, it was easy to mark the end of life by the cessation of heart-beat. Now cardiac is just another potentially reversible organ failure. We now know that life has a fuzzy beginning and a fuzzy end, and we need to think of "the sanctity of life" in terms of potential and reversibility, rather than existence and non-existence.
I agree that biologists can give a scientific account of conception, but the question of whether a human being is conceived at that critical instant depends on what you mean by "human being", and that is a moral problem.

Yes, but science has relevant information to offer, for example, the evidence that twinning, if it occurs, does not occur until a certain amount of post-conceptual development has occurred.

Scientists can offer an opinion on this point of course - but I don't see why their opinions should carry more weight than the views of any other educated person

No, but their evidence may.

So many traditional ethical positions turn out to be, essentially, about hygiene. We know now that disease is caused by pathogens, not sin, and so behaviours that increase risk of disease aren't necessarily wrong, just risky. And the risks can be minimised.
I also question your claim that behaviours which increase the risk of disease aren't necessarily wrong. Well, how could they be "right"? If we can agree that a preventable disease is at the very least a misfortune and possibly a calamity - then any behaviour which certainly increases the likelihood of contracting and/or spreading diseases must be wrong. Should I knowingly take the risk of infecting another person with a disease which I had acquired, surely that would be wrong?

Well, I'd agree in that instance. But is it therefore also wrong to take someone for a car ride, knowing that you expose them to the risk of a traffic accident? Or rock-climbing? Or scuba diving?

Risk=/= wrong.

Of course simply being alive is itself a risk factor, and there are diseases which can't be prevented: it would be absurd to include some of the cancers and genetic disorders, for example, in a generalization about how we can choose to avoid disease.

[Febble]

Of course simply being alive is itself a risk factor, and there are diseases which can't be prevented: it would be absurd to include some of the cancers and genetic disorders, for example, in a generalization about how we can choose to avoid disease.

Sorry missed the end of your post, and messed up the quote tags.

Yes, that's a good point.

[Alex]

@ Febble:

I've enjoyed taking part in this conversation, but I'm clocking off for the Christmas holiday now, and I guess we can just leave these exchanges in "suspended animation".

(One day someone will start a thread on a topic that I know something about. :))

[LukeS]

I recall reading that the term "right" is inherited from the Romans. IIRC It comes from the root 'rectus', and right meant "in line" with convention or morality. The idea was borrowed from carpentry and architecture, where a right angled tool was uses to actertain whether things lined up correctly - if they were out of line, they could be rectified.

[Febble]

I think it's the root of "erect" as well, isn't it? And of course we still "rectify" stuff like AC electric current :)

[Yahzi]

I am suggesting that language that implies a distinction between sense data on the one hand and models on the other can lead us down fallacious paths,
Then you make that error every time you use the word "reality."

I realize you went to great lengths to never actually resort to that word. Which is why I kept bringing it up. In the end you were unable to avoid using the word. Which was the entire point.

You cannot discuss reality without talking about reality. As long as we are discussing science, cognition, or objective moral standards of right and wrong, we are necessarily going to have to refer to reality.

The "sense data" you keep banging on about is the referent to which our models point. A word is a referent which points to something. There's nothing particularly wrong with a double reference, that is, using a word that points to a referent that points to something. When I say "objective reality," that is what I mean: the phrase points to a particular model which points to reality.

I am surprised that this requires explanation, let alone 5 pages of discussion.

So I'm suggesting that a more fruitful way of looking at reality is to consider all our data as models, and that our grounds for assuming that there is an external reality (to which we do not have direct access, as we agree) is the fact that we can build predictive models with such a high degree of inter-consilience.
That is a less useful way. Once you decide your observations are merely models, you immediately open the temptation to simply adjust them in exactly the same way you adjust actual models.

We can profitably adjust models. Non-Euclidian geometry, for example, changes the rules of geometry by deleting or adding elemental rules. The resulting model produces interesting things.

We cannot profitably adjust observations. Modifying observations by deleting or adding parts of them does not produce anything interesting. It produces the opposite of it. (Note: interpreting data through a model to yield a conclusion, which is then fed into a different model, is different. That is not modifying the original observation; it is interpreting it.)

It is important to maintain the difference between observations, which are linked to external events, and models, which are purely internal constructions, if for no other reason than that one can be changed and one cannot. While models should be consilent with each other, observations do not have to be. Observations need merely be consilent with reality.

In the terms of Asimov's essay, models can be changed, but observations cannot. We may refine the exact rotundity of the world, but the observations made by Aristotle remain unchanged and must always be accounted for.

What qualifies it? I mean, I agree that it is an excellent model, and consilient with a great many other models (including the models we make of our measurements) and that working with a cuboid model would be hugely complicated by comparison with working with a spherical model, but I don't know that I agree that "the world is not cubical" is an "objective truth". I think it's just an extremely poor model for most purposes. For some purposes, though, it isn't actually bad, which is of course why Mercator's projection is still used.
First you complain that "qualia holes" are still prevalent, and then you dive head-first into the stickiest relativism imaginable.

Look, it is very simple. There is a real planet. It is one way or another. The way it is is spherical. We know this because of many different observations. It is not cubical. You can pretend it is if that makes some mathematical calculation easier, but that is merely factoring your math out. It is not making a claim about reality.

The inability to tell the difference between Mercator's projection and reality is the problem with your approach. Even 12-year-olds can profitably discuss the difference between reality and models using ordinary terms. How is it your technical jargon makes the discussion so confusing?

You have been reduced to asserting that the roundness of the world is not objectively true. You are leaving room for some advanced model to suddenly discover that everything makes more sense if we assume the world is square. This is the exact opposite of what Asimov's essay tried to convey!

As I said, I think that pitting "objective truth" on one side against "models" on another is misleading.
But there is an objective truth. You agree to this every time you agree there is one and only one reality. However difficult it might be to actually identify objective truths, it cannot help to obfuscate the language to the point where we can't even discuss reality.

My quibble is, that I would say: "Given any two models, we can judge which one is more consilient with our data" with the proviso that the data themselves are models. And so the whole system is infinitely recursive, and so we can never say: "we have discerned truth".
You've already stipulated there is a reality (i.e. truth). You've already conceded our observations are based on that reality. Now you are asserting reality is so remote we cannot actually know anything about it.

These are mutually contradictory positions.

Let me give you an example. I do a lot of work with EEG, and we use Fast Fourier Transforms to decompose our EEG samples into frequency bands. And we do a lot of work, first of all, to clean up the data, so that we eliminate gross artefacts, like eye blinks. That clean-up involves modelling. Then we do analyses on things like the phase or amplitude of each frequency over time, relative to some stimulus. And we see very intersesting things. But we have to be constantly aware that our data are models (even the raw EEG samples are models), and that just because we observe, say, a delta band component, or a gamma band component, that doesn't mean that the brain has delta oscillators and gamma oscillators. It could be that our fundamental data model is wrong.
Having used FFTs myself, I am well aware of both the power to reveal and the inevitable contamination that accompanies it. In my case the particular bugaboo was windowing: the edges of my data always had to be swept under the carpet.

Given what I understand about data manipulation, what you've just said makes no sense whatsoever.

Yes, of course, once you massage your data through a particular model you must be aware that any conclusions might be based on the massaging rather than the data.

This does not, however, mean that the original, untouched data is a model. I don't even know what that means.

Same with virtually any data set. There is always measurement error, and the extraction of data from noise involves modelling.
The extracted data is a model. The raw data is not. The theoretical model is constructed from extracted data models. But the extracted data models are still based on actual, raw observations. This is the anchor that links the theory to reality. Sever that link and you're not doing science, you're doing theology.

Sorting the signal from the noise is exactly what modelling is,
And when I talk about raw data, I am talking about unsorted data. Complete with noise. Why do you keep forgetting that we have data that is not sorted?

I'm saying you adjust the model. That's what "correcting" data, or "refining" it involves. Increasing it doesn't, of course. It's always good to have new data.
For a person who quibbles over terms, you seem remarkably insensitive to the nuances of terms.

Yes, correction and refinement are forms of adjustment. So is "changing shit at random" or "inserting data points that we like." Those latter two, however, are not forms of correction or refinement. This is why it is important to distinguish between forms of adjustment. This is why the word adjustment is inappropriate.

Every time you formulate your position, I see a gaping hole that allows the direct, unprincipled modification of primary data. Please explain how your approach (i.e. your language, terms, and concepts) prevents this.

Well, in the normal usage I'm used to, "direct speech" is indicated by quotation marks. It's true that on bulletin boards we often use quote boxes instead, but I still find it disconcerting to have words in quote marks (as well as boxes) attributed to me when they aren't mine. Couldn't you use italics or something?
I'm sorry, but I'm afraid your tropisms are your own problem. You are, of course, free to ignore my posts if my writing style makes them too difficult to understand.

[Yahzi]

I remain to be convinced that evolutionary biologists can explain the origin of moral principles in the way they can explain something like genetic variation in flowers.
Of course they can. It's not even much harder (but that's because explaining flowers is actually pretty hard.)

Even if they can, I still do not believe that when scientists jump outside their fields of expertise and speak as "mere intellectuals", they are any more fitted to give moral advice to humanity than Joe Bloggs.
That's true, but it is an entirely different statement. "Scientifically discovering the evolutionary underpinnings of morality" does not automatically equate to "giving good moral advice."

Consider language. Language is a product of evolution; the neurological organs in the brain are hard-wired by evolution. They are necessary to processing language. Studying them scientifically will teach us a lot about language. But it doesn't mean we can suddenly speak French.

[Febble]

I am suggesting that language that implies a distinction between sense data on the one hand and models on the other can lead us down fallacious paths,
Then you make that error every time you use the word "reality."

I realize you went to great lengths to never actually resort to that word. Which is why I kept bringing it up. In the end you were unable to avoid using the word. Which was the entire point.

I'm not sure where this is coming from, Yahzi. I am not aware that I have avoided the word reality at all. Indeed I think I kicked off by saying we do not have direct access to reality - all we have access to are models. So I certainly used the word right off the bat. And when you asked me what our models were models of, I said that they were models of reality. That's what I mean. I haven't avoided the word at all. Indeed, I have said that we can infer the existence of a reality "out there" from the fact that our predictive models are consilient.

So I am neither denying reality nor avoiding the word.

You cannot discuss reality without talking about reality. As long as we are discussing science, cognition, or objective moral standards of right and wrong, we are necessarily going to have to refer to reality.

Of course. It is what we model.

The "sense data" you keep banging on about is the referent to which our models point. A word is a referent which points to something. There's nothing particularly wrong with a double reference, that is, using a word that points to a referent that points to something. When I say "objective reality," that is what I mean: the phrase points to a particular model which points to reality.

OK.

I am surprised that this requires explanation, let alone 5 pages of discussion.

So I'm suggesting that a more fruitful way of looking at reality is to consider all our data as models, and that our grounds for assuming that there is an external reality (to which we do not have direct access, as we agree) is the fact that we can build predictive models with such a high degree of inter-consilience.
That is a less useful way. Once you decide your observations are merely models, you immediately open the temptation to simply adjust them in exactly the same way you adjust actual models.

We can profitably adjust models. Non-Euclidian geometry, for example, changes the rules of geometry by deleting or adding elemental rules. The resulting model produces interesting things.

We cannot profitably adjust observations. Modifying observations by deleting or adding parts of them does not produce anything interesting. It produces the opposite of it. (Note: interpreting data through a model to yield a conclusion, which is then fed into a different model, is different. That is not modifying the original observation; it is interpreting it.)

There is no "original observation" is my point. Even our "original observations" are models.

It is important to maintain the difference between observations, which are linked to external events, and models, which are purely internal constructions, if for no other reason than that one can be changed and one cannot. While models should be consilent with each other, observations do not have to be. Observations need merely be consilent with reality.

And that is where we disagree. I do not accept that we have access to reality on one hand, and make models of it on the other hand. That is what I am suggesting is a misleading dichotomy. The rigor of scientific investigation IMO comes not from simply "fitting models to data" although of course that is the simple, and highly useful, way of putting it. But the real rigor comes in constantly checking our models (including, as I would argue the data themselves, which I am arguing are also models) are consilient. That's why I gave the neuroimaging example. Some scientific data are pretty straightforward, and governed by reliable laws - so we can use a thermometer to measure temperature (an example I think you like). But, as you agree, I know, the map is not the territory - the reading on the thermometer is not the temperature, it is a correlate of it. It is, in other words, a model of it. So even there, we have to be constantly aware, IMO, that our data are also models.

In the terms of Asimov's essay, models can be changed, but observations cannot. We may refine the exact rotundity of the world, but the observations made by Aristotle remain unchanged and must always be accounted for.

Not necessarily. Probably in this case. But observations can be in error. Take, for example, all the complicated explanations that used to be advanced to account for the fact that the moon looks larger at the horizon than at the zenith - lensing of the atmosphere etc. The answer turned out to lie in the observational model itself. If you measure the angle subtended by the moon at the horizon and at the zenith it is identical. The whole thing is not an optical illusion but a cognitive illusions. Our observational model was wrong.

What qualifies it? I mean, I agree that it is an excellent model, and consilient with a great many other models (including the models we make of our measurements) and that working with a cuboid model would be hugely complicated by comparison with working with a spherical model, but I don't know that I agree that "the world is not cubical" is an "objective truth". I think it's just an extremely poor model for most purposes. For some purposes, though, it isn't actually bad, which is of course why Mercator's projection is still used.
First you complain that "qualia holes" are still prevalent, and then you dive head-first into the stickiest relativism imaginable.

Look, it is very simple. There is a real planet. It is one way or another. The way it is is spherical. We know this because of many different observations. It is not cubical. You can pretend it is if that makes some mathematical calculation easier, but that is merely factoring your math out. It is not making a claim about reality.

I think you have a false dichotomy here. "Spherical" is a highly useful way of modelling the shape of the earth, because it is highly predictive. That's as far as I'll go. No, I'll go further - because it is such a highly predictive model, and so easily reconciled with other models, we can infer there is a real shape to the earth, and our best model of it right now is more-or-less spherical. I have no problem with that. My caution is against the approach that assumes we check models against reality. We don't, IMO. We check them against each other.



The inability to tell the difference between Mercator's projection and reality is the problem with your approach. Even 12-year-olds can profitably discuss the difference between reality and models using ordinary terms. How is it your technical jargon makes the discussion so confusing?

You have been reduced to asserting that the roundness of the world is not objectively true. You are leaving room for some advanced model to suddenly discover that everything makes more sense if we assume the world is square. This is the exact opposite of what Asimov's essay tried to convey!

Well, directions make more sense if we assume that it is cylindrical (I don't see much useful about a cubical model). And no I'm not asserting that "!the roundness of the world is not objectively true". I'm saying that it is more useful, when push comes to shove, to regard the roundness of the world to be a highly predictive model that also highly parsimonious and consilient with a great many other models.

That's all, Yahzi. I'm not denying reality. I'm not even denying that in everyday language saying that "the earth is spherical " isn't a perfectly true statement. I'm saying that when push comes to shove, what we have are models, not reality itself. And that sometimes matters. What is "obvious" to one generation is often far from obvious, indeed, false, to another. For example, to say the earth is flat, is obviously wrong to us. Asimov points out it is only slightly wrong. But conceptually, a spherical earth is very different to a flat earth, not least because a spherical earth is finite but unbounded.

As I said, I think that pitting "objective truth" on one side against "models" on another is misleading.
But there is an objective truth. You agree to this every time you agree there is one and only one reality. However difficult it might be to actually identify objective truths, it cannot help to obfuscate the language to the point where we can't even discuss reality.

I'm not saying "we can't even discuss reality". I've been arguing fairly vociferously that we can. What I said is we can't access reality, except via models. And some of the time you agree, but say it's trivial, and at other times you say it is fundamentally wrong. I don't think it's trivial. I also don't think it's fundamentally wrong. I think it's fundamentally right.

My quibble is, that I would say: "Given any two models, we can judge which one is more consilient with our data" with the proviso that the data themselves are models. And so the whole system is infinitely recursive, and so we can never say: "we have discerned truth".
You've already stipulated there is a reality (i.e. truth). You've already conceded our observations are based on that reality. Now you are asserting reality is so remote we cannot actually know anything about it.

These are mutually contradictory positions.

Well, yes they are. Fortunately I don't hold the second one, so we're cool.

Have a great day tomorrow, Yahzi, whatever you are doing. Good to see you back in the ring :)

[Febble]

Oops, missed a bit:


Let me give you an example. I do a lot of work with EEG, and we use Fast Fourier Transforms to decompose our EEG samples into frequency bands. And we do a lot of work, first of all, to clean up the data, so that we eliminate gross artefacts, like eye blinks. That clean-up involves modelling. Then we do analyses on things like the phase or amplitude of each frequency over time, relative to some stimulus. And we see very intersesting things. But we have to be constantly aware that our data are models (even the raw EEG samples are models), and that just because we observe, say, a delta band component, or a gamma band component, that doesn't mean that the brain has delta oscillators and gamma oscillators. It could be that our fundamental data model is wrong.
Having used FFTs myself, I am well aware of both the power to reveal and the inevitable contamination that accompanies it. In my case the particular bugaboo was windowing: the edges of my data always had to be swept under the carpet.

Yes.

Given what I understand about data manipulation, what you've just said makes no sense whatsoever.

Yes, of course, once you massage your data through a particular model you must be aware that any conclusions might be based on the massaging rather than the data.

This does not, however, mean that the original, untouched data is a model. I don't even know what that means.

Well, it's a sampling, for a start. It's a model based on the assumption that we can model a continuous wave form as a series of digits. Then there are the assumptions we use to clean up the data from all kinds of contaminants. Again, those clean-up procedures are models.

Same with virtually any data set. There is always measurement error, and the extraction of data from noise involves modelling.
The extracted data is a model. The raw data is not. The theoretical model is constructed from extracted data models. But the extracted data models are still based on actual, raw observations. This is the anchor that links the theory to reality. Sever that link and you're not doing science, you're doing theology.


Well, not totally raw. Fairly cooked, I'd say. We never get our data raw.

Sorting the signal from the noise is exactly what modelling is,
And when I talk about raw data, I am talking about unsorted data. Complete with noise. Why do you keep forgetting that we have data that is not sorted?

I'm not forgetting that. I'm just saying that every time you drill down to another level, you reach a set of assumptions that generates that "raw" data. Nothing is raw. Even when we get to what we receive directly through our sense ("I saw it with my own eyes!") we are modelling.

I'm saying you adjust the model. That's what "correcting" data, or "refining" it involves. Increasing it doesn't, of course. It's always good to have new data.
For a person who quibbles over terms, you seem remarkably insensitive to the nuances of terms.

Yes, correction and refinement are forms of adjustment. So is "changing shit at random" or "inserting data points that we like." Those latter two, however, are not forms of correction or refinement. This is why it is important to distinguish between forms of adjustment. This is why the word adjustment is inappropriate.

Well, use another one then. I certainly don't mean "inserting data points that we like". As I think I've made clear. That would be a totally crap model.

Every time you formulate your position, I see a gaping hole that allows the direct, unprincipled modification of primary data. Please explain how your approach (i.e. your language, terms, and concepts) prevents this.

Because of the rigorous requirement for consilience. Replicability is one test of consilience. And there is no hole for "unprincipled" modification of primary data. Precisely what I am saying is that we need principles - models - and those models need to be consilient. If they are not, we can't use them.

Well, in the normal usage I'm used to, "direct speech" is indicated by quotation marks. It's true that on bulletin boards we often use quote boxes instead, but I still find it disconcerting to have words in quote marks (as well as boxes) attributed to me when they aren't mine. Couldn't you use italics or something?
I'm sorry, but I'm afraid your tropisms are your own problem. You are, of course, free to ignore my posts if my writing style makes them too difficult to understand.

My concern is not with what I understand. Obviously, I know the words are not my own. My concern is that people reading your posts, in which you put words in quotation marks, attributing them to me, may think they are my words. That's all I'm asking, Yahzi - not to attribute words to me that are merely paraphrases, in a manner that indicates to the unwary reader that they are my own words, i.e. by putting them in quotation marks. It seems to me to be a reasonable request.

[Preno]

Well, yes and no. I agree that experiments in natural sciences often have a greater repeatability than, say, social scientific ones. I'm not sure where you're going with this in the context of this discussion, however.What I'm trying to insinuate is my disposition - which I believe is widely shared - to regard the discourse of scientists when talking about physical science as being authoritative. If I heard what a nuclear physicist had to say about the construction of an atom bomb, I'd consider it as trustworthy unless I had reason not to. I would place less trust in what an economist had to say about the causes of and cures for a boom and bust cycle. My trust in what a psycho-analyst might say about the operations of the unconscious mind is close to vanishing point.Well, I agree with all that.
The reason I "privilege" the discourse of physical scientists above all others is because the nature of the data they examine ranges from "hard" to "fuzzy", in my opinion, and because the science involved ranges from empirical to speculative.Well, in what sense do you privilege the discourse of physical science, then? So far, you've said that you trust physical scientists more (when they speak about physical science) than you trust economists (when they speak about economics). That's a perfectly reasonable position, given that physics is a much more exact and well-developed science, at least in certain areas.
However, if a scientist speaks about moral problems and offers advice about what we ought to do, I take no more notice of him than the next person.Well, sure, any more than I take notice of linguists offering advice about climatology. (But at least most of the time, surely it's uncontroversial that for example psychologists can often be in a better position to judge whether someone is responsible for their behaviour or not?)