Sam Harris: Science can answer moral questions

[Dani]

It's as arbitrary as the choice of logic and empirical evidence as the basis for our "scientific" science. You can't make an objective case that empirical evidence is a better value than scriptural conformity, or that we "ought" to prefer logic to iambic pentameter in scientific papers. I can't make an objective case that we ought to value the well-being of conscious creatures. Fine.

"The well-being of conscious creatures" is completely consistent with my own intuitive understanding of morality, so if labeling it arbitrary is necessary for moving the discussion forward, I'm happy to do it.

You're right that logic or empirical evidence is as arbitrary as the moral axiom "well-being of conscious creatures". The difference is that the former axioms belong to the foundations of the scientific method, while the latter doesn't.

 

[bokonon]

You're right that logic or empirical evidence is as arbitrary as the moral axiom "well-being of conscious creatures". The difference is that the former axioms belong to the foundations of the scientific method, while the latter doesn't.

What is the point of this observation?

The science of physics pursues the goal of accurately describing the physical universe, though there is no objective reason we should value accurate descriptions of the physical universe.

The science of medicine pursues the goal of promoting health, though there is no objective reason we should value health.

Harris is proposing a science of morality which pursues the goal of promoting the well-being of conscious creatures. The fact that this goal can no more be objectively justified than the goals of physics or medicine does not, it seems to me, render the whole proposal a priori unscientific.

 

[Dani]

What is the point of this observation?

The science of physics pursues the goal of accurately describing the physical universe, though there is no objective reason we should value accurate descriptions of the physical universe.

The science of medicine pursues the goal of promoting health, though there is no objective reason we should value health.

Harris is proposing a science of morality which pursues the goal of promoting the well-being of conscious creatures. The fact that this goal can no more be objectively justified than the goals of physics or medicine does not, it seems to me, render the whole proposal a priori unscientific.

Science is descriptive. In fact, the scientific method doesn't say that we should value health or the landing on the moon. We use science to diagnose and cure illnesses and to explore space, but science doesn't tell us that we have to diagnose and cure illnesses or that we have to explore space. Our society decides that it is useful.

So yes, you can pursue the goal of promoting the well-being of conscious creatures and use science, but science does not provide this goal.

 

[bokonon]

Science is descriptive. In fact, the scientific method doesn't say that we should value health or the landing on the moon. We use science to diagnose and cure illnesses and to explore space, but science doesn't tell us that we have to diagnose and cure illnesses or that we have to explore space. Our society decides that it is useful.

So yes, you can pursue the goal of promoting the well-being of conscious creatures and use science, but science does not provide this goal.

I think Harris acknowledges this. "Of course, anyone who has an alternative set of moral axioms is free to put them forward, just as they are free to define "science" any way they want." (TML, p. 41). Science doesn't provide the axiom; society uses the axiom and the scientific method to seek a goal.

Harris seems to be arguing that we can use the descriptive power of science to decide that one outcome is better than another outcome. The judgment would be based on which better promotes "the well-being of conscious creatures," just as we might use the descriptive power of science to choose one treatment over another in evidence-based medicine.

 

[Dragoonster]

Without him demonstrating that his chosen axiom of promoting well-being is somehow special, his argument is incredibly trivial, and nothing at all new. Science can be used to better achieve any moral goal, this isn't novel. What would be novel is using science to actually define one goal as empirically superior, which Harris utterly fails to demonstrate. He puts the cart way before the horse.

Choose an axiom and write a book saying how science can help best follow it, and somehow be a new voice on morality? Ridiculous. Humans have been doing similar things for millenia, just with different justifications or sources for their particular bents.

 

[bokonon]

Without him demonstrating that his chosen axiom of promoting well-being is somehow special, his argument is incredibly trivial, and nothing at all new. Science can be used to better achieve any moral goal, this isn't novel. What would be novel is using science to actually define one goal as empirically superior, which Harris utterly fails to demonstrate. He puts the cart way before the horse.

Choose an axiom and write a book saying how science can help best follow it, and somehow be a new voice on morality? Ridiculous. Humans have been doing similar things for millenia, just with different justifications or sources for their particular bents.

Oh dear. Someone writes a heavily footnoted non-fiction book and it contains ideas which are not original? Light the torches...

What I see as new in his book is the framework provided for judging one set of choices more moral than another using the objective "well-being of conscious creatures" metric. This may be old stuff, but I haven't run across the "moral landscape" model before. I guess it wouldn't be the first time an idea conceived by one person was popularized by another. If the problem is that Harris hasn't properly credited his souces, maybe you can share which authors have developed this idea previously.

 

[fls]

What do you mean, without any previously established goal? Why would it occur to anyone that women should be forced to wear burquas in the first place? Doesn't that suggest to you that there is a goal associated with the idea? These aren't random ideas. They arise within a context. Give me an example of a moral statement which arises devoid of any context and then I will understand what you mean by "without any previously established goal".

Don't worry. Since we both agree that science can't give an answer to this question without a previously established goal, we can skip this part. Give me an example of a scientifically established goal.

What's a scientifically established goal? "Avoidance of death" isn't a scientifically established goal. "Presence of water" isn't a scientifically established goal. "The speed of light" isn't a scientifically established goal. Why the need for something called a "scientifically established goal" in this case?

But like I said, science does not treat subjective and objective information differently. That is, regardless of whether you wish to apply these rules about what is subjective and what is objective, it does not prevent science from investigating both of them.

Agreed. I haven't said anything that implied the contrary.

Then why did you bring it up?

Way back I gave some examples of how science uses descriptive information to investigate morals which goes beyond merely describing what subjects believe. I can add the links.

http://www.internationalskeptics.com/forums/showthread.php?postid=6455035#post6455035
http://www.internationalskeptics.com/forums/showthread.php?postid=6458506#post6458506

As far I understand, you describe different outcomes related to different actions, and people are logically inconsistent many times with their choices. I have nothing to add if that's all. If I missed something important, feel free to point it out.

That you call it "logically inconsistent" means that you missed something important. If a model does not match reality, it suggests that the model is wrong, rather than reality being wrong. To scientists, I mean. I realize philosophers probably think of the real world as somehow 'wrong'.

Not really. Science makes use of logic where appropriate, but science is mostly about collecting empirical information, rather than attempting to derive facts from first principles.

Well, you made a derogatory comment about logic: "logic games". We can be logically right or wrong, but logic isn't a game with which one can wind up other people. It's a very serious and consistent method.

I'm pointing out that knowledge about the world does not come from sitting in an empty room and making use of axioms and logic. It comes from gathering empirical information.

You have a narrow view of science. Collecting empirical information without logic is like collecting stamps. Any time scientists decide to collect empirical information, logic is appropriate. Collecting empirical information, with inductive logic, leads to a hypothesis. When a hypothesis is tested, if it doesn't pass the test, scientists logically deduce that it's invalid. Logic is very present in the scientific method.

I'm just pointing out that the information on which science is based is collected empirically. Logic is a tool which can be applied to that information in the process of testing hypotheses, but logic does not generate novel information, nor does science operate by forming axioms from which hypotheses are derived. Axioms are not generated from empirical observations.

It's not of anybody's interest unless woman are actually forced to wear burquas and if coercion is distasteful.

And who decides if coercion is distasteful? Who decides if coercion to women is distasteful? Who decides if, even being distasteful, it's bad? And based on what?

You are really missing my point. Nobody decides if coercion is distasteful, just like nobody decides that pneumonia is sometimes fatal or that crows are black.

Linda

 

[Dani]

What's a scientifically established goal? "Avoidance of death" isn't a scientifically established goal. "Presence of water" isn't a scientifically established goal. "The speed of light" isn't a scientifically established goal. Why the need for something called a "scientifically established goal" in this case?

We're talking about morals. Let's sum up what's been said:


Linda: useful questions such as whether women should be forced to wear burkas can be answered using scientific methods.

Dani: without any previously established goal?

Linda: of course there's a goal associated with the idea. These ideas arise within a context.

Dani: of course, I agree. Show me an example of a goal that has been scientifically established.


So if you don't understand my point, I guess you don't think science can answer moral questions unless we provide a goal.

And this leaves us at "science can answer moral questions... when we have
established a goal". If that's all, it's quite trivial. No one has questioned this.

Then why did you bring it up?

Because, among other things, we're discussing about statements. Subjects do answer moral questions. Objects don't. Moral answers are, by definition, subjective. They express subjective wills and desires about the future. Yes, we treat it like data, but this data comes from subjective minds, and it doesn't describe the universe. Once we have processed this data, it only describes the wills and desires of a subject. If we observe that a sheep is black, we cannot conclude that sheeps are black, and the same follows by observing the moral values of a subject.

That you call it "logically inconsistent" means that you missed something important. If a model does not match reality, it suggests that the model is wrong, rather than reality being wrong. To scientists, I mean. I realize philosophers probably think of the real world as somehow 'wrong'.

What model? People's choices when presented with moral dilemmas? This is not a model. Note that when I say "logically inconsistent", I'm referring to logic, not morals. So people are often logically inconsistent. So what?

I'm pointing out that knowledge about the world does not come from sitting in an empty room and making use of axioms and logic. It comes from gathering empirical information.

Comes from both. And sometimes we can dismiss an argument only with logic. Not what I intended with my "games" anyway, since I was just trying to illustrate the difference between prescriptive and descriptive information within statements, and then move on from what I expected would've been an agreed premise.

I'm just pointing out that the information on which science is based is collected empirically. Logic is a tool which can be applied to that information in the process of testing hypotheses, but logic does not generate novel information, nor does science operate by forming axioms from which hypotheses are derived. Axioms are not generated from empirical observations.

Of course information is collected empirically. And processed logically.

You are really missing my point. Nobody decides if coercion is distasteful, just like nobody decides that pneumonia is sometimes fatal or that crows are black.

Distasteful is a value judgment. Fatal (mortal) is not. And black is not.

There's a qualitative difference. We can objectively observe what's fatal (mortal) and what's black, but what's distasteful? Where has it been empirically observed? We only can objectively provide statistics about moral judgments made by subjects.

 

[fls]

Sorry for the delay fls; i have been offline for a couple of weeks.

That's okay. I'm not keeping up the pace, either.

I am happy to be wrong about this. The problem is that it doesn't help me. If Hume means that there is no logical or emprical basis to move from as is to an ought, he is excluding science's ability to do so from the picture. I'm looking for some sort of justification for this.

View the question from the other way around. Hume points out that so far, no-one has been able to move from is to ought using only is statements. This in itself is merely an observation and Hume would grant that the problem of induction would pertain to this observation - perhaps there is a way to move from is to ought statements, only no-one has cracked it yet. Yet it's like the observation that no human has so far been able to fly by merely flapping their arms. Would I need to justify this claim?

After all, you stated earlier that it is reasonable to say "if we want to improve well-being...". The point is that what "we want" is a factual description of the mental states of conscious beings. It is possible to state whether or not we want to improve well-being to fill in our "if...then..." If Hume claims that this is insufficient, that a description of what we want is not sufficient to indicate what we want, then I need to understand why we are expected to divorce ourselves from science on this point.

If we have Goal A, then Action B may be reasonable. This is not the point of contention. If we agree on our goals and values, science can help us to fulfill them.

Hume's contention was that we do not seem to be able to use logic and empiricism to justify our values in the first place. You use the term, 'what we want' and point out that this is grounded in 'factual descriptions of conscious beings', but this does not avail as those descriptions are not universal. Imagine a world where everyone has identical beliefs about what ought to be done. In this world it would be possible to derive values from facts in the direct way you suggest. Yet it would also be a world devoid of moral discussion.

But if you are admitting that this is really just about real or perceived inconsistencies, then you are admitting that this is amenable to scientific inquiry. This point was brought up by Piggy and myself ages ago. If you are looking for an understanding as to whether apparently non-universal beliefs spring from differing values, or just where it is that inconsistencies lie, then Science is your man.

And you now have just highlighted the apparent vacuity of the insistence on treating "oughts" as a goal to be distinguished from "is". You have just admitted that if it happened to be the case that the descriptions of human values were consistent, it would be possible to treat those descriptions as the "oughts" you and others claim as a necessary condition. You seem to be confirming my suspicions, and confirming the point brought up by Harris, that this is really about the complexity of human values - the task is difficult, so some have decided it is impossible.

If I am wrong about this and Hume intends to incorporate science's purview with respect to useful questions, then this provides a logical and empirical basis for moving from an is to an ought. Either the claim is unsupported by reasoning, or it is unremarkable. Neither option seems worthy of 25 pages of discussion here, let alone several hundred years of discourse. This leads me to suppose that I am misunderstanding some aspect of this.

Let's move from calling it Hume's claim, to calling it Hume's observation. Perhaps that helps you? Hume observes that no-one has logically or empirically justified an ought statement on the basis of is statements alone. As an observation, it stands as good science and I'm not sure why you believe it requires further justification?

I am asking why it occurs to anyone that "ought" statements have anything to do with answering questions about and in regards to human values?

That leaves us with an 'unremarkable' observation. You may consider that the observation itself is unremarkable, but when someone like Sam Harris kinda sounds very much like they are claiming that the observation does not hold, that you can move from is to ought using science, wouldn't you grant that this is remarkable and worth at least 25 pages of a subforum in a relatively obscure internet forum? If I claimed that I can fly when I flap my arms that would be remarkable, no?

I'm not claiming that the observation doesn't hold, just that it is immaterial. It isn't that you can move from an is to an ought using science, it's that oughts have nothing to do with answering these questions.

And since some people here, such as yourself, are adamant that they understand the claim and that it is valid, I am looking to you to provide that justification. What I seem to have so far is something like, "human wants vary which precludes the formation of an objective or absolute sort of statement as to "we want...", and the lack of that objective/absolute statement is what precludes comparison and ultimately judgement."

I don't need to justify Hume's observation about is/ought (which can really be simply stated that, no-one has successfully moved from is statements to ought statements, purely on the basis of is statements), I haven't seen any evidence to the contrary. If you have, then please show us.

I don't care. As far as I can tell, these magical "oughts" have nothing to do with answering important questions about moral behavior.

Linda

 

[AlBell]

As far as I can tell, these magical "oughts" have nothing to do with answering important questions about moral behavior.

Linda

Other than without an 'ought' there's nothing to answer?

 

[fls]

What's a scientifically established goal? "Avoidance of death" isn't a scientifically established goal. "Presence of water" isn't a scientifically established goal. "The speed of light" isn't a scientifically established goal. Why the need for something called a "scientifically established goal" in this case?

We're talking about morals. Let's sum up what's been said:


Linda: useful questions such as whether women should be forced to wear burkas can be answered using scientific methods.

Dani: without any previously established goal?

Linda: of course there's a goal associated with the idea. These ideas arise within a context.

Dani: of course, I agree. Show me an example of a goal that has been scientifically established.


So if you don't understand my point, I guess you don't think science can answer moral questions unless we provide a goal.

And this leaves us at "science can answer moral questions... when we have
established a goal". If that's all, it's quite trivial. No one has questioned this.

The problem, for me, is that you are treating the establishment of goals as though it is an important and a separate step from asking the question in the first place. And that this somehow distinguishes the asking and answering of questions about moral behaviors from other scientific endeavors. Yet as far as I can tell (unless you relent and provide an example) we have no "scientifically established goals" in other fields of science. For example, we can study a drug for its ability to prevent mortality from mortality, even though the prevention of mortality from pneumonia is not a "scientifically established goal".

I'm asking why you think we need to establish goals in this way in the first place. And for you to provide an example of how this takes place in another field of science. Otherwise, it still looks to me like this is an arbitrary rule, made up to provide a justification for "non-overlapping magisterial" which serves religious apologists.

Then why did you bring it up?

Because, among other things, we're discussing about statements. Subjects do answer moral questions. Objects don't. Moral answers are, by definition, subjective. They express subjective wills and desires about the future. Yes, we treat it like data, but this data comes from subjective minds, and it doesn't describe the universe. Once we have processed this data, it only describes the wills and desires of a subject. If we observe that a sheep is black, we cannot conclude that sheeps are black, and the same follows by observing the moral values of a subject.

So what? Since we are attempting to answer questions about which actions are good or bad, characteristics which do not describe the universe, but describe a subjective characterization, your point seems entirely irrelevant.

That you call it "logically inconsistent" means that you missed something important. If a model does not match reality, it suggests that the model is wrong, rather than reality being wrong. To scientists, I mean. I realize philosophers probably think of the real world as somehow 'wrong'.

What model? People's choices when presented with moral dilemmas? This is not a model. Note that when I say "logically inconsistent", I'm referring to logic, not morals. So people are often logically inconsistent. So what?

I'm talking about ideas about why actions are characterized as "good" or "bad" by subjects. If "logic" suggests that an action should be characterized as "good", yet it is chatacterized as "bad", it suggests that our idea needs modification. In particular, it's pretty clear that using logic applied to axioms provides a poor model for characterizing moral behaviors.

I'm pointing out that knowledge about the world does not come from sitting in an empty room and making use of axioms and logic. It comes from gathering empirical information.

Comes from both.

I'm referring to the practice which has been proposed here, where instead of empirically gathering information about why actions are characterized as good or bad, some sort of generally acceptable axiom is put forward and then an attempt is made to use logic to derive which actions should be characterized as good or bad.

And sometimes we can dismiss an argument only with logic. Not what I intended with my "games" anyway, since I was just trying to illustrate the difference between prescriptive and descriptive information within statements, and then move on from what I expected would've been an agreed premise.

I also understand the difference between metaphorical and literal statements. What difference does it make?

I'm just pointing out that the information on which science is based is collected empirically. Logic is a tool which can be applied to that information in the process of testing hypotheses, but logic does not generate novel information, nor does science operate by forming axioms from which hypotheses are derived. Axioms are not generated from empirical observations.

Of course information is collected empirically. And processed logically.

You are really missing my point. Nobody decides if coercion is distasteful, just like nobody decides that pneumonia is sometimes fatal or that crows are black.

Distasteful is a value judgment. Fatal (mortal) is not. And black is not.

There's a qualitative difference. We can objectively observe what's fatal (mortal) and what's black, but what's distasteful? Where has it been empirically observed? We only can objectively provide statistics about moral judgments made by subjects.

So what? We are talking about something which is a subjective characterization. It may interest you to know that colour is also a subjective characterization, by the way. I thought we agreed that science can investigate and answer questions about subjective characterizations?

Linda

 

[fls]

Other than without an 'ought' there's nothing to answer?

Maybe if you're a philosopher and in the habit of looking at the world teleologically.

Linda

 

[fls]

fls,
While I can't show that it is impossible to derive an ought purely from empirical or logical means (without beginning with an ought) I believe I can at least show that there are significant hurdles barring the way which have not yet been scaled.

Your description which follows shows that you are treating 'ought' as 'behaviour which is characterized simply and consistently as good or bad'. That is, if the answers were obvious, then it would be reasonable to treat these subjective characterizations as 'oughts'.

With the concept of ought, we're talking about behavior and its effects. Right away it becomes clear that the effects of any given behavior to be investigated come in multiple flavors and touch on multiple subjects, both human and other.

There are physical benefits and psychological benefits, and many shades of each. Say a certain action took a week off my lifespan, but made me 5% happier emotionally for 6 months? Is that a good outcome, or a bad one? Let's say a certain action I took makes a child 5% happier for 6 months and takes a week off the lifespan of a stranger halfway around the world. What is the objective measure of that trade off?

I'll give you a real world example. PSA screening. There is some debate as to the best age to begin screening. Some experts say the current recommendation results in too large a rate of false positives which call for unneeded surgery and other treatments which can themselves risk death, impotence and extreme stress, all resulting in dramatically lowered quality of life. But for a few it makes the difference between life and death. How do we objectively balance the risk to those with false positives, with the risk to those who need to catch it early? Especially given that unneeded medical expense makes the system more expensive and contributes to people being unable to get health care?

As far as I'm aware, there is no objective way to balance one persons emotional pain with another's physical pain and yet another's lifespan. How much physical pain is equal to the grief of losing a child? How many years off your life is worth feeling physically great every day?

And all that is just among humans. Is a dog's pain worth anything? Why? What if an action caused 10 dogs to die horrible suffering deaths, but gave a man another year to spend with his family?

In order to come to empirical conclusions, we must have consistent measures, and morality deals in incompatible units.

Since your examples deal with issues of health, are you also of the opinion that we are unable to answer health questions using scientific investigation? What non-overlapping magisteria would you suggest serves this purpose instead?

Linda

 

[Paulhoff]

Part 3 of the show.



The whole show.



I have yet to hear a good reason that shows why science can't study the nature of humans and come to a conclusion on what humans seem to believe are moral and I’m sure that range is big but in no way is it infinite.

Paul

 

[AlBell]

Maybe if you're a philosopher and in the habit of looking at the world teleologically.

Linda

We are all philosophers one way or another, and until one decides (unscientifically) one ought to examine some question, nothing changes.

You appear to pretend this fact doesn't exist.

 

[fls]

We are all philosophers one way or another, and until one decides (unscientifically) one ought to examine some question, nothing changes.

Well, despite my appellation, I have come to the conclusion that I cannot be a philosopher because I don't seem interested in declarations of what cannot be done or what cannot be known. I am also grateful that scientists haven't listened to these declarations and have simply gone about the business of making quite wondrous discoveries by violating them.

You appear to pretend this fact doesn't exist.

More like, recognizing that this 'fact' doesn't exist gets me to knowledge that would otherwise be inaccessible to me.

Linda

 

[Dragoonster]

So what? Since we are attempting to answer questions about which actions are good or bad, characteristics which do not describe the universe, but describe a subjective characterization, your point seems entirely irrelevant.

Woah. You mean Harris' "science can answer moral questions" is entirely about subjective experience and subjective/axiomatic questions? So in this case science doesn't describe the Universe?

If good/bad aren't described by science, how can science answer good/bad questions?

 

[Dragoonster]

More like, recognizing that this 'fact' doesn't exist gets me to knowledge that would otherwise be inaccessible to me.

Linda

"when knowledge can't be gained by science, fake that it does so you can feel more secure/righteous in your morality"

 

[Dani]

The problem, for me, is that you are treating the establishment of goals as though it is an important and a separate step from asking the question in the first place. And that this somehow distinguishes the asking and answering of questions about moral behaviors from other scientific endeavors. Yet as far as I can tell (unless you relent and provide an example) we have no "scientifically established goals" in other fields of science. For example, we can study a drug for its ability to prevent mortality from mortality, even though the prevention of mortality from pneumonia is not a "scientifically established goal".

I'm asking why you think we need to establish goals in this way in the first place. And for you to provide an example of how this takes place in another field of science. Otherwise, it still looks to me like this is an arbitrary rule, made up to provide a justification for "non-overlapping magisterial" which serves religious apologists.

That's ironic, because I'm not the one claiming that science establishes goals. If science can answer moral questions in the first place, science can establish goals, since a moral statement is the establishment of a desired goal. I'm not sure if you understand my position, because you're actually supporting my point here.

So what? Since we are attempting to answer questions about which actions are good or bad, characteristics which do not describe the universe, but describe a subjective characterization, your point seems entirely irrelevant.

So science can give an answer that doesn't describe the universe, but instead is a subjective characterization? Science creates its own subjective characterization then? Science doesn't work this way. Precisely my point.

I'm talking about ideas about why actions are characterized as "good" or "bad" by subjects. If "logic" suggests that an action should be characterized as "good", yet it is chatacterized as "bad", it suggests that our idea needs modification. In particular, it's pretty clear that using logic applied to axioms provides a poor model for characterizing moral behaviors.

I don't know what's your point, but it's irrelevant since I'm not arguing about applying logic to axioms, but about the statement "science can answer moral questions". I'm not proposing any specific method of applying moral values, but refuting the former statement.

I'm referring to the practice which has been proposed here, where instead of empirically gathering information about why actions are characterized as good or bad, some sort of generally acceptable axiom is put forward and then an attempt is made to use logic to derive which actions should be characterized as good or bad.

I don't know about others, but I haven't proposed such thing.

So what? We are talking about something which is a subjective characterization. It may interest you to know that colour is also a subjective characterization, by the way. I thought we agreed that science can investigate and answer questions about subjective characterizations?

Well, that's your red herring. I'm not interested in arguing about subjective characterizations in general, which would be a much broader and distracting discussion. Color can be measured. Please provide evidence of objective distastefulness.

 

[Kuko 4000]

I hope Linda and Bokonon have persistence to continue this discussion.

I have privately asked few of the posters here to have a chat session with me about the subject, but no one seems to be bothered.