The Is-Ought Fallacy Fallacy

Neil

New
I'd be interested in opening up a discussion on the Is-Ought fallacy, since it seems to be often cited, yet I cannot see why it has any real force.

I think the is-ought fallacy is itself a fallacy for the following reasons:

1. Science itself cannot even arrive at an "is" without "oughts" since science is not value-free. We value different things like reason, parsimony, etc, so why cannot we value well-being, or things that are beneficial to our development?

2. It's odd that Hume's is-ought distinction gets credit as being a real issue but the same people will disregard the issue of induction. The thing is, logically, induction has no basis, yet that is how science works. It is also true that logically one cannot derive an ought from an is. But since when are we so concerned about logico-deductive bases for different branches of science? If one thinks the is-ought distinction means that there cannot be a science of morality, then they would also have to admit that science itself has no basis. Even mathematics uses induction, which has no logical basis. So why is the is-ought distinction given any real weight? That's not how science works.

It seems that this is-ought fallacy is something that doesn't really make sense and is used to hold back a science of morality.
 
Is-Ought, IMO, shows science cannot set the goals of a moral pursuit.

However, once the goals are set, one can utilize the methods of science to maximize those goals in some fashion. (It doesn't mean that only science will be used.)

On the question of induction...sure, science cannot deal in certainties and has to accept regularities over "natural laws". That's the whole issue of causality, it has no binding. But most people deal with regularities pretty well - there may not be a definite law that says I cannot run through traffic unharmed but I still can see there's little reason to risk it unless someone else is in danger?
 
Is-Ought, IMO, shows science cannot set the goals of a moral pursuit.

However, once the goals are set, one can utilize the methods of science to maximize those goals in some fashion. (It doesn't mean that only science will be used.)

On the question of induction...sure, science cannot deal in certainties and has to accept regularities over "natural laws". That's the whole issue of causality, it has no binding. But most people deal with regularities pretty well - there may not be a definite law that says I cannot run through traffic unharmed but I still can see there's little reason to risk it unless someone else is in danger?

In what way do you think the is-ought distinction says science cannot set the goals of moral pursuit?
 
In what way do you think the is-ought distinction says science cannot set the goals of moral pursuit?

Science is descriptive, not normative AFAICTell.

IIRC even Harris said you have to go with what he called the "Obvious Good" or some such, but then you're deciding what's good before you begin.
 
Science is descriptive, not normative AFAICTell.

IIRC even Harris said you have to go with what he called the "Obvious Good" or some such, but then you're deciding what's good before you begin.

But science is implicitly normative since it uses values in order to establish the descriptive statements. So how can we draw any meaningful distinction? Why would values contribute to descriptive statements, but descriptive statements couldn't contribute to normative statements? This seems especially an odd distinction in light on the methodological primacy of conscious experience and the relation of conscious experience to moral principles.
 
But science is implicitly normative since it uses values in order to establish the descriptive statements. So how can we draw any meaningful distinction? Why would values contribute to descriptive statements, but descriptive statements couldn't contribute to normative statements? This seems especially an odd distinction in light on the methodological primacy of conscious experience and the relation of conscious experience to moral principles.

Can you elaborate on the bold?
 
Can you elaborate on the bold?

Well, there is no logical foundation for science, philosophy of science, or of epistemology, and the same could be said of a science of morality. The example of the is-ought distinction relates to meta-ethical questions, like why should I value well-being over suffering? This extreme skeptical stance illustrates that there is no logical basis to bridge the is-ought distinction, but the exact same stance can be used on science itself. For example, why should I value certain types of thinking, or evidence, or qualities of theories such as simplicity, parsimony, etc? What logical basis could one provide for this? Indeed, Hume's other logical problem is that of induction, since logically you cannot derive a general statement from a statement of particulars. There is no logical basis for this, but that is what science does all the time. It formulates conjectures that must be compared to evidence, but never in a logico-deductive manner, but rather through plausible reasoning using values to determine what is corroborating evidence for a conjecture, what is a reasonable falsification of a conjecture, etc.
 
I haven't done ANY reading into this issue so I'm just going off on my own understanding of the is-ought fallacy from hearing about it from time to time.

The way I see it, the fallacy says that simply because something IS doesn't mean that is what we OUGHT. That's different than saying that the IS should never inform our view of the OUGHT. It just should not be considered automatic, and depends on context, weighing of different (often competing) values, goals, etc.
 
I haven't done ANY reading into this issue so I'm just going off on my own understanding of the is-ought fallacy from hearing about it from time to time.

The way I see it, the fallacy says that simply because something IS doesn't mean that is what we OUGHT. That's different than saying that the IS should never inform our view of the OUGHT. It just should not be considered automatic, and depends on context, weighing of different (often competing) values, goals, etc.

It seems that many contemporaries wish to use the is-ought distinction to prevent science from discussing values and morals in any normative sense.

I agree with you on the last part, but it seems the same is done with evaluating evidence and theories within hard sciences like physics.

For example, Sean Carroll says:

"Attempts to derive ought from is are like attempts to reach an odd number by adding together even numbers. If someone claims that they've done it, you don't have to check their math; you know they've made a mistake."

Yet Carroll believes that the Many Worlds interpretation of quantum theory to be true because of mathematical parsimony. Now, notwithstanding the mathematical problems such as discreteness and basis problems, and the failure to axiomitize MWI or mathematically define "measurement," the idea that I should think that mathematical parsimony is a good reason to think the interpretation to be true is a value judgment.

But my point is that you cannot have a one-way street between values and descriptive statements; if you use value judgments (normative statements) to evaluate and make claims of the truth of descriptive statements, then those descriptive statements can also be used to evaluate and make claims as to the truth of normative statements.
 
I've said most all I have to say on this issue in another thread, but I think you sum things up well, Neil. I'd challenge proponents of the so-called fallacy with this: if "oughts" don't and can't come from "ises", then from where do they originate? And what gives them their obligatory force?

By the way, Arouet, like you I haven't done any reading on this issue other than reference pages like Wikipedia's, but I'm going to be bold enough to suggest that you're misstating it. The supposed fallacy is not so much that "what is is what ought to be" but that one can reasonably start from descriptive statements (of what "is"), and, based on those, proceed in a reasonable or logical way to normative statements (of what "ought" to be (done/chosen)).
 
By the way, Arouet, like you I haven't done any reading on this issue other than reference pages like Wikipedia's, but I'm going to be bold enough to suggest that you're misstating it. The supposed fallacy is not so much that "what is is what ought to be" but that one can reasonably start from descriptive statements (of what "is"), and, based on those, proceed in a reasonable or logical way to normative statements (of what "ought" to be (done/chosen)).

If I've correctly understood you then I'm not sure that we're saying anything different. Mine was more pithy, yours more detailed. By "what is" I'm conceiving "what we identify the 'is' to be" and by ought "how we think we should act".

So it is: Should what we think the "is" is automatically (ie: logically necessitate) how we think we should act. I don't think it should be automatic. That said, "what is" should be one factor that we take into account, which must be balanced by other factors.
 
Arouet, your restatement didn't really clarify for me that we're saying the same thing, although conceivably we are. Best to be sure though. Maybe we can take a couple of examples.

  1. 'What "is" is that humans generally are kind to their families, therefore, humans generally ought to be kind to their families'. This is the sort of formulation I interpreted your statement of the fallacy to refer to i.e. where "what is" is the same as "what ought to be". I hope you can see that your restatement doesn't really clarify that this is not what you are saying. This is, as I understand it, a misinterpretation of that which the supposed "fallacy" consists in, although on second thought, it presumably/arguably (I am not informed enough to know) might be one way in which the "fallacy" might manifest more crudely. More accurately it would though, I think, be referred to as the "appeal to nature" fallacy.
  2. 'What "is" is that without early emotional bonding and physical touch, humans grow up lacking vital relationship abilities, therefore, we ought to ensure that babies get sufficient emotional bonding and physical touch'. This is more the sort of statement that I interpret the "fallacy" to apply to i.e. where "what is" is not identical with "what ought to be" but rather "leads to" or "implies" it in some way.

If you already see things the same way, then sorry for this post, I just wanted to be totally sure.
 
Arouet, your restatement didn't really clarify for me that we're saying the same thing, although conceivably we are. Best to be sure though. Maybe we can take a couple of examples.

  1. 'What "is" is that humans generally are kind to their families, therefore, humans generally ought to be kind to their families'. This is the sort of formulation I interpreted your statement of the fallacy to refer to i.e. where "what is" is the same as "what ought to be". I hope you can see that your restatement doesn't really clarify that this is not what you are saying. This is, as I understand it, a misinterpretation of that which the supposed "fallacy" consists in, although on second thought, it presumably/arguably (I am not informed enough to know) might be one way in which the "fallacy" might manifest more crudely. More accurately it would though, I think, be referred to as the "appeal to nature" fallacy.
  2. 'What "is" is that without early emotional bonding and physical touch, humans grow up lacking vital relationship abilities, therefore, we ought to ensure that babies get sufficient emotional bonding and physical touch'. This is more the sort of statement that I interpret the "fallacy" to apply to i.e. where "what is" is not identical with "what ought to be" but rather "leads to" or "implies" it in some way.

If you already see things the same way, then sorry for this post, I just wanted to be totally sure.

The appeal to nature fallacy was what I was getting at in terms of the fallacy.
I was thinking of it more in the arguments that are sometimes used to support social darwinism, or certain dietary habits.

The argument you present in 1 is fallacious if we're saying is because people are kind to their families they should be kind to their families. People also can be unkind to their families and but we wouldn't want to suggest that therefore they should be unkind.

Rather, humans have the capacity to be kind and unkind. We find that when people are kind, this has certain benefits that we deem good and so value it.

I'm not sure your number 2 represents fallacious reasoning or is an is/ought situation. What you're doing there, I think, is taking your ought (we value the development of social relationships) and suggesting a manner of achieving it (promoting physical touch with babies).

Whether this leaves us on the same page or not, I'm not sure!
 
I'm not sure your number 2 represents fallacious reasoning or is an is/ought situation. What you're doing there, I think, is taking your ought (we value the development of social relationships) and suggesting a manner of achieving it (promoting physical touch with babies).

OK, so you're saying that this is not really a case of getting to an "ought" from an "is" but of getting from one "ought" to another. Nevertheless, this is, I think, the sort of reasoning that is claimed to be fallacious in terms of the is-ought "fallacy", so perhaps you're tentatively agreeing that it is not in fact a fallacy?

Also, stepping back a little: would you extend your interpretation of this example to make the more general claim that we can only get from one "ought" to another? If so, how did we get to the very first, root "ought"? Or is there no "root" ought, and it's all an interconnected web of "oughts"? If so, how could there be any ultimate justification for any "ought", given that justifications would end up being circular?
 
OK, so you're saying that this is not really a case of getting to an "ought" from an "is" but of getting from one "ought" to another.

We have one ought: "value social relationships". Are we saying that we should classify one method of achieving that ought as an ought or is itself? To me that doesn't make sense but like I said, I haven't done the reading so maybe.

Nevertheless, this is, I think, the sort of reasoning that is claimed to be fallacious in terms of the is-ought "fallacy", so perhaps you're tentatively agreeing that it is not in fact a fallacy?


Right, I don't think there is any fallacious reasoning in 2 (noting of course that we're just setting out the argument simply, there are of course nuances that can come in.). As for whether that is an example that people raise in terms of is/ought I can't say.

Also, stepping back a little: would you extend your interpretation of this example to make the more general claim that we can only get from one "ought" to another? If so, how did we get to the very first, root "ought"? Or is there no "root" ought, and it's all an interconnected web of "oughts"? If so, how could there be any ultimate justification for any "ought", given that justifications would end up being circular?

As I don't tend to frame the issue of morality in is/ought terms I'm not sure (I tend to be wary of absolutes though.)

I posted my general approach to morality in previous threads that so I think you've seen this but I take a much more pragmatic approach to morality. I posit three subjective - but I would posit almost universally shared - basic human goals: that we aim to survive, thrive and be happy. We combine these goals with the proposition that humans are social animals - that is, we cannot generally achieve these goals on our own, we need other people. Pragmatic reasoning (not sure if that is a technical term) then starts to apply. We can examine various ways of achieving these goals and weigh the pros and cons based on experience to figure out the best way to live our lives.

Sometimes we might be able to reach conclusions using pure logical reasoning, but often that's not going to be practical.

The pragmatic approach I think will avoid circularity problems (that's not to say that it doesn't come with its own set of problems, but I think it more closely resembles how we actually operate in practice when it comes to morality.)
 
We have one ought: "value social relationships". Are we saying that we should classify one method of achieving that ought as an ought or is itself? To me that doesn't make sense but like I said, I haven't done the reading so maybe.

I'm interpreting your question assuming the "or is" is a typo, and omitting it, because I can't make sense of it as it is. In that case, my answer is yes. The general framing "If we ought to do X, then we ought to do Y" seems functional to me, where Y is necessary to achieve X, which applies in our current example (it is necessary to provide emotional bonding to babies if we want them to be capable of rewarding relationships as adults). One might argue that this is merely an instrumental (cf. your "method") ought, but I think that given its necessary relationship to the first, it actually inherits (at least some of) the moral weight of the first.

Right, I don't think there is any fallacious reasoning in 2 (noting of course that we're just setting out the argument simply, there are of course nuances that can come in.). As for whether that is an example that people raise in terms of is/ought I can't say.

I did some digging, and came across this article in PhilosophyNow: Hume on Is and Ought. Whilst some of the article is opaque to me, I think it justifies my view of how this "fallacy" is defined:

Hume’s idea seems to be that you cannot deduce moral conclusions, featuring moral words such as ‘ought’, from non-moral premises, that is premises from which the moral words are absent.

i.e. the fallacy is about more than merely the "appeal to nature"'s identification of that which "ought to be" with that which "is", it is more generally about "getting to" (moral) "oughts" from (non-moral) "ises". My example seems to fit just fine.

Interestingly, from the point of view of Neil's first post, it seems that, perhaps, Hume, the originator of the fallacy, would have accepted Neil's (and my) view that whilst you cannot derive an ought from a is via strictly deductive logic, it can be derived somehow:

[T]he Cambridge philosopher G.E. Moore (1874-1958) [...] thought it a fallacy, the ‘Naturalistic Fallacy’, to define the word ‘good’ in terms of anything else.

This was not Hume’s opinion. He was not denying that the moral can be defined in terms of the non-moral. He was merely denying the existence of logically valid arguments from the non-moral to the moral. This becomes clear once we note that Hume does not think that he has to argue for the apparent inconceivability of is/ought deductions. It is something he thinks he can take for granted. This is what we would expect if he were making the logical point since it would have been obvious to his readers. For it was a commonplace of Eighteenth Century logical theory that in a logically valid argument the matter of the conclusion – that is the non-logical content – is contained within the premises, and thus that you cannot get out what you haven’t put in. Thus if an ‘ought’ appears in the conclusion of an argument but not in the premises, the inference cannot be logically valid. You can’t deduce an ‘ought’ from an ‘is’ by means of logic alone.

Then again, Hume's derivation seems different to that which Neil (and myself) propose: Hume's is based on "sentiment. It defines virtue to be whatever mental action or quality gives to a spectator the pleasing sentiment of approbation; and vice the contrary", whereas Neil's (and mine) is based on "plausible reasoning" from the nature of sentient experience (emphasis in the original).

As I don't tend to frame the issue of morality in is/ought terms I'm not sure (I tend to be wary of absolutes though.)

I posted my general approach to morality in previous threads that so I think you've seen this but I take a much more pragmatic approach to morality. I posit three subjective - but I would posit almost universally shared - basic human goals: that we aim to survive, thrive and be happy. We combine these goals with the proposition that humans are social animals - that is, we cannot generally achieve these goals on our own, we need other people. Pragmatic reasoning (not sure if that is a technical term) then starts to apply. We can examine various ways of achieving these goals and weigh the pros and cons based on experience to figure out the best way to live our lives.

Sometimes we might be able to reach conclusions using pure logical reasoning, but often that's not going to be practical.

The pragmatic approach I think will avoid circularity problems (that's not to say that it doesn't come with its own set of problems, but I think it more closely resembles how we actually operate in practice when it comes to morality.)

That all seems to be very compatible with my own approach, Arouet, with the addition that I think that what you describe as our being motivated by "basic human goals" can be objectively framed in terms of an "ought". It's really just another way of framing things, a different semantic "thrust" if you like. You say "we (simply) have these goals"; I say, "given what we know about sentient experience, we ought to have these goals".
 
Last edited:
Two potentially relevant essays:

Massimo's critique of Harris' "Science of Morality".

...Harris’s chief claim throughout the book is that moral judgments are a kind of fact, and that as such they are amenable to scientific inquiry. First of all, the second statement does not at all follow from the first. Surely we can agree that the properties of triangles in Euclidean geometry are “facts,” in the sense that nobody who understands Euclidean geometry can opine that the sum of the angles in a triangle is not 180° and get away with it. But we do not use science, or any kind of empirical evidence at all, to arrive at agreement about such facts. At the very least, and without wanting to push an argument for moral realism, this makes the point that “facts” is too heterogeneous a category, and that Harris needs to be much more careful on how to handle it...

Esser's blog post in agreement with Harris (before the book was published)

Sam Harris has a short talk here on why he thinks "science can answer moral questions." A follow-up to some initial criticism is here, and a brief discussion of "getting an ought from an is" is here. He evidently has a book forthcoming related to all this.

While the philosophically more talented will have issues with his arguments (and with my brief discussion of this complex topic below), I agree with his main thrust. There are two key steps here: first, we include first-person experiences in our view of nature, and second, we locate the target for our (natural) moral instincts in the qualities of the experiences of sentient creatures.
 
Ah! Perfect! I have just "finalised" my own review of Sam's book. Now I get to dig into the thoughts of others and potentially "unfinalise and edit" my review before making a total fool of my uneducated self. Thanks, Sci! Digging in...
 
I'm interpreting your question assuming the "or is" is a typo, and omitting it, because I can't make sense of it as it is.

Heh, not as much a typo as horribly written! I'll try to clarify. I was distinguishing between an "ought" and the "process to achieve the ought". I was asking you you were saying that the "process to achieve the ought" is itself an "is" or an "ought". In my view, it isn't really either. We conclude that we ought to do that process only because it helps us achieve the "ought" of fostering social bonds. You got the gist I think of what I was getting at:

In that case, my answer is yes. The general framing "If we ought to do X, then we ought to do Y" seems functional to me, where Y is necessary to achieve X, which applies in our current example (it is necessary to provide emotional bonding to babies if we want them to be capable of rewarding relationships as adults). One might argue that this is merely an instrumental (cf. your "method") ought, but I think that given its necessary relationship to the first, it actually inherits (at least some of) the moral weight of the first.

I agree it inherits the moral weight of the first, which is why I don't find it useful to classify it as an ought in its own right!

I did some digging, and came across this article in PhilosophyNow: Hume on Is and Ought. Whilst some of the article is opaque to me, I think it justifies my view of how this "fallacy" is defined:

I think I'm starting to get what he's saying. If we start from the proposition "early emotional bonding and physical touch helps humans develop X relationship abilities" we can't get from that to "we should therefore promote early emotional bonding" without adding some premises:

  • Early emotional bonding and physical touch helps humans develop X relationship abilities"
  • X relationship abilities help humans to a, b, and c.
  • A, b, and c are desirable results.
  • Therefore we should promote early emotional bonding.
Hume seems to be arguing that the desirability is an emotional rather than logical reaction. I can sort of see what he's getting at. The author suggests that this suggests there is no ultimate objective morality - if that's the case then I think I agree because I don't ascribe to a purely objective morality myself. That said, once you accept the good/ought I think you can use logic and reasoned argument to figure out good ways to achieve that good/ought.

I'd have to spend more time with it to really figure it out but my early conclusions are in line with the author's where he writes: "Why then is No-Ought-From-Is important? The answer is that it isn’t asimportant as many philosophers take it to be."

I get the sense that this is similar to the arguments on causality that we were discussing in another thread: it seems that these kind of technical arguments really only apply if we're looking for absolute justification, 100% certainty, and the like. In practice, I don't think we think about morality or causality in such technical terms. At the end of the day we have to ask: Ok, so what turns on this? How should this influence how we think about morality (or causation). How does it let us make better decisions? How should we employ it in practice?

We can identify clear issues, in my opinion, where concluding an ought from an is results in what I would argue is a bad decision (ie: social darwinism). Does that mean we can never get oughts from ises? I dunno, let's just deal with each case on its own terms and make the best decisions we can!

i.e. the fallacy is about more than merely the "appeal to nature"'s identification of that which "ought to be" with that which "is", it is more generally about "getting to" (moral) "oughts" from (non-moral) "ises". My example seems to fit just fine.

Interestingly, from the point of view of Neil's first post, it seems that, perhaps, Hume, the originator of the fallacy, would have accepted Neil's (and my) view that whilst you cannot derive an ought from a is via strictly deductive logic, it can be derived somehow:



Then again, Hume's derivation seems different to that which Neil (and myself) propose: Hume's is based on "sentiment. It defines virtue to be whatever mental action or quality gives to a spectator the pleasing sentiment of approbation; and vice the contrary", whereas Neil's (and mine) is based on "plausible reasoning" from the nature of sentient experience (emphasis in the original).

I wasn't familiar with the term "plausible reasoning" prior to hearing you guys mention it. Taking a quick look at the wiki it seems pretty standard, so I don't take any issue with it.

That all seems to be very compatible with my own approach, Arouet, with the addition that I think that what you describe as our being motivated by "basic human goals" can be objectively framed in terms of an "ought". It's really just another way of framing things, a different semantic "thrust" if you like. You say "we (simply) have these goals"; I say, "given what we know about sentient experience, we ought to have these goals".

I think the vast majority share these goals pretty instinctively but we can probably argue that we should have them as well.
 
Laird: to be clear my disagreements in the above post are technical but I don't get the sense we disagree on anything significant here in terms of practical application. Let me know if you think otherwise!
 
Back
Top