The Emperor of Fallacies: The Is/Ought Problem

It’s nearly impossible to understand contemporary ethics without understanding the word ‘normative’ as it’s used in philosophy. Since this post is mostly about a kind of argument in ethics, I’ll get the definition out of the way. 

Perhaps the easiest way to understand normative statements is to contrast them with descriptive statements. “She hit the dog,” is a descriptive statement. There’s no judgment–implied or otherwise–of good/bad/right/wrong. Contrast this with “she shouldn’t hit the dog” or “it’s wrong for her to hit the dog.” The latter two are normative statements. They make value judgements about actions or states of affairs.

You can think of ‘normative’ as having to do with judgments of value such as good, bad, right, or wrong. Any assertion that takes the form ‘X is good’, ‘X is bad,’ etc… is a normative statement. Normative statements can also take the form ‘you ought/ought not to do that’. ‘Ought’ statements imply normativity because if I ought to do something, presumably it’s because it’s a good/right thing to do. And if I ought not to do something, presumably it’s because it’s a bad/wrong thing to do.

With the definitions out of the way, let’s look at some normative arguments, shall we?

1. A survey of Western history, anthropology, and current society reveals that, for the most part, men occupy positions of political, economic, and academic power. Women, on the other hand, are typically in charge of raising children, cooking the meals, and generally attending to her husband’s needs. It follows that this is the right way to organize households and society.

2. Across time and across cultures, marriage is between a man and a woman. It follows that couples of the same sex shouldn’t be allowed to get gay married.  

3.The human sex organs evolved for male-female intercourse. That is a scientific fact you can’t deny. It follows that male-male or female-female intercourse is morally wrong. 

I’m going to go out on a limb and guess that most people reading this blog reject the above arguments. Saying you disagree isn’t enough. In philosophy, we’re very interested in why arguments fail. Although they appear different, all three arguments fail for the same reason. Let’s evaluate them and figure out that reason.

Argument 1 concludes that the right way to organize society is according to traditional gender roles because that’s how it was typically organized. For the sake of argument, let’s assume that the premises are true: i.e., that history and anthropology do in fact support the view that societies, institutions, and power relations are typically organized according to ‘traditional’ gender roles.

Question: Does knowing how power, prestige, and work were allocated in the past tell us the right way to allocate those elements now? Not necessarily. Sure, it’s possible that 1950s America stumbled on the one true way to organize society but it’s at least as likely that there was also room for improvement.

Question: Does knowing how power, prestige, and work are allocated now tell us the right way to allocate those elements now? That would be weird. It would imply that whatever the social arrangement at a particular point in history, that is also the most just social arrangement. There would be no grounds to criticize any social practice or institution.

At the most general level, Argument 1 fails because it draws conclusions about what ought to the case from what was or is the case. Facts about how the world is (or was) don’t tell us how the world should be. Arguing from a description of how things are to a conclusion about how they ought to be is called the Is/Ought Fallacy (also sometimes called the Is/Ought Problem or Hume’s Law).

What’s the Problem?

The Is/Ought Problem comes from David Hume. He argued that the above general pattern of reasoning–moving to a normative conclusion from descriptive premises–is a fallacious form of reasoning. Facts that describe social, institutional, and interpersonal relations can’t on their own support claims for how those relations ought to be. And it’s a problem because, “every system of morality, which [Hume had] hitherto met with” commits this error [my italics]. The is/ought problem is a huge problem for moral theory if every system of morality commits it.

I’m gonna make you guys work a bit. Below Hume lays out the nature of the is/ought problem. Read it then I’m going to walk you through it.

In every system of morality, which I have hitherto met with, I have always remarked, that the author proceeds for some time in the ordinary way of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when of a sudden I am surprised to find, that instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible; but is, however, of the last consequence. For as this ought, or ought not, expresses some new relation or affirmation, ’tis necessary that it should be observed and explained; and at the same time that a reason should be given, for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it. But as authors do not commonly use this precaution, I shall presume to recommend it to the readers; and am persuaded, that this small attention would subvert all the vulgar systems of morality, and let us see, that the distinction of vice and virtue is not founded merely on the relations of objects, nor is perceived by reason. (A Treatise of Human Nature)

Let’s use Argument 3 to see what Hume’s talking about in the first sentence. Argument 3 concludes that homosexual acts are wrong (i.e., normative conclusion) from the descriptive premise “human sexual organs evolved for male-female intercourse.”

To make things even more clear I’m going to put the argument into premise-conclusion form:

P1: Human sexual organs evolved for male-female intercourse. (Descriptive)
C:  Therefore, same sex intercourse is morally wrong. (Normative)

The argument begins by describing facts about the world; that is, by describing how the world is. The conclusion, however, refers to a value judgment. Hume points out, you cannot make a deductive jump from how the world is to whether that aspect of the world is good or bad.

This is what Hume means in the first part of the above quote. He’ll be reading some descriptive account of how humans behave, how God made the world, etc…then ‘of a sudden’ the author will jump to a normative conclusion about those descriptive facts. But knowing how the world is doesn’t tell us anything about how it should be. Sometimes the way the world is is bad and sometimes it’s good. Sometimes what people do is right and sometimes what they do is wrong. Merely knowing how the world is doesn’t allow us to draw conclusions about how we ought to behave in it.

To move deductively from descriptive premises to a normative conclusion you need what’s called a linking premise. A linking premise links descriptive facts to normative judgments:

P2: If something is used in a way other than its evolutionary purpose then doing so is morally wrong.

With the linking premise we have a valid deductive argument:

P1: Human sexual organs evolved for male-female intercourse. (Descriptive)
P2: If something is used in a way other than its evolutionary purpose then it is morally wrong. (Linking)
C:  [Same sex sex uses organs in ways other than for their ‘evolved’ purpose,] therefore, same sex sex is morally wrong. (Normative)

The requirement for a premise linking the descriptive to the normative is what Hume refers to when he writes:

For as this ought, or ought not, expresses some new relation or affirmation, ’tis necessary that it should be observed and explained; and at the same time that a reason should be given, for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it [my italics].

He means that in order to move from a claim about how the world is to a conclusion about the moral goodness or badness of the way the world is, we must justify with a reason why the way the world being a particular way is also a morally good/bad thing or why it justifies what we ought/ought not to do. The heart of Hume’s point is a demand for a justifying reason that logically connects how the world is to how we ought to act in it.

By adding the linking premise (P2) to Argument 3 we meet this requirement. The above argument with the linking premise is now valid (it has good form)–but is it sound (i.e., are the premises true)?

P2 is extremely suspect. Counterexamples are pretty easy to come by. Humans evolved to walk on two legs but no one would think it’s morally wrong to walk on their hands. The human skull evolved to protect the brain but few people would think it’s morally wrong to use it to redirect an air-born soccer ball. We evolved to live in small tribal communities but few would argue that from this it follows that it’s morally wrong to live in a metropolitan city. It would be an odd morality that prohibited the use of your life and limbs from all behaviors except those for which they evolved on the African savannah.

When we draw out the implied linked premise we see that it’s not easily defended and admits easy counterexamples.

At this point you might be thinking, “duh, but Argument 3 is an obviously bad argument.” However, I submit to you that it’s only obviously bad because you disagree with the conclusion. When we disagree with conclusions we’re fairly good at identifying weak arguments. However, when we agree with conclusions we’re often blind to the weakness of the reasons supporting our view. Hume’s point is that this error–not providing or defending a linking premise–is everywhere in ethical thinking:

But as authors do not commonly use this precaution, I shall presume to recommend it to the readers; and am persuaded, that this small attention would subvert all the vulgar systems of morality,

Here’s an example you supporters of Obama’s homosexual Muslim agenda have probably made:

People don’t choose their sexual preferences. Gay people are born gay and straight people are born straight. Therefore, gay sex isn’t immoral. 

Notice the pattern of reasoning:

P1. People don’t choose their sexual preferences. (Descriptive)
P2. Gay people are born gay and straight people are born straight. (Descriptive)
C.  Therefore, gay sex isn’t immoral. (Normative)

Here you might agree with the conclusion but the argument commits the is/ought fallacy. For the argument to go through you’d have to defend the view that being born with a particular preference doesn’t on its own imply that acting on those urges is immoral. This is clearly false. Some unfortunate individuals are born with pedophilic desires. That fact that they’re born with these desires doesn’t exempt acting on those desires from moral scrutiny.

Again, descriptive facts about the world don’t on their own imply normative conclusions. If we want to defend the moral neutrality of homosexuality we need to offer reasons that are not mere descriptions of what dispositions people are born with. One alternative is to offer a general moral principle like the harm principle: If a behavior doesn’t harm others then, all else being equal, it’s morally neutral. Now we have an argument that doesn’t depend on mere descriptions of the world.

When to be Alert and Various Subspecies of the Is/Ought Fallacy
The is/ought fallacy comes in many varieties which I’ll cover below. But first I want offer an hypothesis for why and when it’s common to commit the is/ought fallacy. When practices, institutions, and beliefs are entrenched in a society or group their human origins are invisible. Entrenched social arrangements and practices are perceived to be ‘the natural order of things.’

As such, there’s a failure to recognize that these practices, institutions, and beliefs are the product of a combination of habituation, deliberate action, and power relations–amongst other things. Interests are also at stake, especially when one group stands to benefit disproportionately from one arrangement rather than another. As a result the socially constructed nature of institutions, practices, and beliefs is often not only invisible but there are often efforts to avoid, dismiss, and suppress criticism.

History is full of examples but I’ll illustrate with the most obvious. For us the practice of slavery is abhorrent. The fact that it was practiced for thousands of years up until recently in our own country is inconceivable to us. More baffling still is that people not only passively accepted it but many people–throughout world and American history–vigorously defended it.  And guess how it was defended? Usually by committing the is/ought fallacy:

Some variation of

  1. It’s natural for some people to be slaves.
  2. It’s the natural order of the universe for some people to serve others.
  3. Slavery has always existed through human history.
  4. The law says that slavery is allowed.
  5. It’s God’s will.

Appeal to Nature/Naturalistic Fallacy
The first two we might call appeal to nature or naturalistic fallacy. Even if we grant that something is ‘the natural order’ (rather than a social construction) it still doesn’t follow without further argument that that practice is good (or that not doing it is bad).

P1. It’s the natural order of things that some people serve others. (Descriptive)
P2. If a behavior, practice, or arrangement is natural then it is good. (Linking).
P3. Therefore, slavery is good. (Normative)

Notice that even if we’re super charitable and grant P1, the argument still won’t work since it also needs P2 to be true. But P2 is false: It’s natural for humans to go to war, rape, and pillage, and generally to act shitty to each other but that doesn’t make it good nor does it follow that we should do it.

The naturalistic fallacy was/is often used to justify traditional gender roles, withholding the right to vote, hold property, pursue an education, etc…

Knowing what is natural doesn’t tell us what is good or what we ought to do. 

Anytime you hear someone defend a practice, behavior, or social arrangement by arguing that it’s the natural order, poke your finger right in their chest and shout, “it’s natural for people to commit the is/ought fallacy but that doesn’t make it a good argument!”

Fancy Pants Note: (Arguably) G.E. Moore pointed out another kind of naturalistic fallacy but I’ll set that aside for now. My own view is that it’s also a subspecies of the is/ought fallacy.

Appeal to Tradition
Appeal to tradition is also a common subspecies of the is/ought fallacy. The fact that slavery has persisted through human history doesn’t tell us anything about its moral status as a practice. The same goes for any historical practice. The argument from tradition also does a lot of heavy lifting when it comes to defending traditional gender roles. As you now know from being in Obama’s reeducation camps, the fact that women and men traditionally occupied particular roles tells us nothing about whether those were good arrangements or that we ought to adopt/preserve those arrangements.

What was doesn’t tell us what is good or what we ought to do. 

Anytime you here someone saying, “yeah but, that’s the way it’s been done throughout history,” turn your hand into a blade and shout, “oh yeah? In my country it’s tradition to judo chop people who make bad arguments so I guess it’s right for me to chop you.” That’ll show ’em.

Appeal to the Law Fallacy
You can’t justify a practice’s moral value by merely appealing to the law of the land. Laws can be just or unjust. The fact that something is allowed (or prohibited) by a current law tells us nothing about whether it that practice is just or unjust. Obviously, the fact that slavery was permitted by the laws of the day doesn’t allow us to conclude that slavery was therefore morally good.

Similarly, the fact that a law prohibits an act; for example, smoking the pot, doesn’t allow us to conclude that it’s morally bad to smoke the pot.

Appeal to the law fallacy is a subspecies of appeal to authority. And you need to respect my authoritah on this cuz I’m a philosopher:

Knowing what the law is doesn’t tell us what is just it only tells us what the current law is. 

Anytime you hear someone say, “well if it’s morally wrong then why is it legal?” or “if it isn’t immoral then why is there a law against it?” tell them that they just violated Hume’s law and send them these links:

Appeal to God/Bible/Religion
This one might be a bit controversial but I’m going to include it anyway. It goes without saying that holy texts and religion have been used to justify pretty much everything under the sun–both good and evil. Appeal to the Bible was one of, if not the most common justification for slavery (and subjugation of women and demonization of gays and justification of Jim Crow laws and etc…). 

Appeal to religion can take many forms:

  • ‘Cuz it says so in (my One True interpretation of) the Bible.
  • ‘Cuz God designed the universe that way.
  • ‘Cuz God said so.
In the first case, we have a hybrid between argument from authority and argument from tradition. The undefended linking premise would be “if (my interpretation of) the Bible says we ought/ought not to do X then we ought/ought not to do x.” The Bible, like many holy texts, contains plenty of wise moral prescriptions but there are also a great many awful ones.  

However, the good moral prescriptions aren’t good because they just happen to be found in the Bible; it’s not their source that makes them good moral advice. They’re good for independent reasons. For example, the Bible recommends we don’t kill others. 

So deep. 

Truly the product of divine inspiration. 

Surely no human could have come up with this. 

But the prohibition on murdering is not a good moral prescription because the Bible said so. Presumably, there are good independent reasons not to murder even if those words were never written into the Bible

Just as the Bible can offer sage advice it can also be morally bankrupt. Famously, the Bible doesn’t condemn slavery. Infamously, it provides guidelines for purchase and treatment of slaves:

However, you may purchase male or female slaves from among the foreigners who live among you. You may also purchase the children of such resident foreigners, including those who have been born in your land. You may treat them as your property, passing them on to your children as a permanent inheritance. You may treat your slaves like this, but the people of Israel, your relatives, must never be treated this way.(Leviticus 25:44-46 NLT)

Knowing what a holy text says doesn’t on its own tell you what is good or what you ought to do since holy texts contain both good and abhorrent moral advice. Anytime someone says “well, in the Holy book it says…” just find a contradictory passage or interpretation in the same book. You’ll always find one…

Knowing what is written in a book doesn’t tell us what is good or what we ought to do.

The appeal to design is just another version of the ‘natural order of things.’

The “‘cuz God said so” is just like appeal to a holy text (setting difficult epistemological issues aside).

What Was Hume’s Point?
There’s some debate over exactly what Hume was up to in pointing out the is/ought fallacy. Some people interpret Hume as a skeptic with regard to the justifiability of moral claims. If all knowledge comes from observation of the world and observational statements about how the world is can’t justify moral claims then moral claims can never be true (or false).

The standard interpretation, however, is not skepticism but that justifying a normative claim requires reasons in addition to descriptive claims.

So how do we justify moral claims? Unless we have some source of knowledge besides observation, the is/ought problem implies that moral claims can never be justified. So, what is this other source of knowledge?

Classical philosophers along with many of Hume’s contemporaries thought that capital ‘R’ Reason could tell us what we ought/ought not to do or value. But Hume famously argued that reason “is and ought only to be the slave to the passions.” Reason can only tell us how to achieve particular ends or values, it can’t tell us which ends or values to pursue and realize. That is the job of the passions.

There are two related issues floating around: The first is meta-ethical: what is the source of value? Second is epistemological: How do we know what to value?

The is/ought fallacy shows that we can’t discover values by scientific observation of the world. Science can only tell us what is. It can never tell us what to value or what we ought to do about how we know the world to be. Take what seems to be a straight forward case: People who exercise regularly tend to be happier. For most people it seems like a straightforward inference to “therefore, people ought to exercise.” But notice that lurking in the background is an normative assumption about happiness: that we ought to pursue it.

It may seem trivial but it’s consistent with what Hume is saying. You can’t move immediately from descriptive claims to normative conclusions. You always need a premise linking the descriptive claim to the normative claim. In this case: If something makes us happy then we ought to do it. Again, a little reflection reveals that this isn’t obviously true in all cases. It doesn’t mean that we can’t come up with a different linking premise. Regardless, Hume’s point stands that we can’t move deductively from purely descriptive to normative claims. We must always provide additional (non-descriptive) reasons that link descriptive facts to normative judgments.

So, if science (i.e., observation of the world) can’t tell us what to value, what does? As I mentioned above, Hume also rules out Reason. Reason can only tell us the most efficient ways to achieve our values.  So what’s left? If value isn’t “in” the world (and therefore not discoverable by science) where does value come from? The skeptical interpretation is that since value isn’t discoverable by science it doesn’t exist in any objective sense and is (gasp!) unscientific. Thus, normative claims can never be justified as objectively true. But Hume suggests another source for value: Our sentiments.

Our emotions of approval or disapproval ground value. Wait. So this fancy philosopher is telling me that if I like something it’s good and if I dislike it it’s bad? That’s dumb. That’s the answer of a three year old.

As you might expect, Hume’s view is a bit more nuanced. The short version is that those traits of character (or actions that exemplify those traits) that we think are ‘good’ are those that we would approve of. But we have to be careful here because I might approve of someone’s action merely because it benefits me. We shouldn’t conclude from this that the action is therefore good. Hume marks a distinction between ‘interested’ and ‘disinterested’ emotions.

Interested emotions are those that I feel when I’m appraising an act/trait from the personal point of view. Disinterested emotions are how I feel about an act/trait from an impartial point of view. When we evaluate an action or trait we must strive to remove our personal interests and affinities toward the agent contemplate the act from“some common point of view, from which [we] might survey [our] object, and which might cause it to appear the same to [everyone].” In other words, when we assess the moral qualities of a behavior we carefully introspect on how feel about it from an impartial point of view that can be shared by others. The fact that an act would garner shared approval/disapproval grounds that act’s moral properties. If there is widespread approbation from this impartial point of view, then the act/trait is good. The converse is also true.

And so moral knowledge is not grounded in observations of the world, nor is it grounded in Reason. Moral knowledge and knowledge of value come to us through careful introspection of our moral sentiments from an impartial point of view.

Comprehension Test

At the beginning of the previous section I suggested that the is/ought fallacy is most commonly committed when the socially constructed nature of institutions, beliefs, and practices are invisible to the arguer. These institutions, beliefs, and practices are often “extra” invisible to our criticism when we benefit from them since we actively resist evidence and arguments to the contrary.

The testimony of slaves regarding their plight and equal humanity was dismissed as unimportant or plain ridiculous. It was even argued that slavery was in their own best interest. Women’s arguments that they had just as much desire and capacity to participate in political and academic life were often waved off. And today still, the experiences of oppressed groups are trivialized, dismissed as mere whining or as unpatriotic: DON’T YOU KNOW HOW LUCKY YOU ARE JUST TO LIVE IN AMERICA? SHUT UP AND BE GRATEFUL!!!

So here’s a test to see how well I’ve been able to explain the is/ought fallacy. Run through all the arguments you can think of for why it’s morally permissible to raise animals and kill them prematurely for our dining pleasure. How many subspecies of the is/ought fallacy can you find?

Expert Challenge: Here is the argument most commonly overlooked as an instance of the is/ought fallacy:

It’s OK to raise and kill animals for food because we’re human and they’re not.
It’s OK because humans and animals are different.

Can you identify why this is an instance of the is/ought fallacy? (I’ll post some answers at the end).


Loose Ends
A few philosophers deny Hume’s view that facts about the world can’t on their own justify normative conclusions. If you’re interested in this view google “ethical naturalism.”

4 thoughts on “The Emperor of Fallacies: The Is/Ought Problem

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s