Can science lead to moral knowledge? If so, moral naturalism is true.
Naturalism is the philosopher’s jargon for saying “based on natural science.” “Moral naturalism” is the view that morality is part of the reality studied by science (physical reality) and can be known by science, but “moral naturalism” has more specifically become jargon for the view that there are moral facts and they can all be studied by science.1 “Empirical” knowledge (or justification) is knowledge attained through observation and experimentation (the scientific method).2 Naturalism is almost synonymous with “empiricism,” which is the view that we can know everything from observation—and “moral empiricists” would think that all moral knowledge is attained through observation and experimentation.
If moral naturalism is true, then there are moral facts—morality exists as part of the world independently of our opinions. Morality wouldn’t merely be a human fabrication, social contract, or “cultural custom.” Additionally, we would have a way to know about moral facts, and it is quite a reliable method based on our experience with scientific progress so far.
If moral naturalism is true, then there are various questions that need to be answered, which are often taken to be objections to naturalism:
- How do we know when we identify that a moral fact is identical to a nonmoral fact? Whenever we reduce moral facts to nonmoral facts, we should ask, “But what should I do?” This is known as the open question argument.
- How can science give us moral knowledge? Science can only give us facts about objects rather than values—and these are two totally different things. “You can’t get an ‘ought’ from what ‘is.’”
- How can we observe moral facts? Perhaps our observations can be explained without appealing to moral facts.
- Do moral facts explain anything or cause anything to happen? If moral facts are natural, then would everything in the natural world be exactly the same even if moral facts don’t exist? If moral facts are natural, then they might be useless. Morality itself would have no influence or power over us.
I will discuss these two sorts of moral naturalism, the above questions, and briefly discuss how moral naturalism might relate to Sam Harris’s The Moral Landscape.
Two Sorts of Moral Naturalism
Reductionism states that moral facts and properties are nothing other than certain nonmoral facts and properties. For example, “good” might mean “pleasure, and “right” might refer to “whatever produces the most pleasure and the least pain.”
Given reductionism, we can know that an act is right or wrong based on our observations. For example, we could see that torturing a child is wrong because it causes severe pain when better alternate courses of action are possible.
Reductionism has been greatly successful in science so far, so we could suspect that it could be successful for morality for that same reason. For example, H2O is water. We found out that what we call “water” was really H2O.
“Moral naturalism” is often equated with “reductionism” because that has been the historical position within philosophy, but they are not equivalent. Consider that thoughts don’t seem to be equivalent to brain states. Nonetheless, thoughts are “natural” and we study them within natural science (psychology).
Non-reductionism or “emergentism” is the view that moral facts and properties depend on non-moral facts but are not identical to them. It could be that “badness” exists (in part) because pain exists. Whenever there is pain, something bad is happening to someone. However, “bad” doesn’t refer to pain itself. This is much like how thoughts seem to exist (in part) because of brain states, but the word “thought” doesn’t refer to brain states.
Non-reductionism is a view about reality rather than knowledge. Moral non-reductionism is the view that moral facts are “more than the sum of their parts,” and are irreducible to nonmoral facts. Moral non-reductionism doesn’t tell us how we can know about moral facts or justify them.
Some people equate “non-reductionism” with “intuitionism” and “non-naturalism” because all three views have been unified in the past in philosophy. However, these are all different positions.
Intuitionism is the view that we can know moral facts via “intuition” or “noninferential reasoning” or “reflection.” Intuitionism doesn’t tell us whether moral facts are natural (physical) parts of the world, and it does not tell us that moral facts can’t be known though observation.
Non-naturalism takes two forms: One view tells us that facts are part of the reality studied by science and another tells us that facts are knowable by using science.
What I call “non-reductionism” here is actually “naturalistic non-reductionism” because I am only concerned with naturalistic knowledge here—the view that we can know moral facts through observation. Naturalism, non-reductionism, and intuitionism could all be compatible.
1. How do we know when we identify that a moral fact is identical to a nonmoral fact?
It’s not entirely clear how we know that “badness” is nothing but “pain” or the other way around. How could we know such a thing? I don’t have a good answer for this question, but there must be some way that we can know that H2O is nothing but water. Is it even possible that “badness” is nothing but “pain?” I don’t think we can rule it out.
G.E. Moore’s open question argument illustrates how problematic naturalistic reductions (or identity relations) are by saying that whenever someone gives an identity relation X (of goodness, badness, rightness, etc.), it makes sense to ask, “This is X, but is it good?” For example, someone could say that “badness is pain.” In that case we could ask after being burned severely, “This burning feeling is pain, but is it bad?”3
Moore’s open question argument can illustrate our uncertainty—it’s not obvious that “pain” is identical to “bad,” but many people seem to think that the open question argument proves such identity relations to be impossible. This is absurd because the “argument” would even apply to ordinary naturalistic reductions, such as “H2O is water.” Someone could ask, “Yes, this is water in my glass, but is it H2O?” We would simply think that the person knows very little about water for asking such a question.
I agree that we simply don’t know that “badness” is nothing but “pain” and I suspect that there is no good identity relation for moral facts. However, that doesn’t mean that naturalism is probably false because non-reductionism is another option.
2. How can science give us moral knowledge?
In particular, isn’t it impossible for science to tell us what ought to be the case because it can only tell us what is the case? Isn’t what ought to be and what is two totally different things—facts and values? Isn’t it true that you can’t get “ought” from “is?”
One simple answer is that there are moral facts. Facts and values aren’t two different things. Values are a subclass of fact. In particular, pain does seem to be intrinsically bad (bad just for existing). It is a fact that pain exists, so it’s a fact that something bad exists. Therefore, facts about what is the case can tell us something about morality.
Moreover, what ought to be the case seems to relate to intrinsic values. All else equal, it is right to help people or produce something good (such as pleasure), and it is wrong to hurt people or produce something bad (such as pain).
3. How can we observe moral facts?
We can see that someone is doing something wrong when they torture a child, but perhaps that “observation” is nothing other than our own prejudice, cultural indoctrination, emotional response, etc. In that case our observations would be exactly the same even if there are no moral facts. Morality could be nothing other than a cultural tradition.
Although it is possible that moral observation is delusional, it is also possible that nonmoral observation is delusional. We could be dreaming right now, or our minds might be controlled by machines, etc. This sort of skepticism doesn’t seem plausible concerning nonmoral observations, but do we have any reason to doubt that moral observation is based on moral reality?
I think that there is at least one case where moral observation is very plausible—when we experience intense pain. We observe that such experiences are bad. Such an observation seems like it can’t be delusional because even a hallucination of pain would be bad. Delusions and hallucinations are deceptive, but pain can’t be deceptive. We directly experience the reality of pain, but deceptive experiences trick us into thinking something is real or true. (For example, I can dream that I see my hand. I wouldn’t really be seeing my hand, but I would be tricked into thinking I am.)
4. Do moral facts explain anything?
If moral facts are identical with nonmoral facts, then they would explain something because the nonmoral facts explain something. For example, people would want to avoid pain—and “bad” might be nothing but “pain”—so the fact that something is “bad” would explain why we want to avoid it insofar as we want to avoid pain. (The moral property “bad” would cause us to want to avoid touching fire insofar as “bad” is “pain.”) Still, we might wonder if saying pain is “bad” adds anything to the universe. Would everything be the same if “bad” wasn’t “pain?” Does the fact that something is bad explain anything insofar as it is bad (apart from the fact that pain is involved)?
I find these questions troubling for the reductionist, but not for the non-reductionist. The non-reductionist can make it clear that the “badness” of pain explains a lot—we wouldn’t care about pain if it wasn’t bad. In fact, if nothing is bad, then pain couldn’t exist. Therefore a world with no moral facts would be devoid of pain and things would be quite different. (Badness causes things to happen insofar as it is part of the meaning of “pain” and we want to avoid pain precisely because it feels bad.)
What about Sam Harris’s The Moral Landscape?
Sam Harris claims that we can attain moral knowledge from science—but he defines “science” as “rational study.”4 Therefore, Harris isn’t saying that moral naturalism is true. There might be ways other than observation for knowing about certain moral facts. However, moral naturalism, if true, would vindicate Harris’s book. It would not only be possible to know moral facts through reason, but it would also be possible for there to be a moral science similar to psychology and other social sciences.
Non-reductionism is not only compatible with naturalism, but it is more intuitive and initially plausible. The questions that are supposed to be troubling for a naturalist are actually only troubling for the reductionist. Additionally, naturalistic non-reductionism is quite intuitive and is initially plausible given our experiences and the powerful replies it can offer to so-called objections presented against naturalism.
Many people want there to be “evidence” of moral facts and argue that moral facts probably don’t exist given the lack of positive evidence. However, there is evidence of moral facts. For example, we experience pain as bad. The potential fact that “pain is bad” is a highly plausible truism that few people would disagree with. Those who deny that moral facts exist still refuse to be tortured to prove their point. Instead, they often try to explain why they “hate pain” without there being any “moral facts.” I don’t find any such position to be plausible. I have considered such alternatives in my essay, An Argument for Moral Realism.
- For more information about moral observation and naturalistic non-reductionism, you might want to read my review of Moral Realism by Torbjörn Tännsjö.
- For more information about moral facts, you might want to read my discussion, What are Moral Facts?
1 The belief in moral facts—that morality is part of reality itself and not merely a human fabrication—is also known as moral realism. “Moral naturalism” usually refers to a sort of moral realism, but it is possible to be a naturalist and reject moral realism.
2 I will often refer to “knowledge” when the word “justification” might be more appropriate. It is possible to have justified moral beliefs without having “moral knowledge” because justified beliefs—though reasonable—could be false.
3 In this case we might actually think it’s impossible to think that pain isn’t “bad” despite the fact that we do have a good reason to reject that “bad” is nothing but “pain.”
4 “Some of my critics got off the train before it even left the station, by defining “science” in exceedingly narrow terms. Many think that science is synonymous with mathematical modeling, or with immediate access to experimental data. However, this is to mistake science for a few of its tools. Science simply represents our best effort to understand what is going on in this universe, and the boundary between it and the rest of rational thought cannot always be drawn” (Moral Confusion in the Name of “Science”, March 29, 2010.)