Geoffrey Sayre-McCord argues that we can confirm moral facts through observation, and that moral facts can be confirmed in a meaningful way. He admits that there is still some room for doubt. In order to justify moral facts, he takes a close look at epistemology in general. He suggests that theories must be able to explain our observations better than alternatives. In order to do this pragmatic considerations seem relevant, and if so, moral theories could be justified.
In the final section of Sayre-McCord’s article, he suggests a strategy to argue that moral values exist: If we accept epistemological values, then we might be able to prove that we also have to accept moral values.
Observation and Moral Facts
The Is/Ought Gap
We start off with the widely accepted empiricist assumption that theories aren’t acceptable unless they can be confirmed through observation. One reason we might suspect that moral facts have no implications to observation is because “‘ought’ can’t be derived from ‘is.'” This is known as the is/ought gap, and the argument that observation is irrelevant to moral facts could look like the following:
- Moral facts can’t be derived from non-moral facts.
- We only observe nonmoral facts.
- So, we can’t derive moral facts from observation (257).
Sayre-McCord argues that this argument is too strong because we could also claim that there are no psychological facts in the same way:
- Psychological facts can’t be derived from non-psychological facts.
- We only observe non-psychological facts.
- So, we can’t derive psychological facts from observation (258).
This problem could be called the is/thought gap. We don’t observe thoughts, so they don’t exist. We can’t infer what someone is thinking from observing them. However, this is absurd. We do it all the time. Therefore, the is-ought gap can be rejected for the same reason.
Sayre-McCord argues that the problem is that we can’t demand moral facts to be reduced to nonmoral facts by definition, just like we can’t demand that psychological facts be reduced to non-psychological facts by definition (257). Moral facts and psychological facts can be real, even if they aren’t reducible to some other natural kind of fact.
Sayre-McCord agrees that we can’t see various moral facts, such as an action or person’s worth, dignity, or rightness, but this is no different than the unobservable entities postulated by scientific theories (258). Electrons, for example, are not directly observable. (Richard N. Boyd already suggested that the unobservable entities of ethics is no different than the unobservable entities of science, but he argues that these entities can have a causal effect on our observations that imply their existence.)
Verification through Observation
Sayre-McCord argues that there are moral observations in the sense that we can arrive at moral beliefs from perception (259). Defining “observation” as something like “true beliefs based on perception” would be too strong because we still want to figure out how accurate our moral observations are.
It seems theoretically possible to test and verify our moral theories through our moral observations. How exactly a scientific theory can be tested is notoriously complected and requires us to make use of background assumptions (260). For example, we can start with the hypothesis that the action that produces suffering is bad with the assumption that punishing the innocent is bad. Then we can test whether causing suffering is wrong if punishing the innocent proves to cause suffering (260). (Since ethical tests tend to lead to immoral actions, we generally want to rely on personal experience and thought experiments.)
Sayre-McCord argues that many terrible theories are justified insofar as they are observed by people, but moral theory seems to be justified in a much stronger sense. “Disturbingly, just as moral theory survives any reasonable standard for testability, so too do phlogiston theory, astrology, and even occult theories positing the existence of witches. Like moral theories, each of these theories (when combined with appropriate background assumptions) generates testable consequences, and each makes cognitively packed claims about the world… Although testable, they fail the test.” (261)
In other words moral realism might be a claim that we can observe and test, but we don’t know that it’s a legitimate from this simple fact. In order for a testable theory to be legitimate it must be appropriately explanatory. Astrology, on the other hand, is explanatory impotent (or inappropriately explanatory) (261). In order for a theory to be explanatory, we must believe it is true partially because it is true. However, our beliefs in moral facts seem justified for reasons other than their truth.
It seems we make the moral judgments we do because of the theories we happen to embrace, because of the society we live in, because of our individual temperaments, because of our feelings for others, but not because we have some special ability to detect moral facts, not because our moral judgments are accurate, and not because the moral theories we embrace are true. (262)
Sayre-McCord gives an example. Moral facts and scientific facts seem disanalogous. The reason scientists sees a proton vapor trail is because they have a theory and background assumptions as well as the fact that the proton exists. The reason seeing cats set on fire by kids causes us to believe that the kids are doing something wrong is also due to our moral beliefs, but no part of moral reality needs to be referenced. Wrongness is not causing our belief (262).
The Causal Criterion
Sayre-McCord then discusses various forms of justification, and starts with the Causal Criterion. He argues that if a theory is going to be taken seriously, we need to know how we know its truth, and how we can refer to it. One theory of justification is that a theory is justified if we are caused perceive facts because they are true:
The only entities and properties we are justified in believing in are those that we are justified in believing have a causal impact on our perceptual apparatus. (263)
This is related to the causal theory of reference. When we refer to something, such as H2O, we are referring to something that people have been causally interacting with for quite some time. We can then examine it closely to learn more about it.
Morality, however, does not appear to be something that has a causal impact.
Sayre-McCord then reminds us that we can refer to reality by description when we don’t have causal contact with that part of reality. Perhaps morality can be described, even if it has no causal impact (264). If we can refer to moral reality through a nonmoral description and we have evidence that our description refers to a part of reality, then moral realism could be justified.
The Explanatory Criterion
Not all justified theories are justified by the causal criterion. For example, our laws of science, and cause and effect are not justified from the causal criterion (266). These theories are justified because they explain our perception best:
A hypothesis should not be believed if the hypothesis plays no role in the best explanation we have of our making the observation that we do. (266)
For example, we observe objects falling every time we drop them because of the law of gravity.
Notice that the explanatory criterion is not a sufficient justification for any theory. It only gives us a reason to reject certain theories. Some theories might help explain our observations but still be terrible theories. (A theory about witches might be justified by the explanatory criterion.)
Moral realism appears to fail the explanatory criterion. It does not seem that moral facts help explain our moral beliefs (268). He quotes Harman as saying, “[Y]ou need to make assumptions about certain physical facts to explain the occurrence of observations that support a scientific theory, but you do not seem to need to make assumptions about any moral facts to explain the occurrence of the so-called moral observations” (269).
Sayre-McCord argues that in order to be justified, moral facts will have to help explain our nonmoral observations (269).
Explanatory relevance and explanatory potency
We have a challenge: moral realism seems explanatorily irrelevant. He quotes Sturgeon‘s test of explanatory potency, “If a particular assumption is completely irrelevant to the explanation of a certain fact, then the fact would have obtained, and we could have explained it just as well, even if the assumption had been false” (270). Can we have a moral situation where we can understand all nonmoral facts without requiring any moral theory? If so, moral theory is non-explanatory in that situation.
Sturgeon argued that in order to understand children catching a cat on fire without being “wrong,” we would have to change the situation in some nonmoral respect (270-271). Perhaps the children caught the cat on fire on accident. Had the moral facts been different, the nonmoral fact would also be different. In the same way, a scientist who perceives a proton vapor trail might no longer see that vapor trail had there been no proton. Had one physical fact been changed, then other physical facts would have changed.
The difference between Sturgeon and Harman is that Harman seems to assume that we can’t rely on moral theory to tell us if our observations are reliable (272). Harman will argue that we only observe setting cats on fire as wrong because of an unjustified conviction that it’s wrong to set cats on fire. Sturgeon finds our observations of setting cats on fire to be wrong as evidence that setting cats on fire is wrong.
Geoffrey Sayre-McCord then argues that Sturgeon left something important out. Instead of merely talking about explanatory irrelevance, we need to talk about explanatory impotence in the following way:
[A] particular assumption is explanatorily impotent with respect to a certain fact if the fact would have obtained and we could have explained it just as well even if the assumption had not been invoked in the explanation (as opposed to: “even if the assumption had been false”). (272).
We want to prove that our moral theories help our explanation of our observations. Even if our moral theories are relevant to our observations, they might not actually strengthen our explanation (272).
We can’t be satisfied with explanatory relevance because terrible theories can be relevant. A theory about witches might be relevant to our observations (273). We might think a particular woman is a witch for having long hair. We would then have an observation of seeing a witch only if she didn’t have long hair. Not observing a witch would require different non-witch facts. This is an unjustified theory despite being relevant to observation.
Supervenience and lenience
McCord now discusses his new interpretation of the explanatory criterion involving explanatory potency with regard to moral realism. Some beliefs might be explanatorily potent through reduction. The belief that water is H2O seems to help explain a lot about water by reducing water to a molecule. Additionally, some beliefs might be explanatorily potent without reduction: When certain facts supervene upon others.
For example, we believe that roses are red despite the fact that we can explain our color experiences without invoking the redness of roses. Harman agrees that we might invoke the colors of objects “if only for the sake of simplicity” (275) You might say that it is a pragmatic justification. It makes life easier to be able to talk about the colors of objects rather than discussing wavelengths of light, brain activity, and so on. However, unlike colors of objects, Harman argues that moral beliefs are not justified in a pragmatic way.
Sayre-McCord argues that Harman is “patently false” to say that moral explanations are not useful in everyday life (275). It makes life quite easy to tell children that lying is wrong, or that bad moral character makes it difficult to make friends. “Mother Teresa’s goodness won her a Nobel Prize; Solidarity is popular because of Poland’s oppressive institutions; millions died in Russia as a result of Stalin’s inhumanity” (275).
Finally, Sayre-McCord argues that this liberal understanding of the explanatory criterion that allows for pragmatic justifications can justify our moral beliefs (and any stricter interpretation will probably not). It is the pragmatic justification that allows us to justify our moral beliefs and nonmoral beliefs involving regularities and generalizations. We can explain why a particular square peg can go through a particular circular peg hole without a vague pragmatically justified answer, such as the microphysical structures of each object, but we need a pragmatically justified theory to explain why such a peg can go through several similar circular holes (276). “Certain regularities, such as honesty’s engendering trust or justice’s commanding allegiance, or kindness’s encouraging friendship–are real regularities that are identifiable and explicable only by appeal to moral properties” (276).
Although we have found that a pragmatic form of justification can help justify moral beliefs, Sayre-McCord admits that these properties might be merely psychological and we have no reason to see any importance to them. “We will not yet have shown that there is any reason to care about the properties or that some of the properties are better than others” (276).
The Evaluation of explanations
In this final section, Sayre-McCord argues that our beliefs about justification indicate that evaluative facts exist (277). A theory is partially justified if it is the best available, and any way to rank some theories as better than another involve evaluative facts. For example, if “simplicity, generality, elegance, [and] predictive power” are virtues of theories, then they have some kind of value (277).
The assumption that any theory is “best” already makes use of an evaluative judgment. Therefore, if it is possible to find the best theory in a given situation, then some evaluative fact exists. On the other hand if it is not possible, we will be unable to reject moral theory (or any other theory) as false.
It might be that epistemological virtues and values are distinct from moral virtues and values, but the inescapable belief in evaluative facts might help us find a better understanding of moral facts. Sayre-McCord does not provide the argument that moral facts exist, but he does outline how such an argument can be made:
We might argue that higher level epistemological commitments are justified if they help justify lower level epistemological commitments (just like how evaluative judgments are required to accept the explanatory criterion) (279). In this way higher level epistemological values might indicate or imply lower level moral values. For example, the pragmatic justification of theories might imply some kind of pragmatic moral value. (What makes pragmatic justifications matter? Saving time, making people happier, etc.)
I think Geoffrey Sayre-McCord has a lot of important points to make concerning epistemology and justification, but there are still a couple important questionable assumptions that I want to discuss.
Objection 1: We have direct access to moral facts
Sayre-McCord seems to agree too easily with the assumption that we have no access to moral facts. Isn’t pain direct access to the fact that pain is bad?
Objection 2: His final argument is too esoteric
Philosophy is already considered to be esoteric by many people, but his suggestion of linking epistemological value with moral value is very strange and has nothing to do with everyday life. If moral realism is true, then it is hopefully because of everyday experiences people can relate to. Hopefully most people are moral realists before knowing anything about philosophy, and hopefully they already have good reason to be realists.
Relating epistemological value to moral value is completely alien to people in everyday life and seems to suggest that a philosopher has to prove moral realism is true only to other philosophers. What everyone else believes and experiences is irrelevant.