[Metaethics, Relativism] Discussing Harman’s Defense of Moral Relativism

Harman via Russell (Discussing “Moral Relativism Defended”)

The purpose of this essay will be to reason through a certain metaethical position – Harman’s relativism — by way of a sort of metaethical case study.  Specifically, Bertrand Russell’s stated belief that if he were to make a moral claim regarding the introduction of bullfighting to England, he should feel “…not only that I was expressing my desires, but that my desires in the matter are right” will be contrasted against, or viewed through, the moral relativism of implicit agreement and shared intention proposed by Harman in his paper “Moral Relativism Defended”.  The goal will be to use this example to clarify Harman’s position and, if necessity or opportunity demands, to then challenge this position.

Harman’s goal in the paper is to present his moral relativism as a plausible explanatory hypothesis for certain moral attitudes and languages.[1] Underlying this goal is, I claim, the general challenge to present his moral relativism as a more comprehensive explanation of moral discourse than its rivals, namely, moral objectivism and alternative or composite theories of moral subjectivism.  If his hypothesis appears weak, contradictory, or insufficient in comparison to other explanations or in relation to certain objections, it must be rehabilitated or discarded.

To establish his hypothesis, he evaluates our common sense judgments of individuals who are alien or “beyond the pale”.  Our intuition, Harman claims, tells us that individuals who are incapable of understanding typical moral language – aliens, cows, tigers, or a man raised from birth to be a ruthless contract killer – cannot be reasonably condemned for their actions in the same way or to the same degree that a more normal person could.[2] The reason for this, according to Harman’s hypothesis, is that our inner moral judgments, expressed as moral claims, are always made relative to an implicit agreement and belief in shared intentions held by those individuals who are party to the implicit agreement.[3] In short, moral judgments are comparative judgments.

Harman’s relativistic hypothesis is that moral “ought” judgments are logically similar to judgments regarding size[4]; an elephant is large compared to mice, but small compared to Jupiter. It is wrong for us as members of a society to victimize members of our society because we have entered into an implicit agreement by having mutually understood intentions of disdaining harmdoing in exchange for not being done harm. “Wrongness” as a concept is sensible only in relation to these agreements.[5] The sociopath, alien, and tiger are incapable of entering into these implicit agreements; we intuitively understand that they do not (cannot) share our intentions, and that therefore they are not subject to the same moral standards as others who can and are assumed to enter into these agreements of mutually shared intention. We may say that their actions are wrong, for example, when any of these agents kill and eat a human being, but we cannot in any intelligible way say that it is wrong of them to undertake these actions because, according to relativism, wrongness as a property is present neither in the agent nor the act. Intuitively, we know that their actions are not to be understood in relation to our agreement of shared intentions if it is impossible for them to share or make sense of our intentions. Further, when we make the claim that their actions are wrong, we are giving voice to an attitude of aversion, dislike, or abhorrence and not articulating an objective belief about facts in-the-world.

Here we may introduce Russell’s claim. We may ask: When Russell says that he opposes bullfighting, and believes that it is right to do so, will Harman contend that this is true or false? Does it even make sense to say? I think Harman has given himself enough leeway to agree that Russell could be “right” – for citizens of England, it may be entirely correct that one should oppose bullfighting because there are relevant and significant implicit agreements based on shared intentions that preclude the spread of bullfighting as a sport to England.

Harman may inquire: Upon reflection, would the English believe that engaging in the sport of bullfighting requires the senseless infliction of cruel conditions upon animals? Does the English community jointly believe that this is wrong to do? If so, then it can rightly be said that it is wrong of Englishmen to endorse or engage in bullfighting. That is to say that it would be a violation of the implicit moral agreement of the English based on a shared intention to avoid senseless animal cruelty. Englishmen can even sensibly say that bullfighting per se is wrong – in doing so, they give voice to the English agreement.[6] They are not, however, describing a natural fact that exists substantially in-the-world.

Conversely, if the bulk of the English community jointly does not care about senseless cruelty to animals or does not believe that engaging in the sport of bullfighting entails senseless cruelty to animals, then Russell might still be allowed to say that promoting bullfighting in England is wrong without seeming entirely absurd. However, his moral statement might now have as its implicit referent an agreement to which only a minority of Englishmen subscribe. It may even refer to an intention only Russell holds.  Therefore, if Russell says that it is wrong of Englishmen to promote bullfighting when few or no Englishmen believe it to be the case, then Harman might make sense of his statement by explaining that Russell has misunderstood either the agreement or the intentional state of his fellow Englishmen.

Whether Russell would find this satisfactory or comprehensive is another matter. There are many avenues of contention available to Russell; he may outright reject that everyone shares Harman’s intuition of judgment toward sociopathic human beings.  He may simply contend that Hume, Aristotle, or Kant got it more right and Harman is wrong from the outset (although the historical Russell considered himself a utilitarian[7]). Harman may continually reiterate that his thesis was not meant to explain that all moral judgments are relativistic[8]; when pressed, he may say that it was only meant to defend relativism with respect to certain inner judgments in the certain cases where it cannot be contested in a starkly logical way. This evasion would render his hypothesis both unassailable and anodyne, but it is, in a strictly logical sense, an option. Regardless, in all of these cases, we as the hypothetical Russell and Harman come to an impasse in the discussion. In order to advance the dialogue, it is necessary to attempt to meet and oppose Harman as though he were willing to make a strong (i.e., meaningful) claim. What follows is my attempt to do so.

If Harman says that Russell’s assessment is correct, and that Russell’s desires about bullfighting in England are right due to their expressing agreement to a belief that most Englishmen share, therein is implied a sort of objectivity. Russell cannot be correct in a meaningful way if he is not being held to a standard which applies broadly to many agents with similar relevant characteristics – an objective standard. For example, if we agree that weight for certain objects should be measured in ounces, we might call this a weight measurement agreement. When we weigh object x and find that it is appropriate to say that it weighs y ounces rather than z ounces according to our weight measurement agreement, this is not considered a merely subjective evaluation. We are inclined to say that the object objectively weighs y ounces, although it might weigh z ounces on the moon. Perhaps it is right to say that we have established an arbitrary framework of measurement – but our evaluations themselves are not therefore transitively relativistic. We have agreed to a certain weight measurement agreement because it is necessary and useful to do so, but if merely adopting a common terminology and erecting a common framework renders our judgments of length and mass subjective in a substantial way, then objectivity as a concept is farcical. But as we consider objectivity to not be a farcical or absurd concept in measurement merely because our measurement judgments make reference to a common set of measurement agreements, we should not consider morality relativistic on the whole merely because our moral judgments may make reference to a complex moral measurement agreement. When action x has effects y on the well-being of certain animals, Russell might contend, then we can look to our complex moral measurement agreement and say that it is objectively immoral.

Additionally, although we may intuitively assess moral claims in a pseudorelativistic way (according to our personal understanding of what Harman calls implicit agreements), if substantive moral argument always consists in making our implicit understandings explicit, then Harman’s theory that inner moral judgments are entirely predicated on implicit agreements is questionable. To continue the above analogy, Harman’s moral judgment according to implicit agreement may simply be akin to moral estimation.  However, serious moral discourse, even in the everyday sense, requires that we make our moral agreements explicit so that we can evaluate actions and beliefs according to those agreements in an objective or logically sound way.  Russell can be correct without these agreements being explicit, according to Harman, but he cannot know it. It is simply not sufficient to explain moral attitudes as being according to implicit agreements if we must always make these agreements explicit in order to evaluate them for consistency. In fact, that normal people in everyday moral discourse consider consistency in moral belief and action to be desirable is yet more evidence that the kind of everyday moral estimation that Harman calls relativistic is nested in more mechanically objective notions.

There are a few more minor but not insignificant points I would like to touch on. First, it is that Harman should realize that he has in a very complicated way appropriated the language of relativism and the mechanics of objectivism – especially if, as he concedes, these moral agreements are said to have a teleology that implies a hierarchy of moral agreements.[9] For to say that certain moral agreements serve their purpose better than others in working toward an end is obviously to say morality is objectively oriented in the most meaningful sense. Even if Russell is only correct insofar as his actions agree with the moral agreements of his culture, if the agreement that his actions accord to is objectively oriented toward what is good for the health and well-being of his culture, then his actions might be called objectively good.

Second, we may notice that large and small are relative descriptors, but they refer to objective properties that composite the size of things, e.g., mass, density, and so on. If we say that right and wrong are similarly relative descriptors – not, of course, in the same sense that Harman wanted to say – then it might imply that we refer to similarly objective properties which composite the morality of actions when we make our moral judgments. The things that make bullfighting wrong are objective facts about what bullfighting entails. We might judge the act of bullfighting against acts which are not bullfighting (e.g., having cows in a petting zoo) in order to tease out the moral quality of the act as a whole, but our judgments are contingent upon the objective facts of what it means to have bullfighting in Britain.

Third, we must notice that Harman’s analysis does spotlight a very important truth about the way we view aberrant moral perceptions. When we speak or think of the human sociopath, we do not and cannot regard his aberrant moral perceptions to be equally valid to ours, but rarely do we intuitively “write him off” in the sense that we would write off an animal. We recognize something in them as being defective, and recent science supports the intuitive notion that they are objectively damaged or dysfunctional people who might be understood and rehabilitated.[10]

Again, however, this does not imply an essentially relativistic origin of morality; it implies the opposite. When we consider what Harman would make of it, Russell’s statement, though brief as quoted above, broadly opens up the discussion to a fascinating host of objections and replies.




Harman, Gilbert. “Moral Relativism Defended.” In Foundations of Ethics: An Anthology, edited by Russ Shafer-Landau and Terence Cuneo, 84-92. Malden: Blackwell Publishing, 2007.

Russell, Bertrand. Autobiography. New York: Routledge, 2009. E-book.

Meyers, Seth. “Understanding the Sociopath: Cause, Motivation, Relationship.” Psychology Today. April 02, 2013. https://www.psychologytoday.com/blog/insight-is-2020/201304/understanding-the-sociopath-cause-motivation-relationship.


[1] Gilbert Harman, “Moral Relativism Defended,” in Foundations of Ethics: An Anthology, ed. Russ Shafer-Landau and Terence Cuneo (Malden: Blackwell Publishing, 2007), 85

[2] Harman, “Moral Relativism Defended,” in Foundations of Ethics, 85

[3] Ibid, 85

[4] Ibid, 84

[5] Ibid, 84

[6] Harman, “Moral Relativism Defended”, in Foundations, 86

[7] Bertrand Russel, Autobiography (New York: Routledge, 2009), 39, e-book

[8] Harman, “Moral Relativism Defended,” in Foundations, 85

[9] Harman, “Moral Relativism Defended,” in Foundations, 89

[10] Seth Meyers Psy.D., “Understanding the Sociopath: Cause, Motivation, Relationship,” Psychology Today, accessed October 13, 2016, https://www.psychologytoday.com/blog/insight-is-2020/201304/understanding-the-sociopath-cause-motivation-relationship

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s