Dichotomies in Metaethics

Published by timdean on

There are two types of people in this world: those who like dichotomies, and those who don’t. This post is for the former.

Metaethics is riddled with dichotomies. And, unhelpfully, they often cut across each other in unpredictable ways. On top of that, not every metaethicist employs the terms in the same ways, meaning some dichotomies are rendered differently in different texts.

So, here is my understanding of the key dichotomies in metaethics (with my preferred options). I’m not entirely sure I have characterised them all correctly, or that I’m not missing any salient points. Please feel free to criticise or revise this list in the comments:

Realism vs. anti-realism

  • Realism: moral facts exist
    • Often cashed out as objective prescriptive facts
      • “objective, intrinsic, prescriptivity” (Mackie, 1975 – who, by the way, thought these facts didn’t exist)
    • Or facts about a property of goodness in things/actions
    • Example proponents: G.E. Moore, M. Smith, P. Bloomfield, a (disappointing) heap of others
  • Anti-realism: moral properties don’t exist
    • Example proponents: Mackie, Joyce, Greene
  • I fall within this camp as I don’t believe objective, prescriptive moral facts exist, and use evolved moral psychology to show why we might erroneously think the do

Cognitivism vs. non-cognitivism

  • Cognitivism: moral utterances are statements of fact and have truth values
    • Eg: “murder is wrong” is either true or false
    • A descriptive semantic thesis about moral discourse, not a prescriptive or ontological thesis about moral statements
    • Strong cognitivism: as above but the facts are cognitively accessible
  • Non-cognitivism: moral utterances are statements of affect or preference and don’t have truth values
    • Example proponents: Blackburn, Ayer, Stevenson
  • I consider this a spurious distinction as everyday moral discourse is muddled, and most moral utterances have a cognitive and an affective component: moral intuitions (immediate impressions of permissibility/impermissibility of an act) are typically non-cognitivist; while the post-hoc rationalisations of moral norms are typically cognitivist

Naturalism vs. non-naturalism

  • Naturalism: a form of cognitivism and realism that states that moral properties exist and they are natural properties or moral statements are rendered true or false by facts about natural states of affairs
    • Facts about happiness (Adam Smith, Herbert Spencer) or about neurological states (Sam Harris)
  • Non-naturalism: a form of cognitivism and realism that states that moral properties exist and they are non-natural properties
    • Example proponents: G.E. Moore
  • I reject both these formations as they’re dependent on moral realism. Instead I adopt an anti-realist naturalism that says moral phenomena are analysable in purely naturalistic terms, but this alone doesn’t imply any binding normativity, similar to Joshua Greene, Michael Ruse (I think…) and the so-called Duke Naturalists

Internalism vs. externalism

  • Internalism: moral beliefs are intrinsically motivating
    • It’s somehow contradictory to believe that ‘x is good’ and yet have no inclination towards doing ‘x’
    • Example proponents: Michael Smith
  • Externalism: moral beliefs are not intrinsically motivating and require some other impetus to motivate moral behaviour
    • It’s possible to believe ‘x is good’ and not be motivated to do ‘x’
  • This is another dichotomy I reject as I take a softer psychological approach that suggests moral norms are not bindingly prescriptive nor intrinsically motivating, although we are often internally emotionally motivated to act in accordance to the norms to which we subscribe (akrasia not withstanding), and we want people to behave like this, but we often need ‘external’ motivation, such as threat of coercion, to motivate conformity and prevent bouts of akrasia

Absolutism vs. relativism

  • Absolutism: there is one moral standard that is fixed
    • Applies without contingency
    • Can be objective (fixed by facts in the world or God’s will)
    • Can be subjective (fixed by the whim of an authority)
  • Relativism: moral standards are indexed to some value
    • Indexed to culture, group of people, environment etc
    • Can be objective (x group/circumstances always implies y morality)
    • Can be subjective (x group/circumstances can choose y or z morality)
  • I am absolutely not an absolutist, rather more a pluralist along the lines of David Wong, where there I argue there are many ways of solving the problems of social living that morality is constructed to solve, but there are better and worse ways in different environments, and there are clearly some very bad ways

Categorical vs. hypothetical

  • Categorical imperatives: moral norms are binding regardless of an individual’s ends or desires
  • Hypothetical imperatives: moral norms are binding contingent on an individual’s ends or desires
  • I fall into the hypothetical camp, with norms binding hypothetically, contingent on our desire to serve our own interests in a social context, given the assumption (usually true) that acting in accordance with the constructed moral code will advance our own interests better than the alternative of not acting in accordance with a moral code

Not even sure if the last one is strictly metaethical, or more a normative ethical thesis. But hey.


Mark Sloan · 12th November 2011 at 5:11 am

We share confidence in the utility of understanding the evolutionary origins of morality. But I want to argue that, in light of a fact concerning morality’s evolutionary origins, Mackie’s definition of Realism (“moral facts exist” where these moral facts must be “objective, intrinsic, and prescriptive” facts) is misleading and un-useful to understanding morality.

The critical new fact is the universal moral principle “Altruistic acts that increase the benefits of cooperation in groups are moral”. This moral principle can be shown to underlie all biological emotions that motivate altruism (such as empathy, loyalty, guilt, shame, and indignation) and virtually all past and present enforced cultural moral standards. Further, enforced cultural moral standards (whose violation commonly engenders indignation and the idea that the violator deserves punishment) can also be shown to universally advocate altruistic behaviors, at minimum in the sense that they (for example prohibitions against theft, lying, and murder) are binding regardless of individual needs and preferences.

The mathematics of game theory shows that this moral principle is the common core of a set of winning strategies in environments where synergistic benefits of cooperation are available, which includes virtually every environment humans inhabit. These strategies include kin altruism, direct reciprocity, and indirect reciprocity (which share the core principle but direct altruism at different subgroups). The mathematics that makes these strategies winning strategies are facts that are as objective and intrinsic to the nature of physical reality in our universe as the rest of mathematics, but, to my knowledge, cannot be prescriptive. The only kind of ought that can be based on this moral principle is an instrumental ought.

Mackie’s Realism definition is misleading and un-useful because it assumes that the objective, intrinsic, and prescriptive characteristics of moral facts are linked. Evolutionary morality shows there is a moral fact that is undeniably objective and intrinsic, but not necessarily prescriptive.

I find it infuriating when Michael Ruse and others take apparently perverse delight in saying “Morality is an illusion!” when morality is as objective and intrinsic to our universe as mathematics.

Tomasz Herok · 12th November 2011 at 6:31 pm

Why God’s will doesn’t count as a whim of an authority?

Tim Dean · 12th November 2011 at 11:19 pm

Hi again Mark. I wouldn’t say Mackie’s characterisation of moral facts is misleading. Mackie is simply articulating the kinds of facts that moral realists are saying exist. A moral realist wants to say that “the sky is blue” is true, and that “murder is wrong” is true. The strange thing is, while the former fact doesn’t intrinsically bind us to behaving in some way, the latter does. The question is: how can a fact bind us to behave in a certain way; how can a fact be prescriptive?

Mackie, (like myself and, it seems, you) is an anti-realist, because he thinks these kinds of prescriptive facts don’t exist. (I highly recommend his 1977 book, Ethics: Inventing Right and Wrong).

That suggests that you’re advocating an anti-realist stance as well. I’d say you’re in good company. Anti-realists can certainly say there are facts that are relevant to making moral judgements – like facts about harm or cooperation. But there are no facts – like “stealing is wrong” – that can be bindingly, prescriptively true.

So you can say that morality is about cooperation – and I agree – but we still have to desire cooperation in order to motivate us to behave in ways that advance it. Just saying that altruistic behaviour promotes cooperation is a descriptive notion – like saying “the sky is blue”. We have to add that desire in there to make it moral.

So, the upshot is, I think your position is quite close to my own, which says: it’s a matter of contingent empirical fact that just about everybody desires promoting their own interests, and the best way of doing that is to be social and cooperative, and morality is the system of rules we abide by (and enforce) to make that happen. There are no prescriptive moral facts in this story, but there are lots of facts that contribute to whether something is ‘moral’ in this weaker sense. But the whole system isn’t intrinsically binding, only contingently on us desiring to pursue our interests.

As such, moral rules are like the rules of a game. They’re agreed upon and binding to the extent that we’re all playing the same game (and I reckon that, as a matter of contingent empirical fact, we all do want to play the game), but the rules themselves are not written into the laws of nature in the same way physics is. So “murder is wrong” not intrinsically true but only true according to the rules of the (moral) game.

Hope that makes sense.

And Tomasz, I’d say theists believe God’s will is a special case of objective authority. Naturally, I’d disagree. Not least because this particular authority figure doesn’t happen to exist.

Mark Sloan · 13th November 2011 at 5:05 am

Tim, yes I am anti-realist in that I concur that, as you well put it,: “it’s a matter of contingent empirical fact that just about everybody desires promoting their own interests, and the best way of doing that is to be social and cooperative, …. But the whole system isn’t intrinsically binding, only contingently on us desiring to pursue our interests.”

My objection to Mackie’s definition of Realism (“moral facts exist” where these moral facts must be “objective, intrinsic, and prescriptive” facts) is that at least one moral fact concerning the intimate relationship of altruism, cooperation in groups, and morality as an evolutionary adaptation IS objective (even cross culturally) and IS intrinsic to the nature of physical reality as shown by the mathematics of gamed theory. This is to me, an irrefutable counterexample to the logical coherence of Mackie’s definition.

If the community wants to define anti-realism as something like “no morality is intrinsically binding; morality can only provide instrumental oughts” that is at least a coherent position. Mackie’s definition is incoherent in that he falsely links prescriptiveness with being objective and intrinsic (in the sense of it being intrinsic to an object, the universe).

Beyond Mackie’s definition being just incoherent, it is profoundly misleading to say that “Morality is an illusion!” when its universal underlying principle is intrinsic to the nature of reality and has existed, in the sense its mathematics were real, since before “the fusion fires of the first star lit”.

While I’ve just commented on your first dichotomy, the whole post is very helpful to my understanding of moral philosophy. Dichotomies is a nice organizing idea.

Tomasz Herok · 14th November 2011 at 5:45 pm

I don’t think that the question of whether God exists or not is relevant here. I just can’t see how this can be called objective morality in any meaningful sense. ‘Objective’ normally means something like ‘independent of opinions, attitudes or whims of a person or group of persons’. It would be extremely far-fetched to modify it to something like ‘independent of opinions, attitudes or whims of a person or group of persons with the exception of one particular kind of authority, namely God’.

Leave a Reply

Avatar placeholder

Your email address will not be published.