One of the more reoccurring debates I get involved in here on WordPress is whether or not an objective morality is possible for atheists. Most of the time this involves me commenting on posts that posit that secular morality isn’t possible, such that I take it upon myself to show theists where their critique of secular morality goes wrong. Today I will try something a little different: I’m going to write a commentary on the Socratic dialogue Euthyphro, a dialogue that shows why morality does not depend on god for objectivity and that if morality did depend on god then morality could be grotesque and vile due to its arbitrariness. That the dialogue is from a time long before the rise of Christianity is interesting, but not a point I will consider here. Regardless, the theist faces a dilemma, if god grounds morality then, either morality is arbitrary and it could be morally good to murder babies if god said so, or, god is unnecessary to positing the existence of objective moral facts and an atheistic morality is possible.
Socrates implores Euthyfro to provide him with a definition for ‘pious’.
Both men agree that ‘pious’ is a subcategory of ‘justice’, such that all acts that are ‘pious’ are ‘just’. ‘Justice’, I think we can agree, is a morally evaluative term, in the sense that ‘if A is more just than B then A is, at least morally, better than B’. Here, I have endeavored to support this premise: ‘pious’ is a moral term.
So, ‘pious’ is a moral term, and Socrates is asking for a definition of it.
[It is worth noting here that Euthyphro only seems to offer non-moral definitions of piety, such as the psychological states of the gods (dear to/desired) thus suggesting an assumption that moral terms can and must be reduced to non-moral terms in order to be informative (an effort undertaken by Moral Naturalists up until the 20th century when naturalists made use of supervenience to account for necessary coextension without identity). Plato’s theory of the forms is anti-reductionist; the form ‘good’ cannot be reduced into other terms because the form of the good is of the ultimate reality. G.E. Moore argued that ‘good’ is as Plato posited, and is a simple and irreducible property; his view, and that of those who followed him, is called Moral Non-naturalism. The failure of Euthyfro to provide a non-moral definition of ‘pious’ might suggest the impossibility of defining the moral in terms of the non-moral, and thus, count as evidence in favor of Platonism, and non-naturalism.]
The first real attempt at a definition is
- Piety = that which is dear to the gods
So a pious action is an action that the gods desire to be taken.
But the gods desire different actions to be taken, so an action can be pious and impious. If so then this definition fails due to contradiction, a pious act is an impious act.
What this uncovers is that ‘that which is dear to the gods’ is not a sufficient condition for the definition of ‘piety’ because it is not precise enough. Euthyfro notes that there are some acts that all the gods desire to be performed, which suggests that the definition should be altered to be
- Piety = that which is dear to/desired by all of the gods
This spurs Socrates to question the nature of the relation between the gods and what they desire to be done. Do the gods desire certain actions to be done because those acts are pious, or are certain actions pious because the gods desire those actions to be done?
(“The pious or holy is beloved by the gods because it is holy, or holy because it is beloved of the gods”)
They conclude that the gods desire certain actions to be done because those acts are pious. But if this is the case, then saying that a pious act is ‘that which is desired by all the gods’ is circular.
- Pious act = that which is desired by all the gods
- that which is desired by all the gods = pious acts
- pious act = pious act
The point to take away from this is that the explanation would end if we thought that acts are pious because the gods desire them. But if this so then what would be pious would be arbitrary, or worse, could possibly have been what we consider to be impious acts (I’ll discuss later why I don’t think moral properties can attach to non-moral property clusters arbitrarily).
As such, we must tend to the other explanation, which is that the gods desire acts because they are pious. What we find is that this is an insufficient explanation because it is vacuously circular. What is more, it seems the fact that ‘the gods desire x’ is explanatorily irrelevant to what makes x pious. That the gods desire pious acts does not actually describe anything about pious acts themselves. What we really want to understand about pious acts is what makes the gods desire them, not that the gods desire them. Moreover, if we uncover what it is that makes the gods desire pious acts, it is probable that the element would be intrinsically good, and that ‘that the gods desired pious acts for that element’ would need not figure in our explanation of pious acts.
From this Socrates concludes that ‘desired by all the gods’ is an attribute but not the essence of piety. Nevertheless, Euthyphro wants to hold onto the idea that the whims of the gods figure in our best explanation of pious acts. This ultimately leads him to posit a different formulation of the constructivist thesis (that acts are pious because the gods desire them) at the end of the dialogue, which Socrates points out, only for Euthphro to walk away in frustration.
They experiment with the idea that pious acts are those acts that ‘attend’ to the gods. They dismiss this on the grounds that such a hypothesis entails that they can improve the circumstances of the gods. My guess is that they think the gods are causally closed off from the realm of the actions of men.
It is here that they consider pious acts as ministrations to the gods. Taken together with some of what the dialogue says, pious acts are those that cause “learning, how to please the gods by words and deed, by prayers and sacrifices. Such piety is the salvation of families and states. […] [piety is] a science of praying and sacrificing”. So, piety is a science of praying and sacrificing that should lead to the formation of a soul.
Socrates simplifies praying to asking of the gods, and sacrificing to giving to the gods. But both agree that the gods cannot be benefitted by our actions, which means that pious acts please the gods but do not benefit them. If this is the case, then it seems that again we come to the conclusion that pious acts are pious because the gods desire them. But this entails that what is pious could be decided by the gods arbitrarily.
So we come to the same dilemma we faced before, that being: if acts are pious because the gods desire them, then what is pious is arbitrary, or, if the gods desire certain acts to be done because they are pious, then what the gods desire will not figure in our most precise definition as to the nature of piety. As the dialogue notes that piety is a subcategory of just, or morally right, acts, we can take this dialogue to be relevant to the status of morality in general. So if constructivism in ethics is right, then either what is morally right has been decided arbitrarily or ethics is explained better without that constructed position.
What this amounts to is the claim that if the gods did not desire those actions we consider to be morally right, and in fact had desired actions that we currently think to be morally wrong, then those acts would be morally right.
Now, this is a tricky counterfactual to evaluate due to how exceedingly metaphysical it is — how drastically it transcends experience. Obviously if the gods decide what is morally right, then I have been conditioned to think that those acts are right to the effect that it is perhaps impossible for me to comprehend those acts as being wrong. There might be possible worlds where torturing babies for fun (my apologies for the overused example) is morally right because the gods in those possible worlds had different desires, and if I lived in one of those possible worlds I would be equally sure that torturing babies is morally right because I had been conditioned to think it.
Nevertheless, I think considering other possible worlds should lead us to reject the idea that what is morally right can be decided arbitrarily, such as, by the gods. It is commonly thought among metaethicists that if there are moral properties such as good or bad, then they supervene on non-moral properties. The thought is that the properties of a state of affairs constitute its moral status, without that moral property being reducible to those non-moral properties. For example, there are various properties at play in the state of affairs ‘torturing a baby’, the most important to its being morally wrong is the causing of pain. But moral badness is not reducible to the causing of pain, as it seems quite plain that there are instances where a person causes another person pain without doing a moral wrong (dentists cause pain, but their acts aren’t morally bad).
That the moral supervenes on the non-moral entails that if two states of affairs are exactly alike in their non-moral features, then they will be alike in their moral features. But this is exactly what the constructivist counterfactual denies! The constructivist position entails that two states of affairs can be the exact same in non-moral respects, but different in moral status if in one possible world the act is desired by the gods, and in another it is not desired by the gods. So, this constructivist position denies that the non-moral properties of a situation exhaustively constitute the moral properties of that situation. In other words, it is something other than the non-moral properties of a situation that give a situation its moral status. I think we can agree that that is preposterous.
If this is the case then the moral status of an action is not arbitrary because moral properties are necessarily constituted by their subveneing non-moral properties. This gets to a question you asked on a recent voicemail; moral realists think that this is how moral properties attach to non-moral properties and regulate our use of moral terms.
So, this theory gives us good reason to doubt that moral properties are decided arbitrarily, and thus, reason to doubt that the gods decide how moral properties attach to non-moral properties. Thus, we must look to the other horn of the constructivist dilemma, that being, you will recall, that the desires of the gods will not figure into our best definitions for moral terms, including piety. Constructivism, theistic or not, I think, is seen to be defeated. The hypothesis that non-moral properties constitute the moral properties of an action or state of affairs provides testable implications that should lead to theories that have more explanatory power about our moral judgments than constructivist theories, including those constructivist theories based on the whims of a god. God is not needed to ground morality — in fact, we have good reason to believe that god cannot ground morality, as this entails arbitrariness where we know there are necessary supervenience relations.