I have been watching some debates about the existence of god, such as Peter Singer vs. Dinesh D'Souza, or Christopher Hitchens vs. John Lennox. On the religious side, one argument comes up surprisingly often. It says if you take god out of the picture, there would be no concept of what good is. Goodness cannot be explained in an atheistic world, and so everything would be permissible. Of course, as an argument for the existence of another entity, I find it quite bad, but let's leave that aside. My question is:

Isn't a theistic account of goodness just as unsatisfactory as an atheistic one?

It seems to me that the analysis of good would be that which god wants. But how is that comforting? Suppose god would want us to kill all Scandinavians for no apparent reason, then that would be a good thing to do. Or let's say female circumcision is something god wants to happen, etc. It certainly doesn't contradict the idea that god is a good god, as it would just mean that god is in a way in which god wants god to be. So where is the appeal? I guess it lies in the fact that (supposedly) you don't have to think too much about how to act and that being a good person grants you a reward and spares you punishment. But thats the way a legal system could work too. An atheist could just reply, "But that's no problem. Set up a dictatorship that rewards people acting in the way the dictator wants and punishes the others. The rules are those which are chosen by the dictator. According to your (the theists) account of goodness we would have a good system in an atheistic world." (Of course the appeal of eternal rewards and punishment is much greater than it is in this scenario, but that doesnt matter for being able to account for what good is. Also there would be issues of practical nature (e.g. setting up the regime, etc.), but let's leave this aside.)

Am I missing something here? It seems to be quite a popular argument and I'm surprised that it pops up rather often in debates like that. So I want to know, how does this give the theist any advantage over the atheist, when it comes to being able to explain the concept of goodness, from a theistic or an objective perspective?

  • I don't agree with him but Zizek makes some interesting points on Dostoevsky's idea that without God, everything is permitted – Einer Sep 10 '14 at 7:24
  • The problem with trying to answer this is that it isn't clear what you mean by constituting an "advantage" or a "satisfactory account". – RBarryYoung Sep 10 '14 at 20:59
  • Though my suspicion is that the disconnect here is based in two things: first the difference between a supreme being and "God", and secondly, the difference between relativist ideas of "good" and absolutist ones. – RBarryYoung Sep 10 '14 at 21:15
up vote 22 down vote accepted

This is a good question, and it's one people have known and thought about for a long time. The problem is so old that it has gone through several different formulations. What Einer writes above is great, so +1 to him for that, but there's a few interesting bits and possible arrangements that I want to add in.

The Euthyphro version is a debate about whether things are pious because the gods love them or loved by the gods because these things are pious. One interesting feature is that Plato does not have Socrates take a position in the debate but rather uses its to attack Euthyphro's certainty concerning the nature of piety. (For this reason it is often considered an early or early mid dialogue before Plato starts having Socrates give answers).

In the middle ages, the issues come into its prime as a discussion of voluntarism. Voluntary means related to the will so there's been more than one voluntarism, but in this case the question is the relation between God's will and goodness, i.e., whether things are good because God calls them good or God calls them good because they are otherwise good. Several key philosophers voiced opinions on the matter. For Augustine, the answer is that something is Good and the form of Goodness is eternally in and with God and that's how good is decided. For William of Ockham (notice the big jump of about 1000 years here), the answer is that something is good merely because God has decided. Thus, for Ockham if God wanted a world where murder1 is good, he could make one.

There is however a middle view or multiple middle views. Aquinas thinks that God has some freedom in defining the good but is still bound by logic (if my memory serves me well). So in other words, he argues that God could have created worlds that differed in some but not all ways. So God can't create a world that licenses murder, but he can create one where moral rules are different but those moral rules that follow rationally.


A second and more modern debate which is important in Kierkegaard scholarship relates to Divine Command Theory. DCT is basically the modern cousin of voluntarism, which states things are right or wrong for us because God commands them. Since voluntarism is generally non-desirable (for the reasons you capture in your question, there's an interesting workaround championed by Robert Merriwether Adams and C. Stephen Evans -- which is to separate the problem into two problems: (1) a metaphysical problem about what decides what is good and (2) an epistemological problem about how we know what is good.

This gives us at least four horns instead of two:

  1. Good is good because of what it is AND we know what good is because of what it is
  2. Good is good because of what it is BUT we know what good is because God tells us
  3. Good is good because God decides what it is BUT we know what good is because of what it is [after the fact]
  4. Good is good because God decides what it is AND we know what it is because God tells us

[If we add in Aquinas's type of middle position we would get a further type of position on the left and a further type on the right = 9 positions]

1 I say "murder" here, but you need to see that as an ellipsis for reasons @shane could explain if he would ever publish that paper. But basically we mean killing people for practically any reason. (Whether it would still get a word with attached negativity like "murder" is not clear).


Returning to your question, I think the weaknesses on each side thoughtfully considered are:

  1. It doesn't seem to go so far as there's no concept of good for the atheist so much as the atheist now needs to construct one and justify the construction.
  2. For non-voluntarist theists, God is not free to be malevolent tyrant but then that means there's still work to do in constructing the nature of good
  3. Views without God informing us of what is good (regardless of the source of the definition of goodness) have to explain how we can connect that idea of good to reality.
  • 2
    To make it short: Your answer is better than mine. Darn! – Einer Sep 10 '14 at 7:44
  • 2
    @Einer I think yours is quite good here. If only every question and answer involved this sort of problem... – virmaior Sep 10 '14 at 7:47
  • 1
    Thanks, this has been really thorough and insightful, plus a good supplement to what Einer wrote. One question though: what or who do you mean by @shane in the *1 footnote? – Doc Sep 10 '14 at 8:00
  • 1
    Participant here on the forum. Coincidentally (as in we both started using SE and both wound up on the philosophy one -- our writing styles are recognizable so I knew it was him and he I), friend in real life from the same PhD program. I'm referring to an insight I learned from him that I thought was pretty good. It's also an idea you could get from GEM Anscombe if you decipher it in her "bilker" description. – virmaior Sep 10 '14 at 8:03

No, you are not missing anything. This is a rather old problem known as the Euthyphro-dilemma. It has two possible scenarios: Something is wrong because God says so, or God says so because it's wrong.

In the first scenario "before" God said it is a sin, it wasn't. There is no higher source of value, God could have used to model his laws after. He needed to 'invent' them. If a Christian wants to know, what laws to make, he can turn to the bible. God has no super-bible he can read to find out what's right or wrong. Since no higher law exists, god has complete freedom to decide what is a sin - nothing is there that could contradict him, nothing that could support his decisions. If he had decided that wearing gold is a sin, it would be. If he had decided that it is good to always beat other people, then that would be a moral thing to do - as this rule is made by the highest possible authority. It wouldn't make him a 'bad' god since there are no bad rules, unless God says that it's a bad rule. If he says it's a good rule it automatically is and he remains a good god who chose a good rule. But this makes the resulting moral system completely arbitrary. God has no reason to choose one rule over another, he can and must only chose at random.

In the second scenario, there is a higher source of values. God does not have choose at random he has a solid background he can work with. Maybe he has insight in a principle that separates good from evil, or he simply knows what's right and what's not. Now he even is theoretically capable of choosing the wrong rules. He would not do it, as he is assumed to be a good god, but there are wrong rules and God knows it. Now all he needs to do is take all the right ones and try to communicate them to mankind. He has no free choice in the selection of the rules. He cannot take the wrong ones (as he is good), and he must take all the right ones. So basically he is no more than a messenger: The message already stands and he has no influence on it, he just needs to get the message to us. So God didn't make the rules, he just passed them on. His judgments are irrelevant for the message.

  • 1
    This seems to only answer the secondary question, "Am I missing something here?" without addressing the main question, "Is there an advantage to the theistic account?" (or conversely, "Isn't a theistic account as unsatisfactory as an atheistic account?"). – iamnotmaynard Sep 10 '14 at 19:02
  • @iamnotmaynard True. I haven't really thought about that. In retrospect I think that while composing the answer I thought that it's somehow common knowledge how hard it is to find a good foundation without god. And that my sole task is to point out, that with god it ain't easier either. But you are right: It is definitely a weakness of my answer. – Einer Sep 10 '14 at 19:08
  • "In the first scenario "before" God said it is a sin, it wasn't." Yes; incest, for example, wasn't always a sin. (See Abraham's story in Genesis 20, where he was married to his paternal sister.) I will admit that I do not know enough about other religions to know if there are similar instances. – Marcel Popescu Sep 11 '14 at 13:28

I, contrary, think that you are actually missing something, and so are the other answers except CCarter's answer. (That's a pity one needs over 50 reputation to comment, as it would be more suitable I guess....)

What I consider a flaw of any discussion like Eythyphro's dilemma, is the constant change of problem level we are talking on and the lack of linguistic reflection. Let me explain: as we discuss what is good, we are on the field of ethics. Then, you are asking a meta-ethical question (about the source of good in general, its possible creation by God and the implies of that hypothesis), but use arguments and examples back from the lower level: the permission to murder anyone without any reason is certainly something that's not "good" in any ethical system and our ethical intuition treats that as a valid argument against the possibility of creation of ethical system treating it the other way.

Apparently, this can not be considered as an argument: the field of meta-ethics doesn't know anything about why couldn't we name murder good, because that's not in the field of meta-ethical problems. Also, we need to checkout why the example of murder bothers us so much: the cultural and linguistic connotations about that case make an unjustified murder something bad almost by definition. When we enter the field of meta-ethical discussion, we should double-check every definition we use and whether we are even allowed to talk about something.

The concept of bad God is indeed, as CCarter points out in the last paragraph, self-contradictory. I even consider a mistake making an assumption that "there is no God, but objective values do exist", and so I agree with the statement that "without God, everything is permittable". Naming anything objective and absolute without concept of God is, in my opinion, in fact just creating a new God (what's more: if we associate it with a person who knows everything, we are happy to meet a new totalitarian dictator). The fact, that God is often impersonated while values and other abstractions are not is just another example of our cultural background. Therefore, see that using this statement as an argument of God existence is absolutely wrong, as it just tries to prove that there is something that's absolute and objective, often just by our intuition that murder being bad is so obvious it should be objective.

Now, to answer your question is theist's explanation of what's good just unsatisfactory as the atheist's approach. In my opinion, it is not. If we believe that something is objective and not just name it by convention, we have a strong basis and need not to develop further. We are sure. It is just like Newton's laws: as long as you stick to them, you are absolutely sure what should happen, what is possible ("good") and what is not ("bad").

But, at some point someone came up with an idea, that Newton's laws are insufficient and developed something new.

The knowledge we have is never 100% right, because we can never be 100% sure: we are just working with a model of reality, not with reality itself. When we change model, we can achieve something new (like quantum physics etc), but if we change the model of ethics at will, in most cases something really bad can happen, and even if not (we should, again, not use just our intuition) we can not predict what exactly would happen, and ethics is too serious business to play with it. That's why being assured that what is good is good in reality and not just by convention is more satisfactory: it just makes everything simplier. It doesn't help in explaining goodness, because note that goodness, whether God created it or just showed it to people, is, in this approach, an axiom, just like God.

I'm surprised no one has mentioned Immanuel Kant yet! Because he surely understood the theoretical problem I described above: note that Kant postulated the existence of God exactly that way and that is (in my opinion) brilliant: he did not try to prove his existence, but postulated it: he assumed we need objective values, and therefore just showed that we need God to have them. (If I simplify too much or maybe do not understand something, feel free to correct me.) Kant also proved that there can not be any proof or disproof of God's existence.

I hope one day, people from both sides of the debate, theists and atheists, would finally understand that. Or just rename their discussions to something like "Do we need God?", which may be a little less pointless.

  • Is the essence of your answer, "the definition of God includes the postulate that God is Good, so by the time you are a theist you already are logically committed to a system of morality, while if you reject God you have more work to do"? – Rex Kerr Sep 10 '14 at 20:59
  • Main purpose was to show the difference of ethics and meta-ethics and language we use to describe things as they contain hidden assumptions (like the one that God must be Good). I haven't said anything about commitment to a system if you are a theist: the point was only that you have a strong base of morality (containing at least one objective value) you either commit yourself to or develop by yourself. – Bartek M Sep 10 '14 at 21:29
  • I suppose the step that I'm missing is why the Euthyphro dilemma isn't even worse for the theist because if not satisfactorily answered it will not only leave your morality baseless but will destroy your theism as well because your theism is contingent upon there being a good answer. Rather than clarifying things it seems the linguistic distinctions you are drawing are side-stepping the actual issue by means of a linguistic trick. Or are you just making the point that if you assume more you have more to work from? – Rex Kerr Sep 10 '14 at 21:44
  • 1
    Actually, I tried to show that Euthyphro dilemma makes similar mistake to which is made in the liar paradox and tried to deal with first the way Russell did with the latter (you can call it a trick). But why do you think that a) no good answer leaves morality baseless? In my opinion (and the answer), God, as a belief, is the base of morality and is quite indentical to the values we believe. b) theism is contingent upon being there a good answer? I consider being a theist as a choice of faith rather than rational conclusion (as the existence of God cannot be rationally proven) – Bartek M Sep 10 '14 at 22:00
  • I should have said will leave both your morality and theism in doubt. Why? Proof by contradiction: if you assume P and it may be that !P, and P iff Q, then !P => !Q. Also, Russell's type theory is hopelessly restricting since it's impossible to talk about yourself at all; likewise, this approach seems just to throw away the problem without giving it due consideration. The whole point of the concern is that one wants meta-ethics and ethics to line up: if it were the case that following God's wishes left us despondent and heartbroken, saying that "God is Good" would be problematic. – Rex Kerr Sep 10 '14 at 22:16

The theist position is that the source of morality is not arbitrary. In general, and to side-step Eythyphro's dilemma, theists consider God to not just be good, but to be the very essence of good.

Naturalists, on the other hand, have to rely on naturalistic justifications for moral definitions. Does this type of behaviour improve human well-being? Does it reduce suffering? Is it naturally selected for?

Theists consider the naturalist versions to be arbitrary - you can always ask: "Why is that particular moral code the true correct one instead of some other one"?

Is this then an advantage for theism? If the argument goes (as is typical in these debates): "God must exist, because theistic morality is not arbitrary while atheistic morality is", then that is question begging - because the non-arbitrariness of theistic morality is contingent upon the existence of God. If God were not to exist, then theistic morality would be just another arbitrary moral code.

A different version of the argument is: "Only theistic morality is non-arbitrary. Non-arbitrary morality exits. Therefore God exists". The problem with this is the second premise - can we show that non-arbitrary morality exists? How do you go about proving your moral code is the absolute correct one?

  • This really doesn't seem fully-enough developed to answer the question. It's more like points to mull over than an answer. – Rex Kerr Sep 10 '14 at 21:07
  • I point out a line of reasoning that illustrates that even if Eythyphro's dilemma fails, the "advantage" for theism is hardly settled. What aspect of my answer leaves this unclear? – firtydank Sep 11 '14 at 5:20
  • 1
    You point out that getting the existence of God from the desire for a non-arbitrary source of morality is highly questionable--I agree! But you don't tackle the case where someone is already convinced that God exists but is wondering whether that actually leaves them in any better of a position to know what is good--and that seems, to me, more the point of the original question. – Rex Kerr Sep 12 '14 at 18:31

There are some truly superb answers to this question and I feel somewhat under read to contribute but I feel like there's been one point that has not been mentioned in the answers and that is the question of the very nature of right and wrong.

From what I can tell from having listened to some of these same debates, the problem posed by the theists, is not that God has any distinct advantage in knowing right and wrong or choosing it (as some form of dictator might) but is the very definition and source of right and the antithesis of wrong.

This may seem like splitting hairs but I'll try and explain what I mean. When we talk about what is good or bad, we have presupposed ideas on what that is. In an argument we might say that something is not fair and we presuppose that the other person agrees with our definition of fair. The problem is, where did our notion of fair or right and wrong come from.

The traditional atheistic view, as far as I understand it, is either that it comes from societal norms (my family background and culture) or that there is a higher principal of what benefits us as a whole or that it simply is what is good/bad for me as an individual.

I believe that the theistic view has objections to each of these. First the personal view (what's good/bad for me) doesn't give any basis for altruism or any basis for justice - how can you condemn someone for an action they perform if there is no basis for right and wrong that exists outside of the individual - that's every man for himself. Second, the higher principal of society as a whole would have to come from some kind of evolutionary process - we would have to evolve a sense of 'benefiting the whole community being better than purely pleasing myself' over time and seeing the way we all fair better together. This is probably the strongest possibility but it gets very thorny on issues like caring for the elderly or the severely disabled in the community - on a pure survival basis, they do not bring benefit to the community and so the debate over euthanasia voluntary or otherwise would be moot. Third the cultural view - again, this seems to work in some ways but what happens when a society or culture decide that it's right to do something most would consider wrong. How can you condemn a large group in this sense. Basically, the objection boils down to if right and wrong is defined by the majority in a society then we have no basis on which to condemn Nazi Germany and the actions perpetrated in that time by that group.

The theistic view is that it has the advantage because most atheistic views tend toward some cost/benefit model and that usually ends up presenting problems when it comes to judgement. However, if you have God as the definition of right and wrong (as has been said, right is what God is like, wrong is what God is not like) then you have a basis for all right and wrong choices.

It's not so much a reward basis - you'll get some good stuff if you do what God says, as much as it's the nature of the universe that God created.

The question of what if God decided it was good for people to murder or God's nature included murder, has a logical flaw in it. In the theistic view that we're discussing (espoused by John Lennox or CS Lewis for example), the definition of good comes from who God is as he is the creator of everything. If God's nature was such that he created a universe where murder was good how would we have any ability to tell that it was wrong? What basis would we have to say it was not good? In that case, you're back to a personal view of right and wrong in which, it's what benefits me or society the most and you suppose that you are a better being than God. In that case, God is no longer God and the argument breaks down. You're right to suggest that that would make God some kind of weird dictator but that's all he would be.

  • There seems some very serious circular reasoning going on here, or at the very least there is the epistemological question how does any theist know that their deity is actually God and not weird-dictator? Also, given that God doesn't seem to be talking directly and clearly to huge numbers of people, I'm not sure what you mean about the "basis for all right and wrong choices". There are furious disagreements about theology (leading, in some cases, to bloodshed). I agree that in principle this could be an advantage but you seem to be arguing it is in practice, and that needs more support. – Rex Kerr Sep 10 '14 at 21:04
  • I wouldn't suggest that this is circular reasoning (but then it's mine so why would I? :)) I think that the question being asked is lacking something - namely, if humanity is able to achieve a higher or better ethics than God then you have just proven that this scenario is false. Basically, it's saying that the theist position is that all ethics flow from the nature of God so you can't get higher. Also, the fact that humans can disagree on what God is like and get angry about that is hardly surprising and so not a disproof of his existence or his being a basis for morality. – CCarter Sep 11 '14 at 6:59
  • I'm not arguing that God does not exist because humans disagree and fight about God, just that when there are disagreements and anger about what God says it's hard to have a basis for all right and wrong choices in practice. So it seems, even granting that God does exist, that although this advantage is a formal possibility it does not obtain. Separately, however, a reasonable answer must also deal with the "what if God is weird-dictator" issue since atheists do not have to deal with that one and we're trying to do a head-to-head. – Rex Kerr Sep 12 '14 at 18:37
  • I see what you're saying about the possibility of negating the advantage that God brings to the knowledge of right and wrong by including human error. However, by saying that you also assume that God has no ability or desire to make known what right and wrong is. As for the weird dictator issue, I think you're still missing what I'm saying. The weird dictator comes from the assumption that right and wrong exist outside of the nature of God and that he could deliberately disobey them for his own amusement. In the theistic view, absolute right and wrong can only be defined by the nature of God. – CCarter Sep 15 '14 at 7:01
  • I'm not assuming that God has no ability or desire; I merely need to observe that people who profess to believe in God nonetheless can have vicious disagreements about what is right and wrong (to the extent that onlookers are quite sure they are both acting wrongly). I can declare myself a god and say that this is a huge advantage because now I have access to absolute right and wrong, but somehow this would be a rather unsatisfying use of the word "advantage", no? I do agree that if a theist presupposes God has that ability that it is an "advantage", but it's also in an unsatisfying sense. – Rex Kerr Sep 15 '14 at 20:20

I feel underqualified to answer here, given the quality of some of the prose above, but I think that you are all missing something fundamental here, though I must accept that it could be me that is missing something obvious, if so, please forgive me.

There seems to be an acceptance that good is somehow absolute and all the arguments and difficulties mentioned above seem to derive from that assumption. I take the selfish position which tends to make this whole argument meaningless; Good is relative. Something is good if I believe it benefits me, my family, my clan, my tribe, my state, my nation, my environment, my species, my planet or my god. Something might be good if it does not benefit me, but benefits something I can consider someone else thinking of as important. There is no necessity for something that one person considers good to also be good for anyone or anything else. There is no necessity for something that is good to not be bad. There is no necessity for an external arbiter. It is an approach that allows for a god, but does not require one.

If I can convince you that I am your 'voice of god', then I can convince you that the good you do for your god by killing unbelievers outweighs the bad you do for your species by killing.

If I am being altruistic, then the good I feel being able to help others outweighs the cost to me of giving that help.

If I am a legalist, then the good I do for my clan by maintaining societal norms outweighs the costs to my tribe associated with enforcing those norms and the harm I do to individuals who digress from them.

I accept that this approach is sometimes not be popular as it reduces many traditionally profound philosophical concepts down to very personal and selfish motives that can be examined with simple economic style cost/benefit tools. However, as a working theory, it is good.

  • 1
    I believe you have described what the OP would call "the atheistic position" - you define good independently from God. The question was, though, why is this position inferior to the one where good is defined as "what God wants". I don't believe you managed to answer that. – Marcel Popescu Sep 11 '14 at 13:36

Goodness cannot be explained in an atheistic world, and so everything would be permissible.

This is what theists want you to believe. But there is no reason why this should be so.

In fact, thinkers in the tradition of discourse ethics claim that a rational ethic is possible (page 313f).

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.