Dawkins’ Definition of “God” – Part 3

Dilemma for Dawkins
Proof of the existence of Zeus would either verify the claim that “God exists” or it would not. It is not immediately obvious which side of this dilemma Dawkins would choose. If he granted that proof of the existence of Zeus would verify the claim that “God exists”, then he would have to toss out his definition of “God” (as being too narrow). On the other hand, if he denied that proof of the existence of Zeus would verify the claim that “God exists”, then his conclusion that “God almost certainly does not exist.” (p. 189), would fail to rule out the existence of Zeus and Satan, and perhaps dozens of other gods.

Ambiguous Conclusion

Another way of putting this point, is that the main conclusion that Dawkins puts forward at the end of Chapter 4 is ambiguous between a weaker and a stronger claim:

(W) It is almost certain that there is no god who is responsible for creating this universe and everything in it.
(S) It is almost certain that there is no god whatsoever.
Proving the weaker conclusion (W) would not establish atheism, because it leaves belief in non-creator gods (such as Zeus and Satan) untouched.

Proving the stronger conclusion (S) would establish atheism, because it eliminates not only the God of traditional theism (an all-knowing, all-powerful, and perfectly good person), but also eliminates other lesser gods, such as Zeus, Baal, Wotan, and Satan.

There is a third possible interpretation of Dawkins’ conclusion as well. We could simply ignore his somewhat confused attempts to clarify the word “God”, and interpret his conclusion in terms of a more standard definition:

X is God if and only if
(a) X is all-powerful,
(b) X is all-knowing,
(c) X is a perfectly good person.

Note, however, that this definition includes a normative condition: “X is a perfectly good person”. So, this meaning or sense of the word “God” runs contrary to Dawkins’ assertion that the question “Does God exist?” is a scientific question. Since science has no capacity for resolving normative issues (e.g. “Is Jesus a perfectly good person?”), science alone cannot answer the question “Does God exist?” if we use normative categories to define the word “God”.

Nevertheless, since the above definition is closer to the standard meaning or use of the word “God” (among theologians and philosophers in Western thought) than the definition that Dawkins puts forward, it is reasonable to ask whether Dawkins’ argument establishes his conclusion on this interpretation:

(N) It is almost certain that there is no all-powerful, all-knowing, perfectly good person.
Since it is more difficult to prove a stronger claim, I will first consider the possibility that Dawkins intends to only make the weaker claim (W) at the end of Chapter 4. If his argument supports the weaker claim, that will still be a worthwhile philosophical accomplishment (Dawkins would say: “scientific accomplishment”). If his argument does not support the weaker claim, then it certainly does not support the stronger claim (S) either, for the stronger claim implies the weaker one. Finally, I will consider whether Dawkins’ argument supports claim (N).

Jerry Coyne Blocking: Episode II
Religious Experience – Recognizing God
Please Support the Ultimate Counter-Creationism Resource, Troubles in Paradise (TIP)!
The Seven Deadly Sins of Christianity
About Bradley Bowen
  • http://www.blogger.com/profile/16415287407278685717 Lamar

    (1) The problem with defining God/gods in order to see if God/gods exists is that one would first need to show the necessary and sufficient conditions for being called “God”. However, given the way we use the term “God” (for example, we call Zeus God, who, being physical, is nothing like the God of Western Monotheism) there seem to be no such necessary and sufficient conditions. (There may be what Wittgenstein would call a “family resemblance” among the entities we call God/gods, but that isn’t helpful here.) So your criticism of Dawkin’s argument is endless, since one has a never-ending pool of possible entities we all call gods – none of them being the same in systematic ways – which you can draw from to point out how Dawkins didn’t disprove that one, and thus didn’t prove the stronger claim S. In effect, you are demanding the necessary and sufficient conditions for the definition of God, while at the same time taking advantage of the fact that there are no such conditions to begin with.

    (2) We could define God using the 3criteria you set forth, including the normative one, and those would count as sound and meaningful necessary and sufficient conditions. And you are right in saying that science does not end normative debates, so it could not tell us whether or not Jesus, for example, was “good.” However, science can tell us whether we would call something good, based on what we take to be the necessary and sufficient conditions for being good. True, what those conditions are cannot be decided by science, however, what fulfills those already decided conditions – based on empirical data – can. So, if we all agree that it would be “good” to save a drowning baby, science can (despite not being able to tell us whether or not it actually is good) tell us that since babies still drown, an all-powerful God is not doing something that we would call good. Thus, there isn’t a being that we would call all-powerful and all-good.

    (3) I think a more basic discussion lies in whether or not a realm of the supernatural exists. We might not decide that Zeus fits the definition of a god because he is a physical being. So, Zeus could exist, just like the Loch Ness Monster, or aliens, could also exist. But under this view, atheism would say nothing as to the existence of Zeus, but would rather focus on the existence of a supernatural entity (whatever that would be). To be sure, we would need to develop some necessary and sufficient conditions for the “supernatural” also. I think this is what most people think of as atheism anyways, namely, a denial of the supernatural, not of gods per se.

    I think we all need a better understanding of the terms we use, and I think you are right to attack Dawkins on these subtle but important points. He gives philosophically very weak definitions while at the same time making strong and blunt claims.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Response to Lamar…
    (1) Good point. It should be noted that Dawkins himself provides a necessary-and-sufficient-conditions definition of “God”, so if you are correct, then he has painted himself into this particular corner, and I cannot be fairly accused of forcing that kind of definition onto him.

    If “God” is a family-resemblance concept, then we might be able to proceed by identifying various sub-categories under this concept and provide necessary-and-sufficient-condition definitions for the sub-categories. The concept of “games” is a common example of a family-resemblance concept, but we can identify sub-categories of games: board games, card games, video games, sports games, etc. I suspect that it would only take a few sub-categories to cover 90% of the kinds of gods that people have believed in historically.

    (2) I agree. Also, I see that my claim about the limitations of science was too strong:

    science alone cannot answer the question “Does God exist?” if we use normative categories to define the word “God”.

    My statement above is incorrect. There is an interesting asymmetry between atheism and theism here. An atheist could, in theory, disprove the existence of God without having to get into the normative issue (i.e. “Is there a perfectly good person?”), but a theist must address the normative issue in order to prove the existence of God.

    The factual criteria in my “standard” definition could be shown to not apply to any being by proving there is no all-powerful person, or that there is no all-knowing person, or that there is no person who is both all-powerful and all-knowing. If just one of those universal negations can be proved, that would be sufficient to show that there is no God (in this sense). So, science alone could, in theory, be used to disprove the existence of God.

    But to prove the existence of God, a theist must show that there is a person who satisfies all three conditions, including the normative condition. So, science alone is not sufficient to prove the existence of God. However, I agree with your point that normative conditions can be analyzed into factual criteria and then science can take over from that point.

    (3) Another good point! When you subtract the creator-of-the-universe condition from Dawkins’ definition of “God”, what you are left with is the requirement that the being is “a supernatural intelligence”. If Zeus does not count as a supernatural intelligence, then Dawkins might well agree with you that atheism does not take a stand on the existence of Zeus.

    Why do you say that Zeus is a physical being?

  • http://www.blogger.com/profile/16415287407278685717 Lamar

    I say Zeus is physical because that’s the way the Greeks viewed him. (Perhaps I was wrong about this.) He may have had special powers, but he was certainly physical enough to mate with humans. He also lived in a physical location at a certain place and time.

    Zeus may not be physical; perhaps my history is wrong. But I could just as easily make my point with the Mormon version of God. He is a physical being, according to the Mormons.

    It seems to me that atheism would be more concerned with the type of God (such as the Christian one) who is said to be timeless, and outside of space. (I.e., “supernatural”) One that must communicate with humans in a “different,” “non-Physical” (whatever that might mean) way. In other words, one that is antagonistic to science. (Though, I only say it’s antagonistic because its meaningless – I’m kind of a positivist.)

    Science could posit the existence of Zeus (or the Mormon God) if it needed to. For example, if the data could only be explained by positing such a thing. (This is kind of like how no one has seen an electron, and yet scientists acknowledge its existence because such an acknowledgment is the only way to deal with the data received in the lab.)

    Often times atheist literature bothers me because it fails to acknowledge “sub-categories” of the types of deities, and I think it would be a worth-while endeavor to try to do that – both all the logically possible sub-categories, and all the actual ones; explaining what it would take to prove or disprove the existence of each type, and whether or not each type is susceptible to science. It would really systematize the philosophy of religion.