Via Hemant at Friendly Atheist comes a story on the work of Oxford psychologist Olivera Petrovich, who claims in a recent interview that her research has shown that the concept of God is essentially endemic to toddlers, while atheism has to be learned later on. She bases her conclusions on several cross-cultural studies, primarily relying on Japan as a cultural foil to Western theism. Since Japanese culture (by her characterization) “discourages” metaphysical speculation and the idea of a God as a creator, finding children instinctively leaning towards a God-like being as the cause of natural things supposedly implies that children instinctively believe in a God.
As one blogger puts it: Atheism is definitely an acquired position.
Or is it? The main problem I have with her reasoning is that Petrovich seems to conflate the idea of “inherent belief in God as a developmental stage” with “an idea that’s very likely to occur to someone if they are confronted with a particular question.”
That is, she doesn’t actually present any evidence that most, let alone all, children who are not exposed to theistic beliefs as a normal practice, go around regularly and actively believing in God (i.e. seeing a dog, and always then thinking “oh, God made that”) Rather, her research seems to imply that many children will, when presented with the question of ultimate origins, eagerly jump to the offerred conclusion that a powerful, psychological entity would be behind otherwise inexplicable events and causes.
That’s not really the same thing at all.
There are all sorts of ideas that seem highly likely to occur to people out of the blue (i.e. what if the color red to me isn’t the same as red to you), or else are highly likely to occur to someone when asked about a subject they are otherwise unfamiliar with (i.e. eating lots of sugar causes hyperactivity). In the case of God, we may well even be intrinsically inclined to both think of and prefer the idea of a psychological being as the cause of otherwise inexplicable events, simply because we appear hardwired as obsessively social to see everything in terms of personalities and motives.
But that’s still only if we happen to seriously consider the matter as toddlers, and consider it in precisely the same, artificial way that the researchers presented it. There’s no real evidence here that toddlers naturally frame questions that way for themselves, and thus naturally get to the place where they would, on their own, all come up with the idea of a God in response. Indeed, toddlers’ chains of “why?” questions tend to peter out pretty quickly due to boredom when no final “God” answer is given to them.
In short, even if the research here was flawless, it wouldn’t be enough to show that atheism is, as some want to use Petrovich’s work to claim, always “an acquired position.” There are plenty of reasons to think that the sort of theistic assumptions Petrovich references wouldn’t come up until later in life, at which point people might not be so inclined to accept God as an answer.
Furthermore, even if all toddlers naturally thought of the idea of God without provocation, and naturally preferred it, this would still only be evidence that they naturally acquire theism at some point, not that atheism can only be acquired “later.” The lack of a God belief could (and, in my opinion, would) still be the default state prior to the allegedly unavoidable idea of God coming onto the scene.
Given these problems of interpretation, the methodology here is doubly suspect: the studies used forced choice questionnaires in which the children were asked to speculate: with only “people did it” “God did it” and “don’t know” as responses. But when you press a child to answer, especially in the context of a seeming test on a matter that most have likely never considered before, you’ve already essentially downplayed the attractiveness of the “don’t’ know” category (which toddlers may in any case be naturally disinclined to admit to). And the children in question already seem capable of understanding that some objects (natural ones) were not made by people.
This leaves only the vague identity with unknown but potentially relevant abilities as the remaining answer (and while Petrovich talks about Japanese kids not seeing “God” as a creator, Japanese kids are exposed to ideas of spirits and magic creating and doing things, which leaves “God” the only available option amongst the three that’s close to these beliefs). In this sense, the researchers are not so much finding things out from children: they are implicitly, and perhaps inadvertently making an argument for God using the very sort of forced choices and leading questions that creationists often employ. How far off is the study design here from the question begging of “Who is responsible for creation: the creator?”
Then there’s the problem with her talking about “children” as if her findings were intrinsically universal to all children. Yet, when discussing the forced choice results, for instance, she says that they “predominantly” went for God. That doesn’t sound like even the forced choice resulted in any sort of universal preference for God explanations in children, regardless of the cross-cultural similarities. The headlines aren’t reading “most children assume that a God-like being causes things” but rather that “children” do so, period. The null hypothesis here is that God concepts come from external ideas and trains of thought rather than being inevitable aspects of human intellectual development. Thus, anything less than an across the board preference for God seems to leave the alternative wide open.
This weakness is further reinforced in the following passage from the interview:
When you press them, they of course fall back on what they’ve been told, saying things like, “I know he’s a man because I saw him on the telly,” or “He’s just like my daddy.” These are very rational responses, but they’re not natural conceptions formed by children. Rather they’re imposed by the culture in which the children live.
As even Petrovich notes, these are essentially culturally based ideas about God coming from these same kids (albeit seemingly just the English ones). The difficultly here is that you can only question children once they can speak, and yet they learn to communicate by interfacing more and more with the culture around them. A culturally pure child would almost by definition be communicatively mute.
So while these studies are interesting, and could certainly lead to a more complex model of how our causal and metaphysical beliefs tend to develop in modern societies, I just don’t see the warrant for declaring much about anything when it comes to atheism here.
In any case, I think most theists (and some atheists) entirely misunderstand the reason we talk about infants as atheists. It’s not a matter of trying to beef up “sides,” but rather explain what it means to not believe.
But that’s for another post.