Still No Evidence for God?

In recent news, theism has been "devastated" (yet again) by someone reminding us that there is absolutely zero evidence for a supernatural being. (Just when we were this close to finally pulling off that long-awaited theocracy—foiled again by you meddling “no evidence” kids! Guess the jig is up. Pack it up, everyone!)'

Sarcasm aside, this presents a good opportunity to get back to basics—specifically, how people reason about God.

But first, I find the initial claim a little odd. For example, why wouldn’t someone be considered, you know, “stupid” for believing in God if there’s supposedly not only no reason to, but—as many atheists contend—positive reasons not to? Epistemically, they seem to equate belief in God with belief in Santa Claus or the Easter Bunny. (Funny thing: I actually think there’s some evidence for Santa; it’s just that there’s a lot more evidence against him.) Minimally, wouldn’t you want to say someone is acting irrationally for believing in those things—assuming they’re a fully formed, properly functioning rational agent? While I appreciate the (somewhat) friendly attitude, I actually like it more when people are straightshooters—if you think theists are stupid or irrational, just say so. Or maybe—as Mother as used to say—if you can’t say nothin’ nice, then don’t say nothin’ at all!

Alright, like I said: back to basics. Is there really no evidence for God?

Answering that question depends on how we define "evidence." Suppose, for example, someone says something like this (as certain science popularizers have said almost verbatim): "When we look through a telescope, we don’t see any bearded man in the sky, so we shouldn’t believe in God." This, apparently, is meant to show that there’s no evidence for God—because God doesn’t appear anywhere, even when we use powerful scientific instruments.¹ But this is obviously and deeply is confused. Just because we don’t see or detect God directly doesn’t mean there’s no evidence for God—it just means God isn’t a raw observation like a star or an amoeba.

Ironically, the same people who claim there’s no evidence for God because we can’t directly observe Him often argue that other raw observations—like the presence of suffering—do count strongly against the existence of God. Why? Surely because they assume that suffering is less likely under a theistic hypothesis than under a naturalistic one. This shows they understand the distinction between data (raw observations) and evidence (how data supports or undermines a hypothesis), but they apply this distinction inconsistently (only when it suits their pre-existing perspective, say) or haphazardly.

So, if one confuses evidence with data, then obviously there is no evidence for God, since God has traditionally been understood as the transcendent cause for the existence of the physical world—not as a floating bearded man or anything, for that matter, susceptible to direct detection. But that’s a silly understanding of evidence. Again, it confuses evidence with data. Data are just raw observations, while evidence is something that makes a hypothesis more likely. The fact that we don’t directly observe God doesn’t mean there’s no evidence for God, because evidence comes from how well certain facts (data) fit with the hypothesis of God’s existence compared to alternative hypotheses.

Here is a better understanding of evidence: Evidence is any fact that makes a hypothesis more probable than it would be without that information. In other words, something counts as evidence for a hypothesis if it’s more likely to exist or be observed if the hypothesis is true than if it were false. This means the evidence increases one’s confidence in the hypothesis, even if just slightly. Evidence is a form of support.²

Here’s an example, borrowed from Tim McGrew: finding a cup of steeping tea in a cabin is evidence that the cabin is not abandoned. Why? Because I wouldn’t expect to find steeping tea in an abandoned cabin. Importantly, I might not strongly expect to find tea in an occupied cabin either (what are the odds any given person is making tea?). Still, it counts as evidence because the tea is far more likely if the "cabin is occupied" hypothesis is true than if the "cabin is abandoned" hypothesis is true. It’s the ratio of expectations that matters. We can often detect a significant difference in likelihood, even when we can’t assign precise numerical probabilities. It's just obvious that a steeping cup of tea is more likely if someone occupies the cabin than if nobody does.

Notice that we don’t directly observe the person who lives in the cabin—they might be out for a walk or at the store. What we observe is something that is more likely to be present if the cabin is occupied. However, the raw observation—the data—only becomes evidence when considered in relation to how much it confirms a specific hypothesis. Again, raw observations or data points do not automatically qualify as evidence; they gain that status only when evaluated within the context of hypothesis testing.

Anyway, this difference in understanding—that many people seem to have a simplistic, even cartoonish, view of evidence—is why I often don’t like talking about “evidence” for God. Instead, I simply prefer discussing reasons for believing in God. After all, many arguments for God’s existence don’t fit neatly into the “this is evidence for God” category, but are metaphysical in nature, examining the necessary conditions that make various obvious features of the world—like change, composition, or becoming—intelligible. Still, I believe these provide powerful reasons to believe in God—in fact, some of the most compelling reasons—such as traditional cosmological arguments.

Nevertheless, with this better understanding of evidence in place, is it right to say there is absolutely no evidence for God? Of course not. In fact, many intelligent atheists admit there is considerable evidence for God—they just believe there’s more evidence against God. But to give a few examples, features of the world that philosopher have argued as evidence for God include physical fine-tuning, consciousness, rationality, moral knowledge, and, if you’re looking for something more exotic, the wave function.

On the face of it, these features seem far more expected if God exists than on any standard form of naturalism. After all, theism provides a reasonable probability that God could create an orderly world with a great chain of being, featuring conscious agents like us—capable of knowing and willing. Whereas naturalism (run by a principle of indifference) doesn’t.

I’m not going to dive into these arguments right now; I’ve written an entire book (you knew this was coming) that not only argues that features like fine-tuning are evidence for God, but that they are actually quite strong evidence. Before that, I explore other reasons for believing in God, including a contemporary cosmological argument. Ultimately, I think the best case for God is a cumulative one, so I pile up what I believe are the best reasons for believing in God while also addressing what I see as the strongest challenges to classical theism (problem of evil, coherence issues, etc.). The purpose of this post—sufficiently humble, I think—is just to have something to refer to whenever the “there’s absolutely no evidence for God” objection comes up (which I hear all the time, and I’m sure many of you do too). My hope—sufficiently naive, I’m sure—is that this will help move us beyond that tired canard and instead get us to focus on how strong the evidence really is for the various philosophical positions available today.

1. Now, technically, I guess one could say not seeing the bearded man in the sky is evidence against God insofar as one thinks theism predicts that we’ll encounter such an appearance—but that, of course, is a tremendously preposterous notion, and almost certainly not what people have in mind when they bring this silly objection up.

2. I say a form of support because not all support is evidence. That today is fall and that it often rains in fall supports the idea that it might rain today—but that isn’t evidence. If, however, I step outside and discover a large black cloud forming then that is evidence that it will rain today because such a formation is better expected on the “it will rain today” hypothesis than otherwise.