What Isn’t Designed?

“If theists believe that everything is designed, then how could they possibly identify design in the first place? What would they be comparing it to?”

This question comes up constantly, especially online. It’s even been asked in public debates. For example, it was posed to Frank Turek by the atheist Danny, the same atheist who debated Jordan Peterson on Jubilee and caused him great embarrassment over whether he was a Christian or not. You can watch their exchange here:

Danny’s challenge to Frank Turek was this: if theists believe everything is designed, how can we ever tell that something is designed rather than not designed?

After speaking privately with Danny to make sure I have the strongest version of what he’s getting at, he said this:

“The idea is that if God designs everything in the same fundamental sense, then no particular feature of the world can be more evidential of design than any other feature. And if no feature is more evidential than any other, then empirical design arguments can’t really get off the ground, because the evidence can’t discriminate between design and relevant alternative explanations.

In other words, on a very minimal or “bare” form of theism, design might be true, but it would not be something we could justifiably infer from the world.”

To be fair, I understand what Danny and many other atheists are trying to get at. There really are cases where epistemic detection depends on contrast. A good example is forgery detection. You normally identify a forged document by comparing it to genuine ones. Without any authentic documents, it becomes extremely difficult to tell whether a document is fake or not. In that kind of case, contrast with a reference class really does play a vital epistemic role in determining what’s fake and what’s real.

So I don’t believe this objection is coming out of nowhere.

But I also don’t thing the objection succeeds, and this is why…

Disqualifying the Person Is Not Disqualifying the Inference

Let’s grant Danny’s entire argument for the sake of discussion.

Suppose that Danny is right and given a theist’s background belief that everything is designed, a theist is not entitled to infer design from particular features of the world. In other words, suppose that the theist’s worldview really does disqualify the theist, as a person, from being able to make a legitimate design inference.

Even if we grant all of that, it still would not follow that there is no design inference.

Here’s what I mean.

What the theist is trying to claim is simply this: “some things in the world are clearly designed, and the best explanation of that fact is that the world has an intelligent designer.”

That’s the claim.

Danny’s response is not attacking that claim directly. Instead, his response is aimed at the person making the claim. He’s saying, “But you’re a theist. That means you believe everything is designed. And since you believe that everything is designed, you’re not in a position to infer that this thing is designed rather than that thing.”

Now let’s suppose Danny’s right and the theist, given their worldview, can’t legitimately infer design.

Even if that were true, it would only tell us something about the epistemic situation of the theist. It would not tell us anything about whether the design inference itself is true or false. And it would not tell us anything about whether design is in fact inferable.

At most, it would tell us that this particular group of people—theists—given their background beliefs, are not well positioned to make this inference.

It wouldn’t show that the inference itself is illegitimate.

To see why, consider an analogy.

Imagine there’s a group of people who believe that every human being has cancer at some microscopic level. Everyone has cancer in their body, even if it’s not detectable by ordinary medical tests. That’s simply part of their worldview.

These people regularly point to scans, symptoms, and medical evidence and say, “This person clearly has cancer,” and “That person clearly has cancer.”

Now suppose Danny comes along and says to them, “Can you point to anyone who doesn’t have cancer?”

They can’t. Their worldview says that nobody is completely cancer free.

Now imagine Danny concludes from this, “Then no one’s in a position to identify cancer, and cancer doesn’t exist.”

That wouldn't follow.

Even if we granted that their worldview disqualifies them from identifying the concept of cancer in the way most doctors do, it would still not follow that they’re wrong in saying that some people clearly have cancer. And it would also not follow that cancer doesn’t exist, or that cancer can’t be detected, or that there are no legitimate medical criteria for identifying cancer.

At most, what would follow is something much weaker. It would follow only that these particular people, given their own background beliefs, are not in the best position to draw certain diagnostic distinctions.

But that doesn’t undermine the reality of cancer. Nor does not undermine the legitimacy of diagnosing cancer.

In exactly the same way, even if Danny were right that a theist’s background belief disqualifies them from inferring design, that wouldn’t undermine the design inference itself. It wouldn’t show that the theist’s claim is false. And it wouldn’t show that design isn’t rationally inferred in the world.

It would only show, at most, that theists, given their personal worldview, aren’t well positioned to make this sort of inference.

And that’s a completely different claim.

There’s an important distinction here between a claim being true and a person being in a position to make that claim.

A claim can be true or false independently of whether a particular person’s worldview allows them to consistently or coherently affirm it.

To see why, suppose, for the sake of argument, that atheism really does undermine the existence of objective moral values and duties. Now imagine an atheist says, “It’s actually wrong to rape.”

Someone like myself might reply, “Your worldview doesn’t allow you to make objective moral claims like that.”

Notice that even if that criticism were correct, it wouldn’t follow that what the person said is false. It wouldn’t follow that rape is not wrong. It would only follow that this particular person isn’t in a position to justify that claim within their own worldview.

The truth of the claim and the coherence of the person making the claim are separate issues.

The same applies here.

Danny’s argument, even if it were completely correct, wouldn’t touch the actual design argument at all.

At most, it would remove the theist, as a person, from being able to make that argument.

But the argument itself would remain untouched.

Danny’s objection, even if it succeeded, doesn’t even address the actual argument being made, it only attempts to get the person speaking to stop making it because of their personal background beliefs.

The Objection Gets the Direction of Reasoning Backwards

There’s another serious problem with Danny’s objection, and it has to do with the direction of the inference.

The way the objection is framed makes it sound as though the theist starts out by assuming that everything is designed, and only then tries to interpret the world through that assumption. From that Danny says that the theist can no longer recognize design, because they already believe that everything is designed.

But what if that gets the order of reasoning backwards?

What if the reason someone believes that everything is designed is precisely because they first learned how to recognize design?

In other words, what if the universal conclusion comes at the end of the reasoning, not at the beginning?

To see the problem clearly, go back to the cancer example.

Imagine a group of people who now believe that every human being has at least some amount of cancer in their body. But suppose they didn’t begin with that belief.

Instead, suppose they began in exactly the normal way. They studied clear and obvious cases of cancer. They learned how cancer works. They learned what tumors look like. They learned how scans, biopsies, and symptoms reliably indicate cancer. Over time, they developed very good methods for identifying cancer in particular individuals.

Only after doing all of that did they discover there’s something in the air that all humans breathe that gives them cancer to one degree or another.

So now they believe everyone has cancer.

They didn’t begin with the belief that everyone has cancer and then reason their way to particular diagnoses. They began by learning how to identify cancer in clear cases, while it was still an open question whether everyone had cancer or not. The universal conclusion came later, as a result of that investigation.

Now suppose someone comes along and says to them, “Well, because you believe that everyone has cancer, you’re no longer allowed to say that this person has cancer rather than that person.”

That would be deeply confused.

Their belief that everyone has cancer isn’t what allowed them to recognize cancer. It is what resulted from their ability to recognize cancer in the first place.

The universal claim is downstream from the local inferences, not upstream from them.

The exact same structure applies to design.

It’s simply not true that the theist must start by assuming the world is designed and then try to interpret the world accordingly. Many people begin in a position of genuine uncertainty about whether the world is designed or not. They look at features of the world. They study order, information, functional organization, and purposive structure. They learn what intentional systems look like, in ordinary human contexts, in technology, in communication, and in engineering. And on that basis, they conclude that some things in the world are very strongly indicative of intelligent agency.

Only after that do they arrive at the broader conclusion that the world itself is designed.

So when Danny says that the theist can’t recognize design because the theist believes that everything is designed, he’s ignoring the fact that, for many people, the belief that everything is designed is the conclusion of the very reasoning he’s now trying to disqualify.

In other words, the objection assumes that the theist is reasoning like this:

The world is designed.

Therefore, X looks designed.

But very often the reasoning actually goes in the opposite direction:

X looks designed.

X over here also looks designed.

Therefore, the best explanation is that the world itself is designed.

Once that conclusion is reached, it will of course follow that everything in the world exists within a designed system. But that doesn’t erase how the conclusion was formed in the first place.

So the objection fails against this kind of person entirely.

Even if Danny were right that someone who already assumes universal design couldn’t legitimately infer design from particular features, that would still not apply to someone who first learned how to recognize design while the question of universal design was still open.

And that’s a very common path to theism.

Finally, notice the deeper point.

If there really are some things in the natural world that are best explained by intelligent agency, then it follows that our world, as a whole, is the kind of world in which intelligent agency is at work. And if that’s true, then the conclusion that the world has a designer is not an arbitrary assumption. It’s exactly what the local design inferences imply.

Misunderstanding How We Infer Design

There’s an even deeper problem with this objection—it assumes that the only way to infer design is by comparing designed things to things that aren’t designed.

But that’s not how design inference works in practice. Sometimes you do need to have contrasting items to rationally make an inference, like in the case of forged documents. But design isn’t like forged documents, we don’t need negative cases of non design to spot positive indications of design.

To see why, first consider consciousness.

Imagine you’re looking at a statue of a man sitting on a bench. The statue is completely still. Someone asks you, “Is the statue conscious?” Of course, you would say no. But why? You have no reason to believe it is. You can’t say for sure though, it’s possible. We can’t prove the statue has no consciousness, we just don’t have any reason yet to think it is.

But now imagine the statue moves its finger. That would be strange. You might hesitate.

Now imagine the statue starts tapping its finger in Morse code. You decode the message and it says something like, “Hello. I am conscious. Can anyone hear me?”

The inference towards consciousness is now much stronger.

And what if the statue turns toward you and begins speaking. It understands your questions. It responds appropriately. It engages in conversation.

At some point, the correct inference is that the statue is conscious, or at the very least that it’s being controlled by something that’s conscious.

Notice also...

You didn’t infer consciousness from the statue by comparing the statue to something you knew for certain was unconscious. In fact, strictly speaking, we can’t prove that anything is unconscious.

Instead we can infer consciousness from ourselves. We know what it’s like to be conscious. We know how conscious beings behave. When we see those same indicators in other beings, even in objects, we infer consciousness.

That inference doesn’t depend on contrasting conscious beings with unconscious ones. It simply depends on recognizing positive indicators.

Design Doesn’t Require Contrast Either

For Danny’s argument to work, it has to assume that when theists say something is designed, they’re doing so by contrasting it with something that’s not designed.

But that’s not what theists are doing. And more importantly, it’s not what they need to be doing.

The contrast involved in a design inference is not a contrast in ontological status. It’s a contrast in epistemic status.

In other words, the question isn’t whether some things are designed and others are not designed. The question is whether some things display clear indicators of intelligent agency and others don’t.

We’re drawing an inference, not doing metaphysics. Not yet anyway.

A theist doesn’t need to be able to point to something that is metaphysically undesigned in order to recognize design. All they need is to be able to distinguish between things that are clearly designed and things that are not clearly designed.

This is exactly how most inference works in ordinary life.

A detective doesn’t need to find a human being who was not murdered in order to recognize that a murder has taken place. Technically speaking, it’s possible that every human being who has ever died was murdered in some way, and all murders were just made to look like accidents or natural causes. It’s possible. But even if that were true ontologically, it wouldn’t prevent a detective from recognizing obvious cases of murder from cases of murder that looked accidental and natural.

All the detective needs is to know what the indicators of murder look like. Blood spatter patterns. Defensive wounds. Signs of forced entry. Gunshot residue. Witness testimony. Surveillance footage, etc.

The detective learns those indicators by studying clear cases of murder.

The contrast is not between murdered people and non-murdered people. It’s between scenes that clearly exhibit the indicators of murder and scenes that don’t clearly exhibit those indicators.

The same inference applies to consciousness.

If a statue suddenly turns toward you, answers your questions, and carries on an intelligent conversation, you don’t need to have first identified a large class of objects that are definitely not conscious. The positive indicators are enough.

And notice something important here.

The fact that the statue may have sat perfectly still for a very long time doesn’t override the later positive evidence. The absence of indicators earlier doesn’t defeat the presence of strong indicators later.

The same is true of design.

The theist doesn’t need to say, “Here’s something that’s not designed, and here’s something that is designed.”

What the theist needs to say is, “Here’s something that clearly exhibits the kinds of patterns, structures, and organization that we know intelligent agents produce and we know what those patterns look like because we design things ourselves. We know what it’s like to form plans, encode information, organize parts for an end, and build systems to achieve goals.”

So when we encounter information, functional organization, purposive structure, and coordinated complexity in the world, we’re not reasoning by contrast with metaphysical non-design. We’re reasoning by recognizing positive indicators of intentionality.

That’s why some design inferences are stronger than others.

Some things display the indicators of intelligent agency very clearly. Other things don’t.

A clump of sand doesn’t clearly display purposive organization. A sandcastle does. A meaningless pattern of scratches doesn’t clearly communicate anything. A sentence does.

So even if someone believes that the world is designed, they can still correctly distinguish between things that are obviously designed and things that are not obviously designed.

The entire force of Danny’s objection depends on confusing two very different contrasts.

A contrast between designed and undesigned things, and a contrast between clearly designed and not clearly designed things.

But only the second contrast is required for a design inference.

Once that distinction is made, the objection no longer gets off the ground.

Conclusion

I really do appreciate Danny taking the time to work through this with me in private messages. I made a real effort to represent his argument in the way he himself thought was strongest. I also appreciate the fact that this is not a shallow objection. It’s something atheists raise with me all the time, and it’s worth answering in its best and most careful form.

So here is why, even in its strongest version, I still don’t think the objection works.

First, even if we granted Danny’s central claim for the sake of argument, namely, that a theist is not entitled to infer design because the theist already believes that everything is designed, that would still not show that design can’t be inferred.

At most, it would say something about the epistemic position of only the theist. It wouldn’t show that the design inference itself is illegitimate. It wouldn’t show that the features appealed to in design arguments fail to support a design explanation. And it wouldn’t show that non theists couldn’t rationally infer design.

Disqualifying a person from making an inference is not the same thing as disqualifying the inference itself.

Second, the objection assumes the wrong direction of reasoning.

It treats the theist as though they begin by assuming that the universe is designed and then try to read design into everything afterward. But in many real cases the reasoning runs the other way.

People sometimes begin in a position of genuine uncertainty about whether the world is designed. They study clear cases of intentional systems. They learn what the indicators of intelligent agency look like. They notice those same kinds of features in the world. And only then do they draw the broader conclusion that the world itself is designed.

So even on Danny’s own framing, the objection doesn’t apply to the way many people actually come to theism in the first place. The global conclusion is downstream from the local inferences, not the other way around.

Third, and most importantly, the objection misunderstands how design is recognized and inferred.

Design is not inferred by first finding things that are metaphysically not designed and then contrasting them with things that are designed. The relevant contrast is epistemic, not ontological.

We distinguish between things that clearly exhibit the indicators of intelligent agency and things that do not clearly exhibit those indicators.

We don’t need negative indicators in order to recognize positive ones.

This is exactly how we recognize other minds, and it’s how we recognize things like murder. We don’t need to first identify objects that are definitely unconscious in order to recognize consciousness. And a detective doesn’t need to find a body that was definitely not murdered in order to recognize that a murder has taken place. What matters are the positive indicators.

The same is true of design.

As humans we know what it’s like to design things. We know what kinds of patterns, structures, and organizations intentional agents produce. When those indicators are present, the inference to agency is justified. When they’re not, the inference is weak or absent.

That is why some design inferences are much stronger than others.

And once that’s clear, the demand to point to something that’s not designed turns out to be a misunderstanding of how design is recognized in the first place.

So in the end, even in its strongest form, I don’t think Danny’s objection undermines the design inference. It either targets the wrong thing (the person rather than the argument), assumes the wrong direction of reasoning, or builds in a mistaken view of how intelligent agency is detected in the first place.

In short, asking “what isn’t designed?” sounds clever, but it rests on a mistaken picture of how we actually recognize design.