Why I'm Still An Atheist (Though Philip Goff's Version of Christianity Sounds Lovely)
Some thoughts on philosopher Philip Goff's embrace of a "heretical" form of Christian theism.
Durham University philosophy professor Philip Goff often makes waves. I wrote about him last year, when he was tangling with Sabine Hossenfelder about scientific realism. I think he was right about that one. This time, he’s gone and converted to Christianity, and everyone’s talking about that.
A well-known philosopher embracing faith is more than enough to get people’s attention. What makes Goff’s case particularly interesting, though, is the form of Christianity he’s embracing, which he cheerfully refers to as “heretical.”
The philosophical debate about God’s existence tends to operate within relatively narrow parameters. Most philosophers these days are atheists or agnostics, and what the believing minority typically advocates is rock-hard “classical theism.” On this conception, God is definitionally “all-PKG” (all-powerful, all-knowing, and all-good).
The claim that an all-PKG being exists is very hard to defend, given the obvious objection that our world is full of things an all-powerful being would be able to stop and which an all-good being would presumably want to stop. To steal a line from my friend Mark, if I told you that an invisible giant walked around all day every day painting the sky yellow, your first response wouldn’t even be, “Where’s the evidence that?” It would be, “Why isn’t the sky yellow?”
A lot of brilliant thinkers have put all of their cleverness into trying to get around this issue (“the problem of evil”) but, however impressive the effort, the results are pretty underwhelming. At best, you end up with a position called “skeptical theism,” which is at its core an eloquent and sophisticated repackaging of the “God works in mysterious ways” hand-waving non-answer from the Book of Job.
Given that, you might think more philosophers would defend non-classical forms of theism that are less vulnerable to this objection. That’s generally how complicated long-running debates go. If one position has a lot of problems, eventually someone comes with a modified version to get around it.
The fact that this is a relatively uncommon move in “phil of religion” probably needs to be explained on sociological grounds rather than philosophical ones. A great many theistic philosophers, very much including rock stars like Plantinga and Craig, come from Christian colleges where doctrinal orthodoxy is highly prized. They are, in other words, more like the official theoreticians of Marxist-Leninist parties than they are like western Marxist academics who feel free to pick and choose which aspects of classical Marxism they defend.
Goff’s trajectory is very different. In his article in Aeon (“My Leap Across the Chasm”), he says that he was raised Catholic and left the faith as a teenager, for pretty standard reasons—a combination of ethical issues about various stances taken by the church (particularly keenly felt as he was questioning his sexuality) and standard intellectual worries like the problem of evil. The “heretical” Christianity he’s embraced surely comes with better sexual politics. And it also backs away from the “all-PKG” theology:
I remain as convinced as ever that the suffering we find in the universe is powerful evidence against the existence of a loving and all-powerful God. But… I now think the evidence points towards a hypothesis that John Stuart Mill took seriously: a good God of limited abilities. This hypothesis is able to account both for the imperfections of our universe – in terms of God’s limited abilities – and for the things about our universe that are improbably good, such as fine-tuning and psycho-physical harmony. God would have liked to make intelligent life in an instant, or by breathing into the dust as we see depicted in Genesis. But the only way God was able to create life was by bringing into existence a universe with the right physics that would eventually evolve intelligent life. God made the best universe they could.
This also allows him to get around what I think of as the “Hitchensian critique” of Christianity. The late Christopher Hitchens was often eloquent on the absurdity of (a) God punishing anyone for original sin—in his favorite metaphor, the God of Christianity “creates us sick and commands us to be well”—and (b) the idea that anyone can morally atone for someone else’s actions. Goff writes:
Many people assume the essence of Christianity is as follows. We are all sinners and so we deserve to burn in hell for eternity. Fortunately, Jesus took the punishment we deserve and, as a result, if we accept Jesus’ sacrifice on our behalf, we’ll go to heaven to live with God when we die. Everyone who doesn’t accept Jesus’ sacrifice will burn in hell forever. In fact, this is only one interpretation of Christianity, associated with the Protestant Reformation, although a similar view was defended by Anselm of Canterbury in the 11th century. It’s also, in my opinion, one of the most implausible theological doctrines in any of the modern global religions. I don’t think anybody deserves to burn in hell for eternity. And, even if we did, it wouldn’t achieve anything to punish an innocent man in our place.
I’ll pause to grumpily note that it’s a bit much to blame this picture on Anselm and the Protestants. The doctrine of original sin existed long before the eleventh century! But perhaps Augustine no less than Anselm was in error, and the right way to think about Christianity is entirely different.
Using the Hebrew “Yeshua” instead of the more familiar “Jesus,” Goff writes:
[A]ll Christians agree that Yeshua had some central role in the purpose of the universe. But there is no officially agreed view on the mechanics of that.
The views that are more plausible to my mind revolve around love and unity rather than sin and punishment. According to the participatory theory popular in the Eastern Orthodox Church, the Yeshua stuff was all about God becoming more similar to us so that we can become more similar to God. God wants us to share more deeply in their form of existence. But there’s a problem: the timeless, transcendent aspect of God is radically different from, say, a naturally evolved human being. Without God becoming more similar to their creation, the difference between God and creation is just too great for the two to share a common form of existence. The philosopher Robin Collins suggests this is analogous to the fact that ‘a tree branch cannot be grafted into a horse, only another tree; the horse is too alien for it.’ It is only once God, through Yeshua, shares in temporal, physical existence that the gap is bridged between God and creation, creating the potential for human beings and indeed the whole of creation to share more deeply in God’s form of existence.
This view still doesn’t make sense to me if we’re assuming that God is all-powerful. If God can do anything, then they could have created us to share in their form of existence from the beginning, rather than subjecting us to millions of painful years of evolution. But if God is not all-powerful, then maybe they are on their way to creating a perfect universe but are only able to do this in two stages. In the first stage, they create an OK universe, one with the right kind of physics to eventually evolve intelligent life. Next, when creation has evolved enough, God begins to bring the universe to perfection by becoming more intimately involved in it, sharing in its nature that it can share in their nature.
Perhaps this is a process that is still continuing – and maybe needs a bit of help from us – but which took a radical and decisive step forward in the events surrounding Yeshua.
Later, he summarizes this as the view that “Yeshua brought God closer to us not by being punished for our sin but by filling the entire universe with God’s love.”
While I’m a bit unclear on why the torture and death were necessary for this bringing-closer once we move away from the ‘punishing Yeshua as a substitute for punishing the rest of us sinners’ view, I certainly see the appeal of this version of Christianity. My own background means that if I ever became religious again, I would probably gravitate elsewhere, but his version of Christianity sounds…well…nice.
But he is presenting a philosophical argument, one that he claims at least renders his conclusion plausible enough for faith in it to be reasonable. So, as lovely as that conclusion may be, I’m still interested in exploring whether or not he’s earned it.
The existence of Goff’s God can’t be refuted with the problem of evil. But why should we suppose that even his limited deity exists?
Part of his case is about “the fine-tuning of physics for life,” which he describes as “the surprising discovery of recent decades that certain numbers in physics are, against incredible odds, just right for the emergence of life.” As I’ve discussed several times before, I’m not persuaded by this argument. That’s not to say that I don’t feel the intuitive force of it when I start thinking about the sheer scale of the numbers thrown around in such discussions. And, as someone doesn’t feel at home enough in the relevant physics to have an informed opinion about the premise, I’m happy to assume this is accurate. My issue is that, on reflection, I don’t think that the inference makes sense.
One way of seeing the point is to notice that it’s extraordinarily unlikely that I would exist. My parents had to reproduce with one another rather than any of the multitude of other people each one of them could have paired off with, the same is true of all four of my grandparents, all eight of my great-grandparents, and so on through hundreds of thousands of years of human ancestors and an unfathomable number of generations of non-human ancestors going back to the dawn of sexual reproduction in the distant evolutionary past. And all this is before we even notice that, at each link in this mind-numbingly long chain, the chances of any particular sperm-egg pairing (even holding the identity of the sperm- and egg-producers constant) was vanishingly unlikely. And yet none of this makes me even slightly inclined to believe in a Genetic Fine-Tuner who superintended this whole process to make sure that I would be generated at the end of it. Without starting out thinking my existence in particular is the likely goal of any Genetic Fine-Tuner, the fact that I exist doesn’t give me any particular reason to postulate one.
Similarly, if we do go ahead and grant the premise that the chances that the physical constants would happen to fall within a mind-bendingly narrow range that allows for the existence of life, this only counts as evidence for the existence of a Cosmic Fine-Tuner if we assume that any such being would be likely to care about biological life. But that’s arbitrary. Life feels special and important to us, but that’s just a different way of saying that it feels special and important to itself.
Why assume any entity capable of fine-tuning universes would share our priorities? If it turned out that there was a prime number of molecules in the universe, would we take this as particularly strong evidence for the existence of a powerful (but not all-powerful) God whose only goal was to make the number of molecules in the universe the largest prime number He could manage? (Perhaps whether life arose in one tiny corner of an otherwise lifeless universe was a matter of complete indifference to Him, but we were a side-effect of decisions ultimately guided by the need to ensure that prime number of molecules!) Why is this any more or less plausible a goal for an entity whose vast immaterial mind, if it existed, would presumably be very distant in its passions and priorities than our contingently evolved little monkey brains?
If someone shuffles the cards and deals me an Ace of Spades, even if that’s a very convenient card for me to have, this isn’t powerful evidence that, rather than shuffling randomly, the dealer rigged the deal to make sure I’d get it. Sure, there’d be only a 1/52 chance that I’d get the Ace of Spades as a result of a random deal and a far greater chance of getting that at as a result of a rigged deal. But precisely the same would be true if I got the 3 of Diamonds. There’s only a 1/52 chance I’d get it from a random shuffle and something like a 52/52 chance I’d get it from a deal pulled off by a competent cheater who wanted me to get the 3 of Diamonds. And as long as I’ve been hearing about the Fine-Tuning Argument, I’ve never understood what was supposed to change about the logic here if you make the deck so big that the chance of any card being dealt was not 1 over 52 but 1 over some number with a truly brain-melting number of digits.
Goff’s form of Christianity is “panentheist.” (While a pantheist simply equates God and the universe, the panentheist says that God’s being somehow includes the universe.) This seems like roughly the sort of thing people had in mind in a previous generation when they used phrases like “cosmic Christ consciousness.” And in the Aeon article Goff explicitly relates this to the position he was best-known for advocating before his conversion:
[T]here is a close fit with the philosophical theory I have spent much of my career defending, namely panpsychism: the view that consciousness goes all the way to the fundamental building blocks of reality. For panpsychists, the particles or fields that make up our universe have their own very rudimentary form of conscious experience, and the highly complex consciousness of the human or animal brain is built up from these more basic forms of consciousness. Panentheism is more at home in a panpsychist picture of reality, as it’s easier to make sense of the Divine pervading the universe if the universe is filled with consciousness than it is if the universe is a cold, unfeeling mechanism.
Panentheism may also help us to make sense of the idea of a God that is subject to limitations. If God had to create the universe inside themselves, then it could be that the timeless and unchangeable nature of God imposed certain limitations on what could be created. Perhaps the deep simplicity and unity of God’s nature ensured that creation had to begin with a very simple starting point – the Big Bang – and could only progress to complexity over time.
It makes sense to me that he sees these views as a natural fit. But I wonder if his panpsychism actually ends up undermining his assumption that any Fine-Tuner would be motivated to ensure that life could arise. We’re often told that God would want life to exist because He would want there to be beings who can commune with him.
But a version of Neil Sinhababu’s point that an all-powerful God could make electrons or black hole conscious and have them commune with Him seems to haunt Goff’s picture. His God isn’t all-powerful, but surely a being who was at least immensely powerful could fine-tune a universe whose already-conscious building blocks could come together to form more complex forms of consciousness (which could be capable of communing with Him) without having to fine-tune it for life.
Let’s put all my skepticism about the fine-tuning inference to one side. We’ll assume for the sake of argument that the best explanation of the physical constants being set where they’d need to be for life to arise in one impossibly tiny corner of the universe is divine intervention. It still seems to me that, if he wants to get from there to his Limited Panentheist Christ, as the Hitch would say, “all his work is still ahead of him.”
Let’s assume that the universe was fine-tuned for the existence of life, because this was the only way to bring about creatures with the right kind of complex consciousness. That still leaves open the question of whether the creator’s goal in aiming for such creatures to arise was to shower His divine love on them or to amuse Himself by torturing them. And if the limited nature of Goff’s God immunizes Him against the problem of evil, it equally immunizes Him against the problem of good.
My own view is that the improbably good and improbably bad things that exist are both the luck of the draw, and we just have to make the best of all of it. But if we’re going to start speculating that the shuffle is rigged, here’s the problem:
If the existence of genocide and intestinal cancer is a problem for the hypothesis of a morally benevolent creator, the existence of good whiskey and miniature schnauzer puppies chasing tennis balls on sunny days is a problem for a morally malevolent one. Limiting God’s power can solve the first problem, but I don’t see why it would be any less effective in solving the second. Goff wrote, remember, that his hypothesis of a good but limited deity “is able to account both for the imperfections of our universe – in terms of God’s limited abilities – and for the things about our universe that are improbably good.” God can’t get rid of genocide and cancer, but He’s doing His best to minimize such imperfections and fill the universe with His love. OK. Well, why would a bad but limited deity, who can’t figure out how to get rid of the schnauzers and sunshine and single malt, but does his best to fill the universe with death and suffering, be any worse an explanation?
Fortunately for my mental health, I see no reason to take the latter hypothesis seriously. Unfortunately for my prospects for sharing whatever joy Philip Goff has found in his new faith, the point cuts both ways.
Ben,
Your attempt to dissolve the fine tuning problem turns on a confusion between uninformative improbability and informative improbability.
There is a story about Richard Feynman coming to class and telling his students that he had just observed a miracle. He saw a car with the license plate ARW 357. As he enjoyed pointing out, the probability of seeing that plate was 1/26^3*10^3 (ignoring the question of whether the state had particular rules about the order of letters and numbers in license plates).
Now of course this is not miraculous at all because there is no way prior to seeing the license plate of specifying a macro-state that ARW 357 is a micro-instantiation of. It is only once Feynman sees the license plate that he can characterize ARW 357.
By contrast, if Feynman saw AAA 000, then that would indeed be surprising and would call for some special explanation. That is because there is a prior description of AAA 000 as the lowest possible value licence plate (taking A to be low and Z to be high). Of course, we have to be careful here or we get spurious miracles. (For starters, ZZZ 999 would have to be included in the class of instantiations of the macrostate that AAA 000 is also an instantiation of.) If we meet our local pharmacist on a trip to Disneyland, we have to be careful that we don't think of the prior description as "the local pharmacist" but "someone I know" since we would be just as surprised by the local baker and if we do that, then we might discover that there is nothing special to explain here.
So, there is nothing especially miraculous about the fact that the probability that someone would have my exact DNA is very low and I happen to have it. That is analogous to ARW 357. However, there is a need to explain the fact that I have a DNA that allows for a zygote to become an adult who can argue about the fine tuning problem. Natural selection provides that explanation, but it would be wrong to respond that there is just some way a sequence of DNA has to be.
The fine tuning problem arises because of all the ways the universe could be consistent with known physical laws, only a tiny minority are capable of having chemical reactions. This calls for either a designer explanation (usually theism) or a selection explanation (usually many worlds). It is not rationally acceptable to say the world just has to be some way, because there is a description (logically) prior to the instantiation.
Your argument that the designer hypothesis does not work because we don't know what the designer wants is also fallacious, because it confuses explanandum and explanans. A designer who wants there to be consciousness (or, from the panpsychist perspective, complex consciousness) is the explanation for why the knobs of the universe are so finely tuned. If the chances of this happening without either multiple worlds or a designer is 1/10^1000000000 (just to make up a number), then the probability that a designer would want finite conscious beings around just has to be better than this.
The problem with the cards analogy for the fine tuning argument is that for cards you know the underlying processes. For the universe, there is no reason to assume some big "deck of universes" and ours happened to get picked. Maybe this idea comes from the fact that string theory is consistent with a gargantuan number of combinations of physical constants. The multiverse people say: actually all of those universes exist and we are in one with constants that support life. Yes, we won the lottery, but some universe had too! Goff rejects this and assumes some sort of creator for reasons I don't understand (I haven't read his book). But, I don't see any reason to assume a "deck of universes" with a random dealer. That presupposes a probabilistic process for picking universes that precedes the universe, but why would that process exist?