In these rough working notes, I explore Joe Carlsmith's concept of "deep atheism", meaning a fundamental stance of distrust toward the universe. While conventional atheism merely rejects the existence of God, deep atheism maintains that reality itself is fundamentally hostile or indifferent to human flourishing. (I discuss some ambiguities with the terminology.) I discuss how formative experiences -- from personal tragedies to encounters with nature's brutality -- can shift one's stance toward the universe, often at the level of emotional disposition and instinct rather than intellectual or scientific belief. I also explore "optimistic cosmism", a stance that recognizes and values the many a priori surprising properties that enable local order and meaning. Even if the universe does not care for us, it permits tiny physical systems like humans to understand and transform and improve their environment. This fact is very familiar to us, and it's tempting to take for granted, but rests on many profound and surprising facts about the universe.

I wrote these speculative notes to develop my thinking about deep atheism¹, and to explore questions like: do we live in a fundamentally friendly universe? One that looks out for us? Or a hostile universe? If so, just how hostile is it? Why does deep atheism matter, and what else does it connect to? My instinct is that it's an important idea – certainly, I'm surprised how often the idea now arises in my thinking, often in seemingly unrelated areas.

Unfortunately, when I mention deep atheism, people often object that while the implied question is interesting, the term is a misnomer ("that's not about atheism, it's about the character of the universe!"). Someone could believe fervently in God (and thus not be an atheist), and yet also believe in a mean or evil or just plain chaotic God. They'd be a "deep atheist" but not an atheist, as far as I can tell. Or perhaps they believe in a loving God, but a hostile universe, a sort of God-as-only-salvation model. These kinds of beliefs make the term deep atheism somewhat challenging, and are a reason I don't expect the term to catch on, despite the importance of the concept. Still, for this essay I'll use "deep atheism" as a working term².

Carlsmith doesn't precisely define deep atheism. Instead, he explores the concept through many examples, gradually developing the idea rather than engaging in premature definition. My understanding may thus differ somewhat from what he intended. Indeed, much that seems interesting (and perhaps profound) in the essay remains somewhat opaque to me. And while the concept of deep atheism connects to the vast philosophical and religious literatures on the problem of evil, theodicy, providence, and related topics, these working notes only make limited connections to that literature, as I develop my own understanding.

A naturalistic exploration of deep atheism

What might lead someone to believe the universe is fundamentally unfriendly? Through multiple examples, Carlsmith explores this question. How should one respond to losing a child to leukemia? Or to a house fire? Or to the way the ichneumon wasp lays its eggs inside a caterpillar, only for the larvae to slowly devour the living caterpillar from the inside out? He talks about the case of Timothy Treadwell, the "Grizzly Man", a very trusting (some might say naive) person who lived for several years with grizzly bears, often saying things like: "I love them with all my heart; I will protect them; I will die for them. But I will not die at their claws and paws." It was a very strong view of providence, of the universe being friendly toward him. He and his girlfriend were, unfortunately, eventually killed and eaten by one of the bears. As Carlsmith says, quoting Werner Herzog, Treadwell failed to take seriously the "overwhelming indifference of Nature".

These examples are of course not intended as scientific evidence in favour of (or against) deep atheism. If your child dies of cancer, the impact isn't to change your statistical picture of infant mortality rates. Rather, it changes you at a gut level, may even change your fundamental stance toward the universe. You may, for instance, come to believe the universe is a much harsher and more hostile place than you formerly believed. Carlsmith's essay focuses mostly at this level, at the fundamental stances or orientations people have toward the universe, and what kinds of experiences help form or shift those stances. It's not asking "how can we scientifically decide whether deep atheism is correct?" but rather a naturalistic exploration³ of "what are some of the ways in which people's stances are actually formed or changed?"

One example he gives is C. S. Lewis's tremendous anger at God in "A Grief Observed", the diary Lewis wrote after his wife was taken from him by cancer. This book has a lot of personal meaning for me, and I found myself very moved by the use of the example⁴. He quotes Lewis, a man who had placed God at the center of his life after converting to Christianity at age 32, coming to ask if maybe God is evil:

Come, what do we gain by evasions? We are under the harrow and can't escape. Reality, looked at steadily, is unbearable. And how or why did such a reality blossom (or fester) here and there into the terrible phenomenon called consciousness?… If H. [Lewis's dead wife] 'is not,' then she never was. I mistook a cloud of atoms for a person. There aren't, and never were, any people… No, my real fear is not materialism. If it were true, we – or what we mistake for 'we' —- could get out, get from under the harrow. An overdose of sleeping pills would do it. I am more afraid that we are really rats in a trap. Or worse still, rats in a laboratory… Sooner or later I must face the question in plain language. What reason have we, except our own desperate wishes, to believe that God is, by any standard we can conceive, ‘good’? Doesn’t all the prima facie evidence suggest exactly the opposite? What have we to set against it?

Again, this isn't a question of intellectual argument or scientific evidence. This is a man in tremendous pain going through an experience that may well change his fundamental stance toward God and the universe. In some sense, none of us is undecided about how hostile or friendly we default assume the universe is – we have a stance, patterns of judgement and response and action, based on some combination of genetic predisposition, culture and environment, conditioned response, emotional response, habitual patterns, and (often least important) – intellectually-grounded beliefs⁵. Carlsmith illustrates this with the question of what sort of spirituality can be left over, if the theist's God is dead, replying:

To see a man suffering in the hospital is one thing; to see, in this suffering, the sickness of our society and our history as a whole, another; and to see in it the poison of being itself, the rot of consciousness, the horrific helplessness of any contingent thing, another yet…. we could see many forms of contemporary “spirituality” as expressing a form of “existential positive.” They need not believe in Big-Man-God, but they still turn toward Ultimate Reality… with a kind of reverence and affirmation… Mystical traditions… generally aim to disclose some core and universal dimension of reality itself, where this dimension is experienced as in some deep sense positive – e.g. prompting of ecstatic joy, relief, peace, and so forth… Perhaps the Ultimate is not, as in three-O theism, explicitly said to be “good,” and still less, “perfect”; but it is still the direction one wants to travel; it is still something to receive, rather than to resist or ignore; it is still “sacred.”. The secularist, by contrast, sees Ultimate Reality, just in itself, as a kind of blank. Specific arrangements of reality (flowers, happy puppies, stars, etc) –- fine and good. But the Real, the Absolute, the Ground of Being –- that’s neutral. In this sense, the secularist repays to Nature, or to the source of Nature, her “overwhelming indifference.”

Carlsmith considers Eliezer Yudkowsky, one of the first people to take seriously the existential risk from artificial intelligence, as a prototypical example of a deep atheist, quoting him as saying: "You are not safe. Ever… No one begins to truly search for the Way until their parents have failed them, their gods are dead, and their tools have shattered in their hand." Carlsmith remarks: "This, I suggest, isn't just standard atheism. Lots of atheists find other `gods,' in the extended sense I have in mind. That's why I said `shallow atheism,' above. Deep atheism tries to propagate its godlessness harder. To be even more an orphan. To learn, everywhere, from the theist's mistake." And: "Thus: our most basic condition, presupposed almost by the concept of epistemology itself, is one of vulnerability. Vulnerability to that first and most fearsome Other: God, the Creator, the Uncontrolled, the Real. And the Real, absent further evidence, could be anything. It could definitely eat you, and your babies. Oh, indeed, it could do far, far worse."

An optimistic cosmist

Am I a deep atheist? Not exactly. I am an atheist – I don't think there is a God, in the sense of the Abrahamic and many other religions⁶. But I am what might be called an optimistic cosmist. If you consider the fundamental laws of physics, as currently understood, they don't obviously suggest friendliness. The universe in the large seems indifferent or even hostile to casual inspection. And yet we somehow live inside a relatively friendly and safe local enclave. Familiarity makes us take this for granted, but it depends upon the universe having many a priori surprising properties that enable tiny physical systems (like human beings!) to carve out safe and friendly local enclaves, despite that appearance of cosmic indifference.

Those surprising properties include things like: mathematics exists and can be expressed and explored physically by those same tiny physical systems, with astonishing results⁷; it's possible for those tiny physical system to develop simple but powerful scientific theories, and to understand and control and improve their environment, to make it friendly and meaningful; stable, self-maintaining, fault-tolerant, self-reproducing systems can emerge and persist and evolve into entities capable of the above⁸. None of these properties are obvious. We can easily conceive of universes which are incomprehensible and uncontrollable by any system within that universe. Yet the properties I mentioned above (and many others) mean we live in a universe where we can build surprisingly good approximations to utopias⁹, at least for limited periods and in limited regions like cities or planets. Where platforms for ambient friendliness – things like immune systems, legal and judicial systems, and military defense – are not only possible but potentially common. This combination of cosmic indifference and the universe's surprising amenability to local friendliness defines the optimistic cosmist position. It acknowledges the universe's apparent hostility while recognizing the unexpectedness and value of our ability to create meaningful enclaves within it. This is astounding, and it's interesting to ponder universes in which it's not true¹⁰¹¹.

Connecting deep atheism and optimistic cosmism to the Vulnerable World Hypothesis and Artificial Superintelligence

One reason I respond strongly to deep atheism is my interest in the Vulnerable World Hypothesis. This is a term introduced by Nick Bostrom¹², for the idea that (roughly) the technology tree of many civilizations will contain a technology that destroys that civilization. While we humans have survived technological advances so far, the universe offers no guarantee that technology will continue to be manageable. Indeed, arguably we're already sitting on a ticking time bomb, in the form of thermonuclear weapons.

If the Vulnerable World Hypothesis is true, and all or most emerging civilizations eventually destroy themselves, then it's a limit on the friendliness of the universe. Yes, the universe would allow us to briefly create friendly local enclaves. But the same properties that enable such enclaves also ensure that eventually we destroy ourselves. Believing the Vulnerable World Hypothesis is a particular form of deep atheism; and a particular denial of optimistic cosmism. It's also, as Carl Sagan observed in 1978, one possible explanation for the Fermi Paradox – why we don't see aliens everywhere, and (perhaps) why friendliness hasn't gone from local to near-universal:

Of course, not all scientists accept the notion that other advanced civilizations exist. A few who have speculated on this subject lately are asking: if extraterrestrial intelligence is abundant, why have we not already seen its manifestations?… Why have these beings not restructured the entire Galaxy for their convenience?… Why are they not here? The temptation is to deduce that there are at most only a few advanced extraterrestrial civilizations – either because we are one of the first technical civilizations to have emerged, or because it is the fate of all such civilizations to destroy themselves before they are much further along.

This is perhaps why deep atheism has come to mind so often for me over the last year. My habitual and ingrained optimistic cosmism is being tested by the prospect of existential risk from science and technology, especially from artificial superintelligence.

Deep atheism as science?

Carlsmith spends much of his essay exploring deep atheism as a natural phenomenon, and I've done that too. That is, asking about it as a question of psychological and cultural experience. But, of course, one can also treat it as primarily a scientific question: is it true? What evidence and argument could we give for or against it? What would it even mean to be true?

A great deal of work would be necessary to make it well formulated. Just as one tiny example, notions like "friendliness" and "hostility" are surprisingly tricky. They seem very natural to us, as human beings, but that naturalness is the outcome of a long evolutionary chain. Orientations of friendliness are present in many mammals and some other species, but are largely absent in many parts of the biosphere¹³. A shark doesn't feel much we'd regard as empathy or friendliness; much less so an amoeba. In this view, notions like friendliness and hostility seem like largely parochial concerns, a quirk of a particular evolutionary path on earth. The wasp-caterpillar parasitism seems terrible for us, but is it so bad in any other sense? Perhaps that's merely provincial thinking. Now, I certainly believe those notions could be made much less parochial. But it'd be tricky, and this is illustrative of the large tangle of issues that would need to be confronted to make "In what sense is deep atheism correct?" into a well formulated scientific question.

Concluding thoughts

Early in the essay I mentioned some challenges with the term "deep atheism". One thing I very much like about the term is that it emphasizes the relationship of the concept to religion. I gave a long list of ways we develop these kinds of stances toward the universe – genetic predisposition, culture and environment, and so on. An important addition to make to that list is religion and myth¹⁴, broadly construed: powerful, shared stories that help create cultures and shared identity, even a shared consciousness. In his book Dominion the historian Tom Holland has argued that ideas from the Christian gospels actually helped reshape the entire world's ideas of good behaviour, of what it means to live a successful life, especially the role of kindness and so on. It's interesting to ponder which myths carry deep atheism (or its opposite). And, in general, to ponder what stances can (or should) be carried by myth, and how. I've gradually come to believe these shared myths define a surprising amount about the outcomes civilization gets.

I wrote and released these notes with trepidation. It all seems a little galaxy-brained! On the other hand, it seems likely to me and many others that humanity is years or decades away from creating artificial superintelligence. And the friendliness of the universe is going to be tested in a new way. Some galaxy-braining is in order – we need some unusual ideas to rise to this extraordinary occasion!

Acknowledgements

This work is partly personal, but emerged naturally as an extension of my work with the Astera Institute, and I thank them for their support. Thanks to Fawaz Al-Matrouk, Mona Alsubaei, Joe Carlsmith, Ted Chiang, Laura Deming, Andy Matuschak, and Nicholas Paul for conversations about this subject.

Footnotes

Comments