Erik's Reviews > Exhalation

Story-for-story, Ted Chiang is the most award-winning sci-fi writer of all time. No one else even comes close.

Part of it is due to his simple, yet elegant writing style. Another part is due to his ability to take complex ideas and reframe them in a way that makes them easier to enjoy and understand. But I think his most important trait is that he doesn’t moralize. He has no interest in telling the reader WHAT conclusion to draw. Rather, his stories are arenas of thought, in which you are invited to challenge your thinking and draw your own conclusions. Like a good magician, Chiang provides the wonder but not the explanation.

What follows are meant for post-reading ponderance.

The Merchant and the Alchemist’s Gate

One part Arabian Nights, another part Canterbury Tales, a third part Back to The Future made for such a delightful cocktail.

Time travel stories that concern themselves with the actual logic and causality of time travel invariably arrive at the same conclusion: Maintaining coherence in the face of backwards time travel requires the past to be immutable and destiny to be fixed. Such is the case here in Merchant and Alchemist’s Gate. As the Alchemist says, “Coincidence and intention are two sides of a tapestry, my lord. You may find one more agreeable to look at, but you cannot say one is true and the other is false.”

But Chiang (via the Merchant) makes the interesting point that even if you cannot alter the past, you can nevertheless alter your understanding of it.

We have a tendency to learn from experience but to then place that experience in stasis in our memory. Yet, days or decades hence, we might revisit our memories and experiences and revise our understanding of them. Indeed, failure to do so is to risk ossification. I am reminded of perhaps the most important lesson I ever learned from literature, from Anne Rice’s Interview of a Vampire:

How many vampires do you think have the stamina for immortality? They have the most dismal notions of immortality to begin with. For in becoming immortal they want all the forms of their life to be fixed as they are and incorruptible: carriages made in the same dependable fashion, clothing of the cut which suited their prime, men attired and speaking in the manner they have always understood and valued. When, in fact, all things change except the vampire himself; everything except the vampire is subject to constant corruption and distortion. Soon, with an inflexible mind, and often even with the most flexible mind, this immortality becomes a penitential sentence in a madhouse of figures and forms that are hopelessly unintelligible and without value.

To become fixated upon the past, to never revise your understanding of it but to instead worship at the Janusian altar of nostalgia and regret, is to voluntarily incarcerate yourself into such a madhouse.

Exhalation

One day I was bored, so I invented Erik's Law, connecting people running cars in parking lots to Fermi's Paradox.

2nd Law of Thermodynamics is such an ugly name for what is, basically, the core flow of the universe: order goes to chaos. You can create local pockets of order, but you can only do so by stealing from another reservoir of order. In our case, that reservoir is mostly the sun. Eventually, though, there will be no more reservoirs and the Universe will know Heat Death.

But as I’ve grown older, I’ve come to realize it’s a mistake to limit the 2nd Law to only large-scale non-intelligent systems. It can provide great insight into human affairs. The daily struggle to maintain a sense of peace, the cycle of fallen civilizations, climate change, Fermi’s paradox - it’s all related to the 2nd Law of Thermodynamics.

Unfortunately, the 2nd Law is wrapped up in all these technical terms - entropy, free energy, heat - that can make it hard to understand. That’s the genius of Exhalation: by reframing the 2nd Law in terms of air and air pressure, with a universe trapped in an air-tight dome, populated by robots whose brains are powered by air, Chiang grants novelty and clarity to the idea.

The story ends on an optimistic note, as our robot-scientist narrator makes peace with the doom of his civilization. It’s a feeling that we are all going to become increasingly familiar with, as climate change continues to ramp up in the face of our ineffectual response. Many of us will liken it to the acceptance of our own individual deaths - we all know our time here is limited. This is just that, on the civilization level. The religious will, of course, embrace their doomsday prophecies and find comfort in those self-deceptions. Others will rage against the dying of the light, finding value and purpose in the struggle, even if it is futile.

There is so much more to Exhalation, of course. The scientist dissecting his own brain to discover the secrets of the universe is a fun metaphor. And the realization that our “self” is not the brain or the body but rather a transient electro-chemical pattern contained within. And that line equating the Big Bang to a great initial Exhalation. Wonderful, wonderful.

What's Expected of Us

This very short story asks the question, “Is a belief in free will necessary for the functioning of society?” It’s not a philosophical question but rather a psychological one. Chiang answers this question with a massive YES.

In the story, people are demonstrated the non-existence of free will with a simple device: a button that lights up BEFORE you press it. And try as they might, there’s no way around it. People can never fool it. Over-time this degrades everyone’s sense of responsibility and agency, such that large swathes of the population enter an apathetic malaise. The story’s solution - which comes in the form of a message from the future - is simply “pretend.”

That’s where the story falls apart for me. It’s so Western-centric, embracing a strongly individualistic worldview in which every person is an island, ultimately isolated in the universe/ocean.

But there is another solution, encoded in the famous Zen koan "What is the sound of a one-handed clap?": stop thinking of yourself as entirely separate from the world and universe around you. A button that lights up before I press it would prove no problem for me because I don’t consider that button as entirely separate from me. Yes, of course, I have a body distinguishable from that button - but it’s not so sharply different in terms of things like free will, determinism, electric and magnetic fields, flows of knowledge and entropy, the weft and weave of the Universe.

So my view wouldn't simply be that I am pushing a button; rather, by doing so, I am freely participating in the pre-pushing powering of the light. That I am unable to push the button without also participating in the pre-pushing powering of the light represents no more a loss of my free will than does the fact that I cannot randomly decide to ignore gravity and leap into space.

If you’re interested, I write in much greater depth about this view of free will in the bottom third of my Echopraxia review.

Lifecycle of Software Objects

In one of sci-fi’s most famous memes, the Hitchhiker’s Guide to the Galaxy involves a super AI whose answer to the question, “What is the meaning of life?” is the number “42.”

But we might substitute that “42” with something like “to be happy” with minimal loss of vagueness. Douglas Adams didn’t do that because it’s too subtly absurd - too many people wouldn’t get the joke.

I bring this up because I’m pretty sure our interactions with future AI will be similar:

“Dear Oracle Super AI, how do we save the world from climate change?”

CALCULATING… BEEP BOOP BEEP. ANSWER: “Stop using fossil fuels.”

Oh, right. We already knew that. In fact, we already know the answers to most societal ills. It’s not the knowledge we lack. It’s the collective will and/or flexibility and/or selflessness to actually put that knowledge into practice.

In a way, that’s what the Lifecycle of Software Objects is all about: how to create the holy grail of AI: a generalized human-esque intelligence. But as this story gets into… we don’t actually want that, do we? We want a generalized intelligence that is NOT human-esque, so we don’t have to give it human rights. We want one we can put to work in factories and banks and so on, without having to treat it as a human employee. We want intelligent slaves who won't object to their slavery.

And if we did want a human-esque generalized AI? Well, why? We already have a human-esque generalized intelligence... humans, lol.

If this more subtle, realistic approach to AI is new to you, you'll probably like this story more than I did.

Dacey’s Patent Automatic Nanny

Don’t take this story literally. That’s true of every Ted Chiang story, which are rich in allegory and metaphor, so I don’t know why I was so dead-set on reading this one literally. But I did, and I missed the point.

In this story, a Victorian semi-mad scientist creates a mechanical automaton to serve as his child’s nanny. Years later, the son ends up doing something similar. But the story isn’t meant as a machine-version of Tarzan. Rather, it’s exploring the idea that the environment in which children are raised tends to become their preference. That’s where that old joke that people tend to marry their mother/father comes from. Or, less humorously, why so many children raised by abusive parents can end up with abusive partners too.

More particularly to this topic, it asks us to consider the fact that today’s children are, at least in part, being raised by the “machines” of screens and social media. What will that mean for their future?

Even before the pandemic, face-to-face interaction was already dwindling. I had a long running joke with my girlfriend that every time we went out to eat, you could always find a family or a set of friends who basically never interacted but instead looked at their phones for the entire meal.

I only see this becoming increasingly common and emphasized. Lament it, if you will, but I’ve always believed in turning my face to the future and not the past. You can find cause for optimism in this change, if you want to.

Truth of Fact, Truth of Feeling

This is one of my favorite stories in the collection, probably because it explores the role language and writing play in our human experience. And writing is a serious hobby of mine. Not just with these book reviews; I’ve also written two novels, probably about 30 short stories, countless essays. I even had a “writing apprentice,” who is now getting her masters in writing and working on her first book.

So, yeah, big fan. I love that writing allows the writer to condense a lot of work and thought into an extremely rich piece of language. Like a precious gem. Or a depleted uranium round. There’s an impressive contrast between the time to create and the time to consume. A poem that might’ve taken 30 hours to construct can take less than a minute to read. Which is different than other uses of language, like conversation. For me, spoken conversation can be a pleasurable social activity but I tend to find it intellectually dissatisfying.

But getting to the story: Truth-truth is a split narrative, one of which follows a native from a tribe with a strong oral tradition as he learns his letters from a Christian missionary… and how this learning changes him. That’s the part centered on language. The other (main) narrative is set in a future in which most people video-blog every moment of their lives. A journalist, who is a tech skeptic, decides to review a new update that allows easy instantaneous searching and access of this live-blogged history. “Siri, show me my 9th birthday.” Bam, done.

Taken on its own, the main narrative is banal - the journalist discovers just how subjective and unreliable his memories are. Old news. I hope.

But, combined, the two narratives provide sharp insight into our perception and usage of modern technology. Personally, I can be dismissive of those people who feel the need to post their every thought and experience on social media. Especially when they’re not even honest about it - when they edit those experiences to make them seem better than they are. This whole phenomenon has created a social media castle-in-the-sky that so very few of us will ever actually inhabit.

That said, this story made me realize that, really, it’s old news too. Social media may have accelerated the trend, but written language itself often functions similarly. Cause written language is also a technology - we don’t always think of it as one but it certainly is. And it, too, alters our relationship with reality.

Take this book review, for example. It’s highly edited. See in that second paragraph where I put, “So, yeah, big fan.” Wasn’t there in the first draft. I often edit in conversational beats like that in order to make my reviews more approachable. Couldn’t have done that if it were spoken language.

And I think I’ll end the review here. Run out of steam, I’m afraid, so thoughts on the last three will simply have to remain a mystery. But I hope I’ve given anyone reading this some extra food for thought.