Review of The MANIAC, by Benjamin Labatut -- thoughts on logic and the end of play
The book’s main idea
As a creative work, the book resists straightforward arguments, so you have to extrapolate the significance of the events/characters/actions and interpret possible meaning. This is a mode I’m less accustomed to operating in (since I mostly read nonfiction) but nevertheless enjoyed. It’s nice to have a kind of puzzle to solve and events to interpret.
Overall, the author’s core idea is that the pursuit of logic, in the extreme, is running us off a cliff. Or as critic Ed Simon wrote in the Los Angeles Review of Books, the story is “about the ways in which reason, pushed to its extreme, can become its opposite, about how a certain tyranny of unfeeling logic could very well usher in an apocalypse (and soon)” (Nightmares of Reason). Scientists absurdly pursue their research, shutting the world out around them, driven by a mania to discover the ultimate answers, while the rest of the world flails for more practical solutions to pressing concerns. Ultimately, the end result of artificial intelligence is the creation of systems that remove all sense of play and enjoyment in games, paralyzing and choking out human activities.
Let me provide a few notable stories from the book to ground this interpretation. First, the book starts off with a shocker that sets the book’s thesis in miniature: Paul Ehrenfest, a scientist struggling to come to terms with the irrationality of quantum logic in particles, decides to shoot his disabled son Wassik (who has Down syndrome) before taking his own life. Ehrenfest’s decision was catalyzed by the rise of the Nazi regime and their eugenic sterilization laws; he rationalized that killing his son was the only way to protect him from being targeted as a “hereditary disease” and to reduce a burden on his family that would drive them into the ground. Hence, murder.
The opening event is a startling way to begin a book and is the most straightforward delivery of the book’s main idea: the scientist’s rational logic, calmly calculated and arrived at separate from the people around him, can lead to a horrific madness. This is the core tragedy Labatut traces. When logic is dialectically pushed to its absolute extreme, it disconnects from human empathy and instead propels us toward — you guessed it — insanity.
Another compelling example comes from John von Neumann, a scientist considered to be the father of the modern computer, game theory, atomic bomb calculations, and more. Von Neumann is the main profile of the book, but his story is told, strangely, through a polyphonic perspective from more than a dozen people around him. (BTW, if you’re listening to the audio version of the book, the different voices/dialects in these chapters are amazing.)
The most astonishing perspective comes from his wife Klara, who von Neumann mainly ignores while he works in his study on what he feels are the important and pressing problems in the world. His wife wants his help opening the garage door, which is broken, so she can drive somewhere. But von Neumann can’t interrupt his research, and continues ignoring her. Klara waits for him to leave the study to use the toilet, then slips inside and locks the door. She proceeds to stuff the Nils Aall Barricelli’s “universe” printouts that von Neumann was studying into the trash can and sets them on fire before throwing the door open. This scene ends with a psychotic rage and breakdown, and a miscarriage as well.
Figuratively, I think we are “Klara,” and we struggle for practical solutions from scientists such as opening our broken garage doors while they sequester themselves away in a focused, insular mode trying to uncover the deepest meaning and answers to the universe. In his New York Times review of the book, novelist Tom McCarthy points out this exact moral blindness, writing that for the scientists in the book, “even their scrupulously observed reason, their mode of logic elevated to religion, is framed as a form of madness” (The Miracle and Madness of Science That Changed the World). When mathematics becomes a religion, actual human lives become mere static.
The final section of the book describes the famous Go match between Demis Hassabis’s AlphaGo and Go master Lee Sedol. The book paints a detailed inner psychology of both Hassabis and Sedol, but the larger trajectory is to feel the competitive spirit extinguished from Sedol as he comes to grips with an unwinnable machine opponent that simply can’t be beaten except when the AI seems to briefly go haywire. In our book club discussion, one member compared it to playing tug of war against a brick wall. As such, the section might be titled “the death of play” because Sedol’s defeat killed his interest in playing Go; the machine removes the beauty and mutual art that the two humans playing together usually create. When the opponent is an indestructible force that can’t be overcome (the brick wall), we feel a sense of defeat, paralysis, and pointlessness in even trying.
Interestingly, at the end of Part II, the narrative returns to describe von Neumann on his deathbed. The radiation from the atomic testing led to rapid physical ailment and a sharp loss of von Neumann’s mental acuity and sanity. On his deathbed, von Neumann provides some insight about the future of artificial intelligence and how we will reach the highest levels. He says that only when machines can learn to play, as a child, will they get there. This dark irony switches the idea of play from human to machine. Instead of humans possessing the childlike wonder and play that spurs us on with curious questions and explorations, now the machine has usurped this same play, consuming data and iterating with unstoppable force that soon dominates everything around it as it continuously learns.
Overall, The MANIAC doesn’t portray the pursuit of logic and artificial intelligence as a triumph, nor does it celebrate a future of abundance or harmony. Instead, the picture it paints is a dark, joyless world where humanity is eliminated. In fact, while AlphaGo was initially trained on human games to establish “common sense,” Hassabis’s team eventually developed AlphaGo Zero, which removed all human training and influence, letting the machine learn entirely on its own. This resulted in a much more dominant outcome, as AlphaGo Zero defeated the version of AlphaGo that beat Lee Sedol, 100 games to none. The message: to achieve the highest levels of logic, you must strip away humanity entirely. In pursuit of this AGI-like end, we will be like the child (Wassik) that Ehrenfest (the scientists/technologists) eliminate to remove their burden from anyone in the future. Humanity is an illogical burden (no matter how cheerful or happy we may be) in achieving the end goals of logic/reason that scientists so monomanically pursue.
Impact and thoughts
Despite the critical reception of the last third of the book (the AlphaGo match against Lee Sedol, which one critic said “falls flat when compared to the first two parts of the triptych” — see The Unruly Limits of Materialism), this is the section I connected most with, as it deals with the death of play. Has AI removed the sense of play from my own life? I’m conflicted about this.
On the one hand, take an essay like the one I’m writing now. I know the AI can run circles around my interpretation, pulling out exact quotations or moments from a 300-page text that illustrate everything I’ve said, or which might make deconstructive counterarguments if requested, or which make entirely different arguments about the book. AI does this all in a matter of seconds, from consuming the text to interpreting, firing off logical engines that make my own mushy wet brain seem pathetic and weak. Confronting this technology is humbling. What’s the point of writing anymore? Where’s the sense of play when the machine makes anything I do seem so small and meaningless? I can definitely feel a loss of play and enjoyment in writing.
On the other hand, a great many tasks that AI handles are tedious drudgery. Sometimes, AI frees me up to think about larger questions and structures and ideas without getting bogged down in technical details and minutia. I can paint with broad strokes and create much more adeptly without sinking too much time into those areas of writing that are fatiguing — hunting back through pages to find the exact quote, remembering the details of a story accurately, identifying other research about the book, smoothing out the language and making sure it flows/reads well. All of this makes writing a single essay time consuming and physically fatiguing. Does the writing still provide meaning if I’m directing the messaging and larger navigation through ideas — like a director sitting in a chair instead of an actor on stage?
Honestly, I’m not sure, but I have noticed that my writing output on my blog has gradually diminished over the past couple of years, perhaps due to AI. Writing (by this I mean typing it all myself) seems akin to writing longhand, with a pencil — a pencil that I have to periodically sharpen. A pencil that frequently breaks. I can hardly stand the time, the patience it takes, to type out an essay word by word. Not to mention the endless editing and shaping tasks that will follow the first painful draft.
Overall, The MANIAC leaves readers with a dark sense of how everything with AI plays out. In short, it doesn’t end well. It ends in our elimination, pretty much. And with this theme, I’m starting to see a pattern in the books we read in the AI book club. I noticed this same pattern in another book club I ran at work related to automotive and transportation. Briefly, a little backstory. Because I work in Google Maps (APIs in cars), I wanted to deepen my understanding of the automotive domain and relevant industry topics and themes, so I created an “Automotive & Transportation Book Club” at work. After 2.5 years of reading a book a month, what I found was essentially the same theme as the subreddit /fuckcars. (Sorry for swearing.) That is, not a lot of people were loving on the impact of automobiles and society. Pretty much, the recurring theme in everything we read was that cars have negative impacts on society. We would be better off without them.
Same goes with AI. Most books, especially the thought-provoking ones written by deep thinkers, take a similar train of thought: the equivalent of a /fuckai subreddit theme. In other words, critical books tend to be … “critical” of AI. This creates a constant cognitive dissonance as I’m deeply involved in AI as a technical writer. Whether I like it or not, AI generates documentation faster, easier, and more accurately than humans can write it. I’ve written about this cognitive dissonance before. I feel glad not to have to write pages and pages of documentation (but instead to direct AI what to write, and then review and steer the output). I’d much rather spend my creative energy on topics more meaningful to me. This blog remains one of the last spaces where I mostly write by hand.
However, I’m not so sure that I cleanly separate the two realms: using AI to write docs at work, while using my own brain to write blog posts at home. I need to find a way to allow AI to push me to the next level and move past this cognitive dissonance. Ideally, I want to find a way to use AI to engage intellectually in a way that feels like I’m still intensively using my brain. I explored a bit of that in Bakhtin and model collapse: How to use AI with expressive writing without generating AI slop. In that essay, I argued that AI can help identify research that brings in a heteroglossia of voices. Perhaps I’m going in circles, still trying to pin down the right approach.
About Tom Johnson
I'm an API technical writer based in the Seattle area. On this blog, I write about topics related to technical writing and communication — such as software documentation, API documentation, AI, information architecture, content strategy, writing processes, plain language, tech comm careers, and more. Check out my API documentation course if you're looking for more info about documenting APIs. Or see my posts on AI and AI course section for more on the latest in AI and tech comm.
If you're a technical writer and want to keep on top of the latest trends in the tech comm, be sure to subscribe to email updates below. You can also learn more about me or contact me. Finally, note that the opinions I express on my blog are my own points of view, not that of my employer.