52 | Frank Lantz on the Logic and Emotion of Games
What this covers
Frank Lantz, game designer and director of New York University's Game Center, sits down with Sean Carroll to examine games as a distinct art form—one that works not through passive observation but through the deliberate structuring of human choice and action. Rather than treating games as technology to be reverse-engineered or psychology to be exploited, Lantz describes them as cultural practices that emerge from a perpetual conversation between [play and arbitrary structure](concept:Play vs Structure Duality). The episode moves across deep games like Go and Poker, the nature of addiction and what Lantz calls "dark energy," and what games reveal about how minds actually work when given freedom and constraints in dialogue.
Lantz distinguishes between games by their cognitive and social demands. Go exemplifies pure, austere problem-solving pursued for its own sake across millennia, while Poker layers analytical rigor with reciprocal modeling—the convoluted act of imagining what an opponent thinks you think they think. He argues that AlphaGo succeeded partly by abandoning centuries of human intuition, a lesson in humility, while variance in Poker—where the best move can still lose—transforms it into a spiritual discipline about valuing process over outcome. The conversation acknowledges the real problem of dark energy: slot machines and compulsive game design can ruin lives, yet Lantz remains skeptical that addiction follows a simple, engineerable formula because games are culture, not just engineering. He defends games against suspicion by arguing they must live outside ordinary values precisely so they can explore the edges of human possibility. The discussion also touches on storytelling in games, the role of community over individual masterpieces, and how games can make the invisible visible—as when Bennett Foddy's QWOP turns walking into a learnable puzzle by mapping keys to individual limbs.
Lantz argues that games are an art form of interactivity—an ongoing conversation between free play and arbitrary structure—that can range from sublime self-cultivation to dark, addictive energy, and whose value comes from what they reveal about how our minds work rather than from any simple moral or formula.
- Games sit alongside the arts as a creative, expressive form defined by crafting people's interactions rather than by fixed internal traits
- Deep games like Go and Poker function as disciplines for thinking about thinking and for self-discovery
- Addictiveness and 'dark energy' are real but cannot be reduced to a manipulable psychological formula, because games are culture, not just technology
Games let players build mental models, form hypotheses, and gain meaningful insight from loss and exploration.
- Games can make us conscious of processes that are normally unconscious, as in Bennett Foddy's QWOP, which uses a deliberately convoluted control scheme mapping keys to individual limbs to turn the trivial act of walking into a deeply complicated, learnable challenge.
“what he did was create this incredibly convoluted control scheme where you control different limbs, and so it turns this trivial thing into this deeply, complicated thing.”
- A mark of a good game is creating the experience where losing feels meaningful—'I lost, but I understand now what I should have done differently'—so the player builds a mental model of the designed system, forms hypotheses, runs little experiments, and is pulled to play again, making much of game play a kind of toy science.
“what we're trying to do as designers, is create just that experience where, "I lost, but there was meaning to my loss. I understand now what I should have done differently."”
- Part of the deep pleasure of games is the sense that something might be broken or exploitable—the feeling that the designer hasn't thought of a particular strategy (use an axe, wear no armor) and that you can find and exploit the gap—which is 'catnip' to engaged players.
“that feeling of something being almost broken or the potential of it being broken, to me, that's catnip.”
- Any game can be an opportunity for self-discovery and a path of self-improvement and self-overcoming—not just being good at what you do but improving your idea of what 'good' means and what kind of person you are trying to be—though most players are simply, and validly, disappearing into a pleasant experience.
“any game that you play is an opportunity to learn something about yourself, an opportunity to be on this path of self-improvement, self-overcoming”
- Intermittent, unpredictable reward schedules are a major driver of compulsive game play—paralleling experiments where rats that always get a pellet stop when sated, but rats that occasionally get a pellet will keep pressing until they die—though Lantz is skeptical that this psychology can be reliably engineered, since if it were easy everyone would do it.
“what makes them so powerfully compulsive is that the rewards are intermittent... if they push a button and occasionally get a pellet, they'll do that until they die.”
Games are an art form defined by family resemblance and cultural category, not fixed formal traits.
- Games cannot be pinned down by necessary-and-sufficient traits; features like goals, win states, and loss states make something more likely to be a game but are not required, just as music is usually about sound yet can be about silence—games are identified by Wittgensteinian family resemblance.
“having those qualities make it more likely that it's a game. In the same sense that music is about sound, but in some cases, music might be about silence”
- A game is best understood not by precise internal characteristics but as a broad cultural category—a creative, aesthetic form like literature, music, or film, except it is the art form of interactivity: crafting meaningful, expressive experiences through people's choices and actions rather than something they merely observe.
“it is the art form of interactivity, it is the creative form of designing experiences for people. You're still trying to create something meaningful, beautiful, interesting, popular, expressive, but you're doing it by crafting people's interactions”
- Games are primarily a form of culture with a deep relationship to technology, not simply devices that deliver a hedonic pulse like a coffee machine; they are more like hats (a strange social fashion), which is why no formula can guarantee a popular, compulsive, or successful game and design remains a creative, trial-and-error process.
“games are a form of culture that has this deep relationship to technology. But they are not simply technology... instead, they are more like hats.”
Games combine play's unbounded exploration with structure's constraints in productive dialogue.
- Games let us experience the pleasure of voluntarily binding ourselves to arbitrary rule systems while also revealing that rules are constructed and pliable—players can step back after playing, see the characteristics of a rule set, and reinvent it ('what if we tie both hands behind our back?'), gaining a sense of how rules can be otherwise.
“We get to experience the pleasure of being bound by a little arbitrary system, and there is a particular kind of pleasure there. But then we also are given a chance to recognize how rules are constructed.”
- Games derive their power from holding two opposites in conversation: play—wild, improvisational, unbounded exploration of possible actions (like pouring out the bottle instead of drinking)—and structure—rules, arbitrary goals, and constraints (like tying one hand behind your back); freedom and structure are in dialogue rather than opposition.
“in games, you get this idea of play, wild improvisation, kind of unbounded energetic excess and imaginative kind of exploration, and it's also... It's in conversation with the opposite, with structure”
- The interesting thing about World of Warcraft is not that it is a virtual world that happens to be a game, but that it is a game that incorporates a virtual world; the game framing matters because it makes the experience theatrical and performative—players know they are entering a performance—rather than just another social space.
“what was interesting about World of Warcraft was not that it was a virtual world that happened to be a game, it was that it was a game that incorporated a virtual world”
Logic and emotion are woven together in games, not opposed or balanced in tension.
- Logic and emotion in games are not opposed forces in tension with a 'sweet spot' between them, but two mindsets woven together in a fractal rather than dialectical structure: chess is purely logical problem-solving embedded in something entirely illogical, since devoting your life to chess is highly illogical yet done because it is beautiful and meaningful.
“I don't think of those things as being opposed necessarily. I think they're kind of woven together in a game.”
- Professional chess players borrow poker concepts—predicting opponents' openings, and even bluffing by making a move you hope your opponent will think is good when you yourself aren't sure, to rattle them—demonstrating that logic and emotion bleed into each other and that the supposed categorical line between them is illusory.
“you can also have bluffing in chess, right? You can make a move, and hope that your opponent thinks it's a good move when in fact you don't know for sure”
- Poker is beautiful because it combines the deep analytical, mathematical problem-solving of Go with a profoundly social dimension requiring empathy: you must put yourself in your opponent's shoes and model what they are thinking, while they model you in return ('what are they thinking I'm thinking'), producing a convoluted reciprocal modeling that John von Neumann saw as the essence of a true game and that led him toward game theory.
“it involves a kind of empathy, like you can't be good at poker unless you can put yourself into the shoes of your opponent and think, "Oh, what are they thinking?" And then they're doing the same thing to you”
- Even in well-defined problems like chess, the path taken through solution space is irreducible and individualistic—players have distinctive styles (aggressive, stubborn, thorny) and deep heuristics we will never fully reach the bottom of—so style is an expression of who you are as a person even within a purely logical domain.
“even when problems are well-defined, the path that you take through solution space can be totally irreducible.”
Game design should resist moral filtering because games explore meaning itself rather than enforce predetermined values.
- Slot machines are games with a 'dark energy'—they ruin some people's lives while others have a healthy relationship and derive meaning from them—and good game designers feel an aversion to making purely compulsive games, preferring games that leave the player glad they played and feeling something of value was added to their life.
“there is a very dark energy to slot machines. Slot machines ruin people's lives, not everyone.”
- People are right to be suspicious of games, just as they are of art, because games by their very nature must live outside ordinary life and our ordinary system of values; that is precisely their job—to do weird things and explore the edges of what we know, think, and imagine.
“games by their very nature have to live outside of ordinary life. And we should be suspicious of them.”
- We should avoid being too dogmatic in applying a moral filter to games, because the role of games—like the role of art—is not merely to be good or bad but to explore what being good or bad means; a game like Drop7 can be addictive yet be about addiction, giving players a 'toy version' of something like heroin that also yields insight into how their own mind works.
“the role of games, like the role of art, is not just to be good or bad, but to explore what being good or bad means”
Social participation and community matter more to game experience than objective game quality.
- For newcomers, it is far more important to play a game that your friends or a living community are playing—so you can discuss it, learn from others, and treat it as a shared social practice—than to play the objectively 'right' masterpiece, because games are participatory and don't fit the pedestal-and-contemplation model of masterpieces.
“It's so much more important that you play a game as part of a living community than that you play the right game.”
- Given a choice between playing an objectively great game with people who are jerks, or a mediocre game with wonderful, interesting people you love, it is always better to choose the second—because the social company matters more to the experience than the quality of the game itself (a good poker table matters enormously).
“you can go into room A and play this game that you know is a great game, but play with these people who are maybe a bunch of jerks, or you go in this other room and play a game that maybe is not so good on its own, but with these wonderful people... it's always better to do the second thing.”
Go exemplifies deep cognitive inquiry pursued for its own sake, and AI mirrors this kind of problem-solving.
- AlphaGo became dramatically better when its designers stopped seeding it with centuries of accumulated human Go knowledge, heuristics, and teaching, and instead gave it only the rules, a self-play technique, and a preference for winning—an instructive lesson in humility about the value of inherited human intuition.
“it got so much better when we stopped trying to seed it with the centuries of knowledge that humans had developed. With all of our insights, all of our heuristic, all of our ways of teaching Go, we just left that out and instead we gave it the rules and then we gave it a technique for playing against itself, and a preference for winning.”
- Go is a way of thinking about thinking, as is artificial intelligence, so when serious players sacrifice their lives to explore a single precise decision-making problem in the deepest possible way, they are effectively doing the same kind of inquiry as AI research—exploring what it means to apply every fiber of cognitive ability to solving a problem.
“Go itself is a way of thinking about thinking. And artificial intelligence is also a way of thinking about thinking. And when people play Go... They've been doing the kind of Artificial intelligence research”
- Go experts found AlphaGo's surprising moves exquisite and beautiful, making them simultaneously inhuman (because they surprised us and arose from self-play) and deeply human (because the system is the product of thousands of collaborating people, and because it is Go itself expressing itself).
“the Go experts who are watching AlphaGo still found the moves that AlphaGo made that were surprising to be exquisite”
- Go exemplifies a hard, deep cognitive problem of the kind scientists wrestle with, but pursued purely for its own sake rather than as a means to downstream applications—a remarkable art form that people have interacted with in essentially the same way for thousands of years, unlike most works of literature or music.
“In the case of Go, you are just wrestling with that problem, for it's own sake.”
- Games are often the ultimate test bed for artificial intelligence, with Go standing as the iconic, quintessential AI problem and arguably the last full-fledged human game at which humans were really good before computers surpassed them.
“games are often the ultimate test bed for artificial intelligence and Go is this iconic quintessential AI problem which just recently tumbled.”
Games are culturally situated and cannot be predicted; designers actively shape what games become through making.
- Games and sports fandom let us safely experience a 'toy version' of totalitarian or atavistic group identity—plunging into a tribe and shedding the nuance of individual selfhood—because it is bracketed and performative ('not real hate taking place in a virtual space'); violence in sports fandom represents a failure of this ritual transformative bracketing.
“sports fandom gives you this kind of strange toy version of the kind of totalitarian identity that we get when we just feel like, "Oh, I'm disappearing into the mob”
- Drawing on Huizinga's Homo Ludens, play is a deep and ancient behavior that not only exists at the beginning of human culture but predates it, such that human culture and civilization themselves emerge out of the play instinct.
“There's the sociologist Huizinga, who wrote this famous book called Homo Ludens, which is about how play really is the origin point of human culture.”
- The future of games cannot be confidently predicted because culture is too hard to forecast; instead, the job of every game designer is to actively answer the question of what the future of games is and should be through the act of making games.
“Every game designer is trying to figure out what the future of games is and is going to be and what it should be. And so, in the process of making games, we are actively fulfilling that kind of role of predicting what the future games is gonna be.”
- Games are an art form in the same sense pop music is—a creative, expressive aesthetic form—but you cannot dictate to culture how it should interpret things; a designer's contribution is to help outsiders understand what game-lovers actually love (problem-solving, exploring complex systems, forming hypotheses, discipline, social collaboration), which collectively evolves audiences toward demanding better games.
“It's something similar to an art form in the same sense that pop music is an art, is an art form.”
- Unlike a blockbuster movie that everyone knows about whether or not they see it, games are more siloed because a big game like Red Dead Redemption lives and dies as an activity and hobby rather than as spectacle, drawing people into a place and a space rather than imprinting itself as a widely recognized image.
“The Avengers lives or dies as spectacle. Whereas Red Dead Redemption has... Because it is an activity and a hobby almost, lives and dies by a slightly different cultural logic.”
- Some of the most interesting work in games is new forms of storytelling—both traditional rich characters and world-building woven into game experiences, and avant-garde exploration of what a story can be—contradicting the older assumption that interactivity precludes narrative because a story requires a single fixed sequence of events.
“some of the most interesting stuff that's happening in games is these new forms of storytelling... exploring what a story is, what a story can be, and new forms of narrative”
- Doing well at a puzzle game like solitaire is partly a reflection that you are smart, so losing threatens your self-image and creates an urge to replay (almost wanting to take your temperature again to see if you can get 'hotter') or even to cheat.
“when you do well at solitaire it's partly a reflection of the fact that you're smart... it's almost wanting to take your temperature again to see if maybe you could get hotter.”
- By most measures video games are now the biggest pop-cultural industry, such that a launch like Red Dead Redemption 2 is the biggest entertainment launch in history, and people spend far more on video games than on movies—though this remains contentious.
“we spend over twice as much on video games these days as we do on movies... Two-thirds of American households play video games or at least have someone in there playing video games for more than three hours per week.”
Poker combines analytical problem-solving with social modeling and variance, making it spiritually rich yet perilous.
- Poker is a harder AI problem than Go because beating it well is not just finding an unexploitable equilibrium strategy (a hard but knowable problem that at worst guarantees a tie), but building a strategy flexible enough to adapt to and exploit opponents' imperfect play—requiring the AI to infer opponents' strategies from their actions while they do the same to it.
“I also think it's a harder AI problem for that reason too. Because it does involve this kind of reciprocal modeling and guessing.”
- Unlike Go or chess where there is a right or wrong move in every position, in poker you can make the best possible move and still lose—and a table of careless partying players making 'donkey' moves can take all your money repeatedly; overcoming this soul-crushing variance by valuing correct decisions over results is what makes poker a spiritual discipline.
“in Poker, you might make the best possible move and still lose.”
- The mythic archetype of the gambler represents someone trained, or temperamentally disposed, to have a taste for variance and risk—and the deep point is that if you are willing to eat risk, you can profit from it; poker is an art form devoted to randomness and probability, concepts humans still struggle to understand even in physics and philosophy.
“the person who has trained themself or by their very personality has a taste for variance, a taste for risk. If you're willing to eat risk, you can profit from it.”
- Phil Ivey, arguably the world's best poker player, candidly remarked that the feeling of losing so much money you can't breathe—a sickening, painful sensation—is what he is addicted to, illustrating that a real destructive 'darkness' coexists with deep meaning and beauty in poker.
“"You know when you lose so much money that you can't breathe, I love that."”