Game 9 – New Forest

This game with Lukas was played over a week ago, and it gave me much to think about. Although every game of Go is unique, Lukas said that this one’s feel was different from previous games. This happened to be the first game I played since moving to a new city and starting a new job. Coincidence? I think not.

I made the decision to play the 5-4 point. Life began as I leapt into a different forest in search of a new species of Go.

This is how I would like to approach the game from now on: trekking through unknown woods without fear.

In the game below, commented by me with some of Lukas’ suggestions, I played black, and Lukas won by 20.5 points.

Narrative, Games, and Go: An Interview with Professor Marco Caracciolo

Marco Caracciolo is Associate Professor of English and Literary Theory at Ghent University in Belgium. His work explores the phenomenology of narrative, or the structure of the experiences afforded by literary fiction and other narrative media. He is the author of five books, including most recently Narrating the Mesh: Form and Story in the Anthropocene (University of Virginia Press, 2021).

I had the honor of befriending Marco several years ago in Belgium. As I started to explore the relationship between Go, language, and narrative, I realized that his ideas would be an invaluable contribution to this journal. I have fond memories of getting decimated by Marco in games of chess, while I returned the favor by introducing him to Go over coffee. Not only is Marco incredibly thoughtful about literature, he is also an avid gamer: an ideal combination for a conversation about the relationship between Go and narrative. I hope you enjoy learning from him as much as I have over the years.

River: How would you describe narratology to a Go player?

Marco: Narratology is the study of narrative form. It emerged in France in the 1960s as a branch of literary studies, but it has since then expanded to other media: cinema, TV, theater, and video games. Narrative forms are patterns that recur across many instances of narrative, such as embedding a story within another story or “skipping” events in a flashforward. These patterns may resemble strategies or common situations in a game of Go. But while abstraction is important in narrative (and in its narratological study), narrative is—unlike Go—a representational practice through and through: it evokes characters, locations, and events that are in some significant way lifelike. Another way to put the same point is to say that the stakes of narrative depend on its representational content as well as its form; arguably, for some narrative audiences, the content (whether Jane Eyre marries Mr. Rochester or Walter White becomes a drug lord in Breaking Bad) is more important than the form. With abstract games like Go, only the formal patterns (strategies, positions, etc.) matter. Chess is an example of an abstract game that has a representational façade: in essence, chess represents a battle between two armies. But, ultimately, that representational dimension is secondary to the game: it doesn’t matter if the piece that moves diagonally is a bishop or an elephant (as in historical versions of chess); being a good chess player means understanding the abstract rules of the game. With Go, the representational element is even sparser. The game could be understood as a competition over territory, but there isn’t much more than that in terms of representation. So while games like Go foreground abstraction, narrative combines abstract patterns and representational content. Narratology is the study of those abstract patterns, but of course cannot afford to ignore the representational aspect of story.

R: What is immersion?

M: Immersion is a treacherous concept, and the fact that we have so many terms that seem interchangeable with it (absorption, engagement, flow, etc.) doesn’t help, either. But, in short, I would define immersion as a state of focused attention, in which we become intensely aware of a certain object, while losing awareness of our immediate physical surroundings. Depending on the object of attention, immersion can take multiple forms, which perhaps explains the diverse terminology.

R: How would you describe the relationship between “ludic” immersion in games and “narrative” immersion in, say, a novel? Would you even describe immersion in those terms?

M: We can be immersed in games like Go, in narrative, in music, or in many other activities. However, differences in the nature of these activities can lead to considerably different experiences of immersion. While playing Go, we become involved in the abstract patterning of moves and countermoves. There may be an emotional dimension to those patterns, but it seems to me (admittedly, without being a Go player) that the emotional gamut remains fairly limited: expectation, satisfaction, dejection, and so on. With narrative and music, we have a much broader range of emotional expression. These emotional values can give rise to immersive experiences in engaging with music and narrative: think about suspense, curiosity, or surprise (for narratologist Meir Sternberg, these are the emotional universals of narrative). Representation plays a much more significant role in narrative than it does in music, especially if we consider instrumental music, which has an emotional progression but no overt representational progression (at the level of what happens to whom). This has important implications for immersion, too, because being immersed in narrative means being transported to what we often describe as another “world.” With music, this is less often the case: the emotions register in our body, and they may well feel otherworldly, but the world being evoked is less clear-cut and particularized than the kinds of worlds opened up by narrative. With music, we’re transported to a purely emotional world, as opposed to the worlds of narrative, which have well-delineated physical characteristics and are inhabited by human-like individuals. This is in no way a ranking of immersive experiences, though! Subjective preferences are important, too. And, as I said before, these differences and nuances of immersion across various media and practices are perhaps the reason why we have so many different terms for experiences of focused attention.

R: Do we need stories to play games? Do we need games to experience stories?

M: The first question has occupied scholars in the field of game studies quite a bit, especially in the early 2010s. My preliminary answer is “no” to both questions. But it really depends on what you mean by “need.” Humans are storytelling animals—as well as animals that play. (Actually, human storytelling is far more unique in the animal world than play: many nonhuman animals play, only humans seem to have a language that is sophisticated enough for storytelling. Of course, games are not the same as play, but they have common roots. I digress, though.) We don’t need stories to play games, but because we are storytelling animals, we’re likely to reach for narrative in certain situations. Stories may well enrich our experience of games, or may make the rules of certain games more intelligible. Again, to return to chess, the representational façade (a battle between two armies) is not very important to the advanced player, but if you’re new to the game it may be useful to think about that military scenario, because it may help you grasp the differences between, say, a lowly pawn and a valuable rook. So turning the game into a narrative of sorts may have some advantages. I don’t think we need games to experience stories, but—again—games may offer a useful set of metaphors to understand, for instance, the “strategies” adopted by the author, or a particularly astute “gambit” performed by a literary writer.

R: What do you see on the horizons of narratology, particularly with respect to games?

M: I think video games are the most interesting medium from the perspective of the encounter between the abstract patterning of games and the representational dimension of story. Some video games are primarily abstract (think Tetris), others are mostly narrative, with very little interactivity. But between these extremes we find a vast landscape of games in which narrative coexists, productively, with game challenges. Narrative scholars such as Marie-Laure Ryan or, more recently, Dan Punday have already started developing a narratology of video games, but a lot remains to be done.

R: How would narratology make sense of a machine that could generate stories based on a Go game record?

M: I hate to be a killjoy here, but I doubt those would be interesting stories. You see, stories require a skilled storyteller, someone who knows the stakes of human action, can read the audience and anticipate their responses, manipulate their expectations, etc. A machine is no more likely to tell an interesting story than an intentionally funny joke. (“Intentionally” is important here, because, of course, a strange combination of events produced by a storytelling algorithm can be funny, but that humor isn’t deliberate.) By saying this I don’t want to shortchange the work of folks who are working on AI story generation, which I think is a fascinating endeavor—but I suspect its payoff is more theoretical than practical. Humans will always be the best storytellers, because story is intimately bound up with the physical, biological, and cognitive make-up of social animals like us.

R: What, exactly, is happening when we tell a story about a game? (I’m thinking about Yasunari Kawabata’s The Master of Go, or Shan Sa’s The Girl who Played Go).

M: I haven’t read those books, but I’m thinking of The Defense by Vladimir Nabokov, one of the best novels I know about chess… I would say that stories have the power to explore the social and cultural embedding of both concrete games and game-like metaphors. Nabokov’s novel probes the relationship between a professional chess player’s ludic pursuits and his emotional life—how the formal geometry of chess can help him manage anxieties that derive from a messy, socially complex, intimidating world. Unfortunately, it all goes awry. This is also a study in the potentially catastrophic ramifications of games, what happens when immersion borders on addiction. I happen to be watching The Wire at the moment, and “the game” is one of the most powerful metaphors for life in a world shaped by racial inequalities, brutal policing, and corruption. So this isn’t a show about a game, but a narrative that uses a game metaphor to convey the violent competitiveness of the illegal drug trade but also the characters’ powerlessness when it comes rewriting the rules of society.

R: From a narratological perspective, what is the most interesting game you’ve played? Is there something you think Go players can learn from it?

M: So many video games raise stimulating questions for narratology, but I want to mention Mobius Digital’s Outer Wilds (2019) here. This is a time-loop game that asks the player to explore a small solar system in bursts of twenty-two minutes—before the sun explodes and the player is returned to square one. There is so much to uncover about this world and the stories of the extinct civilization that inhabited it, but unlocking these secrets requires precise timing, meticulous attention to detail, and willingness to start over countless times. I can think of no better lesson for a Go player.

Review and Interview: Spindrift (2019) and Planet (2021) by Chen Hsi (Leon) 4P and Xiaocheng (Stephen) Hu

In Spindrift (2019) and its follow-up Planet (2021), Chen Hsi (Leon) 4P and Xiaocheng (Stephen) Hu have created astounding portraits of global life and culture through a unique synthesis of the worlds of Go and narrative. These tsumego collections—in which problems are eclectically paired with microfiction—are not to be missed. In the sometimes bleak landscape of post-AI Go, these works of art will reinvigorate your love of the game. They can previewed here and here on the Online Go Server (OGS), or ordered in full here. I was thrilled when Stephen asked me to review them, and could not think of a better way to communicate my overwhelmingly positive impressions than by writing some poems:


Walk the coast of liminality
With this treasure of philosophy.
Problems solved to sift and cycle
The old tide’s new elixirs.


Read the stardust line by line,
Pour a glass of the finest wine:
Dark shades, jubilant sparkles. 
Absolutely intercontinental. 

I also had the opportunity to interview Stephen about the collections. The depth and sincerity of his responses reflect, I think, the high quality of these galactic creations.

River: How would you describe these tsumego collections’ contribution to the Go world?

Stephen: Planet is our second original tsumego collection, which follows Spindrift (published in December 2019). We hope to continue developing this series and leave our unique legacy; there haven’t been many original tsumego books in the contemporary era that come with a title and a story for each puzzle, but maybe we’ll qualify for that list! 

I’ve always wanted to make a positive contribution to our wonderful Go community, just like all the great tsumego creators and the timeless masterpieces they have produced. Even though it’s hard to imagine that we will ever match the height of these giants in the Go world, I’m sure that we will reminisce fondly about these collections as souvenirs of our youth.

R: How did your collaboration with Chen Hsi (Leon) 4p come about? What was that collaboration like?

I first got to know Chen Hsi 4p as one of the top professional Go players in Taiwan. In November 2019, I accidentally came across one of his Facebook posts, in which he mentioned a desire to compile his original tsumego problems and perhaps publish them as a book. Having always wanted to do something similar, I reached out to him and asked if we could work together and write something to accompany a selection of his favorite puzzles.

That collaboration worked out fantastically, and since then we’ve been able to roll out Spindrift and Planet (the original plan was to release Planet in 2020, but the COVID-19 pandemic and the e-Go Congress made me delay my plans a little bit). I’ve really enjoyed working with Chen Hsi 4p, and I am quite happy to see that he’s also entertained by my stories. Very rarely does someone possess such a special talent to create so many tsumego problems of varying size and difficulty; in addition to that, his passion for the game really stands out among all the players I’ve got to know, and I feel very enriched by the hour-long conversations between us. I look forward to producing new collections with him in the future, and I really hope that more people will know and appreciate Chen Hsi 4p’s tsumego works!

R: What inspired these collections of tsumego/story pairs? Where do the stories come from?

S: My first inspiration came from Xuanxuan Qijing and The Book of Pure Pleasures and Forgotten Worries (Wangyou Qingle Ji), two of the best Go publications in ancient China. The tsumego content was obviously amazing, but I was in fact more impressed by the names beside every puzzle, which often reference a story in ancient folklore or compare the shape to a person, an animal, or a mythical object. In more recent times, the late greats Maeda Nobuaki and Hashimoto Utaro also made a name for themselves by publishing copious volumes of original tsumego, and it’s amazing to observe their meticulous attention to small details, as if they were trying to “carve” each tsumego into a worthy sculpture.

Even though all of the aforementioned authors have sadly passed away, their admirable craftsmanship still lives on in my heart. In Spindrift and Planet, I try to create a similar experience with the title and story text. The name of each puzzle can come from their original shape or the correct solution – of course, they won’t overtly point out the correct answer to the problem, but rather serve as a subtle hint. The stories themselves might be inspired by literary works, songs or movies, or personal experiences retold by me or Chen Hsi 4p with varying degrees of fictionality.

R: How would you describe the relationship between Go and narrative? Do we need storytelling to play Go?

I wrote the stories in Spindrift and Planet mostly for enjoyment, with the hope that they would perhaps paint a lively picture of each puzzle. A compelling narrative brings vigor to a stationary Go position, and enables the reader to unleash their imagination beyond black and white stones on a two-dimensional array of gridlines. It also allows the puzzle to stay longer in the reader’s memory; for instance, I can still recall the shape of all 272 puzzles (so far) in this series, and just thinking about some of the names might ring a bell about a particular shape on the goban, or a precious moment off the goban. It will be very interesting if I run into someone at a post-COVID Go congress, and they suddenly remind me of a particular puzzle and its narrative; that will most definitely make my day!

I would say that people learn to play Go in many different ways, just like they do in any trade. Some learn by intense reading and calculating, some learn by listening to a strong player’s commentary, and some can do just fine by simply imitating the moves that they observe. It’s not just about tsumego – even when you learn a new joseki or mid-game technique, you can still uncover a riveting plot of what’s actually going on. Maybe someone is trying to pull off an attack to psychologically intimidate their opponent, and the next response also tries to put up a resilient fight after weighing the pros and cons of different strategies… If you can achieve a deep comprehension of the story in an actual game, that might work out much better than any sort of pure, reasonless memorization.

R: How would you describe the role of narrative in your personal Go development?

S: Although I grew up in Beijing and lived close to a few of the Go academies, I’ve never really intended to attend full-time training in those places. However, I’m very lucky to have read and heard many brilliant narratives from stronger players; as a result, I have managed to improve over the years and climb to a 6.7 AGA rating, which I feel decent about for a modest amateur level. In my occasional teaching with fellow Go players, I’ve also tried to emphasize the importance of storytelling and creative imagining to them. To me, it’s just amazing how an ostensibly simple position can generate so many intriguing variations; that kind of mindfulness has helped me keep my enthusiasm in Go and carried me through the highs and lows of my career.

For Chen Hsi 4p, the perspective of a tsumego creator is far more unique than mine; I can think of very few people in the entire community who can grow a five-figure collection and work out more than a dozen new positions on a 30-minute train ride! Ever since his first original tsumego in 2017 (a.k.a. The title puzzle A-1 of Spindrift), he’s embarked on many new adventures and created all kinds of novel shapes. Sharing that kind of joy with fellow professional players and students is very special, as you get to participate in conversations that are just pricelessly enlightening. Of course winning competitive tournament games and taking cash prizes is also nice, but curating something for the benefit of the Go community will produce a longer-lasting moyo!

R: What is your favorite story/problem pair from the series, and how does it illustrate the relationship between Go and narrative? 

S: This is going to be a very difficult pick! In the interest of promoting the newer collection, I suppose that I should choose something from Planet. We typically select a couple of puzzles from each section (5-7 in total) and release them for free in the OGS preview, but I thought I could offer an exclusive sneak peek at one of the puzzles not in the preview. The story is based on an ancient Chinese parable, and those who speak the language might recognize it:

“There was an old man who lived in a small border town and unexpectedly lost his horse. When his neighbors came to console him, he replied, ‘this might not be a bad omen, but in fact a blessing in disguise!’ A few days later, the lost horse returned home safely and brought back a new, quicker horse.

Whether on or off the goban, we can always expect some unexpected gains or losses, but the old man’s wisdom has taught us to not gloat when capturing the opponent’s stones, and to not despair when getting captured by the opponent.

It’s nice that Go is a perfect information game, which allows one to precisely evaluate the worth of every trade. Please try to predict Black’s fate in this position.”

This puzzle is called “Blessing in Disguise” (Planet, D-19), which should be quite manageable for dan players. I’ve chosen it over some of the more difficult puzzles, because it’s nice to see everyone enjoy an achievement of solving a tsumego problem. The story will speak for itself – I won’t spoil the solution for you, but do note that there will be some gains and losses in every variation. As long as you’ve gained more than you’ve lost, things will turn out just fine!

R: You mentioned that elements of many of the stories resemble certain shapes appearing in the problems or their solutions. Is this the only way that the stories interact with the problems? 

Some of the stories do come from shapes in the original problems or their solution sequences, so it’s possible that the interaction becomes clearer after you solve a problem. The connection might not seem super obvious at first, but a deeper understanding of the variations will reveal the subtlety of the story text. At times, the story can be also related to the “manner” in which the puzzle is solved: sometimes a violent combination of tesuji moves will be required, but in other circumstances the calm move might be the only viable option. It’s like examining different techniques in the field of martial arts, which are also referenced in a considerable number of stories. Of course, if the reader is curious to know more about the origins of each tsumego/story pair, I am very happy to take questions about any of them!

R: What do you want readers and solvers of these collections to learn about Go? About stories?

I am not knowledgeable enough to impart wisdom on the readers and solvers of this collection, but I do hope that they will get the most out of this book by taking time with every tsumego problem; in fact, it will be nice to do the same for any Go position, not just the ones included in this series. When a puzzle has many interesting branches apart from the solution, it really multiplies and evolves into many similar puzzles with what can be completely opposite outcomes. Therefore, I believe that one should not skim through a puzzle too quickly, even if they manage to solve it from a quick glance; rather, they should treat every position with thorough care and reserve some extra time for self-reflection. 

A well-constructed tsumego will have a unique optimal solution, and every other method with different placement or order of moves will be countered by the opponent and led to a slightly worse result (e.g. unconditional vs conditional kill, and direct-step ko vs multistep ko). Knowing the tiny differences between success and failure will prepare you well for victory in a real game, or save you a few seconds of thinking time when the opponent strikes with an off-path response in byoyomi. In this spirit, a tsumego might not only help you improve your reading skills, but also become an intellectual dessert of life.  

I am constantly reminded of the fact that English is my second language when translating the stories, so my only hope is that people will still be amused or enthralled by them. Maybe it will be cool to see this series translated into more languages in the future!

R: How should we tell the story of AI’s ascendance in the Go world to young and beginning players?

I think that we should be optimistic about the future of AI’s influence in the Go world. It’s certainly a bittersweet moment for our generation, as we did enjoy all the great battles and teachings offered by our predecessors. Certainly, things like opening theory and mid-game strategies will be inevitably changed by AI recommendations, and the current generation of young players will enjoy the benefits of learning from a program of superhuman strength from the very beginning, without having to go through a “mental reset” like us in 2016! It will be very interesting to see how future talents from around the world will surpass us and write their unique narratives in the upcoming decades. 

Lucky for us, our tsumego solutions seem to have survived so far, but maybe we will get to see newer interpretations of them, just like the way Xuanxuan Qijing, Gokyo Shumyo and Igo Hatsuyoron have evolved. It’ll be an honor if they still remember our names by then!

Xiaocheng (Stephen) Hu is an AGA amateur 6 dan player who won the 2017 San Francisco Jujo Ing Cup and North American Collegiate Go League as captain of the UC Santa Barbara team. Aside from competition, he is also an executive producer at the AGA E-Journal broadcasting team and a regular organizer of the U.S. Go Congress. He is currently based in Beijing, China as a full-time Go content producer.

Chen Hsi is a Taiwanese professional 4 dan player, who took 3rd place in the Korean Prime Minister Cup at the age of 11. A regular quarterfinalist in multiple tournaments, he is frequently ranked among the top 10 Taiwanese professionals with the highest annual match winrate. He is a graduate of the NTU College of Law, and now runs a successful Go academy and a YouTube channel with his brother Chen Feng 5p. He is also a regular Go coach at multiple schools in Taipei.