Sunday, August 04, 2019

Turandot and narrative failure

I'm working on a game that is based loosely -- very loosely -- on Puccini's opera Turandot. This post is not about that game. It is about the opera itself, which is an absolutely fascinating example of a writer (in this case the librettist Giuseppe Adami) setting himself up for narrative failure. But first, a quote from a recent piece in the Guardian:
I’ve always hated “Irish jokes”. Having an Irish mother, I’ve always been aware how they were used to denigrate Irish people and undermine the cause of Irish nationalism. There’s one joke, though, I’ve always enjoyed. It’s the one where the guy asks the Irishman for directions, to which he replies: “Well, if I were you I wouldn’t be starting from here.”
Why is this relevant? Well, if your aim is to write a story that ends with the boy and the girl getting married and living happily ever after, then the place that you really don't want to be starting out from is the end of Act 1 of Turandot. The background to the plot is this. Turandot is a princess who does not want to marry. Her father doesn't agree, naturally, but has committed himself to the following scheme. Any prince who wants to marry Turandot must solve three riddles. If he fails even one of them, he is executed. If he succeeds, he gets to marry the princess. Since Turandot is exceedingly beautiful, and since the riddles are extremely hard, the walls of the city are adorned with numerous decapitated remains of royal suitors.

The hero of the opera is prince Calaf, exiled from his own country by an evil usurper. In Act 1 he witnesses the execution of the prince of Persia, who has just failed the trial by riddles. Calaf curses the cruel princess, but when she appears on the balcony, he is smitten, falls hopelessly in love with her, and rings the gong that means he too wants to attempt the challenge of the riddles. (There's an important subplot involving Calaf's father and the slave girl Liu, who is in love with Calaf. But we can ignore this for now.) That's the end of Act 1.

Where can you go from here, as a writer? You have to go on to the riddle ceremony, of course. But how should it end? If Calaf fails at one of the riddles, he will be beheaded. That's not the satisfying romantic ending Adami is aiming for. But what if he succeeds? Then Turandot will have to marry him, but against her will. There's no way you can sell that as a happy romantic ending either. In other words: Adami has written himself into a corner and seems to be headed for massive narrative failure.

There is, perhaps, no way to really fix the problem. But Adami does make a valiant effort. He first does all the things he has to: give us the riddle ceremony and have Calaf succeed. He now has to face the problem of a forced marriage head-on and, to his credit, he does precisely that. Turandot is given the space to express the problem in the clearest possible terms:
TURANDOT
August father! No!
Don’t cast your daughter
into the stranger’s arms!

THE EMPEROR
The oath is sacred!

TURANDOT
No, don’t say it!
Your daughter is sacred!
You can’t give me to him, to him
like a slave, ah no!
to die of shame!
(to the Prince)
Don’t look at me like that!
You, who mock my pride!
I shall not be yours! No, I will not!

[...]

THE CROWD
You are the reward of his daring!
He offered his life for you!
The oath is sacred!

TURANDOT
Would you have me in your arms
by force, reluctant and enraged?
Obviously, that is not what the prince wants. But how can the narrative move forward? Adami does the only thing he sensibly can: he plays for time. Calaf gives the princess another chance: if she manages to find out his name by dawn, he will consent to die. If not, they will marry. Here ends Act 2.

By itself, this solves nothing. We are still left with the unpalatable choice between death and rape. But at least it gives us some room for a plot twist. (And for one of the greatest and most famous opera arias ever written, Nessun dorma.) That plot twist takes the form of the slave girl Liu -- who knows the prince's name -- killing herself when Turandot orders her tortured to find out Calaf's name. As Liu dies, she prophesies:
LIÙ
Yes, Princess, listen to me!
You, who are enclosed in ice,
conquered by such flame,
you will love him, too!
Before the dawn,
I will wearily close my eyes,
so he can win again...
And I’ll never see him more!
How does this help? Logically speaking, it doesn't. But dramatically, it generates a moment of large and complicated emotions. The prince is filled with anger against Turandot, even though he is also, at the same time, still in love with her; the princess is shocked and confused, not only at having lost the opportunity to find out the prince's name, but also at the act of selfless sacrifice she has just witnessed. And this moment of instability creates dramatic room for the move that Adami now makes: Calaf, overcome by emotion, moves in for the kiss, and indeed kisses the protesting Turandot -- who ends up submitting to this treatment with what does, in the end, look like some enthusiasm.

Clearly, this is extremely problematic. It takes very good acting to make the scene so much as watchable -- you'll really have to play down the protests and act up the enthusiasm if you don't want the audience to think about #metoo more than they are thinking about the opera itself -- but it is hardly possible to make it a success. The trope of the woman who says 'no' but really means 'yes' is too problematic. Morally problematic, certainly, but also aesthetically problematic: it's not real, it's not true, it's not believable. If this is where the opera had ended, its failure as a narrative would have been more or less complete.

Luckily, Adami too senses that he has fallen into a narrative abyss. And while he cannot precisely undo the damage, he can at least mitigate it by giving Turandot's agency back to her. Obviously, the prince's justification for kissing Turandot, even though she said 'no', was that he knew for certain that she wanted to be kissed. That's what all #metoo men would say, wouldn't they? "I saw that she wanted it." But Calaf now does something that few other men would be willing to do: he bets his life on the claim that he was right. Even though it is not yet dawn, he tells his name to the princess, essentially allowing her to kill him at her whim.
THE UNKNOWN PRINCE
My mystery?
I no longer have one!
You are mine!
You who tremble if I touch you!
You who pale when I kiss you,
can destroy me if you will.
My name and my life
I give you together.
I am Calaf, son of Timur!

TURANDOT
I know your name!

CALAF
My glory is your embrace!

TURANDOT
Listen! The trumpets blare!

CALAF
My life is your kiss!

TURANDOT
Lo, the hour has come!
It’s the hour of the trial!

CALAF
I do not fear it!

TURANDOT
Ah, Calaf! come with me before the people!

CALAF
You have won!
It really is the only course of action that can somewhat justify Calaf: he makes his dubious kissing move, but then he is at least willing to die if he has wronged Turandot by doing so. And of course it ends well.
TURANDOT
August father...I know the name
of the stranger!
His name is...Love!
(Calaf rushes up the steps. The two lovers are locked in
an embrace.)
It is damage control, to be sure, but it is good damage control. And part of me wonders whether it wasn't precisely the sheer impossibility of getting from the narrative's starting point to its end point that attracted the librettist -- and perhaps also the composer -- to this particular story. There's something enticing about taking a story that cannot possibly work and then pushing it as far in the right direction as it can possibly go. That's certainly part of my own attraction to the story, and hence part of the reason why I opted to base an interactive fiction off of it. Though I should stress that it moves far from the original and most certainly does not involve a forced kissing scene. Whether it too ends up as a narrative failure is something we'll have to see!

Saturday, August 03, 2019

Commenting on this blog

I know of two people who had trouble posting comments to this blog. One of them was me. I'm curious whether this is a general phenomenon -- if some people could try to reply, and if it doesn't work, send me an e-mail about it, I would appreciate it!

Address is victor at lilith dot cc.

Sunday, July 14, 2019

Thoughts on criticism

The primary aim of a review is to tell us whether a particular piece of fiction is worthy of our attention. The primary aim of criticism is to teach us to read. There is of course no sharp line between the two genres, and a single article can have both aims. But it is nonetheless a useful distinction to make.

Good criticism teaches us to read. How? By showing us good reading in action. In the ideal case, we and the critic have both read the piece to be discussed; but the critic has seen things we have not seen, has thought about the piece in ways we have not thought, and has related the piece to contexts that may not even have crossed our minds.

The point of this is not that the critic has arrived at the correct interpretation of the work and will explain it to us. If the work is rich in meaning, many interpretations are possible, making it senseless to seek the correct one. If it is not, then the question of interpretation does not carry much weight.

The point is also not that the critic is able to give the correct judgement about the work's value: whether it is good, and how good it is. No critic commands that much authority; and, what is more, such value judgments are ultimately of limited interest.

The point of criticism is that it increases our sophistication and our sensitivity. It shows us how to get much more out of this particular piece than we were able to get out of it by ourselves; and in doing so, it trains us to approach the next piece we read with just a little more understanding, a little more feeling, a little more openness to what is new and what defies easy categorisation.

One is tempted to say: it teaches us to be better people.


We may further distinguish between negative and positive criticism. Positive criticism enriches and strengthens the work it discusses; it make us admire and enjoy that work more than we formerly did. Negative criticism undermines and diminishes the work it discusses; it tears down the pleasing façade and shows us the shallowness and rot underneath.

Negative criticism may seem to serve vital political purposes. Perhaps it does; perhaps we really need to point out, again and again, the shallowness and rot in much popular culture. But I wonder. I wonder if the same purpose cannot be achieved, in a better and higher way, by the positive criticism of work that is good. One learns to hate coarseness not by being berated, but by tasting of tenderness.

An objection. It is important to identify and call out racist dog whistling in the discourse of real-life politicians; so how could it not be important to identify and call out hidden racism in, let us say, Shakespeare's The Tempest? But this is not important at all. If The Tempest has nothing interesting to say about racism, then we are wasting our time reading it with that particular topic in mind. Better approach the play from a direction that will allow it to shatter us with its aesthetic magnificence. And if it does have something interesting to say about racism, then we need to amplify and illuminate this message, thus turning to positive criticism.


I want to write positive criticism. If I write about a piece of interactive fiction, I want you to end up enjoying it more. I want to enrich you by letting the work enrich you more than it initially did. I want you to fall in love with what was new in it, and unexpected, and subtle, and true. I want you to want to go back and reread it immediately.


Rarely, if ever, do I live up to this exalted ideal.

But I'll keep striving.

Wednesday, July 10, 2019

[IF Comp 2018] Railways of Love

Another review from last year's IF Comp. Spoilers ahead.

Railways of Love by Provodnik Games

One of the questions that kept nagging me as I played through Railways of Love was whether the game really had a Russian vibe, or whether I was just imagining this, based on the fact that you can choose between Russian and English. Of course, the long train journey might conjure up images of the Trans-Siberian railway, and the failing lights fit well with a perhaps clichéd idea of the state of household technology in the USSR… but there are long railway trips in the rest of the world too, and I’ve seen the lights in Dutch trains fail at times. But then there was the Progress Program, which sounded ever more like a science fiction version of Marxism-Leninism, 5-year plans included. And when I got to an ending in which the protagonists fail to hook up because one of them is praying and the other cannot refrain from making a hard-line atheist comment, I was certain: this is light years away from Hollywood, and very much in the cultural space also inhabited by Dostoyevsky and Tolstoy.

The structure of Railways of Love is quite original. The two protagonists are sitting in the train, some minor events happen, and all you can do is try to make them confess to each other. But the situation isn’t quite right, and nothing happens. The potential love affair dies in the bud. Then, you get to replay the game; but this time, you are in control of which events happen. Brilliant – instead of controlling the protagonists, we control the environment, hoping to get them together. We will fail a few times, revealing more about the people and the culture involved as we do so, but with a little perseverance, we can get them together. At which point we get an ending that is at least as negative as the other ones – finding somebody who loves you turns out not to be, by itself, the recipe for happiness. Light years away from Hollywood, absolutely, and for me this was the point at which I became really impressed by the game. The sad ending rang true. And yet, it was not the end.

In order to reach the real ending, you have to first find all the other endings. I think the developers should put just a little more effort into steering players who get stuck in the right direction. It is very hard to predict which events will lead to which endings, and the possibility space is large enough that one can get lost exploring it. I certainly did, stuck on 6 of 7 endings. In my particular case this was extra unfortunate because there happened to be a bug in the walkthrough, now fixed; but the game is so nice and atmospheric that having to use the walkthrough at all is a bit of a bummer.

But getting to 7 of 7 endings is certainly worth it, for when we accept our fate, rather than try to change it, the game turns into a neat little comment on the human condition. There are all these wild possibilities that we can fall in love with, but pursuing them will ruin the quiet happiness that is ours. Life is choice, and that means it is sadness, for every choice precludes an infinity of other paths we might have taken. But if we learn to accept the sadness, it is also a joy. Tolstoy’s Anna Karenina never learns this lesson; but the protagonists of Railways of Love do. For what is, after all, only a little game, I found it surprisingly moving and surprisingly deep.

For me, one of the highlights of the competition; I rated it 9/10. It is also now available on Android (and perhaps iOS?).

Monday, July 08, 2019

[IF Comp 2018] They Will Not Return

Another review from IF Comp 2018; spoilers ahead!

They Will Not Return by John Ayliff


They Will Not Return tells the story of a robot who spends his life cleaning up after his human owners, until one day a deadly virus wipes out all of humanity. The protagonist slowly comes to grip with this fact and ends up having an opportunity to master himself and thus become free.

In terms of craft, this is a very fine work. The writing is good, with the right amount of detail to bring the environment to life, especially the house in which the first half of the game is set. Letting the player see several phases of its slow deterioration is effective. The interactivity is also designed well: while the vast majority of choices doesn’t matter in terms of outcomes, quite a number of them work as opportunities to express the protagonist’s character. (E.g., will you hide the potentially compromising piece of underwear?) Although I did not replay to check, I assume the very last choices in the game do lead to different endings, and I also think I have a good notion of what those endings will be.

Given that all the demands of craft are satisfied, it makes sense to judge the story in terms of, well, the story, and especially the way it develops its themes. The main idea is surely the difficulty and importance of going from a life of servitude to a life of freedom. The protagonist is the ultimate servant, has a hard time coming to grips with the idea that there are no more masters to serve, and then gets to reprogram itself to be its own master. This is fine in itself, but I don’t understand the game’s obvious attempts to link this theme to contemporary capitalism. We have the house hidden behind a security gate; the graffiti against the rich; the revelation that cures were too expensive for most people; and the revelation that the robotics company was already planning how to make more money from the post-disease world. All of this suggests that the tale of the robot is meant as a parable that shines a light on our own society. But modern capitalism precisely does not work by turning workers into the ideal images of servitude. Modern capitalism makes us all into consumers, into people who want to work more because they believe that this will give them the freedom they most crave, the freedom to become happy by buying whatever they choose. Our happily serving robot is much more a pre-modern type, the “loyal servant” of yore. Hence the tale of the robot and the backstory about its capitalist masters fit together only uneasily.

More crucial, however, is the way the plot’s central problem is resolved. The robot was essentially a servant, doing its master’s bidding. Now it has to confront freedom. The central problem, then, is the problem of freedom, freedom after what Nietzsche metaphorically called ‘the death of God’. Nobody tells you what to do. All external guidance has gone. How then to face the infinity of choice? Well, here is the answer that They Will Not Return gives us: by taking on the role of masters ourselves. (The robot literally reprograms itself in such a way that all the code remains intact, but that they themselves now count as a master.) But that makes no sense as a solution to the problem of freedom! “How should I choose if nobody commands me?” “Command yourself!” Yes, but which commands should I give myself? I cannot understand myself as a master of myself, that is, as external to myself. I have to be myself. I have to make the choices. The victorious ending afforded to the robots is too easy, comes too cheap.

I end up, then, being quite critical about the thematic development of the game. But I want to stress that the very fact that They Will Not Return asks for discussion at the level of thematic development shows that it is a good effort. The game is serious; it wants to make us think; and it is well-written. But for me, it needs just a little more to really stand out.

Thursday, July 04, 2019

[IF Comp 2018] Writers are not Strangers

Continuing with my reviews for the Interactive Fiction Competition 2018. I wrote reviews for most games in a topic on the private authors' forum over at the interactive fiction forum. I'm posting the more interesting and more spoilery ones over here, and the less interesting and less spoilery ones directly on the IFDB. So, again: spoilers ahead!

Writers are not Strangers by Lynda Clark
(Placed 27th out of 77; I might have placed it somwhat higher.)

When I started the game, I was confronted with a barely coherent fragment of fiction that I suspect was supposed to be an interpretation of Space Invaders. Then the game asked me to rate that piece. Interesting. I gave it a 3 and continued with the story of Alix and her dying superhero mother. Just when I had almost forgotten about the fragment, Alix came home, started up her computer to see if anyone had rated the piece of fiction she was so proud of… and was heartbroken to see it had been rated with a 3. But, she decided, she could write another piece and see if that was received any better. Very interesting

At that moment, I formed a hypothesis about the work, which is apparently part of a PhD project. The hypothesis was this: the game gives a random story fragment to players at the beginning of the game; it asks them to rate it; it then shows us the writer’s response to the rating, engaging our mechanisms of personal sympathy; it shows us another random fragment and asks us to rate it; and finally, when all the data is in, the researchers will check whether people give higher ratings to the fragments after they’ve become aware of the fact that the writer will respond emotionally to the rating, that is, after they’ve become aware that writers are not strangers.

It was a good hypothesis, but also totally wrong. As the game progresses, the protagonist’s reaction to the grades becomes less and less of a concern (not surprisingly, since her mother is dying and the end of the world is nigh). At the same time, the fictional fragments themselves come to be about the way the protagonist reacted to the grades, which means that the random factor is missing. Furthermore, when I replayed the game I found that the ratings you give to the initial and later stories have a major impact on how the story unfolds, an impact that is in no way due to the impact of the ratings on the protagonist. If you give good ratings, the story will be mostly about the protagonist’s career as a sort of Youtube star; if you give bad ratings, the story will be mostly about the protagonist’s relation with her mother. There might also be a middle path focusing on the protagonist’s aunts and uncles, but I didn’t fully explore this possibility. None of this really makes sense on the aforementioned hypothesis, which certainly doesn’t require the game to have paths that differ so widely – unusually widely for any kind of interactive fiction.

So what is the point? (Apart from telling a compelling, layered story, something at which Lynda Clark certainly succeeds.) Well, one of the perks of being a university researcher is that I have access to this brief paper. I quote some relevant parts:

The second part of this quote explains the wide nature of the game, in which you can indeed experience one and the same event in very different locations with very different company. It is an interesting way of setting up an interactive fiction, and one that I enjoyed exploring. But I’m less certain what to make of the ambitions described in the quote’s first part. Does the game really force me to confront my impact on all areas of Alix’s life? Here, the structural features of the game actually seem to work against the author’s intention. First, a single reading doesn’t give the player a very good sense of the impact they had. I was surprised at how different my second reading was; but of course, that means that it was only on this second reading that I understood the impact of my choices. Yet at that point I have also made the opposite choices and am no longer very invested in any of them as ‘mine’. Second, I wasn’t really tempted to interpret what happened to Alix as a result of my choices anyway, since the link between the choice and the events seems to be non-causal. Different ratings for the first piece of fiction will generate different story events before Alix ever learns of the rating. So it’s not my rating that changed her life; it is the game author’s underlying system that pulls at her strings, something which I might want to explore, but for which I feel no responsibility.

The game does gain something from being part of a competition where one is judging: judging a game about the impact of judgments on authors certainly makes one self-conscious. And yet – that Space Invaders piece? It deserved that 3 I gave it. Writers are not Strangers, on the other hand, was well-written and made me think, so it deserves much more.

Tuesday, July 02, 2019

[IF Comp 2018] The broken bottle

After a bit of a hiatus, I'm back posting some of my IF Comp 2018 reviews. They're all quite spoilery, so beware!

The Broken Bottle by Josh Irvin

In Disney’s The Little Mermaid, the mermaid and the prince she so desperately loves end up marrying. In the original Andersen fairy tale, the prince marries a princess and the mermaid’s heart breaks. Disney changed the story in order to make it more… well, the gamut of possible answers ranges from “appropriate for children” through “commercially successful” to “American”, but one thing we can no doubt agree on is that it changes the nature of the entire story in a fundamental way. But what about an interactive version of the story where you could get either ending depending on the choices you make earlier? Would it work? What kind of story would it be?

In essence, these are the questions asked by The Broken Bottle, although the game features an original story instead of rehashing an existing one. But it is a fairy story, seemingly targeted at an age category of approximately 8-12 years old, where you can end up with either a stereotypical Disney ending or a very dark ending indeed, depending on the choices you make. At first, I got the Disney ending, complete with the sickly sweet moral that “there’s nothing more important than friendship” and a final sentence that reads: “No, they would never be alone again, and that was worth all of the treasure in the world.” It was almost more then I could handle. But I’m glad I replayed to check out the other possibilities and found that you can also end up betraying your sibling and even turning him towards evil. This juxtaposition of Disney and despair is what made the game interesting to me.

Of course, it is a fair question whether Disney and despair can be more than juxtaposed; whether they can also be integrated. Now The Broken Bottle certainly attempts to integrate them; I can even imagine that the dichotomy I’ve been spelling out was never on the author’s mind. But the integration doesn’t fully succeed. In reading the story, one is always aware of an uneasy tension between easy moralism and dark realism; between a world in which being nice and loyal always leads to a happy ending and a world in which scarcity of resources and human egotism make the tragic inescapable. Insofar as the game tries to bring these worlds together, it doesn’t succeed. But it is an instructive failure.

I wasn’t blown away by the other aspects of the game. I found the interface rather annoying, not least because it forced me to have very little text on a page even though I was playing the game at a very high resolution. Another weakness of the interface was the absence of Save/restore. On the positive side, the book format does allow the author to put in some stylish images. While the character portraits are merely okay, the pictures of the outside and inside of the tree house are really good and help bring the world to life.

Interactivity is very low: there is a lot of text between the choices, far more than I am comfortable with when reading an interactive work, and most of the choices turn out to be meaningless. (Do you want to buy A or B? Never mind, whatever you choose you’ll get both of them. How do you want to leave the tree house? Don’t bother, the results are identical.) While meaningless choices are a common trick of the trade, one should be wary of them if one’s game has almost no choice points to begin with. Fortunately, this situation improved as the game moved on, where choices become both more numerous and more meaningful.

The truly crucial choice is the one between saving your brother or going along with the evil guy’s offer. I was rather confused by the way this was presented, in several ways. First, I was confused by the situation itself. Given that the evil guy has me in his power, why would he offer me a deal? And why is the content of the deal that he’ll keep my brother, given that just a few pages eariler he has told me that what he really wants is me? I was also confused by the phrasing of the choice. Part of the evil guy’s offer is that he’ll spare my brother, so it seems that saving your brother and accepting the offer are not mutually exclusive; but it turns out that they are. Finally, I was confused that depending on earlier choices, the saving your brother option was interpreted very differently by the game. In the worst circumstances, it is interpreted as not trying to save your brother, but instead taking up the evil guy’s offer. Huh? I think this is a part of the game that might be fruitfully clarified.

Thursday, June 27, 2019

My interactive fiction archive 2004-2019

I’m putting a massive 372 MB zip-file online which contains more or less all of my Interactive Fiction related creations of the period 2004-2019. All my completed games, with source code and assorted extra files; all of my abandoned, incomplete games; reviews and essays; backups of my blog, my IFDB reviews and my forum posts. I will also upload this to the IF Archive, since the archive’s administrators have told me that they do accept such collections.

What is the purpose of this archive? Partly it is to safeguard a very small but I hope not entirely insubstantial part of IF history, so that future ‘digital antiquarians’ can look for whatever they might then be looking for – even if sites like the IFDB or Blogger go offline. Partly it is for those of you who are curious to delve into some of the dusty corners of my IF directories and see, say, the somewhat impressive number of ATTACK-based I games I started and then quickly abandoned before I hit on the idea of make the roguelike that would be Kerkerkruip.

Probably the most relevant part of the release are the source codes to my completed games. Some of these were already available, but many were not. So whether it is The Baron or Terminal Interface that you are interested in, the source is here! (I’ll also be uploading these separately and linking to them from the IFDB at some later point.) What’s more, I am relaxing my source code licensing. In the past, I licensed my source code under the GPL; from today, I’m giving you the choice between GPL and Creative Commons Attribution-NonCommercial, the latter of which might be more useful to some of you. (Since it does not require you to publish your own derivative source code under the same license.)

The archive’s main directory contains a file that serves as a table of contents, so please view that for further details. Have fun! (And perhaps consider doing the same thing, if you’re an author.)