IF Comp 2021: The Last Night of Alexisgrad

The Last Night of Alexisgrad is a choice-based game for two players. This is an unusual but venerable format – the first I’m aware of is the Duel Master series, first published 1986, which were sold as boxed sets of two books. The players in Duel Master were competing with one another, sometimes in asymmetric contexts (the second in the series, Blood Valley, is about a hero being Most Dangerous Game’d by a dark lord) and exchanged code keys in order to pass state information back and forth while keeping most of what they were doing secret.

At first glance, Last Night works kind of like this. You play leaders on opposite sides of a conflict in something roughly resembling 19th-century central Europe; a monarchist General who’s doing a revanche, and the recently-appointed Dictator whose Republic is crumbling under invasion. Your characters don’t directly interact at first; you each have your own narrative, which you read without sharing information until you come to a choice point, at which stage you exchange codes, input the code you’ve been given, and continue play. (I played over voice chat with the inimitable Astrid Dalmady.)

The first thing is that it starts out with a lot of wall-of-text. This means that you and your friend have planned to do this thing together, and now you’re sitting silently and reading instead of doing something together, but you’re reading kind of faster than ideal because you don’t want to make your friend wait around. The plot does start out with you already in the thick of the action, which is a solid choice, but even so, you get several choices deep before your characters are even aware of one another’s existence. And a lot of this isn’t necessary – the General gets several paragraphs of peroration about the mindsets of soldiers versus killers and what the attitude of a real hardened veteran is and blah blah blah, and like, I guess this fits in if I’m reading a fuckin’ Bernard Cornwell novel or whatever, but I’m not trying to fill time on a long plane flight, I’m playing a game with someone and would appreciate something much more tightly edited.

The thing here is that the Duel Master books were kind of, well, crunchy. They had maps you could traverse and re-traverse, inventories with spells and items that could be used (and expended) at many different points, combat systems. They had relatively brief text, partly because of the space constraints of a physical book but also because of the need for efficiency. The point of having a second player was so that you would have a Worthy Adversary, someone who you legitimately competed against and were trying to out-think. This wasn’t always very satisfying, in practice – the ratio of fiddly book-keeping and waiting around to actual Doing Shit wasn’t the best – but in theory there were actual, y’know, tactics. Last Night is not crunchy, and it isn’t really a tactics game. The Dictator always loses, and the game is mostly about the how rather than the if, about what sort of leader you’re willing to be: do you send in a surgical infiltration team to take key prisoners, or burn the whole Senate building to the ground? Do you offer your captive enemy terms or just have her shot? Are you willing to use your people as human shields? What terms are you willing to surrender under?

But the problem here is that ‘what are you willing to sacrifice’ doesn’t really feel like a weighty choice. The General always wins, so choosing to do so in a more brutal manner doesn’t feel like a cost-reward calculation so much as an aesthetic preference. The Dictator always loses, so why make terrible sacrifices to stave off the inevitable? And, OK, there are some consequences here – but those costs are all very hypothetical, offscreen costs, especially for the General. The King might be less happy about how I handled things. I might have to pay someone some money. Yeah, sure, but I don’t really know how much leeway I have with the King in the first place, or how easy it’ll be to round up that money, and it’s not going to matter within the scope of the game. The result is that the stakes feel very marginal and theoretical, and I felt pretty off-hand about choosing them.

There are several people other than the author credited with worldbuilding for this game, so I have my suspicions that this is a story set in an established shared world of some kind – my immediate thought was an RPG campaign world, but it could be other things. That kind of comports with the sense that the outcome of the conflict is mostly inevitable – this story has already been written and you’re filling in the details – but it really clicks with the sense that the stakes here are about the consequences of details that aren’t explored onscreen. 

There are definitely some advantages to this being a world that’s already well-fleshed-out; this does feel like a real place. Astrid said ‘this author definitely played Disco Elysium,’ and I’m not 100% on that, but it does have that sense of grimy post-disillusionment Europe.

So I’ve played quite a lot of gamebooks as a group activity, and in general it’s great, but it’s mostly great because of what you’re sharing. One person reads, and the rest yell commentary, heckle, discuss choices, gasp in horror, and the game doesn’t have to be good; in fact it’s pretty great if it’s terrible because that’s funnier, and you’re all in it together. When a narrative game asks you to play with someone else, but it also says you shouldn’t share information with the other players, I always feel immediate distrust – very often it’s one of those things which was in the designer’s original Vision, and at no point in the dev process did anybody ask them ‘hey, so, is this element actually contributing anything or are you just attached to it?’ Because, hidden information, if the gameplay doesn’t deeply rely on it, can cost you a great deal without adding very much. If I’m playing something multiplayer I want it to encourage talking with my friends! If what the other player is contributing is meant to be about the uncertainty of war – well, it would be pretty easy to make an AI player take their moves.

One of the big things that feels different between playthroughs: in some versions of the story your characters end up meeting face-to-face and having a conversation. And that’s kind of weird, because you’re just alternating handing off codes to someone and then going silent and reading, but it feels more fun because it’s more immediate, you have permission to peanut-gallery about it, it’s more social. I’d be interested in a whole game that was about this.

Here’s the other thing about playing with a partner: you’re a lot less likely to play exhaustively. If the game doesn’t absolutely grab both of you, you’re probably going to call it after one or two playthroughs, because – look, sometimes I’ll stick with a game I don’t love for longer than I think it deserves, but no way am I going to ask a friend to put up with that. In fucking around randomly on my own afterwards, I managed to find a scenario where the General can make some obviously-rubbish decisions and kind of get to a situation where he accepts a lacklustre victory that, from the Dictator’s perspective, is basically a good outcome. But if you’ve played a couple of games with one of you as General and one as Dictator, it’s probably going to look as though the only real thing you’re deciding is the details of how the Dictator gets her ass kicked, and that’s mostly up to the General. Which – if this is meant as a competitive, tactical game – isn’t much fun for either player, and especially not the Dictator. Well, OK, so probably this isn’t meant as a Fun Game about competitive tactics – but all of the choices you’re making are presented as tactical and strategic considerations, so it’s going to feel as though it is. Moment-to-moment, the General mostly gets to decide how much brutal suffering he wants to inflict, and the Dictator mostly gets told about how much of a failure she is, and – if you consider this as a role-playing game, an enacted thing rather than a static text, that kind of sucks, actually?

Another thing about exploring this solo: those walls of text, frankly, are a lot less obnoxious when you’re playing alone. And I think it’s kind of a major problem when a story reads better when not read in the intended manner. That suggests pretty strongly that the writing isn’t well-adapted to its format. This is perfectly capable as prose, but it’s not a good fit as game writing, at least for the particular game that it’s in.

Which, y’know, is OK? This is not a new form, by any means, but it’s not a heavily-explored one, and anything in this space is going to be something of an experiment. I think this would be an unusual thing to design for, and would in in particular present some special challenges for testing. I’m not sure that there’s a solution to the problems it’s poking up against – see also Aspel, I think, in the sense of small-group multiplayer IF where the multi-player part mattered without just becoming a MUD – but it’s an interesting challenge and I’m glad to see it explored.