Words: 2273 Approximate Reading Time: 15-20 minutes
One game I finished recently was Tunic, which was a delightful game. There were some issues with combat, as the game demands the player to accomplish tasks within a system that isn’t exactly suited for those tasks. But aside from that problem, it’s a lot of fun to explore, and the game offers a bunch of puzzles that can really wrack your brain.
One of the interesting elements of the game is how it teaches the player. While you naturally experiment with the buttons and figure things out on your own, as you play through the game you encounter small white diamonds that are actually pages to an instruction manual. By which I mean a literal manual within the game itself. The manual generally reveals information as it becomes relevant – every mechanic in the game is available from the start, but the buttons are set up so that you may well not know what to do until information is revealed to you through the manual. It’s a really clever way to teach the player and layer in mechanics, while also allowing more knowledgeable or experimental players to do things out of sequence.
The manual is also interesting because it contains small notes written into the pages. Usually these notes clarify small pieces of information (such as explaining how much health is healed when you upgrade your potions), but sometimes they reveal tips for how to solve some of the game’s final puzzles. It captures the feeling that you’ve purchased a used game and are trying to decipher notes left by the previous owner.
But the thing I want to focus on is the game’s language. The manual uses both standard letters and a newly constructed written language, composed of a series of obscure symbols that are undecipherable at the very beginning. Significant chunks of the manual are written in this constructed language, especially the backstory to the game. This places demand on the game’s narrative to be minimalistic – the player has to figure things out from how characters interact and how the world is set up, because there’s no text to provide further context.
As you continue through the game and pick up more pages, you start to get little clues on how to decipher the symbols of the language. I am not really one for these kind of puzzles, but others have put in a lot of work to translate the language. What they’ve found is that the words are themselves composed of symbols that denote sounds that then combine to form English words. So once you figure out the different symbols, you can then translate the words and thus everything in the manual.
Now as I said, this kind of puzzle is not the kind of stuff I normally interact with. But it is interesting. And putting together a new written language that is more than just “here’s a symbol that represents an English letter” is quite difficult.
But I imagine there are quite a few players who also aren’t interested in or have the ability to figure out the language. And that’s what I wanted to talk about. Because as interesting as this system is…there’s so much demand placed on the player that it discourages players from wanting to engage with it.
So I wanted to examine the topic of constructed languages in video games and how to deal with them from the perspective of a player.
Puzzles and Time
To put this bluntly, puzzle solving is a subject that revolves around time. If a player has enough time and patience, they can eventually try everything to ultimately reach a solution, whether it is through careful thought or brute force. But that it’s possible does not mean that players will actually want to spend that time. So puzzle design is about figuring out a sort of sweet spot in terms of time.
This is one reason why more self-contained puzzle games work so well. Games like Baba Is You and Patrick’s Parabox work so well because each puzzle takes place within a single level. And it is not necessary to complete every level to get through the game, even to “beat” the game. Which means that if a puzzle is taking too long, the player can just stop and tackle a different puzzle.
I mention this focus on time because there are a few possibilities for how to solve this time problem from a player standpoint. One is to just keep going, obviously. If you have enough time, then eventually you’ll get to the end. Another is to simply give up: if you don’t want to keep trying to solve puzzles, you can just stop playing – either in the sense of “I’m not going to get to the end of the game” or “I’m not going to complete 100% of the puzzles.” And yet a third solution is to just look up the solution: to see what someone else on the internet says you should do – either in the form of a hint or in the form of the puzzle’s solution.
Unfortunately, all of these solutions are somewhat…bad. A player who sinks hours upon hours trying to solve a single puzzle, or feels like just giving up, or who just looks up the solution is going to wind up feeling frustrated for one reason or another.
Now there’s not a perfect way to eliminate these problems. Arguably the only sure way is to make your puzzles so simple that people couldn’t get stuck on them, but then the result would be to make a puzzle game that’s boring. So walking that tightrope is always going to be a problem of puzzle design: how do you make a puzzle that is difficult, but not too difficult? I won’t be trying to solve that problem here, because I’m not sure if it is an entirely solvable problem to begin with.
Instead, I just want to focus on how game design can provide shortcuts to these kinds of problems, especially as we get further and further away from the “self-contained” puzzle system.
Which is ultimately what Tunic’s language is. It’s a puzzle system that requires pulling information from all these different places to put together a key which is then used to decode the text you find throughout the manual.
But that means that the language puzzle is a significant timesink. If you’ve never tried doing any form of translation, it’s a very meticulous process that’s easy to mess up. And there’s quite a bit of text to translate in Tunic – not only is about 40-50% of the manual’s text written in this constructed language, but all dialogue within the game is written in this same language, of which you’ll find about a dozen or so characters speaking various lines in this language. So there’s quite a lot to go through.
And that much work means a lot of players are going to look at the puzzle and just immediately say “nope.” It’s not clear whether the best case scenario is for them to just play the game normally and then never bother with the language (it is at least a testament to the game’s design and narrative that decoding the language is not necessary to get through the game), or for players to just look up a translation of the manual and dialogue (plenty of other players at this point have done the work, and you can even find a full translation of the manual online).
Which should get us to ask how a developer could try to address this problem internal to the game itself. I want to offer a couple of solutions, though by no means should we regard these ideas as exhaustive.
The first idea is to construct a language decoding tool within the game. The most simplistic version of this can be found in Final Fantasy X. The game contains a group of people called the Al Bhed who speak in a language that merely involves switching around English letters (such as “A” becoming “L” and “M” becoming “W”). The player can find small booklets which reveal the substitution cipher one letter at a time, and as those letters are revealed the subtitles to the Al Bhed characters are changed until eventually their dialogue is fully readable. Now of course this is a very simplistic method, but provides the basic idea. By giving the player some tool for translating the language given to them and thus transforming the symbols into readable words, players have a method within the game to translate, rather than needing to do everything by hand.
The obvious benefit to this system is that it strongly preserves the puzzle element of the game. It encourages players to take clues and apply them, and provides players with a feedback mechanism to let them know that they are succeeding. And that feedback is really useful for encouraging players to keep playing.
But putting this kind of system together would be incredibly difficult. I’m not even entirely sure it would work with Tunic’s language. For one, you’d have to figure out a system that couldn’t simply be brute forced, or at least not brute forced so incredibly easily that it defeats the purpose of the puzzle.
The second solution is to reward players for getting through the game by doing the translation work for them. If players want to do the translation puzzle themselves, the information is given to them. But upon beating the game and starting up New Game Plus mode, the game could simply do the translation for the player (or let them know that a translation option is available). This system is used in The Legend of Zelda: Wind Waker. On your first playthrough of the game, you’ll encounter several characters who speak in an ancient Hylian language, which is represented by a bunch of unreadable symbols. But on your subsequent playthrough much of that language in dialogue is translated for you, allowing you to get a bit more context. It doesn’t change a lot, because that dialogue isn’t terribly important, but it does provide a bit of encouragement to keep playing a bit longer to see what you missed.
The reason we’d want to apply this system is that while translating the language in Tunic is not necessary, that’s not a good measure for these things. You can play the game and get through it without knowing the backstory. But still, having that full context is something the player wants. The fact that the language is solvable and can be translated means that the player isn’t simply guessing based on clues in the world and characterization, but there is literal information that is obscured to the player. So it’s not necessary…except that it kind of is.
Now just having the game translate for the player isn’t all that interesting. But it doesn’t really need to be interesting. Tunic has quite a few complex puzzles that can be fun to solve. The game already has enough content to fill up about 10-20 hours with a fun experience, which is definitely worth its asking price. So it doesn’t need to make the translation puzzle something that adds more time to the player’s engagement. Again, this is something that can ultimately be left to the player’s choice. Do you want to solve the puzzle? Go right ahead. And then if you try and give up, the game can allow you to just get the translation. Do you want to just know what information was being obscured from you? Then you can choose to get the translation as soon as it is made available.
This goes back to what I said about time. Translating takes time, and the system provided by Tunic means that it maximizes the amount of time required. But that also maximizes the number of players who are going to wind up frustrated – and especially the number of players who just resort to using the internet to find an online translation.
Concluding Remarks
I find Tunic’s constructed language fascinating. It’s a clever system, the symbols look really strange in a way that makes the language feel unique. And more important is what this constructed language means as a constraint on game design. Because so much of the game is effectively locked behind this language that the player cannot read, the game needs to communicate its systems and narrative without being able to rely on readable text. And I think the game does a good job with this communication.
But while I like the language conceptually, I find its execution falls short in one key way. While it’s possible to solve, the language feels almost like it is discouraging people to solve it. People will – and have – powered through and done the work, but there’s little incentive for your average player to actually bother. Why should I bother translating the language when I can be confident that someone else has already done it for me? Especially when figuring out the cipher and then applying it is going to take hours of my own time, versus five to ten minutes by using the internet? Or why should I bother translating it when I can just…give up entirely?
Which is why we should think about not just how players can solve the problems of time and frustration, but how game design can address these issues. The more easily players can get feedback for these kinds of puzzles and use those systems to directly interact with the game, the more engaged players will feel. The more that players will want to solve those puzzles, and solve those puzzles themselves.
Abstracting Games 