A game about linguistics

[FlatAssembler]

Hey, guys!

I've just made some game about linguistics in JavaScript. You can see it here:

http://flatassembler.000webhostapp.com/etymologist.html

Do you have some ideas on how to improve it that are easy to program?

I don't think I'd be able to add anything complicated to it. I am not a programmer and I can barely manage the 2000 lines of code it has now.

[hats]

Can I ask what your goal is? Do you have any linguistics background? This is more coding than I could do, but I do have notes about its contents.

First, your framing is misleading. It doesn't work to show only one potential word pair at a time. Remember, linguistics determines whether two words are actually cognate by identifying 1) corresponding meanings and 2) recurring phonetic correspondences. One instance of an apparent phonetic match (especially without considering meaning) cannot show that two words are cognate. It can't even show that when every letter is the same! For example, Malagasy and Russian are not related just because Malagasy has vorona 'bird' and Russian has vorona 'crow'. You'd need more examples of v:v in both languages to show that that sound generally corresponds.

My simplest suggestion for re-framing would be to say something like, 'I have a set of cognates that have been shown through independent linguistic analysis to be cognates. This is a game to see how well your intuition matches those conclusions.' That seems like a harmless thing. If anyone walks away from this game thinking that's how cognates are established, you're feeding people an inaccurate sense of what linguistics can do, which I can't advise. Doing a kind of matching thing could be fun for people, if you're clear that they're guessing, not following a real method of analysis.

Even better, although it sounds like a complicated coding project, would be to have people pick out recurring correspondences in a set of words. Like: cook (English) : kook (Dutch) and milk (E) : melk (D) shows a k : k match between those languages. If you did that—perhaps using data from some of the historical linguistics problem set type books, if they're in the public domain or this qualifies as fair use—this could be a good resource for linguistics 101 classes.

[FlatAssembler]

Look, the Part #1 is supposed to give players a sample of 26 words in each simulated language for them to be able to study the phonologies of those languages. Having to choose the correct cognate out of four random words (which is usually quite easy) is a way to make the players think (even subconsciously) about the phonologies. I realize that can be somewhat stupid and tedious to someone who has a strong background in linguistics, but I don't see any alternative. We need to somehow give the players data to solve the Part #2 and Part #3. Showing them a large table and saying "study the phonologies" would be even worse.

[TakeruK]

I have no expertise in this area at all. But I am not sure if I am the intended audience of this game. If you're seeking feedback from outsiders to linguistics then I'll provide my thoughts! Just let me know.

[FlatAssembler]

Well, I haven't really tried to make the instructions clear to people who know nothing about linguistics, and that could be quite hard to do. If you have something other than "The instructions are unclear to someone who knows nothing about linguistics!", feel free to say that!

[TakeruK]

4 hours ago, FlatAssembler said:

Well, I haven't really tried to make the instructions clear to people who know nothing about linguistics, and that could be quite hard to do. If you have something other than "The instructions are unclear to someone who knows nothing about linguistics!", feel free to say that!

Sure, I think the initial page instructions make sense. I like the examples you gave there and even though I have no idea what a "phoneme" is, what I understood was that you are trying to get us to match up letters that have the same position and sound in two languages (if this was not right, then oops!). I also like the example/test case of twig/zweig because it reminded me of the last screen. I was surprised to see so much text pop up after clicking on that answer, but it provided more instructions, which was nice. I did not really understand what "cognate" means though. And what I got from the explanation was that sometimes letters appear in the same place and have similar sounds(?) but aren't cognate and instead are "borrowed". This distinction is made me lost but I forged ahead anyhow!

Then, all of the made up languages really really confused me. I ended up just clicking on the choice that seems to be similar in length and/or have similar vowels. It seems to work most of the time. The explanation that pop up after an answer made little sense to me. For example, the prompt was "rorir" and I chose "rorur" since it looked the most similar and got a message saying, "Correct! The root was *rorr". I have no idea what is meant by "root" and what the * means and where "rorr" came from (since neither word had rorr in it). 

To be honest the first time, after 5 or 6 of these, I got too lost and gave up. However, after returning here and seeing that there is a part 2 and part 3, I went back to went through the 26 questions in part 1 just to see what part 2 is.

Part 2 with English and German made sense because I know English and not German! But I am not sure how I would transfer the skills from that example to the actual task. I recognize those * things from Part 1 but I have no idea how I would distinguish which word came from Language A vs Language B. I just clicked a few things and hit SUBMIT.

You can probably guess that Part 3 was hopeless for me. I don't understand the symbols at all. I just guessed.

In the end, I got 29/55. I guess that's a passing score? lol

Note: I only went back to do parts 2 and 3 after your response to my question. So I guess if I had stuck to the end of Part 3, I would realise that the intended audience is for someone who actually knows these linguistic symbols. Oh well. I still had fun going through it once knowing that I would be seeing things completely out of my realm of understanding. Maybe some overview at the beginning to let people know what to expect in each part would be nice? 

[hats]

It just seems like you're aiming for an extremely narrow audience, if it's true that @TakeruK knows too little about linguistics and I know too much. I've only taken two linguistics courses! Did you make this because you're teaching a linguistics 101 course? I could imagine it as an assignment the week before you start the comparative reconstruction unit, but after you've done phonology.

[Yanaka]

I don't know anything about linguistics, but when I played the game I often had lucky guesses based on how some words looked like. I was just wondering why there wouldn't be something else such as little explanations or something to give it a goal?