AGM MOTHER 3 Translation Interview

A few years back I was asked to do an interview for Active Gaming Media, which focuses a lot on game localization. It was kind of surprising since it seemed to focus on the professional side of things, but I was asked to discuss the MOTHER 3 fan translation. Not only did they post the interview, they had it professionally translated into Japanese for Japanese readers to check out 😯

In the past few years since, the site has gone through changes and that old interview is gone. I managed to track down a copy of the English one on archive.org here, but I’ve included it below for easy reading.


(start of article)

Japan-only titles fail to carry the air of mystery that they once did. Games which once existed in gamers’ imaginations as only the rough scans offered by the popular game publications of the 80’s and 90’s can now most likely be viewed in full video with the help of no more than a PC and a search engine, rather standard tools nowadays in most gaming households.

Which is why a game like Mother 3 is such a special case. A frustratingly special case. EarthBound, the series’ second installment and the only game in the series to be released outside of Japan (for the Super NES), back in its day seemed to come and go with little fanfare or recognition despite a reasonable marketing push on Nintendo’s part. Yet as time passed, both the game and the series have grown to be regarded as truly special titles which are held dearly by a passionate fan base, which arguably goes unrivaled in terms of “dedication” when held up against nearly any other game or game series. However despite passion and petitions, Mother 3 ended up as one game that seemed destined to find its audience with the third installment but was never given the chance to shine on Western store shelves.

That is until Clyde Mandelin, likely better known by the Mother 3 fan community as Tomato, took responsibility into his own hands, setting out to bring, while technically “unofficial”, one of the most highly praised and appreciated game translations in the medium’s history.

In the following interview, Clyde shares with AGM what he’s learned from working on Mother 3, his general philosophies on translation and localization, some of the technical details related to working on a fan-translation as opposed to something commissioned officially, discusses his personal relationship with the popular series, and tells us why localizing from Japanese to English is like reproducing the Mona Lisa with broken crayons.

AGM: It’s unquestionably your work on the Mother 3 translation patch that has garnered you a great deal of well-deserved attention from throughout the game community, which is what I’d like to focus on, but first would you mind giving just a brief background about your work in translation/localization leading up to your work on Mother 3? What sorts of projects have you worked on and in what capacity?

Clyde Mandelin: Prior to Mother 3, I’d worked on many unofficial game translation projects, many of which were console RPGs that came with all of their unique quirks and challenges. A few of the more well-known ones were 16-bit games like Star Ocean, Bahamut Lagoon, and Live-A-Live. I did the majority of this translation work on the side during college, partly out of love for the old classics and partly to gain real translation experience. I also studied computer science for a few years during college, which comes in handy from time to time to this day.

Once I was out of college, I was able to use that experience to translate for a living, mostly working on Japanese anime but with an occasional game here and there. I’ve been lucky and honored to work as the translator on some pretty popular series, things like Dragon Ball, One Piece, Tenchi Muyo!, and Crayon Shin-chan. Naturally, each series brings its own set of challenges to wrestle with, and with each passing project I feel almost like I’ve “leveled up” as a translator. The kinds of things I work on now, I’d never have been able to do when I first started out, for example. It’s usually tough work, but it’s also gratifying with each passing project.

AGM: What from those previous experiences did you take with you which you feel lent to you being able to produce the quality of end product that Mother 3 turned out to be?

Mandelin: My experience with the long, complicated process of fan translating RPGs definitely helped out with Mother 3. I knew what needed to be done on both the translation side and the programming side, and how it should and shouldn’t be handled. Being able to handle a lot of the programming myself also meant I could be extra-picky with the presentation side of things.

Through my professional work, I learned how to multitask between multiple projects and between different parts of individual projects. This helped me a lot with Mother 3, where I was constantly juggling translation, programming, and localization duties all at once. Also, most of my professional work involves subtitles, which requires trying to pack as much meaning and information into as few words as possible, which is something Shigesato Itoi is famous for.

Most importantly, I learned – through both fan translation experience and professional experience – how to work effectively in a team environment, something that’s usually pretty tough over the Internet. The Mother 3 translation project was basically a result of good teamwork and love for the series.

I also had a lot of experience with the Mother / EarthBound series prior to translating Mother 3. I helped found Starmen.Net and I’d translated news updates about Mother 3 for years. I had dug into the Japanese versions of the Mother games and compared them in detail to the English versions. I had also translated Mr. Itoi’s interviews and articles from long before Mother 3’s cancellation and followed along every step of the way, up to the Mother 3’s GBA release. I think all of that insight together lent the final English translation an extra spark.

AGM: What variety of material have you worked on for translation outside of games and game-related media? Is there anything in particular that you’ve gained from grappling with another form of content that you feel really helped you with working on games, or maybe changed your approach to working on games? In your personal opinion, what sorts of experiences do you think translators/localizers of games and gaming media should have in order to help them become better at their craft?

Mandelin: Like I mentioned before, the majority of my professional work is with things like Japanese anime and movies, translating scripts and doing subtitles. Subtitles have their own limitations that need to be worked with, namely the length of time a subtitle can appear and the amount of space a subtitle can take up on the screen. Games have different constraints, like memory usage and programming limitations. Sometimes limitations are shared between mediums; both subtitle text and video game text have screen space limitations, for example. Learning to work with different constraints has made me more flexible as a translator and made it easier to handle new constraints I run into.

Besides the obvious skill of being able to translate, I think good game translators need good writing skills in the target language, a respect for and an understanding of the source material, a respect for the audience, lots of creativity, the ability to be detail-oriented, and most of all, patience. Most of those skills come over time, with experience and maturity.

Localizers need much of the same. They need to be able to write in the target language much better than translators. They also need heavy-duty reading comprehension skills – it’s all too easy to gloss over or change an important nuance because you weren’t reading the raw translation closely enough.

Sometimes the translator is the localizer too. Translator-localizers need all the above, but they also need courage – the courage to take the text beyond a super-literal translation. It sounds kind of dumb, but I find it’s actually tough to do in practice – a translator is supposed to recreate text, but a localizer is supposed to make changes to text. But those who’re able to pull it off make some of the best game translations out there.

For aspiring translators, I’d recommend trying to translate all kinds of different media. Try translating books, comics, films, games, radio programs, TV programs, articles, interviews, and anything else you can find. Try to do as many different things as you can. Try to do localization too, whenever you can. Try to pull your text away from a super-literal translation and give it life. If you can do that, you’ll be a success at this!

AGM: How did working on Mother 3, something done on your own time as a labor of love as opposed to being a contracted project, change how you approached handling the localization? What was your strategy for tackling a project of that scale on your own?

Mandelin: The most obvious difference with Mother 3 was that I wasn’t bound by a deadline – this meant I could spend a lot more time on the translation and localization, and could work at more or less my own pace. Of course, that pace was still pretty quick because of fan demand, but it gave me a lot more freedom.

Being able to work on the game in this setting also let me get around some of the red tape that comes with professional work. Names and terminology is probably the biggest example of this. With professional work, you often have to get every single name and term approved by many different people who often work in completely different companies. Or sometimes you’ll be given a list of names and terms at the very start. But in both cases, you’ll almost always find all kinds of mistakes and problems with the names, and it causes a lot of headache trying to get things fixed – which you often can’t. With Mother 3, I was able to bypass these sorts of bureaucratic obstacles.

For Mother 3 in particular, I used a combination of methods from fan translating and professional translating to speed up the process and to improve text quality. One example being a custom text editor my partner Jeff put together. This was far more useful than the Excel spreadsheets most professional game translations are done on – it was a snap to navigate between text from different parts of the game, it allowed me to change multiple instances of text throughout the entire script in just a few clicks, it let me track my progress, and it let me preview in real-time the text as it would appear in-game. It had many more features, too. It’s really useful and I wish I could use it in all my professional work!

What was cool was that whenever I needed a new feature of some sort, Jeff would implement it in almost no time. We had really good tools and really good teamwork. It almost reminds me of the really top-notch translation teams that do everything in-house.

Without the luxury of all of this, it would’ve taken a LOT longer to localize Mother 3. I almost cringe at the idea of having to translate the entire thing in a spreadsheet, which is kind of funny now that I think about it, since apparently Mr. Itoi typed all the lines into a spreadsheet when writing the game.

AGM: According to the Mother 3 fan translation site, translations of the game are currently at various stages of progress in 7 different languages (9 if you include the multiple Spanish and Portuguese versions). What kind of involvement have you had in organizing these works, if any? Have you been in communication with the people working on these and has it been a collaborative effort? Have you been a “project manager” of sorts as well? Are all of the translators working from the original Japanese text, or are some of them working off of your English translation?

Mandelin: Surprisingly, the Mother series has a very international audience. So we specifically designed our tools and our files with multilingual translations in mind. I posted about this on the project’s site one day, and before we knew it we had dozens of people eager to localize the game into their native language.

I waited until after the release of the English version to really coordinate the other translations. It was easier this way and it gave the other translators a chance to really dig into the game and understand everything. After that, I assigned a project manager for each language team and let them have the necessary tools and files to work on their own. Jeff and I interact with the project managers whenever questions or issues come up, but for the most part we let the translation teams manage themselves.

The other translators are working off of the English script, so it is a translation of a translation. That’s always a wary prospect, so I left notes in the script for the other translators while I was translating the Japanese script into English. Every so often they’ll ask for clarification about lines, too. I think these things will help improve the quality somewhat. One team did ask to work off of the Japanese script at first I think, but then decided to go with the English one. I’m not sure why.

AGM: The caliber of writing in the Mother games is something special, something entirely unique compared to the offerings found in many other Japanese RPGs (or other games in general), and is an essential element conveying the game’s charm and lending to its appeal. What’s your personal view on narrative in games and how do you perceive and engage with the narrative of Mother 3?

Mandelin: I do think most games – and most entertainment – has pretty generic writing. I’m not a writer myself, so I’m in no position to complain. But it usually feels like game text is purely functional – like the generic RPG townsperson who stands in one spot and tells you the name of the town; or like filler text – “Okay, we have some characters in this part of the game, we need to make them say something now.” It’s hard to find REAL writing in games.

Sometimes it’s hard to explain what that means, but it’s kind of like acting (and voice acting) in games – it’s usually not very good, and if a movie or a play used that same level of acting they’d be ridiculed to no end. Game writing is a lot like that, it’s just not as obvious and most of the time it doesn’t matter, unless a game is purposefully text-heavy. Action games don’t need good writing, for example.

With Mother 3, it’s hard to find purely functional or filler dialogue. Almost every character in the game – even the non-playable townsfolk – has a back story and many changes throughout the game, to the point that each character can have a timeline written up for them. They all have their unique qualities to them too – from their names, their speech styles, and even to their different bleep-bloop “voices”.

Even when Mother 3’s gameplay does call for purely functional text, Mr. Itoi approaches it in an unusual way. For example, there’s a point near the end of the game where the player is supposed to be blocked off from going to certain areas. Mr. Itoi accomplishes this by putting crowds of people in the way and giving them EXTREMELY filler text. Things like, “Blah blah.” and “Uh oh.” It’s like he’s saying, “You know what this really is and I know what this really is, so let’s not patronize each other.”

More than anything though, the writing in the game feels real – apparently Mr. Itoi would actually say every line aloud until he was happy with it, which is also why the Mother series is primarily kana only. To the dismay of his colleagues, Mr. Itoi wouldn’t hesitate to throw out entire blocks of text later on if he wasn’t happy with it – he was very picky about what people said in the game and how it was said. The end result was a very polished script, though.

AGM: One main reason that I’m interested in your answer to the previous question (above) is because translation is very much a writing skill, something more than just a translation skill. I would think that your feelings about the game’s narrative would impact how you approached the game’s localization. Can you elaborate on that idea? Is it something that you were conscious of and paid attention to? Why do you feel that this is significant (if you do)?

Mandelin: Shigesato Itoi is well-known as a creative writer, but I’m very much NOT a writer. So my strategy was to first approach the text the way I would work with most everything else, as a sort of first draft. After this, I put the translation aside for many months (a time which I used to work on the reprogramming) and then returned to it with fresh eyes. I was actually surprised, because for the most part, Mr. Itoi’s writing still shined through – the “what” of the text was still there and very charming in its own way, it was the “how” of the text that was missing.

Since I have so much familiarity with the games and with how EarthBound (Mother 2) was localized, I was able to use that to polish the script into a second draft. This put a lot of that style back into the text.

I did have a third draft planned, where a much more skilled writer would go in and spice my writing up, but there was a conflict of schedules and in the end only a few key lines got the third draft treatment. In the end though, I think this approach, together with the presentation side of things, helped the game’s script feel a lot like the official EarthBound localization.

AGM: Was there any area of the project where, prior to beginning work on Mother 3, you had decided that you were going to work on the localization differently than you would have if, say, it had been a project assigned? Naturally there was no deadline or paycheck, but was there any element of the process where you consciously said to yourself, “I’m going to do it this way because it’s in the best interest of the final product, and I can do that because it’s a luxury that I have given my situation,” or something along those lines?

Mandelin: The biggest example I can think of is with naming. It turns out the official EarthBound localization used some wrong name romanizations, though no one found out for sure until 10 years after EarthBound’s release. This caused the dilemma, “Should I use the names fans know and love, or should I use the correct name that the creator intended?”

In a professional setting I probably wouldn’t have been able to make that choice, but for Mother 3 I was able to say, “Okay! I’m going to do it this way!”

There was also the issue of, “Should I localize some of these new names?” Being able to go in and say, “We’re doing this!” was nice.

Basically, whenever the question of, “Should we try this?” came up, we tried it because we were able to. We even added many things to the game that Japanese players didn’t get.

AGM: To say that the Mother/Earthbound series has a passionate fan-base would be something of an understatement. Most official localizations are handled under a rather thick veil of anonymity from the end-user, but not Mother 3. Did you feel a different sort of pressure as a result? How did you cope with it? What kind of pressure, if any, do you feel when working on something of less notoriety or outside of the public eye?

Mandelin: As far as unofficial translations go, I don’t think there’s really a thick veil of anonymity, just obscurity. Most projects ARE right out there in the open, usually on message boards and small blogs. Usually the more popular games will get their own project sites, as Mother 3 did.
Mother 3 was the first time I’d done such a big “see inside the localization’s on-going development” project. What I found was that the expectation of regular updates got me to work on it even harder than I normally would have. The pressure to have constant, large-scale progress was tough. Looking back, that was probably one of the biggest factors in the project’s success.

When I work on less popular projects out of the spotlight, there’s very little pressure. I get all the time I want, and I don’t even have to produce a final product in the end if I decide not to. But sometimes projects I leave alone for too long start piling up, which can result in a different kind of pressure.

AGM: Would you say that your work on Mother 3 is, from a quality standpoint, the product that you’re most proud of? If not, what would it be? What is it that you feel most lends itself to contributing to the production a quality translation or localization?

Mandelin: Yes, but probably only because it’s the most recent large-scale project I’ve worked on. I’m definitely proud of my earlier projects, it’s just that my skills weren’t as honed when I worked on them. That, and I put in a lot of extra time with the localization and the reprogramming for Mother 3.

I’d say I’m probably only 70% happy with the final result – the writing could still be improved a lot and there are various reprogramming issues that could be fixed. But with stuff like this it’s pretty much impossible to ever be 100% happy. It’s important to know when to say, “Okay, that’s enough.” for things like this.

In the end, the audience is what matters most to me. Does the player like the text? The text isn’t pulling the player out of the game due to awkwardness, is it? How obvious is it to the player that this wasn’t written in English originally? Does the localization accomplish what the original text was intended to do? Things like that are important to localization quality.

AGM: Do you feel that translation/localization is an art? A skill? A craft? How do you view and define it? Why?

Mandelin: This is always a tricky question. My opinion is that translation is a craft or skill, and localization is an art.

When I translate a sentence, I see the original sentence almost as a painting that the mind creates as the sentence progresses. The most important parts of the sentences – usually facts, nouns, or the point of a sentence – have an emphasis in the painting. Things like tone of speech and nuances add little flairs to the painting too as the sentence goes along. There’s probably also a layer of feelings and thoughts in there that we can’t express easily in words. All together, a sentence of text has a detailed, flowing, painting-like quality to it.

From that view, I see translators as people who take these paintings and try to recreate them as closely as possible, using different tools than the original. Some languages are very similar, so the “brushes” are similar, meaning the translations will be pretty close to the original. Some languages are very different, so their brushes will be very different. So translating something from Spanish to English might be like painting a copy of the Mona Lisa with watercolor paints, but translating the same thing from Japanese to English might be like trying to draw a copy of the Mona Lisa with crayons.

Besides the tool differences, the skill of the translator/copy painter is also important. This is why I feel translation is a craft, but not necessarily an art form.

Localization, on the other hand, requires the creativity to create something new. So even if you’re stuck trying to draw the Mona Lisa with a broken crayon, you can use some creativity to get a better result than you would have otherwise. I would say that act of creating something new is an art form.

AGM: Could you explain a bit your attachment to Mother 3 and the Mother/EarthBound series in general? I think that it’s rather unique and significant and really may help to frame the context within which you worked on this product.

Mandelin: I bought EarthBound when it was released in 1995, and I liked it a lot. A few years later, I helped start EarthBound.Net, now Starmen.Net. I did all sorts of things there, from article and interview translations to reprogramming the games. Basically, I happened to know a lot about the series and I had done fan translations in the past, so it was a good match.

AGM: Can you share a bit about the technical side of Mother 3’s localization/translation? What sorts of tools, software or resources did you use? What was the process and sort of general work flow of translating, getting the text in, doing text debugs, etc.? Did you make any alterations aside from just the text? What were some areas that you struggled with?

Mandelin: Almost everything we did was custom-made from scratch. The text editor, the text convertors, the text inserters, the voice clip editor… One team member even wrote a THUMB assembler from scratch to help organize and speed up the project!

Basically, the entire process was something like this:

  • We got the text out of the game and put together our text editor
  • I spent about a month translating the text into its first draft
  • We spent many months learning how to reprogram Game Boy Advance games and then applying that knowledge
  • I spent another month on the second draft of the text
  • I spent a couple weeks playing through the game with the text, polishing things up and tweaking things here and there
  • We did last-minute programming fixes
  • Final testing
  • Release

Our text editor had a real-time preview function, so that saved us a lot of time with text formatting. When we needed to test text in-game, we just needed to run the text inserter utility and then start the game. Luckily for us, the original programmers created a “debug room” that allowed us to jump to any point in the game. We used it religiously.

Besides the text, we also altered some graphics. Some of this was graphical text, and some of it meant to mimic the way Mother 2 had been officially localized into EarthBound. For example, In Mother 2, there was a building with the name “MONOTOLY” written on it, but it was changed to say “MONOTOLI” in EarthBound. The same building makes a cameo in Mother 3, so we went in and changed it to say “MONOTOLI”.

We also replaced two very short Japanese audio clips into English, un-dummied some old content, added a few simple extras to the gameplay, and things of that sort.

By far the most difficult task was reprogramming the game. It became clear very early on that localization hadn’t been on the mind when Brownie Brown programmed the game. It didn’t help that practically every system in the game had been written by a different programmer. We pretty much had to fight every step of the way for more memory and more space on the screen to achieve something akin to a professional release.

The online readme file has a lot more details: http://mother3.fobby.net/or/

AGM: How would you like to see the craft of localization evolve in the near future so that more players in more regions can experience more games in a way closer to the original intentions of the creators?

Mandelin: In terms of professional localization, the easiest answer to this would be to get developers to design games with localization in mind from the very start. This includes things like allocating enough memory and being able to utilize screen space effectively. Translation-wise, things like notes from the writers and developers are always extremely useful to have.

In terms of unofficial localization, it’s probably naïve to expect game companies to specifically design games to be easier for fans to translate themselves. In fact, it’s probably becoming even more difficult thanks to DRM and other lock-out mechanisms.

AGM: Lastly, upon the final release of the Mother 3 patch, you were able to have a kind of communication with a very grateful collection of players worldwide who have been playing and enjoying the game that you worked on in a much more personal way than most localizers may ever get to experience. How has that experience been?

Mandelin: It’s been very nice, and the positive response has been amazing. It was an awesome feeling seeing a topic on a message board called, “Post your favorite quotes from Mother 3!” and then seeing so many people posting lines that made them laugh and cry. I wasn’t the game’s writer, but there’s still a lot of joy in seeing that the game could still touch people like that, even after crossing the language barrier.

It’s also been neat in that people can send in bug reports and typo reports and we can respond to them personally if we need more clarification. We still get reports all the time, almost two years later! We’ve also held contests and done a lot of polls after the translation’s release, and the response there has been great too.

A couple people have also contacted me since the translation’s release, saying that they’ve gotten into translation or gotten into programming after watching the project come together. It’s a really nice feeling hearing that. Who knows, if a Mother 4 is ever released and it’s another Japan-only game, maybe one of those people will be responsible for fan-translating it. Or if it DOES get an official translation, maybe they’ll be the ones who translate it!

AGM would like to thank Mr. Mandelin (Tomato) for taking the time work with us on this special interview. Those interested in playing Mother 3 in their native language are highly encouraged to visit the Mother 3 Translation Site and to acquire an original Japanese version of the Game Boy Advance title as well. Fans of all-things-Mother/EarthBound should also visit Starment.Net and EarthBound Central.


(Yep, they called it Starment.Net! Seems like everyone gets it wrong in unique new ways all the time 😛 )

Both comments and pings are currently closed.

3 Responses to “AGM MOTHER 3 Translation Interview”

  1. Scottt says:

    Very interesting. I have to say that as a player, I’m much more than 70% happy with the final product!

    Where are the English audio clips, and who provided them? I don’t seem to remember that.

    • Mato says:

      I don’t have the English clips handy (I’ll probably post them in an upcoming post) but they were “Ready, set, go!” and “Look over there!”

      I had Reid, Meeellla, and Plo record a whole bunch of takes and then I chose one to use for each clip. It looks like I posted about it on the dev blog but only slightly because I was expecting backlash from sub/dub arguments: http://mother3.fobby.net/blog/2008/08/24/timbrrrrrrre/

  2. […] usually translate stuff from Japanese to English, so it was an odd feeling when a few years back an interview I did about the MOTHER 3 translation was translated into Japanese! What was even more interesting was that the translator had contacted […]

Subscribe to RSS Feed Follow me on Twitter!