Video game localization is suddenly a fiery subject of debate. Spurred by alterations in games like Xenoblade Chronicles X, Fire Emblem Fates, and others, a subset of gamers has engaged in a heated rally to have video games brought to Western countries with as little alteration from their Japanese versions as possible. In some cases, the campaign has gotten extremely ugly: The perceived involvement of former Nintendo employee Allison Rapp in the changes to several of their titles subjected her to a prolonged attack as people went out of their way to dig up personal dirt. She said she was let go from the company as a result, though Nintendo officially denies any outside involvement in its choice to fire her.
Many of those involved in this movement, however, fail to understand that game localization isn't merely a process of getting the rights to translate something and then plopping an English script into it. It's a challenging process that involves a lot of back-and-forth between translators, editors, marketers, and developers, with an occasional bit of interference from local ratings boards. In fact, it's nearly impossible for a game to make the transition overseas without some alterations beyond simply replacing foreign-language text with English.
We reached out to folks involved in the localization business to get an inside look at what goes on behind the scenes when Japanese games—both mega-million-sellers and niche titles—get brought overseas. All of the people we contacted for this piece have requested anonymity, as even divulging a small bit of what goes on behind the scenes can threaten their careers. Their names have been changed.
The first step of localization is deciding whether or not a game is even worth bringing overseas. Recently, Koei-Tecmo made headlines when they opted not to bring out Dead or Alive Xtreme 3, a fantasy spin-off of the popular fighting game series featuring the series' women in skimpy swimsuits and suggestive activities.
I asked Mike, a veteran localization producer at a large publisher, about how companies decide which games to localize and how they approach the project. "[It] completely depends on the project and the company's priorities," he explained. "Where's the funding coming from? Was the game built for Japan-only? Does the series have existing localization history? What's the framework you are building on?"
For many companies, the decision process of whether or not to localize a game is handled while the game is still in development, and oftentimes the process begins before the game even releases in its native market of Japan. Jenny is an employee of a translation agency that handles localization work for many of these games in various stages of their development. "It's pretty well understood that the worldwide games market is much bigger than Japan alone, especially with regard to console and handheld games, so if you want to give your game its best shot at success, you need to bring it to the US and Europe."
In many cases, the localization team works with the developers to best approach making the game saleable in markets outside of Japan. "[Whether we give feedback] depends on how early a Western release is decided, and how much the dev team wants that feedback," Mike said. "Some developers put a lot of thought into overseas markets—we like those. Some teams consider overseas releases as just extra change on the side. But a localization group has an obligation to review the game and make suggestions on how to improve the game for a Western release."
The concept that the Japanese side of things is completely "hands-off" when it comes to localization is also untrue in several cases. "The level of engagement between Western and Japanese production entirely depends on how much the Japan team members want to engage on the English product," Mike said. "Some Japanese teams simply don't care. Some care a bit too much. The best mix is when the Japanese teams care enough to review the text and bring up concerns and encourage the localization team to bring out the fun in the game."
Jenny agrees. "Some games, the client is pretty much hands off beyond some basic guidance, and other games we're working hand-in-hand with the devs, asking questions and bouncing ideas back and forth on a daily basis," she said. "We of course prefer the latter, but budgets and schedules and lots of other factors mean it's not always possible to have that kind of involvement. We love it when we can, though, because it almost always leads to a better end product."
Jenny recounted a story of a particularly memorable project where the creator was enthusiastically involved. "We were translating the script of a game in which the story took place over a short period… we asked for a little background info on the world, and the developer set up this huge whiteboard and proceeded to outline a massive timeline of the entire history of the world they'd created—from thousands of years before this game's story to thousands of years beyond it—just to give us a big-picture view of what was going on in this tiny little two-year sliver of the saga. It was pretty epic, and even though none of this info ever made it into the actual game, it still helped a ton with writing the characters and... minor bits of side dialogue."
Mike points out, however, that it's easy for the Japanese side of things to over-manage and cause headaches. "In the worst case scenarios, Japanese team members will try to make quality judgement calls based on their own limited knowledge of the language, which almost never works out."
Some consumers have rallied behind the notion that localization should aim to alter as little of the game's content as possible. The folks we talked to, however, said that simply doing a direct translation isn't what their job entails. "Localization is, by definition, an act of changing content," Jenny said.
Why do certain content changes happen, then? "There's always a legitimate reason, it's just that reason may never be made public," Mike said. "Maybe it's a PR risk to leave it. Maybe it's a ratings risk. Maybe there's a licensing issue. Who knows, but removing or changing content is never a casual conversation. Why add onto your workload when you don't necessarily have to? It's not done to troll the intrawebs or anger people… sometimes small changes may take a lot of work, so you have to consider if the work to make the change is really worth the effort."
Larry, a localization expert who works at a smaller game publisher, agrees on this point. "I'm fairly certain that most companies that deal in localization do not want to censor or alter any content," he told us by email. It takes time to send video content and explanations to rating boards to get their feedback, and then the localization team needs to determine what exactly needs to be removed—or, if altered, how to alter it. Changes aren't taken lightly either.
"Extra care and attention that must be paid in debug because when you alter ANYTHING in a title, you run the risk of introducing new bugs," Larry said.
There's an opportunity cost, as well, since writing new code costs time and money of the above. Finally, the negative backlash from fans—and the potential impact it would have on sales and PR—are carefully weighed.
"Within those points, the latter two cannot be understated," Larry said. "While there is some debate about the legal aspect of this, it's generally understood that executives of publicly traded companies have a responsibility to maximize profits."
One of the most damaging things that can happen is being presented with the potential of an Adults Only rating. The AO rating is akin to an NC-17 rating for film: most mainstream outlets won't sell your game, and in the case of console manufacturers like Sony and Nintendo, they outright bar AO-rated titles from appearing on their platforms. As a result, the number of AO-rated games is absolutely miniscule. Some folks may remember that when the Hot Coffee sequence code was discovered in Grand Theft Auto: San Andreas, the game was temporarily re-rated AO until it was removed.
Larry has seen a game his company has localized threatened with a dreaded AO rating before. "Spending an inordinate amount of time figuring out how to release a game with problematic content certainly does not help [with profitability], and no company can afford a lengthy back and forth with releasing a game only to have it get an AO," he said. "In almost every situation, licensing a game requires money upfront and in every case is a risk, so minimizing risk is paramount."
Even games that aren't staring down an AO rating without cuts might be looking at some issues with the Electronic Software Ratings Board, which assigns ratings to games released in North America. I asked Mike if the ESRB factors into localization decisions at all. "A good bit, actually," he said. "Certain games are obviously targeted at certain age groups, though what is considered the social norm or appropriate for said age groups varies from region to region. So you just try to make sure things line up with those expectations. If a game needs to be E10+ due to the obvious target of the audience, then there's a strong chance that content that would keep it out of that range is going to be tweaked. If a 25 year old feels offended by that, well, that's sort of tough because they weren't the target audience."
Marketing plays a key role, as well. "PR is always consulted before changes happen," he continues. "Usually change requests are raised by Production, since they tend to know what's in the game, and what will cause an issue. It's reported to marketing and a discussion takes place. In my experience, changes are only made if the change improves the game in a real, tangible way, or if there is a sales/PR risk from leaving it unchanged."
What might surprise some is that almost every alteration is signed off on by the original team in Japan, and not every alteration is made to remove content. Many things introduced in localization are done to improve the game experience for a wider audience. "We always work closely with our clients and often directly with the original dev teams to make sure any changes that get made in the localization process have the original creators' blessing," Jenny notes. "[These can be] simple changes like the spelling of character names to more big-picture adjustments like increasing the speed at which battles play out to improve a game's overall pacing."
I asked Jenny if developers opt to alter their own games for overseas versions. "Oh, all the time," she replied. "[Localization] is very much a collaborative process. Sometimes we'll suggest things, other times clients will make requests. None of it happens in a vacuum, the original developers are almost always involved with and sometimes even behind any changes that take place."
Even more surprising is that the Japanese development team are usually the people implementing these alterations, as well. Since they were the staff programming the game initially, they are equipped to make necessary changes. This can sometimes cause complications, especially if a localization wasn't initially planned. Something simple like changing the alphabet and text box display can be a nightmarish undertaking. "Localization is not always being taken into account [from development outset], and localizing games that aren't built that way is not an easy task... the J2E [Japanese to English] localization space is very matured, and J2E localization produces some amazing writing and content, but the development side lags behind. In general, the tools are pretty basic."
What the localization folks I interviewed want people to know is that they put a lot of work into trying to bring games over from Japan in a form that audiences in the West will enjoy, and don't want to upset players.
"Our employees love games and specifically the titles we work with—otherwise they wouldn't be here," Larry said. "No one on staff really wants to alter a game in any way. Sometimes, you have to take the lesser of two evils. Furthermore, we strive to release titles that have a numerous positive points beyond the more prurient elements, so while, yes, it's disappointing that a game may have been altered, the title should offer a lot to the player even with that material altered."
Larry also flat-out stated that, from his experience, internet-led boycotts were largely ineffective, as they didn't appear to have a significant impact on his company's or others' sales. "At the end of the day, consumers need to make choices based on what's best for them. If someone feels strongly about something, they should most certainly vote with their wallets."