1.Introduction

For the past two decades or even more, transliteration has been a conventional approach in translation of names or appellations. The origin of this convention is too far ago to be examined, but it seems that it was after 1989, the Guidebook of Translation of English Names” and 1993 the Guidebook of Translation of Foreign Landmark Names” by The Commercial Press that this principle came to public attention and was documented in script source for the first time [1,2]. The guidebooks were later adapted by The Ministry of Foreign Affairs of the government of The People’s Republic of China, which led them to be considered the most authoritative instruction of foreign language translation in China. In the guidebooks, it was clearly instructed that when translating names, titles, or appellations, transliteration should be considered as a first option, while free translation should be cautiously dealt with [1-3]. The principle soon became widely spread and accepted by both governmental and nongovernmental translators and has been used as the guideline of English-to-Chinese translation for over three decades.

However, despite the practical usage of these guidebooks in translations, in reality, it may bring up some concerns when this strategy is applied to fictional translations. In reality, it is effective to keep the translations neutral when translating names or titles to avoid misconception, but its reliability becomes questionable under fictional backgrounds such as novels and, the focus of this essay, video games. In fictional artworks, plots and cinematic are usually intentionally designed to associate with the terms and texts so that players can be guided or inspired to predict or reflect upon the gameplay performance [4]. To better achieve such goals, a notable greater amount of coinage is used in the word-forming of games, which means that made-up proper nouns in video games, such as names or titles, are usually composed by meaningful roots/etymas that are clearly understandable to players who are native-speakers. Another concern that led to the usage of transliteration in reality is that information could be invalidated as circumstances change, mostly due to time effect [3]. Whereas for video games, most of the game contents are finalized once the work is published, which allows players’ perceptions to remain consistent over time. Given the backgrounds above, the effect of transliteration to avoid misconception could also have blocked those “rightful” conceptions when it is applied in video games.

Despite all the potential flaws of insisting on transliteration in video games, such a topic does not seem to be widely discussed or researched academically due to the multiple dilemmas video games have been facing in the Chinese market. First, video games are typically restricted by strict policies and auditing in China. China's government performs no rating system in entertainment auditing, which means for adults and teenagers alike, contents of violence, sex, and politics are prohibited indiscriminately [5]. Considering the high frequency of the above contents in video games, it became extremely difficult for most video games to be published legitimately in China mainland, compelling video game producers to sell their products unofficially [4, 5]. Although most video games are “forbidden” in China, the government does not prohibit citizens from purchasing digital products on overseas platforms such as Steam, Epic, and Origin [6]. However, since the companies are not registered within China mainland, players from mainland have to rely on Hong Kong or Taiwan services, and their experience cannot always be guaranteed when encountering issues related to gaming quality. This fact, accompanied by other factors, has contributed to the third problem video games face in China: lower purchase capacity [6]. Given the inconvenient purchase method, it is imaginable that Chinese consumers have less purchasing willingness and acknowledgment of video games, and there are even more factors, such as income issues, game-style preference, and social orientation that drove the situation even worse. Since video games are not typically published in China, most video games are priced based on the income levels of Euro-American players, which makes them less affordable for Chinese players considering the relatively lower average income, even when the game is lower priced based on location and income. One other problem is the abuse of in-game purchases in China video games [7]. The majority of play style in China is mobile video games, which apply in-game purchases at high frequency in their game design [8]. Considering the mainstream of the worldwide gaming market is a one-time purchase, this actually separated the Chinese player community from the rest of the world. At last, the traditional social orientation of China has been against video games. Entertainment has been a questionable concept in the traditional Chinese moral system for thousands of years (e.g., 玩物丧志), and video games, possibly because of their novelty, have faced the most intense objection among the aged people in China [9]. Under such social orientation, potential players could detest engaging video games because they fear being seen as “undignified,” and teenagers and students, who have occupied a large portion of the gamer community, would have no approach to purchasing when their parents refuse to support their consumption [9].

For all the reasons mentioned above, video games have somewhat been a “gray zone” in China and many of its problems, such as the potential abuse of transliteration, have never been treated seriously. Translators Only little research has been done about video game localization in China, so most of the translators would simply apply what the guidebooks taught mechanically to video games. Moreover, more than half of the video games were not published without official Chinese translation and relied on third-party translator groups who are non-commercial, the quality of translations is quite varied [6, 10]. The credibility of traditional translation methods in China is yet to be examined in the area of video games, which is the purpose of this essay.

2.Methodology

A statistic research and a survey were conducted to measure the use of transliteration in the Chinese localization of video games and the effective of transliteration to potential or actual video game players. For the statistic part, 15 reputed video games published within the past 20 years, which include eight commercial games and seven indie games, were chosen based on their user score on Metascore and the sum sales volume on all video game platforms to ensure their representativeness. From each of these 15 games, three nouns of phrases that function as names of characters, locations, objects, or other categories will be randomly picked out and analyzed. The original text and Chinese translated text were documented, as well as the type of these words, the producer of the game (commercial/indie), the identity of the translator (official/third-party), the type of the text (character name/object name/etc.), the translation method (transliteration/free translation), and the text format (proper noun/common noun/phrase/etc.).

For the survey, a questionnaire was distributed to a group of randomly selected participants, who would be presented with both the original text and the Chinese translated text of a series of location names drawn from the video game Elden Ring and asked about their impression toward these texts [11]. Several widely discussed topics on English native-speaker players' forums were also adapted into the questionnaire as advanced tests to measure if participants could notice the association between texts and game contents on a deeper level. Participants would be asked a question about to what extent they know about Elden Ring at the beginning of the test to ensure their performance is not influenced by their previous experiences, and a brief language test was posted at the end of the questionnaire to measure if the participants can understand the original English text. Considering participants may change or stick to their previous answer once they realize they were asked twice about the same location, two different versions of the questionnaire were randomly given when distributing, with one version presenting translated text first and vice versa for the other. The questions of original text, translated text, and language test will be separated on three different pages so that participants will not be able to go back to their previous questions and change their answers accordingly. Participants who answered “Actually played but never finished” and “Finished playing” in question 1 were marked as invalid responses except for question 10, “If you were to choose a nickname for the game “ 艾 尔 登 法 环 ( Elden Ring ),” which would you choose”. Participants who failed to answer correctly two out of four in the language test would be marked as invalid responses.

3.Results

According to the statistical result, it seems the use of transliteration is largely influenced by the identity of the translator. 11 out of 16 translations were transliterated for texts translated by official translators, while only 5 out of 29 translations were transliterated for texts translated by third-party groups, showing that transliteration is used significantly more frequent among official translators compared to third-party translators. The rate of transliteration showed no apparent correlation with the producer of the games, although it seems that official translation is affordable only for commercial producers. All four pieces of texts that are freely translated by official translators are phrases including at least one common noun. Official translators would only free-translate separated common nouns in phrases, while all proper nouns and isolated common nouns would be transliterated. If a term is used as a game title and in-game text, it would usually have two different versions of translation, but it is not a universal pattern.

25 participants are marked as valid responses (except of question 10) out of the sum number of 42. According to the response, it seems that participants’ impressions significantly changed once they are exposed to the original text. Nine participants chose the option “A peaceful countryside,” and five chose the option “A blooming flower field” for the transliterated text “ 宁 姆 格 福,” but the numbers significantly dropped to three and one once they were exposed to the original text “Limgrave.” Contrarily, the numbers of option “A horrifying torture house” and option “ A ruin of slave uprising” rose from one to six and zero to four once exposed to the original text. Five participants have chosen the “countryside” option for the transliterated term “史东薇尔,” and seven have chosen the “flower field” option, both of which dropped to zero once exposed to the original text “Stormveil,” and the option “A windy fortress” rose dramatically from zero to thirteen. The responses between “ 摩 恩 城 ” and “Castle Morne” seem to be less differential, but participants do seem to shift their options from “Countryside” to “Capital,” possibly due to their different terms “ 城 ” and “Castle.”

However, as for advanced questions such as “What do you think the area BOSS (final enemy for a campaign) for ‘Limgrave’ would be like” or “Where do you think ‘Castle Morne’ would be located,” participants’ performance seems more randomized and less meaningful, possibly due to insufficiency of text comprehending (e.g., participants chose even less “limb hunter” option after being exposed to the original text “Limgrave”, indicating that they might have failed to acknowledge the relationship between middle English vocabulary “lim” and modern vocabulary “limb”).

At last, in the nickname question of question 10, participants with previous experiences of Elden Ring were removed from invalid responses. All participants are then grouped as the “game experienced” group and “non-experienced” group to measure the differential in their responses. According to the result, a notably higher portion (8/ 10) of the experienced group would choose a more comprehensive nickname, “ 老 头 环,” compared to the simple abbreviation “法环”, while the amount is much less (6/ 19) in the non-experienced group.

4.Conclusion

Judging by the statistical results, it seems fair to conclude that transliteration is widely applied in official video game translations of Chinese, especially in the translation of proper nouns. Official game translators tend to indiscriminately transliterate all the proper nouns even when the roots/etymas are meaningful and notifiable. On the other hand, third-party translators are much more likely to use free translation whenever it is possible. Sometimes, they even do the translation based on their understanding of the gameplay rather than the text alone (e.g., for the translation of “Deca & Donu” from Slay the Spire, the translator translated “Deca” as “八体”, which can be roughly translated back to English as “octagon” because the character Deca looked like an octagon in the game. However, according to the author of the game on his blog, Deca was named so because it is malformed and there are six faces rather than four on its unrevealed back side, which makes it a decagon rather than an octagon).

Whereas the questionnaire results, although not much information could be gathered from advanced questions, participants have notably changed their answers to the original text of simpler terms such as “Limgrave” and “Stormveil.” Participants performed notably better in associating “Stormveil” with “a windy fortress” once they were exposed to the original text, indicating that the transliteration in Elden Ring has, at least to some extent, distorted the message conveyed by the original text. Beyond that, due to the insufficient amount of neutral onomatopoeia in Chinese, it seems non-neutral vocabularies are inevitably included in transliterations. For example, participants’ responses to “ 宁 姆 格 福 ” and “Limgrave” sifted dramatically from positive options to negative options, indicating that vocabulary “ 宁” (which means peace) and “福” (which means happiness) have impacted participants’ conception and misguided their understanding.

Considering the representativeness of Elden Ring (the “Game of the Year” award winner of TGA in 2022), what happens in Elden Ring could be observed as an example of the overall contemporary video game business. Transliteration is widely used by official translators of commercial games, but it is not always accepted by the player community. in China, third-party translators typically originated as players themselves, which means their performance in translation could have also represented the demands of the player community. Likewise, participants’ performance in question 10 shows that players would prefer to use more comprehensive nicknames as long as they have some basic understanding of the game content, possibly indicating their demand for more information that can be drawn from translations. The evidence so far has shown that transliteration could be a deficient method in video game localization, and translators, especially those who work as official translators of a gaming company, should be alerted and cautious in their future works.

5.Limitation

Several limitations have defected the applicability of this research. First, the sample capacity of the survey was lower than expected. A more detailed language test was initially designed to measure participants’ language ability, but with low sample capacity, it has to be shortened and used as a filter (to testify only if they have a basic understanding of the text) rather than a rating system, which makes it impossible to conduct correlation research between participants’ performance and their language ability. This has also led to ambiguity in specifying participants’ performance in advanced questions. Considering these advanced questions typically include quotations of non-standard English or a third language (e.g., “lim” is a vocabulary of middle English, “morne” is a French vocabulary and a homophone of English vocabulary “mourn”), the differential between responses of participants and native-speaker player community is highly likely due to the insufficient comprehending to the text, but it is impossible to conclude without detailed language test. At last, the sample games/translations listed in statistical data were chosen by the researcher alone in a limited amount of time. Therefore, the coverage of data is not extremely broad and could risk being subjective. In future studies, interview-based research on Chinese players with high language ability and linguistic knowledge would be required to specify further the effectiveness of transliteration/free translation in video games.