Translation of Video Games in the Context of Polish Localizations

This article presents the notion of translation in the context of localizations of video games. It introduces a theoretical background as well as analysis of chosen video games and studies the process of translation. The author considers the obstacles which can be encountered by translators as well as methods which are used in the process of localization. The paper is a critical analysis games from different genres: Ace Ventura (1997, adventure), Baldur’s Gate (1998, RPG) and Warcraft 3: Reign of Chaos (2002, RTS), where the plot (and as a consequence its translation) is very important. The author studies fragments from these games and tries to explain and critique (if necessary) the choices taken by translators.


Introduction
The idea of localizing video games is relatively new. Video games have become very popular and they have undergone a great metamorphosis during recent years. With this newly gained popularity, they have started to be perceived not solely as a kids' toy but as something serious with great potential. Due to this new perspective, video games started out as a global business worth billions, which operates in many countries. To maintain their worldwide range, it is necessary to localize them.
The paper differentiates between translation and localization. Firstly, we should define clearly what we can understand under these terms. Translation is "rendering the meaning of a text into another language in the way that the author intended the text" (Newmark 1988, 5). On the other hand, localization is more complex. The simplest definition can be the same as the one provided by Hoft (1995, 11) -"the process of creating or adapting an information product for use in a specific target country or specific target market". However, Kolko and Thayer (2004, 1) underline that the term "information product" is not adequate today, because it leaves out the entertainment market, which should be added to the definition.
In the definition, words adapt and product are crucial. Localization is a translation of a product, not only a text, but also visuals, music and all the ways to interact with them. Localization is not only the process of translation itself (however it is the main part), it is also an attempt to prepare software for new languages, dubbing (if it is necessary) and all marketing actions. The complexity of the task of localization is not the only problem -video game translation is a process which requires certain knowledge. There are many game genres -and every one of them requires a translator's special set of skills: for example, simulators games differ drastically from fantasy games. Various cultural elements are also very important -many jokes or cultural references cannot be simply translated.
What is more, a translator has to have in mind that a text should fit perfectly to the story and to the game's mechanics -a bad translation can cause major problems for players, and in some cases it can even break a game (Dietz 2006, 121-134).
The popularity and complexity of video game localizations make them an important part of contemporary translation studies, and a worthy focus for this paper. The article consists of three parts. The first one presents the definition of the concept of localization and describes in detail different approaches to translating video games and difficulties which can be encountered in this process. The second one presents the methodological assumptions of the paper and characteristics of methods used in the process of analyzing texts. It also explains the criteria of selecting games to the analysis. The third one is the analysis conducted on games, which, in the opinion of players and journalists, are representatives of good localizations: Ace Ventura (1997), Baldur's Gate (1998) and Warcraft 3: Reign of Chaos (2002). The paper aims to compare localizations and the original scripts of games in order to examine strong and weak points of the translations.
The idea of a game which should be understandable for players, directs us to the fact that a translator must always remember playablility. A translator should be cautious while localizing video games, it is evident especially in nonlinear games, where a player has freedom to go everywhere and do everything s/he wants. Dietz (2006, 125) stresses that a piece of information must be coherent in a different context and "linguistic plot-stopper[s]" must be avoided. By linguistic plot stoppers we may understand situations as the one provided by the author. In Torin 's Passage (1995) one of the tasks is to assemble magic crystals which are supposed to create a magic sentence; however, in the German localization this sentence has the different word order, while the crystals order was left unchanged. As a consequence, German players could not finish the game without an instruction from the producers (Dietz 2006, 125).
Having in mind playability, Kolko and Thayer (2004, 478) stress the importance of technical issues. The authors highlight the crucial role of proper preparations of the software development level in which developers should have in mind different target languages. For example, while localising games, translators have to take into consideration space. As Dietz (2006, 126) states, it is visible in the context of interface which is designed to not interfere with the player's immersion. Different language versions have a grave impact on the visual effect of it. The author gives an example of "Quit," which translated to German is "Abbrechen" (5 letters longer). Costales (2012, 391) remarks that the situation is even more complex if we take into consideration animated scenes where the text should fit perfectly into the spoken dialogue. As Costales (2012, 390) writes, cinematics "can be easily compared to the scenes of animation movies like Monster, Inc. or Finding Nemo and pose an additional challenge in the translation of video games". More and more games use lip-sync, which means that translated text should not only be accurate, but also fit as much as possible to the movement of the lips of the heroes. Voices are also required to be fitted into the context. For example, sports commentators should be realistic and their voices have to be adequate to the situation (Dietz 2006, 124). For the series FIFA -one of the most popular sports games, it is customary to see real sport commentators -in the Polish version these are Dariusz Szpakowski and Jacek Laskowski (Olber 2017).
There is a need to remember that a localization process does not concentrate solely on technical adjustments, but also on cultural nuances. As O'Hagan and Mangiron (2013, 210) underline, some gestures or jokes cannot be transferred without some loss in meaning, and there is a need to compensate for it. For example, sneezing in Japan is connected with a belief that someone is gossiping about you, and in the USA it is not. This situation occurs in the case of the game Final Fantasy XI (2002) -one of the most popular Japanese series of RPG (JRPG). American translations had to be inventive and they explained the situation as a part of Elvaan people's (a nation in the game) tradition. A similar case is with names and holidays. Dietz (2006, 131) gives the example of Animal Crossing (2001). Changing typical Japanese holidays and substituting them with American as well as renaming characters took 6 months. Nevertheless, cultural differences can concern less obvious issues than traditions, customs and names -they can be more subtle. O'Hagan and Mangiron (2013, 212-214) analyse the preferences of two major gaming markets representing East and West -Japan and the USA, dealing with the look of characters. They conclude that Western players prefer adult, more realistic characters, while Eastern players prefer stylised and cute. Authors give an example of the Japanese horror game Zero (2001), where the main (Japanese) character is a 17 year-old girl wearing a school uniform. For the purpose of Western markets, the girl was redesigned -she has become older, without the uniform and even her pattern of speaking has changed. On the other hand, Crash from the Crash Bandicoot was redesigned to look friendlier for the purpose of the Japanese market (Kolko and Thayer 2004, 481). Cultural differences are not only connected with esthetical preferences, traditions and religion but may be related to law. A good example is Germany. After a shooting in one of the schools, which the authorities connected with the influence of the game Counter Strike (1999) "Germany has tightened existing laws concerning the depiction of violence in computer and video games" (Dietz 2006, 129). In the game Return to Castle Wolfenstein (2001) humans were changed into robots and Nazi paraphernalia were replaced by neutral ones in order to meet the standards of German laws forbidding the use of certain representations (Kolko and Thayer 2004, 486).

Strategies of translation
Having in mind difficulties which are connected with the localization process, let us look at the strategies which are undertaken to create a translation. It is almost impossible in some cases (especially of humour and puns) to translate them without the loss of some meaning (Di Marco 2007). However, Costales (2012, 395) argues that in the case of video games the "concept of 'right' and 'wrong' translation does not apply" and the most important is "preserving the game experience". Venuti (1998, 240) has forged the terms domestication and foreignization. Domestication is an adaptation of the text for the standards of a target language -making it more familiar for the reader, while foreignization is "aimed to keep a 'foreign flavour'" (Costales 2012, 395). Costales (2012, 395) gives an example of Assassin's Creed (Ezio trilogy) as a good instance of foreignization. In the games there are many Italian names and cultural references, which are preserved. What is more, actors recording voices for target versions have Italian accents, and from time to time they are using Italian words. On the other hand, the series Final Fantasy (originally from Japan) tries to merge into the culture of target language. Mangiron and O'Hagan (2006) state that effect of domestication in this series is achieved "by the use in the target text of idiomatic and colloquial language, the adaptation of jokes, sayings and cultural references, and the re-creation of new cultural references and plays on words" (Mangiron and O'Hagan 2006, 16). Another good example of a domestication strategy is the series Fifa. As stated before -in every new release of the game, commentators are voiced by local sports presenters.
There is a need to know when it is necessary to translate, and when it is not. For example, there is a trend to leave game titles in the English version, in which they are recognizable around the whole world (Costales 2012, 398). However, this practice is not only restricted to the titles. As Costales (2012) underlines -in some series players are used to particular terms and are expecting them, to the point that translation would introduce confusion and disappointment among the fans of the series. The author gives an example of Street Fighter, where names of attacks such as Shoryuken or Hadouken are left without translation. However, at the same time, the author underlines the fact that this strategy cannot be connected with the fighting genre as a whole -for example in Super Smash Bros Brawl all combos and special moves are translated (Costales 2012, 397).
In the so-called literal translation, grammar of the source language is translated to the closest equivalent in a target language; however, words are interpreted one by onewithout a context (Ordudari 2007, 3). Costales (2012,) states that while in games with a rich plot this method can be less efficient, this approach can be very helpful in the case of games with many technical elements. The author gives examples of simulators such as Forza Motorsport 3 (2011), Gran Turismo 4 (2011) or Microsoft's Flight Simulator (2011) -these games are rich in the technical elements, which are easier to translate in a literal manner (Costales 2012, 400). Mangiron and O'Hagan (2006) applied the concept of transcreation to the field of video games. It is a creation of completely new terms, which are not connected with the original text. It leaves a great deal of free will to the translators. The example of such practice is Final Fantasy X (2001). In the game, translators from the USA had to be creative due to space restrictions for the names of objects and had to create completely new terms for armour and weapons. While new expressions do not have to be completely separated from an original, Mangiron and O'Hagan (2013) state that in this approach the preservation of the experience of a game is the most important. The term should give the same feeling as in the original form, not necessarily be similar in the meaning; the translator should have freedom to choose between the most suitable alternative (Mangiron and O'Hagan 2013, 159-160).
As it was argued earlier, translation of culturally specific notions can be very difficult -especially humoristic elements, which may be the hardest to interpret. There are two major approaches to translating such problematic fragments -compensation and omission. Compensation is the translation of the text to the closest equivalent in a target language, which very rarely has anything in common with the original literal meaning. It is used when there is a chance of visible loss of the humour or the meaning. Omission is used in these rare cases in which a text cannot be effectively translated. Costales (2012, 390) provides the example of Monkey Island 2: Lechuck's Revenge where the name for the bar is SCUMM Bar-this is a play on the word scum and the name of the Script Creation Utility SCUMM used by developers of the game. This pun was not translated to any language.
In brief, localization and translation of video games are complicated processes. In the case of such complex medium, it is not enough to simply translate the text. There are many factors which should be taken into consideration while localizing a video game -culture, censorship, technical issues, specifications of the market and many more. Obstacles which occur for translators need to be dealt with caution and creativity -there is no one good method of translation; this is captured well by the quotation: "translation strategies cannot be strictly associated to a single genre or textual type but they are usually combined to preserve the game experience of video game"(Costales 2012, 392).

Subject and aims of research
The aim here is to analyse approaches to translation of video games and study how they function in the context of popular game localizations. The subject of research is translations of fully localised video games. The article analyses fragments from popular games and how they are translated. There are many aspects, which are taken into consideration in the process of choosing games for analysis: • A game must be fully localized.
• A mode of a game: the author takes into consideration solely single player games, or games in which a single player aspect is the most important. • A game has to be mentioned as a good example of translation either by experts (game journalists) or players themselves. While conducting research about good and bad localizations (and more precisely translations) of video games, it has become evident that there is not nearly enough research materials provided by ordinary players themselves. While their opinions can be found on websites devoted to particular games (especially in the case when localizations are very good or bad), there are not many general discussions started by players or even much useful information in general. It is pictured very well by Google and its search results.
If we input the phrase "najlepsze tłumaczenia gier wideo" ("the best translations of video games") into Google, among 20 first propositions, we can find 4 websites of companies which are connected with game translation industry, 2 hits which are not connected with video games at all, and 12 articles by different journalists. The last two search results are originated from translators themselves -however, they are the only results which left the space for discussion (rather than a presentation of a topic from the perspective of a journalist). The situation is similar if we change the phrase to "najlepsze lokalizacje gier wideo" ("the best localizations of video games"). In this case, 7 positions out of 20 are websites of companies, and the rest are articles. If we search for these terms in English, the situation is similar. However, it changes if we start looking for the worst translations and localizations. Among the first 20 research positions there are 3 which are started by gamers. It can lead us to the assumption that gamers tend to acknowledge translations when they are not good, and in most other cases translations tend to be invisible.
The translations of video games in the minds of players tends to be invisible, unless it is of very poor quality. After a pilot analysis of comments on the threads left on the forums of portals CD Action and Mlingua and PC Labs it can be seen that attention is mostly paid to games which are connected with strongly unpleasant memories. If we take a closer look at the thread initiated by a user kul (2009) on the forum of portal CD Action, we can find an interesting view on the topic. A comment left by SilentMan (kul 2009), for example, states that the translation can be very poor, as long as it does not break anything. He is not the only one who thinks that way, as we can see examples of Wizardy 8 (2001) in which the lack of Polish letters makes some puzzles impossible to answer, or Konung (1999), where some dialogue options where so similar that it was very difficult to pick the proper one. Many commentators come back to the memories of the games from the 1990s. As we may see in the video material of , it was the time when games were not translated into Polish; however, there was a demand for such translations. This led to the outcome that many amateurish translations of low quality were created and sold (illegally), and it was not until the late 1990s that official Polish localizations started to be created more often.
Leaving the extreme cases, translations tend to be left in a shadow of dubbing. For the question on the forum of gaming magazine CD Action: "Jakie gry z gatunków RTS i RPG uwazacie za najgorzej przetlumaczone na jezyk polski [sic]" (Which RPG and RTS games do you believe are translated badly into Polish?) (Kul 2009), most people write about dubbings. Out of 112 comments, only 32 (29%) are about translations. What is worth mentioning is the fact that some of them are about amateurish translations from the 1990s and have little value for research conducted in this paper. 17 (15%) comments mention both translations and dubbings. I do not classify 11 (10%) comments to either category because they are either not connected with the topic or they state: "Game XXX is good/bad" without giving a reason. As many as 52 (46%) comments concentrate solely on good and poor dubbings.
As we can see, voice acting is very important in the case of the experience of players. One of the main reasons of the disproportion between translations and dubbings may be articulated by one of the CD Action's forum users with the nick Diex (2009) "Najgorsze spolszczenie? Jak nie mam oryginału to nie mogę porównywać" -"The worst polonization? If I do not have an original version, I cannot compare" (kul 2009). Very often, if a player is not provided with the English version of a game (or simply s/ he cannot speak English sufficiently enough to understand the text fully) s/he cannot compare their quality. Such mistakes as misspellings or grammar errors are quite easy to spot for a Polish native speaker; however, if we take such nuances as jokes (which can be omitted), names or many other elements -the situation is more complex. On the other hand, poor dubbing -improper and emotionless voices -are quite easy to spot for everyone, to the point that even some game journalists closely connect the word polonization with dubbing only. In his article titled "Mainstreamowe TOP 5 -Ulubione polonizacje gier wideo" (" Taking into consideration the importance of dubbings, there is a need to consider them in this research. In his paper, Sajna (2013) contrasts movies and games, and concentrates on the ways of translating them. As the author states, in the case of a movie we can talk about 3 kinds of translations -subtitles, dubbings and voiceovers; however, the situation is different if we are talking about video games. The author underlines the fact that in games, there are only two methods of translation -dubbings and subtitles (Sajna 2013, 230).
Sajna's division can be completed by Chandler and O'Malley Deming (2011), who distinguish different levels of localizations. The authors ring-fence 4 levels: • No localization: a game is taken as it is -without translation (it is mostly common in the case of low budget games produced by small companies, which are sold through big online platforms like Steam). • "Box and docs" localization: a game itself remains not localized, however packaging and manuals are translated. • Partial localization (also known in Poland as "wersja kinowa"): a game is translated only partially. In most cases text and packaging are translated; however, spoken dialogues are solely subtitled. We may say that because of differences in preferences, it is very difficult to identify good and bad translations. As comments to the analysed threads show, some games which are perceived by others as poorly localized, may be good for others; or some games which are clearly badly translated may be funny for some players, to the point that they like them (and can we say that the translation of an entertainment product is bad if it entertains customers?). Even in the case of such classics as Baldur's Gates (1998) -one of the first fully localized games, and by many given as a good example of both dubbing and translation, there can be controversies. Chojnowski (2013), one of the authors of Polish localization of Baldur's Gate, states in an interview that he believes that their localization is just mediocre.
The paper analyses the aspect of translation in the selected fully localised video games and tries to answer the question how translations look in the context of popular video games. The paper tries to answer the following research questions. I believe that in many cases (especially in the case of games highly concentrated on a story) the most common practice of video games is domestication, which is used to make games more familiar. However, very often, a good translation does not mean flawless and flaws are masked by general quality of a story, gameplay and a decent level of translation. Mistakes, frequently, are connected with a busy schedule and insufficient additional information, not with an incompetence of translators or impossibility to translate a given fragment. Because of that, many aspects of translations can be corrected and, regularly, are corrected -officially with different patches or unofficially by fans.
The starting point for research in this paper is a content analysis of forum posts. It is a form of quantitative research, and it is based on "counting phenomena in text" (Stokes 2003, 56). The main core of the research is based on qualitative analysis -more precisely, semiotic analysis. Semiotic analysis is useful if someone wants to look closer at the meaning of the text (Stokes 2003, 70). This kind of technique is frequently used with images. However, in the case of the paper and translations research, the meaning is a key factor, which encouraged me to use this method. As a result, the research methods used in the process of conducting analysis are both quantitative and qualitative. The research techniques used in the paper are content and semiotics analysis. Analysed fragments are from different genres of games (adventure, RPG, RTS). Games selected are: Ace Ventura (1997), Baldur's Gate (1998) and Warcraft 3: Reign of Chaos (2002).

Analysis
The aim of this part is to compare English and Polish versions of selected games and differences which occur between them. It presents case studies of three games, which represent different genres -RPG (role playing game), RTS (real-time strategy) and adventure. Games analyzed in this chapter are: Ace Ventura (1997), Baldur's Gate (1998), and Warcraft 3: Reign of Chaos (2002). All of these games are chosen based on opinions of experts, players and my personal impressions. They are perceived as products with very good polonizations and of good quality.

Ace Ventura
Ace Ventura is an adventure game which was released in 1997. It is perceived as one of the first properly and fully polonized games . The story is a spinoff of very popular movie Ace Ventura: Pet Detective (1994). In the game, we, as the eponymous hero -Ace Ventura, find a trace of people who are using endangered animals to make profits. Our goal is to stop them. I have analyzed the whole gameplay and decided to provide few fragments which, in my opinion, proves the state of the translation best.
While the game is constructed in the universal way -there are not many puns connected with the culture -there are such elements.

Example1
Jacques Demonkey ex-ocean documentary host, who has got the network acts because the Nielsen families could not understand a word he was saying Jacques Demonkey były prowadzący w telewizji dokumentalny program oceaniczny, wylany, bo nikt nie rozumiał słowa z jego gadki As we can see above, there is a notion: Nielsen families, which means a tested audience for new TV programs. While this term may still be untranslatable (despite the fact that since 2001 Nielsen families are used in Poland, the term alone is not so commonly known), translating it simply as nikt (no one) and leaving out any cultural addition is not the best option. I would like to propose: Jacques Demonkey były prowadzący w telewizji dokumentalny program oceaniczny, wylany, bo statystyczny Kowalski nie rozumiał słowa z jego gadki.
The root of the sentence is the same; however, statystyczny Kowalski adds a nice cultural connotation (Kowalski is a popular Polish name, which is used often to exemplify a statistical Polish citizen).

Example 2
• You're serving a lovely dinner of death Jacques. • I see, you don't digest denizens of the deep.
• You've got that right frog breath, besides this is a wrong wine. • But they are such delicacy.
• Yeah? Try spelunking through your poop chute sometimes. What you see may help you make a switch to whole grains.
In this fragment there is an interesting term -frog breath, which corresponds with the nationality of Jacques -French. Polish translators translate it as żabojad (frog eater), which is a very nice familiarization.
Another cultural reference in the fragment presented above is a poop chute; it is a colloquial name for an anus. In the game it is an obvious wordplay, which is connected with the fact that at the beginning of the level, the hero is supposed to swim through a pipe (chute) with waste. It is translated simply into zbiornik odpadów (holding tank), and the wordplay is lost completely. While it is really difficult to construct a proper phrase in Polish, I can propose the equivalent: Tak? Obadaj zadnią część swojego statku. Zadnia część swojego statku literally means -at the back of your ship (however zad also means haunch). There are other mistakes in this fragment, but they are of different nature than cultural and I analyze them in the latter part of the chapter.

Example 3
-Guess what Day it is Ventura! -Groundhog Day? No, Warthog Day? -Try rent day.
In the original the text is connected with animals -groundhog and warthog. What is more, Groundhog Day in its figurative meaning means repeating a day over and over, the meaning which was popularized by a movie Groundhog Day (1993). In the case of Ace, Mr. Shickadance (the dialog partner) visits him after every big mission in the same manner -he wants his money.
In the Polish version, the motive of animals is substituted by a family motive -Mother's Day and Children's Day. I believe that the original Groundhog Day should be preserved. The change in the meaning is probable because translators were worried that the Groundhog Day and its figurative meaning is too foreign. However, in the day of a release of the game, it had been 4 years from the world's premiere of the movie Groundhog Day (1993), which is enough time for the term to be familiar -taking into consideration that our country was very open to American products and movies in the 90's. For the comparison, it had been 3 years in the case of the movie Ace Ventura: Pet detective (1994) (which the game is based on).
I strongly believe, that the other day -Warthog Day is there for maintaining a rhythm and melody in the sentence. Polish Dzień Świstaka would greatly fit with the phrase Dzień Futrzaka (Futrzak means an animal with fur, which means that the animal motive can be preserved). My proposition is: Mantooth is an American actor, who was very popular in the 70's, he is known for many roles in action/adventure TV series and movies. I believe that translators had a great idea to compare him with our own TV action/adventure series from the 70's Stawka większa niż życie -Bruner and Agent J23 are characters from the series. The translation, while good, is not ideal -there are some mistakes, which are worth analyzing. In some cases, they are almost unnoticeable; however, sometimes, they can change the sense of the fragment.

Example 5
• You're serving a lovely dinner of death Jacques. • I see, you don't digest denizens of the deep.
• You've got that right frog breath, besides this is a wrong wine. • But they are such delicacy.
• Yeah? Try spelunking through your poop chute sometimes. What you see may help you make a switch to whole grains.
Let us look at the fragment, which was analyzed in the previous section, one more time. This time I have underlined different parts, which highlight the mistakes. The phrase it's a wrong wine is translated into to nie jest wino (it's NOT a wine), and in the place of whole grain, Polish text states: winogron (grapes) in the place of produkty pełno ziarniste. While produkty pełnoziarniste may be a little too long for the sentence and to maintain the rhythm -pełne ziarna should be a better and more logical option than winogron (grapes). Including my former adjustment -the whole text can look like this:

Example 6
Current a scourge to the seven seas and no friend of marine life, who still hasn't figured out that elocution lessons are now available on books and tapes.
In the fragment presented above marine life is translated as marynarski styl życia (a lifestyle of a sailor). If we look at the context -Jacques is a captain of a submarine who kills fish -we may say that he likes being a sailor; however, he is an enemy to marine life (fish). My proposition: Aktualnie jest plagą siedmiu mórz i wrogiem życia morskiego, który wciąż nie odkrył, że podręczniki sztuki wymowy wydawane są na kasetach. In the fragment I added a term fire extinguisher -it means the moment when Ace is cooled by it -which brings his temperature back to normal. In the translation, the fragment after the part with fire extinguisher states -control environment has been destabilized -so here the situation deteriorates, not stabilizes. The proper term in this case should be:

Huston kontrola klimatyczna środowiska się stabilizuje
In the next part of the analysis there are fragments, which are not translated wrongly -rather differently, but because of that the nature of the original is lost.

Example 8
Take me to your heater and then take me to your leader.

Weź mnie do grzejnika, a potem weź mnie do siebie
The text presented above in the Polish translation loses both a rhythm and, to some extent, a meaning. The term weź mnie do siebie-means simply -take me to your place rather than to your leader. While it is not a mistake, I felt obliged to mention this fragment. My propositionis: Weź mnie do kaloryfera, a potem do lidera The word lider may not be strictly a Polish word; nevertheless, it is so popular, that it is completely understandable by most people.

Example 9
Tragic cross the tundra in search of gas to the Arctic avenger or seize command of the not so slow looking slowtinian sleigh ride Nie ma sensu przedzierać się przez tundrę w poszukiwaniu paliwa dla Arktycznego mściciela. Muszę zatroszczyć się o jakiś nowy środek lokomocji.
In the translation presented above, the wordplay with the word slow is completely lacking -my proposition is: Rozpaczliwa podróż w poszukiwaniu paliwa do Arktycznego mściciela, czy uwolnienie tego nie wyglądającego powolnie skutera. As it was stated in the previous parts of the article, technicalities are also important in the process of making translations. In the game Ace Ventura, we can observe some mistakes in this matter. In the game there are no Polish signs, and consequently, we can see pictures as provided above. What is more not all the abbreviations on the computer screens are corrected, for example: S stands for Security, while in Polish it is Ochrona, so it should be O. While it may be the case of technical restrictions (shortcuts might be Pobrane z czasopisma New Horizons in English Studies http://newhorizons.umcs.pl Data: 14/06/2020 17:01:14 U M C S hardcoded), some of them are corrected like Księga Inspekcji, where there is I rather than L or W rather than M. This is worth mentioning, because it can be confounding for players.
The overall translation of Ace Ventura is good, as well as the voice acting. Authors tried with great care to familiarize all cultural nuances and jokes. There are some mistakes, which can be very difficult to spot, if there is no access to the original. The source of these mistakes can be different -a short amount of time or not enough information about the context (it is evident in the case of translation of marine life or the fragment with extinguisher). In some cases, it appears as if the text (or at least its parts) was translated from English audio, which in the case of Ace Ventura is extremely fast, and sometimes almost illegible and prone to misunderstanding. For example: Whole grain -all/only grapes wrong wine -not a wine Of course -it may be not the case, but taking into consideration that the game was one of the first Polish localizations and had many difficulties to overcome, the possibility cannot be completely discarded.

Baldur's Gate
Baldur's Gate is an RPG game which was released in 1998. It is perceived as the first polonization on such a big scale . Despite the old age, the game is still popular, the fact which is proved by the recent re-master of it (in 2012). The game tells a story of the child of Bhaal (the god of destruction and slaughter), who doesn't know his/her origins. S/he (the player can choose the gender) looks for the murderer of his/her stepfather. Taking into consideration the massive amount of text, I have decided to analyze only cutscenes at the beginning of the chapters and dreams (which are meaningful in the context of the story).
While Baldur's Gate (in my opinion) is translated very well it has some flaws, which are unavoidable in a production on such a scale.

Example 11
Imparting hurried instructions, for you to equip yourself for travel, Pospiesznie udzielił ci kilku wskazówek odnośnie ekwipunku, While in the original, the hero's stepfather orders him/her to equip for the journeyin the Polish version he gives him/her some advice about equipment. This may be not a big mistake; however, it is quite visible and it may sound awkward. My proposition is: Pośpiesznie przykazał, abyś przygotował wszystko, co niezbędne do podróży.

Example 12
You must find and expose the one who is responsible for your predicament, you must find Sarevok Musisz ich zdemaskować, niech odpowiedzą za wszystkie twoje nieszczęścia, oraz musisz znaleźć Sarevoka This fragment in its original form states directly that it is Sarevok who is responsible for the problems of the main hero; the certainty is gone if we look at the translation. I believe that the difference between the original and the translated fragment is too big to perceive it just as a mistake. Probably it is a conscious decision of the translator to change the meaning; however, for me it is a mistake and an unnecessary addition. My proposition is: Musisz odnaleźć tego, który odpowiada za twoje nieszczęścia, musisz odnaleźć Sarevoka.

Example 13
When light returns, you do not find its presence comforting.

Światło znowu rozbłyska, co nieco cię uspokaja
The fragment presented above has completely different meaning than the original. In the Polish version the light is comforting -in English it is not. My proposition: Powrót światła wcale cię nie uspokaja.
It appears that in Baldur's Gate there are many fragments, which are translated slightly differently, and consequently, they can change the atmosphere of a given part. Let us look at the fragments from the Prologue.

Example 14
It is an imposing fortress, kept in strict isolation from the intrigues that occasionally plague the rest of the Forgotten Realm. It is secluded, highly regimented, and it is home.

Example 16
With your hurried flight from Candlekeep barely behind you, the troubles facing the Sword Coast seem an unfamiliar blur to your fractured nerves.
In my opinion, the fragment presented above is one, which undergoes the biggest change. Let us have a look at my translation of the Polish text: You have just flight from Candlekeep, and yet you have to face all of the dangers which await you in the Sword Coast… Your mind is confused. The spirit of the translation is completely different than the original. The English version underlines the fact of the lost and the confusion of the main hero, rather than emphasizing the part of the challenge and fight (as it is in the case of Polish translation).
As it is presented, in the game Baldur's Gate (1998) there are many fragments which are different than the original. However, I do not believe we can call them mistakes. The pattern and the number of them may suggest that it is a conscious decision of the translators -to create their own story, their own atmosphere -a little bit darker and full of an ominous presence.

Warcraft 3: Reign of Chaos
Warcraft 3: Reing of Chaos is an RTS game which was released in 2002. It is perceived as an example of a very good polonization (Krajewski 2014). The game tells a story of a fantasy land, Azeroth, which is polluted by the plague of undead. In this paper I have analyzed the first chapter of the game -in which a player commands the human fraction.
In Warcraft 3, a player can notice the fact that some names are translated, and some are not. It is an inconsistent mix of foreignization and domestication. In the polonization there is a pattern that geographical names are not translated -for example, towns or names of countries, but there is at least one exception to this rule, King's Road -Królewski Szlak. We can find major characters without translated names such as Jaina Proudmoore or Uther Lightbringer, while a few missions later we meet a hero whose name is Muradin Bronzebeard -Muradin Miedzianobrody. While some may say that the situation is similar to the game Baldur's Gate, where some geographical names are also translated and some are not, it is important to notice that in the case of Baldur's Gate, translators had to cope with the existing lore of the paper RPG. While as we may read on the webpage https://www.gry-online.pl Warcraft 3 was the first part of the series which was polonized and probably translators had greater freedom in the context of translating proper names (in the case of previous installments only the "box and docs" translation was available in our country). For example, the German version of Jaina Proudmoore is Jaina Prachtmeer, and in Spanish the same character is called Jaina Valiente, so the author of this paper assumes that the translation of proper names (at least some of them) was possible at that time; now, because of the rich lore of Warcraft, it is probably impossible.
Nevertheless, it would be much better if translators had been more consistent in their choices. My proposition of translations: Jaina Proudmoore -Jaina Wrzosodumna Uther Lightbringer -Uther Światłonośny Not only is inconsistency connected with translation of the game, we can also find simplifications in comparison to the original texts, which are unrecognizable if we play the Polish version, only. They do not have a great impact on the overall quality of the translation; however, it says a lot about the attitude of translators toward fidelity to the original text.

Example 17
If this grain can spread the plague, there's no telling how many villages might be affected.
Jeśli to ziarno jest zakażone, plaga ogarnie wkrótce setki wiosek While the Polish translation is not bad -it is quite good; however, it gives us the number -hundreds of villages. The English version is more careful -it can be dozens, hundreds or thousands -and I believe it is better, because it leaves to the player's imagination how grave situation is. My proposition of a translation is: Jeśli to ziarno jest zakażone, nikt nie jest w stanie powiedzieć ile wiosek zostanie ogarniętych plagą.

Example 18
Prince Arthas! During the night, a vast army of undead warriors emerged and began attacking villages at random.
Panie! Zeszłej nocy pojawiła się w okolicy potężna armia nieumarłych. Atakują nasze wioski In the case of this translation, the Polish version lacks random part -again it is an unimportant difference; however, it can somehow influence the whole atmosphere of the text -for example: when we are attacked at random -we do not know where or when an enemy can attack -it can be anywhere. On the other hand, the translation is much simpler -they attack our villages. I believe that a substitute for at random should be included: This is a notification after the finish of the main campaign (before the final cutscene in which Arthas kills his father). As we can see one sentence is completely omitted by translators -the one which definitely states that the old prince Arthas is no more.
Warcraft 3: Reign of Chaos (2002) is an RTS game, in which the story itself is not so important, and it is visible in the fact that the translation is less precise than in the case of the other two analyzed games. We can see many simplifications and inconsistences (as in the case of the names). However, even in this game, translators decided to tell the story in their own way -including only information, which they wanted.
It appears that there are no perfect polonizations, even if we take into consideration highly regarded titles. While they all include mistakes, which are sometimes difficult to explain, all analyzed games are examples of very good translations. Taking into consideration the difficulties and the Polish game market, which was relatively young when these polonizations were created, translators have done a tremendous work. However, while the titles try to juggle between styles -foreignizations and domestications, and they try to be as funny, atmospheric or mysterious as they can be, something else is more prominent than this. Translators try to tell their own story -they omit and alter fragments to obtain the goal and the vision they decided to pursue. This can leave us with the impression of how much an original can differ from a translation and to what extent an impression we have after playing games depends on translators, rather than creators.

Conclusions
Translators encounter many problems of different nature during their work. In order to properly translate the text it is necessary to know the game very well -many titles are based on already existing lore or technicalities. Cultural and legal differences between countries are also very important in order to make a game understandable or, in extreme cases, allowed in a given country (for example, Germany, where there is very strict policy about violence in games). However, there is a need to remember that the most important goal of a translator is to make a game enjoyable for players. To obtain this, there is a need to use different techniques such as domestication, foreignization, transcreation, compensation or omission.
Nevertheless, problems in the localization process are not solely connected with texts. We should consider the process of translation itself as a possible obstacle, for example, in the case when there is not enough time or information about a game. There are also restrictions connected with the software (like limited space for translation or a lack of some target language letters).
As the research proved, a proper localization is important for players; however, they tend to forget about translations as a separate entity -they perceive the game as a whole. It is highlighted by a small number of discussions between players about the quality of polonizations. By analyzing available research material (various forum entries), it appears that in the case of games, dubbing seems to be more visible than translation. It may be connected with the fact that for players who do not have access to original versions of games (or their knowledge about an original language of a game is insufficient) it is more difficult to point out a mistake in the translation than in the quality of dubbing. Nevertheless, it does not mean that translations are irrelevant. Players tend to remember games which are translated poorly and without translation we cannot say anything about the quality of a dubbing or polonization in general.
While analyzing translations of selected video games, it seems appropriate to take opinions of players and experts into consideration as well as acknowledge the influential role of a dubbing. Having this in mind, I decided to analyze three highly esteemed (for dubbing and gameplay) games -Ace Ventura (1997), Baldur's Gate (1998) and Warcraft 3: Reign of Chaos (2002).
Games analyzed in this paper are translated very well, with some minor faults. These translations are similar to original English versions, but not the same. Translators tend to use as many familiar aspects as possible -they translate jokes and cultural elements (with an exception of names and some geographical places, which are left untranslated). In some rare situations they use omissions to deal with problematic fragments. In most of the cases, teams which were responsible for translations tried to make them as close to the original as possible, or at least to make them preserve the original spirit. However, when we analyze given fragments with utmost care, it is possible to see a common tendency in two of the analyzed games -when we compare Baldur's Gate and Warcraft 3: Reign of Chaos, it can be observed that Polish and English versions have specific differences, and most probably it is a conscious decision of the translators.
These changes are very subtle. They are mainly connected with an omission of some fragments, or in a change of meaning of some words. In Baldur's Gate it can be observed that translators tend to make the mood a little bit darker and ominous in comparison to the original text. In the case of Warcraft 3: Reign of Chaos, altered fragments are connected with the future of the human prince Arthas, and with the fact that he undergoes an inner transformation. The regularity and consequence of the changes may suggest that they are planned. These alterations can cause a difference in the reception of these two games between Polish and English players.
It can be said that the role of a translator is to make a game for players. While doing this s/he does not have to be faithful to the original version of a game -the translator is to some extent a creator and should be treated as one.