Crowdsourcing y traduccion colaborativa: ?fenomenos de masas o amenaza silenciosa para los estudios de traduccion?
1. INTRODUCTION: A BRAVE NEW WORLD
Globalisation is a widespread phenomenon that has reshaped the way people communicate in current society: the geo-political changes of the 20th century and economic deregulation (market liberalisation, free trade etc.) have significantly increased the flow of goods, people and information. In other words, globalisation has shrunk the world by removing barriers and allowing access to information from (almost) anywhere in the world (Cronin 2003: 43). On the other hand, the progress achieved in the field of telecommunications allows us to consider the information society as the heir of the industrial society (Castells 1996: 46). As a matter of fact, new technologies have promoted the exchange of information and knowledge transfer amongst people by providing them with suitable tools and resources (Schaffner 2000: 1).
Arguably, technological changes and the establishment of the Internet as the standard communication tool for millions of people have contributed to the configuration of today's global panorama. The Web has clearly influenced our lives and some authors have already highlighted the possible dangers of the excessive dependency on information technology (or 'googelizatiori) of our society (Carr 2010). Regarding translation, the current panorama was foreseen more than a decade ago when O'Hagan (1996) described the forthcoming industry of the teletranslation, a global network where the paradigm of the information society would reshape the way people communicate (and translate) in the new era. Broadly speaking, these predictions were fairly accurate; in the last two decades we have witnessed the consolidation of the GILT industry (Globalisation, Internationalisation, Localisation and Translation) as a response to the need for the internationalisation and localisation of products (mainly software and websites) which has been approached from different points of view (Esselink 2000, Fernandez Costales 2009, Pym 2004, O'Hagan and Ashworth 2002, Singh and Pereira 2005, Yunker 2002).
The discipline of Translation Studies has evolved in order to meet the continuous flow of challenges posed by multimedia products and modern genres and textual types: high-speed modes of interaction among users in the so-called 'Web 2.0' have opened new horizons, as have they fostered the dissemination of new phenomena in Translation Studies. Fan translation, or the adaptation of multimedia products by amateur translators, has been clearly propelled by new technologies and is a phenomenon that can hardly be avoided. Indeed, collaborative translation can have an impact on the way the translation industry develops in the coming years; as big corporations (e.g. Google, Microsoft etc.) rely on crowdsourcing for translating some of the content of their websites and applications, new business models and synergies can arise in the short term.
Relatively recently, linguistics and translation scholars have begun to work in the field of amateur translation, an activity usually frowned upon by academia and professional associations alike. A relevant question to pose here is whether it is actually important to find out about the work being done by nonprofessional translators. In other words, should scientific research be conducted on an amateur area where no quality standards or criteria have been established? Is it possible that by exploring this new field we are in fact promoting low-quality activities and improving their visibility? As there are no empirical data on the number of non-professional translations (here including interpreting, which is usually ignored in the study of collaborative translation), this seems to be a rather undefined and obscure area to analyse.
This paper underlines the idea that Translation Studies have to approach this field in order to gain a clear picture of the status and the real situation of translation today. Also, for the industry and the professional practice of translation, a knowledge of 'who is doing what' must be on the radar of both associations and freelancers, since it may supply valuable data and information to promote best practices and aid the struggle for recognition of the profession and the role of (human) translators. The article's main hypothesis is that a line must be drawn to separate crowdsourcing from amateur or collaborative translation, as the rationale and motivation triggering both phenomena cannot be grouped under the same category.
The paper is structured as follows: the following section (section two) defines the object of study being analysed, and the different activities developed under the taxonomy of collaborative translation; section three reports on the main features of collaborative translation and also discusses the different possibilities regarding the terminology to be applied; in section four, the grounds for defining the boundaries between crowdsourcing and fan translation are explained, while section five discusses the implications of web- based and collaborative phenomena for Translation Studies and the translation industry. Finally, section six comprises the main conclusions of the paper.
2. DEFINING THE OBJECT OF STUDY: THE iTRANSLATION CROWD
As an emergent mass phenomenon, collaborative translation has yet to be sufficiently explored and a large body of specialised literature relating to the field does not exist. However, it is worth mentioning that this web-based process has been recently addressed by several scholars (Cronin 2010, Diaz- Cintas and Munoz Sanchez 2006, Ferrer Simo 2005, Garcia 2010, O'Hagan 2009, Perez Gonzalez 2006, Perrino 2009) and is being widely discussed at international conferences and congresses.
The coming of age of the Internet with Web 2.0 and further developments aimed at achieving the so-called 'ubiquitous Web' have triggered the emergence of new collaborative phenomena whereby individuals and groups of people share information on a free basis, collaborate on different projects, and edit and create new digital content such as wikis, forums, podcasts or websites as a community. In addition, the interaction between users and the technology itself has been triggered through allowing individuals to produce rather than only consume electronic content, leading to a shift from consumers to prosumers (O'Hagan 2009). Virtual communities have been created and the number of Internet users has grown exponentially in recent decades; the setting up of global networks has also supported the belief that collective intelligence or "the wisdom of crowds" (Surowiecki 2004) may be collectively more powerful than the efforts of single individuals. In a nutshell, the Web has allowed users to participate and collaborate in the creation of a system where "much like in a social insect colony, individuals contribute to the bigger picture by adding or editing the elements that make up the building" (Perrino 2009). Similarly, Brabham (2008) considers that the rise of Web 2.0 is the main contributing factor to defining a new paradigm based on the overwhelming potential of the crowd, which is able to outperform the staff of multinational corporations and provide solutions to problems encountered by many companies.
It is worth mentioning that the Web may also have affected and reconfigured fan cultures (Hills 2002: 170). Even though 'fandoms' were set up and begun a long time ago, before computers were available to wider society (Jenkins 2006: 137), the impact the Web has had on the promotion of fan culture seems quite clear due to the number of devoted sites that have appeared on the Web. In fact, the involvement of more and more users in any kind of Internet-based fan activity is moving the fandom "from cult status to cultural mainstream" (Jenkins 2006: 142). For the scope and interest of this paper, fansubbing, romhacking and scanlations can be regarded as the three most relevant and influential types of "collaborative translation" which exist today.
Fansubs have been already approached by several scholars (Diaz-Cintas and Munoz Sanchez 2006, Ferrer Simo 2005, O'Hagan 2008, Perez Gonzalez 2006), as they are a growing phenomenon and a fashionable activity for amateur translators to engage in. Fansubbing, derived from the expression fan subtitling, dates back to the 1990s and can be specifically located in the community of anime followers (Diaz-Cintas 2005). As originally no translations for many manga series were made into Western languages, some fans started to adapt their favourite stories from Japanese into English, aiming to grant access to these materials to the community of non-Japanese speaking manga followers across the world.
It is worth highlighting the increasing number of websites and forums devoted to the subtitling of American TV shows from English into Spanish, with a particularly high number of fansubbers from South America (Argentina and Mexico are especially active as far as fansubbing is concerned). The striking fact is that a new episode of a TV show released on an American TV channel is subtitled and uploaded to certain sites within hours of being aired. This is easily explained by the number of fansubbers working on the translation of one particular episode, as shown in Figure 1:
As can be seen in Figure 1, episodes are not translated and subtitled by a single user. On the contrary, amateur translators adapt particular strands or small portions of each show. In this case, 37 different users have translated this episode of House, which has been edited 971 times. Every user can edit any of the 975 sequences or segments into which the episode has been divided, providing his or her own proposals for the subtitles. This can be used as an example of the type of 'workflow' normally created on this type of website, whereby a group of users works as a community in the adaptation of a product to a particular culture.
Fansubbing is an activity which falls under the scope of audiovisual translation, as fans use specific software (in most cases Open Source and freeware applications) (2) to subtitle a series and, as in any subtitling process, sync techniques are then applied to synchronize the subtitles with the image on the screen. Indeed, fansubbing is primarily concerned with interlingual subtitles where the script is adapted from the source into the target language, as opposed to intralingual subtitles (or captioning) in which the on-screen text is in the same language as the dialogues (Gottlieb 1998: 247). Besides the translation of the dialogues and lip-syncing, other extra-textual elements must be taken into account: cultural references, the use of colours, cumulative subtitles, dialogue techniques etc. (Diaz-Cintas 2005).
Another type of fan translation is so-called 'rom hacking' (Munoz Sanchez 2008, O'Hagan 2009), where players adapt their favourite games to a particular locale. In this case, gamers also adapt a digital product into a specific culture in order to allow other fans to play it in their own language. However, in some cases the motivation is slightly different to that of fansubbing, as many 'rom hackers' adapt the ROM with the aim of playing an old game from an extinguished platform on a modern system like the PC. Also, on many occasions the translational approaches are rather different to those of fansubbing because rom hackers adapt the game to a target locale without adhering to the source text. This can involve changing the dialogues, modifying proper nouns, creating new scenes from scratch or altering the story line. These can all be considered as proof of 'translating for pleasure's sake', or could even be regarded as examples of the transcreation or carte blanche that translators are alleged to have by some in the case of video game localisation (Mangiron and O'Hagan 2006). Besides rom hacking, in the case of video games other alternative practices relating to user-generated content can be observed: there are large communities of players translating content on the Web, modding and undubbing video games, or creating walkthroughs to help others advance in a particular game, or even writing so-called 'fanfics' or "creative extensions of an original game" (O'Hagan 2009).
Finally, the adaptation by fans of comic books into other languages is a phenomenon known as 'scanlation', an expression formed by the words "scan" and "translation" (Ferrer Simo 2005). In this case, followers of comics translate their favourite stories and scan them, creating electronic files that are subsequently distributed via the Web. As in the case of rom hacking, space restrictions are one of the most important challenges needing to be tackled by amateur translators adapting comics to other cultures. As in the two previous cases, a large number of fans are translating comic and graphic novels from Japanese in to Western languages. Therefore, a good cultural command of Japanese customs and traditions is an expected requirement for amateur translators engaging in scanlations.
There are no empirical data on the total amount of information translated by people on a voluntary basis. Similarly, the number of documents adapted by the masses through the process of crowdsourcing cannot be estimated, even when some companies partially reported their figures regarding the massive outsourcing of translation to volunteers. In other words, we have reached the era of the translation or 'itranslation' crowd. The differentiation between amateur translation and crowdsourcing is an essential issue approached in this paper, as there are visible and underlying features separating the characteristics of each that need to be explained.
3. COLLABORATIVE TRANSLATION REVISITED
The previous section offered a brief outline of some of the most representative and relevant cases of collaborative translation. Several differences between these activities are evident since they belong to different fields and scopes. Nevertheless, there is one common feature underlying all these processes which means the basis for a translation can be considered "collaborative": the fact that they are all adaptations made by fans for fans.
We are now dealing with the motivation of amateur translators. It seems that the personal desire to contribute to the spread of information or to allow other people to read and access materials in their own language are both underlying features in all these phenomena. Additional nuances can be explored, as there are many different grounds to explain why people get engaged in collaborative translation. They do so in order to promote a particular field and let other people read about it, simply for the pleasure they get from translating (i.e. 'translation for art's sake'), to improve their writing skills in a foreign language, (3) or even for more altruistic and determined purposes (for example, some people translate on an unpaid basis for NGOs or non-profit organizations such as UNICEF and Save the Children, and there are volunteer translators adapting texts on rare illnesses, marginal or peripheral communities etc.). A rather interesting case in point is the Rosetta Foundation, a spin-off of the University of Limerick which aims to relieve poverty and promote justice by removing language barriers. (4) More specifically, this non-profit organisation aims to make information available to people all over the world regardless of their level of income, cultural background or the language they speak (Anastasiou and Schaler 2010).
In any case, the different motivations of amateur translators can be categorized under the label of 'translated by fans for fans', as this generalization also constitutes the main feature common to all of them. One of the most famous examples of a rapid "fan" translation assignment was the adaptation of the last book in the Harry Potter series into Chinese in less than 48 hours by a group of amateur translators and volunteers (Munday 2008: 190).
As for the terminological issue, the adaptation of content on a voluntary basis has been given different names and labels: "wiki-translation" (Cronin 2010), fan translation, amateur translation, collaborative translation and community translation. Albeit all of these newly-coined terms seem applicable to some extent, for the purposes of this paper the last two seem to be most accurate compared to the alternatives. Wiki-translation would be appropriate to the adaptation of 'wiki' websites, but it would be unsuitable if applied to other contexts; fan translation can define processes like fansubbing, romhacking or scanlations, but is somewhat vague when used to deal with translations carried out by amateurs working in the NGO field, as for example people who translated information voluntarily after Haiti's earthquake in 2010 cannot be regarded as 'fans'. In addition, the term amateur translation ignores the fact that there are professional translators involved in some of these processes. Indeed, some professional translators feel engaged in humanitarian causes, social issues or political activism and willing to work for free in certain circumstances. Also, some translators who are starting their professional careers translate online content on a voluntary basis to obtain experience and perhaps also gain some visibility or publicity, or even build a portfolio for future employment. In fact, some researchers make the point that crowdsourcing and collaborative initiatives can benefit those users seeking to acquire new skills or to build on their experience when searching for a job or when establishing themselves as freelancers (Brabham 2008).
Collaborative translation and community translation cover all the cases previously mentioned and include reference to the translations performed by the community of users on a voluntary basis. These two terms would therefore seem to be more descriptive and accurate from the point of view of the subject being analysed; however, due to the advantage already provided by the expression 'fan translation', it is possible that this expression is more likely to be used by the practitioners of this phenomenon themselves. The acronym CT is used by some authors (DePalma and Kelly 2008) in reference to "community translation", "collaborative technology and processes" and "crowdsourcing", as they estimate all of these to be within the framework of user-based initiatives. However, this paper argues that although all these phenomena share common core features, some distinctions should be pointed out between "spontaneous" collaborative initiatives and those other processes initiated by companies or third parties.
The term 'User-Generated Translation' or UGT is preferred by O'Hagan (2009), and Perrino (2009) who considers collaborative translation to be "undoubtedly vague and tautological". As has already been stated in the introduction, one of the aims of this paper is to establish the possible barrier between the activities that have been commented on thus far and the process known as crowdsourcing. Therefore, accepting the suitability and appropriateness of the term User-Generated Translation for all translational activities observed on the Web (including crowdsourcing), the concept of collaborative translation is used in this paper to refer uniquely to those projects and initiatives started by users for users, leaving aside those processes promoted by companies or institutions outsourcing their services.
4. CROWDSOURCING: THE OBLIVION OF TRANSLATORS?
The term 'crowdsourcing' (a word compound created from "crowd" and "outsourcing") implies the massive outsourcing of tasks to volunteers and is intended to reduce the gap between professionals and amateurs. Jeff Howe, one of the promoters of crowdsourcing as a mass phenomenon (2008), provides the following definition on his website: "Crowdsourcing is the act of taking a job traditionally performed by a designated agent (usually an employee) and outsourcing it to an undefined, generally large group of people in the form of an open call".5 This definition is quite interesting as it mentions the fact that this task was previously assigned to an employee and is now outsourced to a non-specified pool of people. However, this statement does not comment on the possible use of the information by the companies performing crowdsourcing or the implications this may have for the people engaged in the process. On the same theme, Garcia (2010) states the following:
Of recent web developments, often conflated under the Web 2.0 tag, a major one is linked to crowdsourcing, roughly meaning the delegation to (unpaid) volunteers of tasks previously reserved for professionals. Journalism and photography are given as typical examples of areas in which the crowd is replacing the professional. Translation is another.
This observation adds a couple of relevant details to the prior definition, one of which being the inclusion of the adjective unpaid and the other the use of the word professionals. From reading this explanation, it could be suggested that crowdsourcing can lead to a loss of quality and professionalism in order to save money or cut costs (an idea supported by some researchers such as Bogucki 2009). Although some scholars working in the field consider crowdsourcing to be a beneficial process able to provide solutions to particular problems, we need to bear in mind that "problems solved and products designed by the crowd become the property of companies, who turn large profits off from this crowd labor" (Brabham 2008). Indeed, Brabham reports on well-known examples of organisations (such as Converse, Sony and Chrysler) which have implemented successful crowdsourcing strategies in order to develop their business and increase their market share. Obviously, the Web has turned out to be the perfect tool for the promotion of any activity which aims to reach the greatest number of people possible, so it is not surprising that web-based crowdsourcing has become tremendously successful. This process has been widely criticised as its ethical, social and economical implications are subject to debate (Diaz-Cintas and Munoz Sanchez 2006, Leonard 2005, Munday 2008: 190), and the true interest of many companies is somehow hidden behind the idea of improving interaction amongst users and consolidating the freedom of the Web.
For the interest and the scope of this paper, the question to be addressed at this point is who benefits from the efforts of volunteer translators (or shall we coin the terms crowdsourcees to refer to those performing the tasks and crowdsourcers to the people outsourcing the job?). In other words, provided that crowdsourcing is not run 'by fans for fans', the desire of public and private companies to cut costs and save some money can be the ultimate and underlying reason to promote it. Indeed, it would be interesting to have empirical data on the corporations and "dot-coms" whose revenues have rocketed thanks to the support of the 'collective intelligence of the crowd'.
Notwithstanding that the massive outsourcing of a task is contextualized in the framework of Web 2.0 and the promotion of user interaction, it is feasible to think that companies are using volunteer translators to keep the money that is not then invested in the translation industry (including here professional translators, proofreaders, post-editors or even machine translation systems). Also, the alleged "community or collective sense" defended by companies that rely on crowdsourcing is somewhat overshadowed if we pay attention to the fact that many of them have staff devoted to selecting and filtering the best alternatives of those submitted by the crowd.
This is the main difference which must be established between collaborative translation and crowdsourcing. While the former is based on the relationship between supporters of a certain field or volunteers in particular settings and scenarios, the latter cannot be included in the same category as some opacity can be detected in the workflow of the process. In other words, we have a relation among equals on the one hand and a hierarchical structure on the other. This is in tune with the differentiation made between crowdsourcing and "commons-based peer production" (Benkler 2002) in which users cooperate and work collectively without being managed, directed or sponsored by any organisation, an example of this being the creation of articles in Wikipedia (Brabham forthcoming).
In addition, issues of quality may not be considered and assessed on the same basis: collaborative translation relies on the free exchange of information as its main grounds; however, in crowdsourcing companies profit from the work done by people and therefore a certain degree of professionalism--i.e. quality standards or criteria--could be expected and requested by the final users. Many well-known companies such as Google, LinkedIn and Facebook have relied on crowdsourcing to have their online content translated into different languages (Garcia 2010, O'Hagan 2009). To illustrate this with an example, the user interface of the social network Facebook has been translated into 23 languages by volunteers who have agreed to participate in the crowdsourcing process launched by the company. Even though there are professional translators and agencies working on this project (and some of these can even gain some recognition or pursue promotional goals by translating pages or applications), most of the people adapting the content can be considered to be amateurs or fan translators. In the case of Spanish, Facebook has received 147,797 translations by users and the number of active translator accounts is 2346--this can be checked in the translation section of the website (shown in Figure 2):
The example of Facebook may have influenced other companies to follow suit. In 2009, LinkedIn sent an e-mail to more than 41 million users with a survey asking about their availability and willingness to translate online content on a voluntary basis. This provoked bitter protests from professional translators (Kelly 2009). Similarly, Second Life fans adapted the content of its site into eleven different locales, as three quarters of the registered users are not native speakers of English (Ray 2009). The process undertaken by these sites may serve to illustrate how crowdsourcing is working on the Web and the way in which it can affect the practice and industry of translation.
As commented on in previous sections, crowdsourcing and collaborative translation are commonly included in the same category. O'Hagan (2009) considers there is an "evolution from unsolicited fan translation to solicited community translation now called crowdsourcing" and includes both phenomena within "the framework of user-generated translation (UGT)", this concept being a hypernym of the previous terms:
Drawing on this description, I use the term user-generated translation (UGT) in this article to mean a wide range of Translation, carried out based on free user participation in digital media spaces where Translation is undertaken by unspecified self- selected individuals. The user in UGT therefore is somebody who voluntarily acts as a remediator of linguistically inaccessible products and direct producer of Translation on the basis of their knowledge of the given language as well as that of particular media content or genre, spurred by their substantial interest in the topic. (O'Hagan 2009).
In the same vein, Perrino (2009) also uses the expression User-Generated Content as an "umbrella term used to define translation practices made possible by various online services". However, the author of this paper considers that it is important to distinguish between user-based collaborative efforts (even when they may be affecting the translation industry in a negative way, as discussed in the next section) and those processes initiated by companies or corporations. In fact, we are contributing to the promotion of ambiguity and anonymity in the creation of digital and online content, and the concept "user" would be somehow obscured if private interests were involved.
As a concluding remark, the difference between crowdsourcing and collaborative translation can be settled by the fact that the first is a market- driven phenomenon while the second a user-centred process. Beyond the social or sociological connotations this statement could have, there are also nuclear questions regarding the status of the translator and the consideration of translation as a profession, as crowdsourcing invites "anyone to be a translator" (Bogucki 2009). At the same time however, the recognition of this activity is somehow ignored since "the crowdsourcing trend shows that the web values translation as a skill, if not as a profession" (Garcia 2010). The same label could be applied to the case of collaborative translation but the key difference here lies in the promoter of the activity: as long as the users are the ones who start translating, no economic interest can be directly linked to this initiative. In other words, while crowdsourcing is a top-down managed process, collaborative efforts such as open source methods and commons-based production rely on bottom-up approaches (Brabham forthcoming). This can also be explained using a more academic approach by applying the concept of the commissioner or the initiator in translation (Nord 1991: 93). While in one case, the translational process is started by private or public companies in the other, it is initiated by the users themselves, who also happen to be the target audience and the consumers / producers of the texts.
It is worth mentioning that there are no published studies exploring the motivation of people who translate on a voluntary basis: as we explained in the previous section, within the category of "collaborative translation" further sections or sub-groups can be established. As for the motivations of the companies, these might be even more difficult to define, since companies listed on the stock exchange such as Facebook should not impose budget constraints in order to avoid designing and implementing the localization of their website and applications into different locales. Following this, it could be hypothesised that other motivations lie behind the idea of crowdsourcing (e.g. using Web 2.0 as a tool intended to supply companies with market surveys and other information, or simply trying to make people use their website as much as possible). However, if the terms and conditions of most crowdsourcing-based activities are studied in detail it seems quite clear that the established policy seems to be rather unprofitable for the users who collaborate as part of the community, as can be observed in the following statement (Figure 3):
Figure 3: Screenshot of the terms and conditions applied to translations on Facebook (www.facebook.com/translations). Condiciones aplicables a la traduccion de Facebook Fecha de la ultima revision: 29 de junio de 2009 The Translate Facebook application collects translations, comments, suggestions, ideas, feedback and other information ("Submissions") from you and other users in connection with Facebook's language translation project to provide access to Facebook and applications and websites that use Facebook's platform in multiple languages (the "Project"). You understand that your participation in the Project is for the benefit of the Facebook user community as it will allow users whose participation is currently limited by language to more fully participate. You acknowledge that your participation in the Project is entirely voluntary, and you understand that no monetary or other compensation will be given to persons, including you, for Submissions. You may provide as much or as little input into the Project as you wish and can cease contributing to the Project at any time. In consideration of Facebook's permitting you to participate in the Project and the benefits to the Facebook user community of which you are a member, you acknowledge and agree that any Submissions that you provide to Facebook will be owned by Facebook. Accordingly, you irrevocably assign to Facebook all right, title and interest, including all intellectual property rights, in and to all Submissions, and Facebook is entitled to the unrestricted use and dissemination of these Submissions for any purpose, commercial or otherwise, without acknowledgment, consent or monetary or other tangible compensation to you. To the extent that the foregoing assignment is or becomes invalid or unenforceable to any degree or for any reason, you grant Facebook an irrevocable, perpetual, exclusive, fully-paidup, royalty-free, worldwide right and license, with the right to sublicense, to use, reproduce, display, perform, create derivative works of, distribute and otherwise exploit the Submissions in any manner.
Taking the example of Facebook, according to its terms and conditions of use (Figure 3), amateur translations "irrevocably assign to Facebook all right, title and interest, including all intellectual property rights". This includes the use and possible dissemination of all content submitted by users for any purpose and without any acknowledgment needing to be given to the authors.
The difference established between crowdsourcing and collaborative translation may seem to be idealistic, but is indeed based on very basic concepts of Translation Studies, and to some extent on the ingenuity of the translators. These translators keep offering the defence that translation is a human activity occurring on the eve of a possible paradigmatic shift towards a pattern of machine translation assisted by humans, where the post-edition of texts is, at least according to some scholars, called upon to be the future of the translation profession, acting as a replacement for 'translation proper' (Garcia 2010). As with any professional activity, translation has to be paid for, and therefore the involvement of companies and institutions is needed. However, the apparent goodwill of corporations when asking people to develop a task must be scrutinized, as they might be avoiding the proper recognition of professionals.
In a similar way to collaborative translation, Open Source solutions provide free software applications developed by computer engineers to users. In addition, there are doctors who work for humanitarian causes on a voluntary basis. By contrast however, neither Google nor Microsoft ask computer engineers to develop the code for their websites for free, and neither are surgeons invited to design medical equipment on a purely voluntary basis.
Having said that, it is reasonable to acknowledge that the differentiation between the two concepts dealt with in this paper is not seamless and can be eroded with time. On the one hand, the legality of fansubbing is questionable from the point of view of intellectual property and distribution rights (7) and fansubbers operate in a legal vacuum (Diaz-Cintas and Munoz Sanchez 2006, O'Hagan 2009). But also, the lack of legal action by companies producing TV series against fansubbers might be explained by the promotional campaign that fansubbers indirectly carry out when translating digital content into different languages. It could therefore be argued that companies are also engaged in collaborative translation by intentionally permitting people to adapt their multimedia products in order to gain more visibility and reach a wider audience.
Also, some cases can be too close to the line which separates both concepts and difficult know on which side they should be placed. Many institutional websites are not translated by professional translators but by the institution's own staff: European university websites for example are mainly translated thanks to the work of administration and teaching staff (Fernandez Costales 2010). Indeed, only 10% of the websites of universities listed in the ranking published in the Times Higher Education Supplement have been adapted by professional translators. In this case, the categorization of the situation as collaborative translation could pose some problems, but nor is it possible to consider it a clear example of crowdsourcing comparable to those described in this paper.
The specific nature of the Web as a high speed and dynamic communication tool suggests that the number of people creating, editing and sharing online content will continue to grow in the coming years. The amount of information on the Web increases on a daily basis and this will have a clear impact on the number of documents needing to be translated. It is worth mentioning that English is currently the international lingua franca, but in the case of the Internet, its growth is far lower than that of other emerging languages such as Chinese, Portuguese or Spanish. Indeed, statistics show that those using English only account for 27.3% of the total number of Internet users. (8) As more and more countries and territories have gained access to the Web, the number of users reading online content in different languages has grown exponentially (Cronin 2003: 14).
Under this framework, the need for translation seems to be quite clear and there is reason to believe that the volume of untranslated documents will drastically increase in the coming years if the number of--human--translators remains steady. This idea is frequently used to defend the development of machine translation systems and their use in sectors where human translators are not usually required. Thus, another hypothesis of this paper is that collaborative translation could also be subtly introduced into certain areas where companies do not need (or wish) to invest in professional translators or even in expensive automated translation systems. This could lead to a progressive erosion of the borders established setting out the differences between crowdsourcing and collaborative translation in this article.
5. TRANSLATION IN THE NEW ERA: NEW WINE IN OLD WINESKINS?
The new setting or scenario described in the previous section may have a clear impact on the translation industry, as new models and workflows are being established resulting in major changes in the professional practice of translation. However, do these new patterns lead to a shift in the translation paradigm? In other words, are Translation Studies as an academic discipline being affected by the demands and constraints of the new panorama? Indeed, we are currently living in the "technological turn", which has followed the linguistic turn of the 1970s and the cultural turn of the 1980s (Snell-Hornby 2006).
The question as to whether Translation Studies as a discipline should be concerned about these phenomena has been indirectly answered in the previous sections. As a relatively young subject, translation has been engaged in achieving recognition of its status as an academic field and in improving in its visibility amongst other branches of knowledge. The so-called fragmentation of Translation Studies (Munday 2008: 197) should not be regarded as a handicap but as proof that it is a dynamic and flourishing area in which many sub- disciplines can be effectively integrated. In addition, the interaction with other fields supports the idea of multidisciplinarity, a key concept necessary to address the challenges of the new era.
From an academic standpoint, it is necessary to study the new phenomena, even when they are not specifically addressing professional translators. Although no empirical data are available due to the lack of research in the field, it can be considered that the output of collaborative translation is poor, or at least that it is "often below par" (Diaz-Cintas and Munoz Sanchez 2006): from assessing the quality of many articles on Wikipedia and the typographical and other errors in many fansubbed series, it is clear that mistakes are relatively frequent. Indeed, in Spanish fansubbed versions, grammatical errors, wrong syntactic structures or literal translations of cultural elements are quite usual (Ferrer Simo 2005). One of the most common typographical errors in Spanish concerns the omission of the opening question mark (?) in many questions (an obvious case of negative transfer from English). Also, it is usual to see the use of Mexican Spanish expressions (e.g. carro, lindo) in Castilian Spanish subtitles. The figure below (Figure 4) shows several examples of incorrect translations in the last episodes released of the TV shows Person of Interest, Fringe and The Vampire Diaries.
Figure 4: Examples of incorrect translations into Spanish of TV show fansubs. TV Show Person of Interest Season 1 Episode 8 00:31:50 00:31:52 Person of Interest Season 1 Episode 8 00:32:02 00:32:04 Fringe Season 4-Episode 3 00:02:52-00:02:54 Fringe Season 4-Episo de 3 00:09:32-00:09:35 The Vampire Diaries Season 3 Episode 9 00:04:10-00:04:14 The Vampire Diaries Season 3 Episode 9 00:37:44-00:37:46 English You are ... familiar with poisons? Then the STASI came calling. --You haven't sleep in how long? --Since I got here. What are you saying. that he was scared to death? I miss you. I'm miserable here. So pull the plug on the plan. Spanish Eres. familiar con venenos? Entonces la STASI vino llamando. --?Cuanto hace que no duermes? --Desde que estuve aqui. Quieres decir. que se asusto hasta que se murio? Te echo de menos. Soy miserable aqui Asi que tira del enchufe en el plan.
Beyond typographical errors and other "minor" issues, failing to comply with the standards of (professional) subtitling is also commonly observable in fansubbing. For instance, certain conventions and subtitling techniques are sometimes not fully respected in fan subtitles, which do not stick to the golden rule of two lines per subtitle or 72 characters on screen (Pedersen 2011: 19). Although not a usual tendency, we can find examples of fansubs in which three lines cover an unusually large proportion of the screen. This might be more relevant, as failing to meet standards will also lead to a failure to meet user expectations.
Figure 5: An example of a three-line subtitle in a TV show fansub. House MD Season 8--Episode 6 00:03:58--00:04:02 Mucho. La eco nos mostrara cualquier crecimiento de la valvula cardiaca. Lo mas probable es que sean suficientes unos medicamentos.
Nevertheless, the possible lack of quality standards does not seem to be solid grounds for dismissing the exploration of this field from a scientific point of view. On the contrary, the study of collaborative translation and crowdsourcing can be used in order to improve best practices, defend the role of (professional) translators as the real experts in any communication setting far beyond simply the role of a technician who transfers language strings from "A" to "B"--and even to train our students. In any case, as most collaborative translation or user-generated content relies on the Web as the main communication tool, these phenomena could be effectively included within the paradigm of 'constrained translation' as described by Roberto Mayoral in his seminal paper:
We cannot translate the text without understanding how the other communicative elements add to or modify the meaning: and, on the other hand, the non-linguistic elements of the message not only constitute part of the meaning but also, on occasions, impose their own laws and conditions on the text (Mayoral 1988: 363).
It is interesting to note that the segmentation of texts in some crowdsourcing and collaborative translation contexts (e.g. fansubbing) do resemble some of the scenarios where professional translators have to adapt language strings without having the context of the message they are trying to convey. In this sense, some scholars warn about the dangers of forgetting about translational issues (the target audience, the customer, the translation brief etc.). When translators transfer information strings in an automated way without reflecting on the text as a whole, they are ignoring the selection of appropriate translation strategies. Thus we arrive at the concept of 'word-for-word transfer'. This idea is supported by Pym (2003) and also by Cronin, who comments on the threat of translators becoming "transnational cyborgs who can no longer be conceived of independently of the technologies with which they interact" (2003: 112).
The other side of the coin concerns the translation industry. Should professional translators be worried about the growing number of amateurs doing their job for free? As Garcia explains "the amount of work available to the professional translator on the web appears to be shrinking, due to the pincer effect of both machine translation and crowdsourcing" (2010). In this regard, it is interesting to point out that the possible comparison between crowdsourcing and machine translation has already been made (Anastasiou and Rajat 2011) on the basis of the cost reduction and time saving features shared by both processes. On the other hand, research is being conducted to assess the possible ways in which professional quality can be achieved by means of crowdsourcing translation. In order to do this successfully, Zaidan and Callison-Burch (2011) state that "although many of the individual non-expert translators produce low- quality, disfluent translations, we show that it is possible to get high quality translations in aggregate by soliciting multiple translations, redundantly editing them, and then selecting the best of the bunch". Perrino (2009) establishes a comparison between the recent developments in translation and major changes in the music industry, and uses the example of Napster--an underestimated initiative that planted the seeds the subsequent revolution in the way people consume audiovisual products in the new millennium. Gouadec (2007: 312) also states that "the industrialisation of translation tools and procedures is largely responsible for the appearance and the development of the rift between three separate translation worlds, i.e. 'industrial' translation, 'craft' translation and 'amateur' translation". Finally, Brabham (2008) comments on the impact of crowdsourcing in professional sectors, focusing on the example of iStockphoto, which has provoked a situation in which "professional stock photographers have become obsolete".
However, as has been previously mentioned, the dynamism and the interoperability of translation leave it a perfectly placed to overcome the challenges of the new era. From the very beginning of its recognition as an academic discipline, Translation Studies has been evolving and integrating new patterns and processes (e.g. audiovisual translation or interpreting studies) and the professional practice of translation has been reformulated and reshaped on the basis of new technologies. More recently, the consolidation of the localisation industry in the 1990s led to the debate as to whether or not it should be included under the umbrella of translation, or if it should in fact be regarded as a different discipline (Fernandez Costales 2009). In any case, the very essence of translation as a human activity has been preserved and every researcher and practitioner working in this field must be aware that the ingenuity and the audacity of (human professional) translators will be key to overcoming the linguistic and cultural barriers of a multicultural setting. In other words, the more automation there is, the more important the role of authentic experts should be in the communication and translational process. Indeed, research in the field of machine translation is not intended to replace human translators but instead provide them with useful tools and resources (Melby 2006). It is therefore the role of academia and the translation industry to analyse the new process and promote best practices at the same time as training the translators of the future.
This is in tune with some less pessimistic views suggesting that there is real power in the hands of amateur translators who are using Web 2.0 and the translation technologies provided by the boom of localization to turn crowdsourcing into an "instrument of human political intervention" or the "strategic use of technical resources to further human concerns or agendas" (Cronin 2010).
Modern trends and tendencies must be studied in order to gain a realistic picture of the current situation in the world of translation. The knowledge and the results obtained can then be used to take advantage of the new phenomena and provide solutions and best practices. The very essence of translation will remain, even when it is labelled with tags such as 2.0 or 3.0.
This paper highlights the importance of the emerging phenomena associated with user-generated content and tries to establish the difference between crowdsourcing and collaborative translation on the basis of the initiators of the translational process and the rationale promoting both processes. Where the former may bow to the interests of private companies, the latter is driven by the collective sense of collaboration traditionally associated with Web 2.0. Beyond legal and moral issues, and leaving aside the question of the poor output quality of non-professional translation, it is a fact that the amount of collaborative translations has been growing in recent years. This is a concerning issue for Translation Studies and more research is needed in this field, as it is necessary to have a comprehensive and realistic view of the current panorama of translation. Finding out 'who-does-what' can contribute to promoting the recognition and enhancement of the figure of the translator as the expert in multicultural scenarios. Further research could be conducted in the field of fansubbing and web translation, and the results could be used to train students or even to design methods of language teaching and learning through translation exercises.
The involvement of scholars and researchers in this field will contribute to the humanization and professionalization of the community discourse, not only with regards to user-generated content but also concerning language and translation technology. In the era of collective intelligence and crowdsourcing it is necessary that translation scholars and researchers are able to interact with professionals from the industry in order to provide solutions and best practices. The ingenuity and ability of human translators and researchers have allowed advancements in the field of Translation Studies and contributed to the never- ending process of overcoming linguistic and cultural barriers. These two features, together with the knowledge and expertise of scholars and professionals from other related areas, must be employed in order to analyse emerging phenomena resulting from progress in the field of technology.
Anastasiou, Dimitra and Reinhart Schaler. "Translating Vital Information: Localisation, Internationalisation and Globalisation". Syntheses Journal, 3 (2010): 11-25.
Anastasiou, Dimitra and Rajat Gupta. "Comparison of Crowdsourcing Translation with Machine Translation". Journal of Information Science, 36.6 (2011): 637-659.
Benkler, Yochai. "Coase's Penguin, or, Linux and The Nature of the Firm". Yale Law Journal, 112.3 (2002): 369-446.
Brabham, Daren C. "Crowdsourcing as a Model for Problem Solving: An Introduction and Cases". Convergence: The International Journal of Research into New Media Technologies, 14.1 (2008): 75-90.
Brabham, Daren C. "Crowdsourcing: A Model for Leveraging Online Communities". The Routledge Handbook of Participatory Cultures. Eds. Aaron Delwiche and Jennifer Henderson. New York: Routledge, forthcoming.
Bogucki, Luckasz. "Amateur Subtitling on the Internet". Audiovisual Translation: Language Transfer on Screen. Eds. Jorge Diaz-Cintas and Gunilla Anderman. New York: Palgrave Macmillan, 2009. 49-57.
Carr, Nicholas. The Shallows. London: Atlantic, 2010.
Castells, Manuel. La Era de la Informacion. Economia, Sociedad y Cultura. Vol. 1. La Sociedad Red. Madrid: Alianza Editorial, 1996.
Cronin, Michael. "The IT Crowd". Tradumatica, 8 (xxxx). <http://www.fti.uab.es/tradumatica/revista/num8/articles/04/04art.htm> [16 March 2010]
--: Translation and Globalization. London: Routledge, 2003.
De Palma, Donald. A., and Nataly Kelly. Translation of, for, and by the People. Lowell, MA: Common Sense Advisory, 2008.
Diaz-Cintas, Jorge. "Back to the Future in Subtitling". Marie Curie Euroconferences MuTra: Challenges of Multidimensional Translation. Xxxx: University of Saarland, 2005.
--: "El Subtitulado como Tecnica Docente". Vida Hispanica,12 (1995): 10-14.
Diaz-Cintas, Jorge, and Pablo Munoz Sanchez. "Fansubs: Audiovisual Translation in an Amateur Environment". The Journal of Specialised Translation, 6 (2006): 37-52. <http://www.jostrans.org/issue06/art diaz munoz.pdf> [16 March 2011]
Esselink, Bert. A Practical Guide to Localization. Amsterdam: John Benjamins, 2000.
Fernandez Costales, Alberto. Traduccion, Localizacion e Internacionalizacion: El Caso de las Paginas Web Universitarias. Unpublished PhD Thesis. University of Oviedo, 2010.
--: "Translation 2.0. The localization of Institutional Websites under the Scope of Functionalist Approaches". Translation and the (Trans formation of Identities. Selected Papers of the CETRA Research Seminar in Translation Studies. Ed. Dries De Crom. Leuven: Katholieke Universiteit, 2009. <http://www.kuleuven.be/cetra/papers/ papers.html> [17 March 2011].
Ferrer Simo, Maria Rosario. "Fansubs y Scanlations: La Influencia del Aficionado en los Criterios Profesionales". Puentes, 6 (2005): 27-43.
Garcia, Ignacio. "The Proper Place of Professionals (and Non-Professionals and Machines) in Web Translation". Tradumatica 8 (xxxx). <http://www.fti.uab.es/tradumatica /revista/num8/articles/02/02art.htm> [17 March 2011].
Gottlieb, Henrik. "Subtitling". Routledge Encyclopedia of Translation Studies. Ed. Mona Baker. London and New York: Routledge, 2001. 244-248.
Gouadec, Daniel. Translation as a Profession. Amsterdam: John Benjamins, 2007.
Hills, Matt. Fan Cultures. London and New York: Routledge, 2002.
Howe, Jeff. Crowdsourcing: Why the Power of Crowd Is Driving the Future of Business. London: Randomhouse, 2008.
Jenkins, Henry. Fans, Bloggers, and Gamers: Exploring Participatory Culture. New York and London: New York University Press, 2006.
Kelly, Nataly. "Freelance Translators Clash with Linkedin over Crowdsourced Translations". Common Sense Advisory, 16 (2009). <http://www.globalwatchtower.com/2009/ 06/19/linkedin-ct3/> [March 2011].
Leonard, Sean. "Progress Against the Law: Anime and Fandom, with the Key to the Globalization of Culture". International Journal of Cultural Studies, 8.3 (2005): 281-305.
Mangiron, Carmen, and Minako O'Hagan. "Game Localisation: Unleashing Imagination with "Restricted" Translation". The Journal of Specialised Translation, 6 (2006): 10- 21. <http://www.jostrans.org/issue06/art ohagan.php> [16 March 2011.]
Mayoral, Roberto et al. "Concept on Constrained Translation: Non-Linguistic Perspectives of Translation. Meta, 33 (1988): 356-67.
Melby, Alan K. "MT+TM+QA: The Future is Ours". Tradumatica, 4 (xxx~x). <http://www. fti.uab.es/tradumatica/revista/num4/articles/04/04art.htm> [18 March 2011]
Munday, Jeremy. Introducing Translation Studies. 2nd Edition. New York: Routledge, 2001.
Munoz Sanchez, Pablo. "En Torno a la Localizacion de Videojuegos Clasicos Mediante Tecnicas de Romhacking: Particularidades, Calidad y Aspectos Legales". The Journal of Specialised Translation, 9 (2008): 80-95. <http://www.jostrans.org/issue09/art munoz san chez.pdf> [17 March 2011]
Nord, Christiane. "Skopos, Loyalty and Translational Conventions". Target, 3.1 (1991): 91109.
O'Hagan, Minako. "Evolution of User-generated Translation: Fansubs, Translation Hacking and Crowdsourcing". Journal of Internationalisation and Localisation, 1 (2009): 94-121.
--: "Fan Translation Networks: An Accidental Translator Training Environment?". Translator and Interpreter Training: Methods and Debates. Ed. John Kearns. London: Continuum, 2008. 158-183.
--: The Coming Industry of Teletranslation. Clevedon: Multilingual Matters, 1996.
O'Hagan, Minako, and David Ashworth. Translation-Mediated Communication in a Digital World: Facing the Challenges of Globalization and Localization. Clevedon: Multilingual Matters, 2002.
Pedersen, Jan. Subtitling Norms for Television. Amsterdam: John Benjamins, 2011.
Perez Gonzalez, Luis. "Fansubbing Anime: Insights into the Butterfly Effect of Globalisation on Audiovisual Translation". Perspectives, 14.4 (2006): 260-277.
Perrino, Saverio. "User-generated Translation: The Future of Translation in a Web 2.0 Environment". The Journal of Specialised Translation, 12 (2009): 55-78. <http://www.jos trans.org/issue12/art perrino.pdf> [17 March 2011].
Pym, Anthony. "Localization and the Dehumanization of Discourse". The Linguist, 41.6 (2003): 168-170.
--: The Moving Text: Localization, Translation and Distribution. Amsterdam: John Ben jamins, 2004.
--: "Translation Technology as Rupture in the Philosophy of Dialogue". Translation Technologies and Culture. Ed. Ian Kemble. Portsmouth: University of Portsmouth, 2007.1-9.
Ray, Roger. Crowdsourcing: Crowd Wants to Help You Reach New Markets. Romainmotier: Localization Industry Standards Association, 2009.
Schaffner, Christina, ed. Translation in the Global Village. Clevedon: Multilingual Matters, 2000.
Singh, Nitish, and Arun Pereira. The Culturally Customized Website: Customizing Websites for the Global Marketplace. Burlington: Elsevier Butterworth-Heinemann, 2005.
Snell-Hornby, Mary. The Turns of Translation Studies: New Paradigm or Shifting Viewpoints? Amsterdam: John Benjamins, 2006.
Surowiecki, James. The Wisdom of Crowds. Why the Many Are Smarter than the Few and how Collective Wisdom Shapes Business, Economies, Societies and Nations. New York: Doubleday, 2008.
Talavan, Noa. "Using Subtitles to Enhance Foreign Language Learning". Porta Linguarum, 6 (2006): 41-52.
Yunker, John. Beyond Borders: Web Globalization Strategies. Indiana: New Riders, 2002.
Zaidan, Omar F. and Chris Callison-Burch. "Crowdsourcing Translation: Professional Quality from Non-Professionals". Proceedings of ACL 2011, Portland, Oregon: 1220-1229.
Articulo recibido: 18/03/2011
Articulo aprobado: 14/12/2011
Alberto FERNANDEZ COSTALES
University of Oviedo
(1) This research is supported by the grant from the Spanish Ministry of Science and Innovation FFI200908027, Subtitling for the Deaf and Hard of Hearing and Audio Description: Objective Tests and Future Plans, and also by the Catalan Government funds 2009SGR700.
(2) The type of software used by fansubbers has not yet been analysed and no empirical data can support this statement. However, in many cases fans rely on free software (such as Sub Station Alpha or Virtual Dub) instead professional subtitling applications. Indeed, Open Source Software is one of the first scenarios in which amateur translation started to develop.
(3) Research can be conducted in this field, as the translation of materials can be effectively used as part of language and translation courses to improve students' linguistic competence. To this end, some papers have been published on the use of subtitling techniques in translation classes and in language learning environments (Diaz-Cintas 1995, Talavan 2006).
(4) See the official website at: http://www.therosettafoundation.org. Accessed on November 20th 2011.
(5) Source: http://www.crowdsourcing.com/. Accessed on 16th March 2011.
(6) Source: www.facebook.com/translations. Accessed on 17th March 2011.
(7) Indeed, it can be argued that many amateur translators are not complying with the Berne Convention for the Protection of Literary and Artistic Works, governing copyright laws: http://www.wipo.int/treaties/en/ip/berne/index.html. Accessed on 18th March 2011.
(8) Source: www.internetworldstats.com. This website provides updated statistics on the top ten languages on the Web. Accessed on 18th March 2011.
Caption: Figure 1: Screenshot with the percentage of subtitles submitted by the different fansubbers of an episode of the American medical drama House (www.subtitulos.es).
Caption: Figure 2: Statistics on the number of translations of Facebook content into Spanish.
|Printer friendly Cite/link Email Feedback|
|Author:||Costales Fernandez, Alberto|
|Date:||Jan 1, 2013|
|Previous Article:||ANALISIS CONTRASTIVO DE LA NEGACION GRAMATICAL EN ESPANOL Y EN AGNI MOROFUE.|
|Next Article:||Interpretes en el peregrinaje de Xuanzang en el Asia del siglo diecisiete.|