Where do Borders Lie in Translated Literature? The Case of the Changing English-language Market

Translated literature is a growing industry in the UK. If the figures are to be believed (“First Research on the Sales of Translated Fiction in the UK”), then translated fiction is becoming more and more popular amongst British readers. In some quarters this has been attributed to the high quality of the translations produced, as well as some trailblazing texts (and TV and film) that have opened audiences’ eyes to what is available from beyond English-speaking territories. Yet as the industry grows, its traditional methods are being challenged. In this article I shall focus on the increasing tendency for agents and other figures on the source side to commission their own translations before translation rights have been sold, and determine reasons for this. In particular I shall focus on how this affects the notion of borders in translated literature, where these lie in the industry, and the figures that function as gatekeepers.

Translated literature is a growing industry in the UK.If the figures are to be believed ("First Research on the Sales of Translated Fiction in the UK"), then translated fiction is becoming more and more popular amongst British readers.In some quarters this has been attributed to the high quality of the translations produced, as well as some trailblazing texts (and TV and film) that have opened audiences' eyes to what is available from beyond English-speaking territories. 1et as the industry grows, its traditional methods are being challenged.In this article I shall focus on the increasing tendency for agents and other figures on the source side to commission their own translations before translation rights have been sold, and determine reasons for this.In particular I shall focus on how this affects the notion of borders in translated literature, where these lie in the industry, and the figures that function as gatekeepers.
Publishing Trendsetter sets out the "Life cycle of a book in translation", and in its standard model the foreign publisher "buys translation rights for a title and works with a translator to develop a new version to publish, market, and sell in the given territory" ("Life Cycle of a Book in Translation -Publishing Trendsetter").In this traditional scenario, the translator is part of the lifecycle once the text has arrived in the target locale-and furthermore, the translator is an important agent in deciding the brief for the new translation, not simply a pen for hire.Likewise, Deirdre Smerillo in her recommended contract for translation rights suggests that the translation is the responsibility of the target publisher, and that " [e]xcept in the most extraordinary circumstances, the foreign [i.e.target] publisher should be granted discretion in these matters" (Smerillo 289).However, in recent years the translation industry has seen a new model challenge this traditional way of thinking; rights holders of the source text are increasingly commissioning full-length translations of their works before the rights have been sold to foreign publishers-a phenomenon I call here source-commissioned translations.Sometimes this (mostly English) translation is used to whet publishers' appetites, but often, when dealing with English-language publishers, the translation itself is sold as part of the deal.This clearly disrupts the chain of production for translated literature, and raises questions as to where borders lie, and when and how they are crossed.I shall use Anthony Pym's notion of borders as locations where transfer takes place (Pym,Method in Translation History 93) as well as Chris Rumford's notion of networked borders (Rumford 157) to argue that these changing practices disrupt the borders in translation, creating a segmentation of the agents involved in the circulation of literature.This disruption in This work is licensed under a Creative Commons Attribution 3.0 License turn has serious consequences for the quality of translations, as well as the role of the translator in the chain of production.
Where are Borders?Pym suggests that when we envisage borders in translation history, we should scrap "the underlying cartographic assumption of a world divided into nations" (Pym,Method in Translation History 93).Rather, we should look "where discontinuous movements become especially intense, where transfers have been achieved thanks to points of change or changeover (yes, points of translation)" (Pym,Method in Translation History 93).So, in at least one respect, borders are crossed when (and where) translation takes place.This means that traditional maps cannot represent the reality of translation networks.For example, in the case of a translation from Spanish to French, we cannot count the Pyrenees as the border between Spanish and French literatures, since that is not where the exchange takes place.Nor can we draw a line on a map between Madrid or Barcelona and Paris to represent a relationship between Spanish and French literatures, since a single translation does not indicate any sort of lasting relationship.Rather, to map a translation network, we must map "the transfers that should give a more or less adequate representation of some underlying but unseen network" (Pym,Method in Translation History 92).Thus geographical location is important in recreating networks and defining borders; we need to determine where transfers take place (that is, where was the translator, where was the editor, and so on).Equally, we need to know whether a transfer is a one-off, or whether there is frequent activity that suggests a regular and lasting relationship.So, the dimension of time also needs to be added to such a map (Pym,Method in Translation History 93), and from here we can begin to judge the importance of such links, and the strength of networks.Neat, single lines on a map are out, since "cultures rarely correspond to this assumption" (Pym,Method in Translation History 93).Such a map would highlight points of greater and lesser activity, and here Pym comes to the conclusion that it is cities that function as borders, phenomena "of the air that freed the peasant, of the rapid circulation of texts and people, of nonsedentary culture.And so we find translators working in cities, or in the telecommunicational extensions of cities" (Method in Translation History 105).This understanding of cities as borders populated by translators relies on one assumption; that a cultural product enters in one form, is altered by the agent of change in the city (the translator, in Pym's case) so that a new form can be distributed in and from the city.Traditionally, in the UK translated literature market, we assume that this point of changeover is London, home to so many publishers, organizations and events.However, if English translations are no longer commissioned from London, but rather Stockholm, Barcelona or Beijing then London ceases to be a border from this point of view of translation.
From a sociological viewpoint, Rumford states that "[w]e have become accustomed to a world where borders wax and wane, as it were, and the important borders in our lives do not remain fixed" (Rumford 156) and that rather than think of either a "borderless world" or "rebordering", an alternative is "thinking in terms of the existence of 'networked borders"' (Rumford 157) where international networks can function as borders.He uses the example of the UK's arrangements to perform border checks in Lille, Calais, Paris and Brussels; we are not in UK territory when we have gone through those checks, but we have effectively crossed the border.Likewise, when we need to provide identification for services in our own towns and cities, we are effectively undergoing border checks, "where border control takes place at different points in society not simply at the territorial limits" (Rumford 158).This ensures that we are entitled to a particular service, or that we are indeed allowed to be in a particular area (and to perform certain activities, such as work, in that area).This raises the important question of who performs these border controls and, in our case, who allows literature to pass and where translation takes place.

The Existence of Source-commissioned Translations
Full-length manuscript translations, to publicize a work, have been carried out for years.In 1993, Maarten Asscher talked about the problems for smaller languages using the example of the young Dutch writer Marcel Möring: [H]is novels have to be translated in manuscript first, partly or wholly, before they can even be judged properly by an editor abroad.This is perfectly possible of course (in fact, a sample translation of his second novel has been made), but it takes time and it costs money, and there are only a handful of good literary translators from Dutch into English.That a translation is necessary for a work to be "judged properly" follows on from the fact that the majority of UK and US editors are monolingual (Ban 163) and so cannot compare the translation to the source in terms of judging the quality of the translation, nor can they read books in the source language to judge their quality (Marcus 54).These translations are not only used to try to sell the book to English-language publishers, either; English translations are frequently used to make a book "more accessible to editors around the world" (Blake 136).A glance at the rights catalogues of agents gives plenty of evidence of the existence of sourcecommissioned translations aimed at a worldwide audience.Frequently agents will list the reading material that is available for a given title, and in many cases a sample English translation (or synopsis) is a selling point, as is an existing published English translation.2 The Barcelona-based agency Pontas has six titles that mention their published English translation (although not all of the other available languages) amongst reading material in their December 2016 list as well as another five titles with English samples ("Pontas Literary & Film Agency.Rights List December 2016").They also have one title with an English translation available despite the English rights not having been sold (the title is Het Stempel by the Dutch writer Judith Uyterlinde).This is by no means an isolated case; in its Autumn 2016 rights list the Swedish agency Bonnier Rights also has a full English translation available for Den skeva platsen by the Swedish/Spanish author Caterina Pascual Söderbaum ("Bonnier Rights Autumn 2016.Rights Guide, Fiction & Narrative Nonfiction"), and in Spring 2015 it offered a full "unedited" English translation of Clinch by Martin Holmén ("Bonnier Rights Spring 2015.Rights Guide, Fiction & Narrative Nonfiction"), which went on to be bought by Pushkin Press and was published in English in May 2016.Finland's Elina Ahlback Literary Agency appears to use this tactic extensively; in their rights catalogues for the Frankfurt Book Fair, they have five young adult and children's titles with unpublished English translations ("Elina Ahlback Literary Agency.Frankfurt Rights Guide 2016, Children's and Young Adult.") and their lead title in fiction, Hammurabin enkelit by Max Seeck is offered with an English translation described as a "manuscript" ("Elina Ahlback Literary Agency.Frankfurt Rights Guide 2016, Fiction.").Interestingly, the book was sold to the German house Blanvalet in October 2016 at auction (Hoch), although not yet to an English house.So, source-commissioned translations are part of the marketing machinery of selling rights abroad.

A Fragmentation of the Border: How Editors Lose Gatekeeping Control and Translators are Side-lined
So why do such translations affect borders?This is partly to do with where gatekeeping power lies, which for Franssen and Kuipers is with editors, not translators: Editors are the main gatekeepers in the acquisition process and the only ones involved in all stages of the decision-making.Their boundary-spanning function-between managerial and creative branches in their publishing house, and between the house and transnational field-gives them a great deal of autonomy.(Franssen and Kuipers 56) Clearly translators have an important role to play at least in crossing linguistic borders.However, Franssen and Kuipers focus on the decision-making process in acquiring translation rights (and so how a book actually comes to be translated, and that translation published), and this underlines that translators are just one of the many agents involved in the chain of production, the same as linguistic borders are only one type of border in the circulation of global literature.Franssen and Kuipers analyse the Dutch market and determine that the main problems facing acquisition editors are "an excess of new titles; uncertainty over the nature and quality of new titles; and strong competition" for translation rights (Franssen and Kuipers 70), and that the same steps are followed by all publishers to approach these challenges and acquire the rights for translation. 3Firstly, they need to select from everything that is out there (Franssen and Kuipers 57).Frequently this work is outsourced to agents, friends in the industry and informal networks such as foreign editors, translators and literary critics, and also literary scouts, which causes a certain level of globalization: "The organizational innovation of hiring scouts-intended to tackle uncertainty, abundance and competition in the Dutch literary field-has directly affected literary output, leading to an increasing orientation towards the global centers, especially New York" (Franssen and Kuipers 59). 4 Secondly, they need to position a manuscript (Franssen and Kuipers 59) -essentially, they need to decide what to read themselves.These first two stages deal with the problem of excess supply.The third stage is to read the manuscript (Franssen and Kuipers 60), and this stage deals with evaluating the aesthetic (literary) quality of the work.Fourthly, they need to "sell" the work to others at the editorial board meeting (Franssen and Kuipers 60), which combines an evaluation of the aesthetic quality of the work with the business case, including (importantly) whether or not the work fits the publisher's list.Finally, they need to purchase or compete for the translation rights themselves (Franssen and Kuipers 66), which This work is licensed under a Creative Commons Attribution 3.0 License often includes selling the target publisher to the rights holder, especially when there are other interested parties.
In a "traditional" chain of production, translation normally takes place after the final step above -once the publisher has acquired the rights, and then got the best translator for the job.This is backed up by Michal Shavit when she was editorial director at Harvill Secker: Once a book is acquired, I then have to think of who I'd like to translate it.Every author is different as is every translator.Matching translators with books is an intuitive business a lot of the time on the part of the editor, but it's also dependent on whether or not a translator likes the book and feel that they can work well with the text.(Shavit 49) Note that this traditional model also includes translators as part of the informal network helping to narrow down the excess supply at stage one.However, with source-commissioned translations, translation takes place at or before the first step outlined above, to try to give the work an advantage over the excess supply.This stage is where editorial decision-making is traditionally decentralized-and with source-commissioned translations, the act of translation is decentralized from the editorial process too, and the close relationship between (target) editor/publisher and translator is diluted (since there may be contact later, after the translation has been acquired) or erased entirely.This is not a case of outsourcing work, as with dealing with excess supply above.Rather, the editor has no control over the process, and it is akin to the scenario voiced by Rønning and Slaatta where there were once fears the book trade would be an industry like any other and "Schiffrin's worst fears would have come true: a decline in the role of the editor and outsourcing of the publisher's spiritual centre-the editorial desk" (Rønning and Slaatta 2).This has serious consequences for quality, since it is normally the editor who assesses the quality of the translated work, and the working relationship between translator and editor is of great importance.The translator Kate Ashton feels this strongly: More than 35 years working in the field have convinced me of the superlative skills of editors in judging quality of writing and translation.The most elevated academician is no better qualified to assess the sensibility and skill of a literary translator than an experienced, working commissioning or desk editor.(Ashton 66) Source-commissioned translations thus diminish the editor's role as a gatekeeper, as defined by Franssen and Kuipers above.When editors are not involved in the translation of a work, they no longer span the border between languages; rather, this part of the border is pushed back to the 'source'5 side of the chain of production, and is managed only by the translator, not the translator and editor together.Furthermore, linguistic borders are removed from the borders between national literary fields.Applying Pym's notion of translation and localization as forms of distribution, in such cases London is not a centre of translation but rather determines what is distributed.So, from London there is a position of "non-localization" (Pym,The Moving Text 26).It is important to remember that within paradigms of localization, translation is seen merely as a linguistic problem, which demeans the position of the translator as an expert across various stages of the chain of production.We could look at this phenomenon positively, and hope that it could help foreign literature to enter the English-language market, getting around the problem that the UK/US market translates little (and that publishers rarely have a range of linguistic ability in-house to be able to read manuscripts in the source language).Pragmatic as this may be, however, it throws up a whole series of problems for the position of the translator, which is especially serious when we consider that translators are frequently not awarded the position and recognition they deserve.The key problems in this new approach is that translators are further removed from the ultimate distribution of the translated work, and are working at a point where the status of the translated work in terms of distribution is unknown: there are no guarantees it will be published at all, and if so, where and by whom.

Is English Dominant?
One possible hypothesis for the existence of source-commissioned English translations is the anecdotal claim that English is the dominant language in translated literature around the world.This does seem to be supported by evidence.Rønning and Slaatta report that: In the 12 months between April 2008 and March 2009, 7 of the 40 bestselling titles were identical in four or more Western European countries.Fourteen of these were originally written in English, indicating that the search for global bestsellers among international publishers currently favours books in English.(Rønning and Slaatta 7) This position of English as a dominant source language (SL) in the translated literature market is also reported by Sapiro, who says that one third of translations in large French houses come from English (Sapiro, "Translation and the Field of Publishing" 162), and, following Heilbron, that in the 1990s 59% of all translated literature in the world came from English (Sapiro, "How Do Literary Works Cross Borders (or Not)?" 87).This gives English a "hyper-central position" (Sapiro, "Translation and the Field of Publishing" 158) in the global literary field, placing English firmly at the dominant core.In one respect this justifies the view that all languages apart from English are in fact "small literatures" when it comes to the global literary field: Franssen and Kuipers, in their study of translation in the Dutch literary field, class translation from any language other than English as "smaller niches" (Franssen and Kuipers 67). 6s well as being a dominant source language in translated literature around the world, English is a key language of communication in the industry, too.Bourdieu states that: Foreign literature agents representing the world's major houses on the international scene create vast information-sharing networks where "everyone speaks English" (says the same literary director), where everyone knows and helps each other with the quasimagical phrase: "Hi! X, I have a book Y that is absolutely the book for you."(Bourdieu 150) In this regard publishing is like many other fields where English has become the de facto lingua franca (Pym, "Globalization and the Politics of Translation Studies" 746). 7However, it must be remembered that in the global literary field, products must be translated for them to travel beyond their national literary field.Since English is the language of communication of agents (in the widest sense of the word), then clearly it is an advantage for the product to be translated into English at an early stage, too, since then it can be read by all of those agents,8 and the transnational circulation of books (as far as the target side is concerned) begins to resemble other cultural products that do not require translation, such as art and music (Ginsburgh,Weber,and Weyers 228).
So, English is in a dominant position as a source language for much of the world's translated literary content, and also as the language of communication for the global publishing industry.In this situation, it makes perfect sense to have an English translation of texts from other languages, so that it can circulate more widely and more freely.However, this does not mean that publication in English will be the ultimate goal of such an approach, if only because the UK and US notoriously publish few translations.In terms of percentages, the mythical and widely quoted figure is that they make up 3% of books published in the UK and US.Venuti claims that the figure has hovered between 2 and 4% since the 1950s, being as low as 1.4% in the UK in 2001, compared to between 10 and 25% for France, Italy and Germany (Venuti 11).Literature Across Frontiers backs up the 3% figure for all publications, but says in the case of translated literature the figure is consistently above 4% in the UK and Ireland, reaching 5.23% in 2011 (Büchler and Trentacosti 5).This is still far below the figures for other Western European literatures, and anecdotally the feeling from abroad is that English literatures are not interested in translations: "'The English buy nothing except American products.As for Americans, they are only interested in themselves, that's all,' says the female literary director of a large house" (Bourdieu 151).All of this challenges commonly held notions about translated literature and the role of English-and whether translation into English is indeed the gateway to success in other languages.The former director of the Catalan cultural funding body the Institut Ramon Llull says that "One of the main challenges faced by European literatures is translation into English" (Bargalló and Allen 11); in the same report, Allen and Torner state that "Today, it is the person who does not speak English who risks exclusion-not merely social exclusion but exclusion from the ability to survive in the global economy" (Allen and Torner 13).Yet does this mean that English as a language, and the UK and US as its principal publishing centres, are at the centre of global literary movement?That London and New York are the key frontier cities, following Pym's concept of borders?
This hypothesis relies on the English-language market (and the US/UK market as its primary exponent) being dominant, more successful than all others and exerting an influence on all others.However, as well as the aforementioned proportions of translations published in the UK and US, we also need to bear in mind the well-documented aversion of US/UK publishers to translation, and so this initial hypothesis has an immediate problem. 9This is exacerbated by the fact that UK and US markets are known for being reactive in terms of translations, wanting to see (and know) what works have done in other markets before committing to buying rights.So, although English is the lingua franca of publishing, that does not mean that any participants in this dialogue have to be UK or US based, or in any other English-speaking territory for that matter.Globalization has not necessarily led to the centralization that many feared-writing in 1999, Bourdieu stated that indications pointed to publishing "evolving into a sector of mass production (of profits) like any other" (Bourdieu 146).With hindsight, Rønning and Slaatta recall 2001 when "book publishers were on the brink of adopting the same commercial logic that had previously spread through other types of media" (Rønning and Slaatta 1).Even in 2016, Sapiro offers a damning verdict: "In the era of globalization, the publishing industry has been increasingly dominated by large conglomerates that impose fierce criteria of commercial profitability and operation to the detriment of literary and intellectual criteria" (Sapiro, "How Do Literary Works Cross Borders (or Not)?" 87).Yet Rønning and Slaatta's analysis concludes that although this was the fear at the turn of the millennium, it has not materialized: "The publishing industry no longer bases its business models on the idea of integration and synergy," but rather "[w]ithin the framework of corporations, publishers function more like independent publishing units" (Rønning and Slaatta 2).One of the reasons for this relative autonomy is that books "represent a certain symbolic capital and status for media companies involved in publishing" (Rønning and Slaatta 3) and for this reason "publishers are allowed to operate with lower profit margins than elsewhere in the company" (Rønning and Slaatta 4). 10ranssen and Kuipers go a step further, claiming that the "increasing segmentation of the of the field is propelled by growing globalization of the literary world" (Franssen and Kuipers 67).Markets are not becoming increasingly centralized and focusing on best-sellers, since quality is paradoxically cheap, although it does not necessarily sell: According to one literary agent, "the cheap*, crowd-pleasing books are very expensive because everyone in the world wants them, because they are worldwide successes.On the other hand, you have good novels and very good authors who are unlikely to sell in large numbers, but who offer quality.And these books are often taken on by smaller publishers who accept them because of their quality.[...] The financial, economic or commercial side of things is not their priority."(Bourdieu 138) 11   This means that there is "a growing trend towards niches and specialization, both inside and outside major publishing groups" (Rønning and Slaatta 2), with a healthy level of communication between different national literary fields.Franssen and Kuipers discovered that for the Dutch literary field, informants from foreign literary fields are "especially important" (Franssen and Kuipers 57), and chief amongst these are editorial counterparts in other countries.With this cross-border communication, they find that as well as the movement of products, "the structure of the field itself-including entire catalogues and taste repertoires-is becoming increasingly similar between nations" (Franssen and Kuipers 69).What is more, given the nature of this communication, traditional centres such as New York or London are bypassed, and instead networks directly link fields on the periphery.This is where Rumford's notion of a network This work is licensed under a Creative Commons Attribution 3.0 License border is more applicable than Pym's notion of cities as borders-it is the networks of agents (in the widest sense) across different countries and cities that form the border into a particular national field, with gatekeeping control not held by a single editor, but by the various players who form the networks of which editors are just one part.

Relay and Indirect Translation
In this scenario, source-commissioned translations begin to take on a mediatory role, as outlined by Ringmar ten years ago: [I]n today's world, translations into English tend to be (relatively) few and marginalised in the Anglo-American cultural context, but at the same time English translations are used, more often than other translations, as mediators in ITr [indirect translation].And even in cases of direct translation previous translations into English often serve as indicators that this particular work/writer is worth translating.(Ringmar 12) Different scholars use different terms for the phenomenon of when a translation is carried out from a translation rather than the "original" source.In his criticism of the "widespread and growing" practice, the translator Don Bartlett calls all such cases "relay translation," although his emphasis is on when this is done without the translator's permission, and his examples are from when published English translations are used (Bartlett 60).Dollerup distinguishes between relay translation (where there is an audience for the intermediary translation) and indirect translation (where there is not) (Dollerup, '"Relay' and 'support' translations" 2000), taking his cue from conference interpreting practices.Although he believes its use is rare in translated literature, he relates one such case: The second example derives from an experienced Danish translator of children's books (Mette Jørgensen).She was asked to do a translation of a Dutch children's book and received an English text from the publisher.She found this translation too tame for the illustrations and concluded that the English text had been made only in order to promote sales of the book with publishers.(Dollerup,"Relay in Translation" 8) Dollerup also follows Dinda Gorlée in that "relayed translations will be published with 'exotic language combinations' when no translator can undertake a direct translation" (Dollerup,"Relay in Translation" 5). 12 In his study of the reasons for relay and indirect translation, Washbourne also mentions that this might be because of a lack of translators in a particular language pair, or the prestige of the intermediary language, amongst others (Washbourne 611).However, he does not mention that relay and indirect translation might arise from English as a cheap(er) and (more) accessible lingua franca of the translation business. 13Interestingly, Washbourne also asks "whether [intermediary translations] are written in such a way as to facilitate translation; in other This work is licensed under a Creative Commons Attribution 3.0 License words, a particular kind of writing for translation: translating for translation" (Washbourne 615), raising a serious concern.Indeed, Dollerup's differentiation is interesting in considering how relay and indirect translation are used in translated literature.In the situation that a translation is commissioned by the source author's agent, and this translation is to sell the work to target-language publishers (and the translator is paid a one-off fee), it is entirely conceivable that such a translation will differ from a translation commissioned by a target-language publisher.In this case of indirect translation, according to Dollerup's terminology, the ultimate audience of the translation will be (to use Holz-Manttari's terminology) a user of texts (editors), and not a receiver of texts (the book-buying public).Would the target-language editor want a raw version of the features of the text, or a polished final product?Indeed, would there be a polished final product without editorial input?It is interesting that Dollerup uses the term relay to highlight "the dynamics in the interlingual movement of translation" (Dollerup, '"Relay' and 'support' translations"), whereas the editorial dynamics of the translation process are lost in indirect translation.This latter term focusses on the static product of the process itself, taking attention away from the process and the translator.
Another question raised by Bartlett concerns the quality of relay translations: For editors, it [relay translation] means that there could well be distortions of the original text.Editors and translators adapt a text to suit a certain market and then this text is further adapted to suit another market.Who knows, with the lack of consultation, what can have happened in between?(Bartlett 61) Note that this concern, though, is about a lack of communication along the chain of production, rather than an assumption that indirect and relay translation is bad because translation is not directly from the source.Indeed, Dollerup talks (although not in depth) about the fact that relay translation does not have to create linguistic errors, but rather that there are errors in how a translation is used, especially in the field of interpreting.This is backed up in a report on an experiment in relay translation between Dutch and Indonesian: [R]elay translation can be a blessing-that is if the first translator had taken the care to explain some cultural nuances that may be lost to readers from outside the country of origin.So instead of "lost in translation", we found more contextual explanation within the first translation.Of course this depends very much on the care and quality of the first translator.(Vitri Handayani 71) This is a refreshing take on an activity that has traditionally been a taboo in translated literature, and shows that with good communication along the chain of production, indirect translation can be a benefit.
It is interesting to consider how relay and indirect translation affects relationships between national literatures.Dollerup (Dollerup, "Relay in Translation") talks about relay translations as causing a delay in literary fame, since a work needs to become established in one territory before it can move on to the next.This time is reduced with a single intermediary language since the chain of communication is much shorter, and so again we can understand why the use of English as a lingua franca might grow.This in turn will cause isomorphism across national literary fields, as noticed by Franssen and Kuipers above, since Anglo-American tendencies will be transported to many other national literary fields.However, the use of English indirect translation and a network of agents on the periphery (as mentioned in Franssen and Kuipers 69) might actually lead to "faster" translational relations, there is no need for a work to be published in the UK or US and become established there.
These developments fuelled by globalization are not without their dangers, and it is in this regard that we must remember that translation is necessary for literature to circulate globally.Even Franssen and Kuipers' peripheral networks take editors as their centre-and the centre of these networks is where the boundaries between national literary fields lie (Franssen and Kuipers 50) as well as between individual writers (and translators) and their distribution.It is worth bearing in mind the three functions of publishers that Rønning and Slaatta identify through the German term Verlag, English publisher and Romance editeur, editore, editor, etc.: publishers serve to bridge private and collective economies, to disseminate work and make it available to the public, and they have the cultural function of improving quality .With sourcecommissioned translations, editors and translators are detached from each other, and so roles normally performed or guaranteed by the editor need to be filled.Who chooses the translator, and on what grounds?Who ensures quality during the translation process, as well as assessing the quality of the final translation?And importantly, what is the situation of the work regarding ownership and future distribution, when it is not even certain whether the work will be published or not?

Copyright
In the final question, the issue of copyright needs to be addressed, especially since indirect and relay translation is frequently not mentioned explicitly in translators' contracts.In her advice to translators, Kate Pool makes this assumption (Pool 62) and Don Bartlett recommends that contracts should include a clause about relay translation (Bartlett 61), although both note that existing copyright law makes it clear that unauthorized relay or indirect translation is a breach of copyright.However, both deal with the assumption that the intermediary translation in such a case has been published.What happens when the terms of distribution are not part of the original contract?
Copyright and the position of the translator regarding the translated work is a muddled area at the best of times.Yet in the "standard" model introduced at the beginning of this article, translators deal directly with editors, who represent the purchaser of the foreign language rights of the source text, and so translators are free to accept, reject or renegotiate whatever the editor offers in terms of rights and royalties. 14No work should be done until a contract is in place, and in as much as a translator is able to refuse work and earnings, he or she may choose not to take on the task at all.However, source-commissioned translations significantly alter the chain of production and the nature of negotiations.The translator's client is no longer the target-language editor, but rather the source-language author, editor, or agent.The translation takes place before 14 For example, the fifth of six "commandments of fair-play in literary translation" from the European Council of Literary Translators' Associations (CEATL) is as follows: "5.Share in profit.The translator shall receive a fair share of the profits from the exploitation of his/her work, in whatsoever form it may take, starting from the first copy" ("CEATL Hexalogue") the foreign-language rights have been purchased, so at the point of the translation taking place (and of the contract between the translator and the client), there is no certainty that the book will even be published.Then, when the translation is used as a sample for target-language publishers, there is no obligation on the target publisher to buy that translation-just the rights.Should the publisher buy the translation and rights as one, what happens if they want some changes made to the translation?Do they approach the original translator?Does the translator even get paid for amendments?And finally, does the translator retain copyright and receive any royalty payments for the translation, or is the original translation classed as work for hire with copyright assigned to the commissioner?Anecdotal evidence appears to confirm the more pessimistic end of this spectrum.At the Translators' Association event What Do Publishers Look for in a Book? on 11 th November 2015, Richard Milbank from Head of Zeus referred to their Swedish crime novel Victim Without a Face by Stefan Ahnhem, which they acquired, translation and rights together, on the basis of a completed translation provided by the source author's agent. 15This has received a series of positive reviews, and so could be seen as a success, yet he provided an opposing experience: a French book proposed by an American publisher, which seemed like an excellent idea.Head of Zeus took it on after reading one chapter of the translation, but it was only when the rest of the translation came through that Milbank perceived problems, and then went back to the Frenchand found out the problems were with the translation.Since the translation itself was part of the deal, they had to pull out of the contract.

Conclusion
It is clear that source-commissioned translations are already a significant phenomenon in the translated literature market, with consequences for the global circulation of literature.Such translations remove one of the significant borders between national literary fields, and the border that makes the circulation of literature different to the circulation of cultural products such as music and art-namely, the border between languages.In such cases translated literature takes on traits seen in paradigms of localization, since translation takes place before a work aims to cross into a different literary field, creating a form of internationalized text; however, without a clear destination, there is no clear localization of the literary text either, since the translator is not sure who the ultimate audience of the text is supposed to be.On the one hand this sees translators as purely involved in linguistic transfer rather than cultural transfer, as is also often the case in paradigms of localization.On the other, this erodes the position of editors as gatekeepers, since although they determine what can and cannot be distributed in their own field, they are not part of the initial process of translating the text for their field.
However, in terms of borders, in many respects literature can be seen to be following other globalized phenomena, including supra-national organizations themselves.With editors having less and less time to edit, and so more (traditionally) editorial decisions being devolved to other members of formal or informal networks, the border for any notion of an Englishlanguage literary field does not correspond to national borders, but nor does it correspond to the traditional publishing centres of London, New York, and a few others.Rather, the border is more complex, and is managed by a network of agents, who perform the literary equivalent of border checks wherever they are; and clearly other members of these networks, including acquisitions editors, need to rely on these decisions.Ultimately, without communication along the chain of production there are many dangers in this practice, which may fulfil the negative depictions of a globalized publishing industry devolved from editorial control, as voiced by Bourdieu and Sapiro.