{10} Translation contracts in the postwar period have in fact varied widely, partly because of the ambiguities in copyright law, but also because of other factors like changing book markets, a particular translator’s level of expertise, and the difficulty of a particular translation project. Nonetheless, general trends can be detected over the course of several decades, and they reveal publishers excluding the translator from any rights in the translation. Standard British contracts require the translator to make an outand-out assignment of the copyright to the publisher. In the United States, the most common contractual definition of the translated text has not been “original work of authorship,” but “work made for hire,” a category in American copyright law whereby “the employer or person for whom the work was prepared is considered the author […] and, unless the parties have expressly agreed otherwise in a written instrument signed by them, owns ail the rights comprised in the copyright” (17 US Code, sections 101, 201 (6)). Work-for-hire contracts alienate the translator from the product of his or her labor with remarkable finality. Here is the relevant clause in Columbia University Press’s standard contract for translators:
You and we agree that the work you will prepare has been specially ordered and commissioned by us, and is a work made for hire as such term is used and defined by the Copyright Act. Accordingly, we shall be considered the sole and exclusive owner throughout the world forever of all rights existing therein, free of claims by you or anyone claiming through you or on your behalf.
This work-for-hire contract embodies the ambiguity of the translator’s legal status by including another clause that implicitly recognizes the translator as an author, the creator of an “original” work: “You warrant that your work will be original and that it will not infringe upon the copyright or violate any right of any person or party whatsoever.”
Contracts that require translators to assign the copyright, or that define translations as works made for hire, are obviously exploitative in the division of earnings. Such translations are compensated by a flat fee per thousand English words, regardless of the potential income from the sale of books and subsidiary rights (e.g., a periodical publication, a license to a paperback publisher, an option by a film production company). An actual case will make {11} clear how this arrangement exploits translators. On 12 May 1965, the American translator Paul Blackburn entered into a work-forhire arrangement with Pantheon in which he received “$15.00 per thousand words” for his translation of End of the Game, a collection of short stories by the Argentine writer Julio Cortázar.[6] Blackburn received a total of $1200 for producing an English-language translation that filled 277 pages as a printed book; Cortázar received a $2000 advance against royalties, 7.5 percent of the list price for the first 5000 copies. The “poverty level” set by the Federal government in 1965 was an annual income of $1894 (for a male). Blackburn’s income as an editor was usually $8000, but to complete the translation he was forced to reduce his editorial work and seek a grant from arts agencies and private foundations— which he failed to receive. Ultimately, he requested an extension of the delivery date for the translation from roughly a year to sixteen months (the contracted date of 1 June 1966 was later changed to 1 October 1966).
Blackburn’s difficult situation has been faced by most freelance English-language translators throughout the postwar period: below-subsistence fees force them either to translate sporadically, while working at other jobs (typically editing, writing, teaching), or to undertake multiple translation projects simultaneously, the number of which is determined by the book market and sheer physical limitations. By 1969, the fee for work-for-hire translations increased to $20 per thousand words, making Blackburn’s Cortázar project worth $1600, while the poverty level was set at $1974; by 1979, the going rate was $30 and Blackburn would have made $2400, while the poverty level was $3689.[7] According to a 1990 survey conducted by the PEN American Center and limited to the responses of nineteen publishers, 75 percent of the translations surveyed were contracted on a work-for-hire basis, with fees ranging from $40 to $90 per thousand words (Keeley 1990:10–12; A Handbook for Literary Translators 1991:5–6). A recent estimate puts the translation cost of a 300-page novel between $3000 and $6000 (Marcus 1990:13–14; cf. Gardam 1990). The poverty level in 1989 was set at $5936 for a person under 65 years. Because this economic situation drives freelance translators to turn out several translations each year, it inevitably limits the literary invention and critical reflection applied to a project, while pitting translators against each other—often unwittingly—in the competition for projects and the negotiation of fees.
{12} Contracts since the 1980s show an increasing recognition of the translator’s crucial role in the production of the translation by referring to him or her as the “author” or “translator” and by copyrighting the text in the translator’s name. This redefinition has been accompanied by an improvement in financial terms, with experienced translators receiving an advance against royalties, usually a percentage of the list price or the net proceeds, as well as a portion of subsidiary rights sales. The 1990 PEN survey indicated that translators’ royalties were “in the area of 2 to 5 percent for hardcover and 1.5 to 2.5 percent for paperback” (Handbook 1991:5). But these are clearly small increments. While they signal a growing awareness of the translator’s authorship, they do not constitute a significant change in the economics of translation, and it remains difficult for a freelance translator to make a living solely from translating. A typical first printing for a literary translation published by a trade press is approximately 5000 copies (less for a university press), so that even with the trend toward contracts offering royalties, the translator is unlikely to see any income beyond the advance. Very few translations become bestsellers; very few are likely to be reprinted, whether in hardcover or paperback. And, perhaps most importantly, very few translations are published in English.
As Figures 1 and 2 indicate, British and American book production increased fourfold since the 1950s, but the number of translations remained roughly between 2 and 4 percent of the total—notwithstanding a marked surge during the early 1960s, when the number of translations ranged between 4 and 7 percent of the total.[8] In 1990, British publishers brought out 63,980 books, of which 1625 were translations (2.4 percent), while American publishers brought out 46,743 books, including 1380 translations (2.96 percent). Publishing practices in other countries have generally run in the opposite direction. Western European publishing also burgeoned over the past several decades, but translations have always amounted to a significant percentage of total book production, and this percentage has consistently been dominated by translations from English. The translation rate in France has varied between 8 and 12 percent of the total. In 1985, French publishers brought out 29,068 books, of which 2867 were translations (9.9 percent), 2051 from English (Frémy 1992). The translation rate in Italy has been higher. In 1989, Italian publishers brought out 33,893 books, of which 8602 were translations {13} {14} (25.4 percent), more than half from English (Lottman 1991:S5). The German publishing industry is somewhat larger than its British and American counterparts, and here too the translation rate is considerably higher. In 1990, German publishers brought out 61,015 books, of which 8716 were translations (14.4 percent), including about 5650 from English (Flad 1992:40). Since World War II, English has been the most translated language worldwide, but it isn’t much translated into, given the number of English-language books published annually (Table 1 provides the most recent data).
[6]
This account of Blackburn’s Cortázar project draws on documents in the Paul Blackburn Collection, Archive for New Poetry, Mandeville Department of Special Collections, University of California, San Diego: Letter to John Dimoff, National Translation Center, University of Texas,
Austin, 6 May 1965; Contract with Pantheon Books for the translation of
[7]
The 1969 translation rate is taken from the “manifesto” that concludes the proceedings from the landmark PEN conference held in 1970 (
[8]
British statistics are drawn from