In August 1929, the New York-based Hispanic journal Alhambra published two English translations of poems from Lorca’s Gypsy Ballads, ‘Ballad of Preciosa and the Wind’ (‘Preciosa y el aire’) and ‘Ballad of the Black Sorrow’ (‘Romance de la pena negra’), in an edition published to coincide with Lorca’s visit to Manhattan. The translations appeared on page 25 of volume 3 of this ephemeral publication, accompanied by two photos provided by Lorca of his stay in Cadaqués with his friend and putative lover, Salvador Dalí. This edition of Alhambra also included an article about Lorca signed by one ‘Daniel Solana’, along with three more photographs provided by the poet himself: one in Granada, one more at the beach in Cadaqués, and a third in which he appeared ‘writing a manifesto with Dalí’. According to Andrew A. Anderson and Christopher Maurer (2013: 183), and following the suggestion of Andrés Soria Olmedo, the name ‘Daniel Solana’ was the pseudonym of Lorca’s friend Gabriel García Maroto, who was responsible for the graphic design of Alhambra, and who had been born in the town of La Solana in La Mancha. If this was the case, the text would also have been translated from Spanish into English, as García Maroto, like Lorca, had a very limited command of English and no intention of learning it (see Maurer and Anderson 2013: 178). Just before the piece by Daniel Solana, this edition also included a text by Francis Dickie entitled ‘The Alhambra: Ancient Moorish Grandeur in the Light of Modern Eyes’ which, according to Evelyn Scaramella (2017: 432), served ‘as if to establish a context for Lorca’s work’, one in which Dickie ‘waxes poetic about Granada’s Moorish past’ and ‘highlights the continued allure of Andalusia, its “strange” Oriental past, the “queer mixture of races” that dazzled Irving and the Romantics a century before’.
Alhambra was founded and funded by the Hispano and American Alliance to promote Spanish culture in New York and only four editions were published between June 1929 and January 1930, when it disappeared in the aftermath of the Wall Street Crash. During this brief existence, the journal also offered the first English translations of texts by Gerardo Diego, Ramón de Basterra, José Moreno Villa, and Edgar Neville. The Puerto Rican translator and literary critic Ángel Flores (1902-1992) was chosen to run the Hispano and American Alliance by its patron, Charles Jean Drossner, a Hispanophile millionaire and Spanish art collector, and was also placed in charge of one of the main projects of this newly created Hispanic institution in New York, the literary journal Alhambra.1 This is how Flores described the institution, its patron, and its house journal:
Yo dirigía en esa época una revista literaria que se llamaba Alhambra. Era una revista muy lujosa, bilingüe; el editor de la sección de arte era Gabriel García Maroto. La revista la patrocinaba un señor que resultó un pillo: no le pagaba a nadie. Este señor había querido reproducir el ambiente de la Alhambra en el piso onceavo de un edificio que se hallaba en la esquina de la calle 42 y la Quinta Avenida: importó mosaicos de España y mando construir una fuente. Quiso hacer una especie de Ateneo: había tertulias y ambiente español (Herrero-Senés 2019: 211).
According to Josefina Cornejo (2012), Drossner chose Ángel Flores to run the Hispano and American Alliance and the journal Alhambra as ‘le precedía fama de eficiente traductor y crítico, que se acrecentó en esos años con la traducción de The Waste Land, de T.S. Eliot, y la publicación de varios estudios literarios, por ejemplo, The Anatomy of Don Quixote’. She also categorically asserts that ‘para ella [Alhambra] tradujo también dos romances de Federico García Lorca -con quien había coincidido y trabado amistad en la sede de la alianza’. Flores was a notable Hispanist and translator, and throughout his life he played a very significant role in building bridges between English- and Spanish-speaking cultures. As a translator, perhaps his most notable feat in mediating between these two literary traditions was his aforementioned 1930 translation of T.S. Eliot’s The Waste Land (Tierra Baldía), which is usually considered to be the first Spanish version of Eliot’s totemic poem, although another version by Enrique Munguía appeared almost simultaneously in the Mexican journal Contemporáneos.2 However, Flores’s translation was the one that Lorca read in New York, and it most certainly exerted a notable influence on his own poetic reaction to the alienating modern metropolis in the form of Poet in New York. According to Howard T. Young:
Mientras Ángel Flores hacía su traducción iba mostrándosela a Federico García Lorca, visitante asiduo a la tertulia que hospedaba Flores en su pequeño despacho en la calle 42 en Nueva York y a la cual asistían Ángel del Río, Federico de Onís, y León Felipe. Así, tiene que ser Lorca uno de los primeros lectores de The Waste Land en español. (Young 1993: 277)
This translation was published by the Barcelona literary firm Cervantes, where Flores also collaborated as an advisor on American literature. Indeed, throughout his life Flores was an extremely prolific literary critic and translator, and in 1930 alone he published three books. The first was a critical biography of Lope de Vega written in English and entitled Lope de Vega: Monster of Nature.3 The second was an English translation of a book by Ramón Gómez de la Serna, which was originally entitled Cinelandia and in Flores’s translation became Movieland.4 The third one was an English translation of Miguel de Unamuno’s Tres novelas ejemplares y un prólogo. It is clear, therefore, that Flores’s credentials as a proficient Spanish-English literary translator were impeccable. Furthermore, during this same period Flores also published his literary criticism in prestigious anglophone publications such as The Bookman, the New York Herald Tribune and Books Abroad, the first academic journal in the USA devoted entirely to foreign literature. Flores eventually succeeded León Felipe as a Professor of Spanish at Cornell University, where he founded the Dragon Press, and co-translated Fuenteovejuna with Muriel Kittel. Along with Dudley Poore, he published the first anthology in English of Latin American fiction,5 and is also widely considered to be the first person to introduce English-speaking readers to the term ‘magical realism’. In total, Ángel Flores was responsible for more than 80 books, including studies of Latin American writers (Octavio Paz, Jorge Luis Borges, Pablo Neruda) and several Spanish literature manuals (such as An Anthology of Spanish Poetry from Garcilaso to García Lorca, published in 1961 by Doubleday), as well as various bilingual editions aimed at university students, which included texts ranging from the Middle Ages up to the contemporary period.6
Flores met Lorca during his time in New York through Federico de Onís and Ángel del Río, and later declared that he saw Lorca very frequently as the latter was a habitual visitor to his office at the Hispano and American Alliance, where the aforementioned tertulia was regularly held. Flores was also the instigator of a significant poetic encounter in New York when he introduced his friend Hart Crane to Lorca at a party held at the former’s home in Brooklyn.7 Regarding the probability of Ángel Flores being Lorca’s first English language translator, in a letter that Herschel Brickell wrote to Lorca from the Henry Holt Publishing Company in New York on 7 October 1929 (i.e., after the publication of the poems in August of that year), the former mentioned that Flores had confided in him that he was planning to translate some of Lorca’s poems and send them to Seward Collins of The Bookman, which Brickell describes as ‘la mejor de nuestras revistas literarias mensuales’ (Maurer and Anderson 2013: 69). He also states in the same letter that he has spoken about Lorca to Mr Collins and that he hopes the poems can be used. This would seem to be further proof that Flores was undoubtedly seen as a viable translator to render Lorca’s work into English. And in a letter to his family, provisionally dated by Maurer and Anderson on 22-23 October 1929 (2013: 75), Lorca made a further reference to these translation projects, stating that ‘Ahora empiezan a moverse algunos amigos míos ingleses aquí en New York para ver si consiguen que se ponga mi teatro aquí’.8
The fact that the bilingual Puerto Rican Ángel Flores was a translator not only of T.S. Eliot but also of Pablo Neruda would also suggest that he was a very likely candidate to have provided these two translations for the journal that he himself edited. Indeed, in a 1991 interview for the Mexican newspaper Proceso, just before his death the following year, Flores seemed to confirm that he was the first translator of Lorca into English, telling the journalist that ‘yo fui el responsable de la primera traducción al inglés de Lorca’ (Flores 1991), although one must admit that being the editor of the journal would also make him ‘responsable’ for the first translation, and this was the exact word he chose. If Flores was in fact the first translator, what remains a mystery is why he did not sign this text, given his personal involvement with the journal, and his proficiency and prolificacy as a English-Spanish translator in both directions. Perhaps the fact that the secondary text was signed with a bogus name (Daniel Solana) would indicate that Flores was happy to remain anonymous and share a private joke among friends.9
Despite the strong circumstantial evidence and the biographical coincidences that point to Ángel Flores as the author of those first English translations of Lorca, some notable lorquistas such as Andrew A. Anderson and Christopher Maurer have suggested that the translations may have been made by Lorca’s friend and perhaps also lover, Philip Cummings (1906-1991), who collaborated with Lorca in the summer of 1929 on the translation of Canciones during their shared vacation at Eden Mills in Vermont. This particular translation, which was not published until 1976, was made in this same month of August 1929, during the approximately ten days that Federico spent with Cummings after the conclusion of the Columbia University summer session. According to the editor of the 1976 edition, Daniel Eisenberg ‘The translation was made over the following week, with continuous discussion between Lorca and Cummings over the meaning of the Spanish and the appropriate English equivalents […] Although Federico’s English was limited - non-existent - he was able to assist Cummings by exegesis in Spanish, which he was under other circumstances always so reluctant to do, or by attention to the phonetics of the English and the syntax and vocabulary as described to him by Cummings’ (Eisenberg 1976: 13). In a letter to Eisenberg, Cummings also recalled that ‘we read all of these poems and argued as to their meanings which I gained very well. He could get the nuances of my translation even if he didn’t get the exact words […] he liked their sound’ (1976: 13-14).
As stated previously, this translation was not available until 1976, when it was published by the Duquesne University Press with the full title of Songs. Translated by Philip Cummings with the assistance of Federico Garcia Lorca. Perhaps the prime interest of this text lies in the fact that is the only documented instance of Lorca’s personal involvement in the translation of his own poetry into English, notwithstanding his collaboration with José Weissberger’s ill-fated version of Bodas de Sangre (Bitter Oleander) in 1935, or his participation in the translation of his Seis poemas galegos, which appeared in the same year. In his Introduction to this 1976 edition, Eisenberg explains in detail the vicissitudes that led to such a long gap between translation and publication, revealing that Cummings left instructions for the text to be destroyed if it had not appeared at the time of his death. This request is remarkably similar to the one that Lorca himself supposedly made to Cummings, and which the latter claimed to have subsequently complied with.10 Nevertheless, in all matters related to Cummings’s relationship with Lorca it is necessary to exercise extreme caution. According to subsequent declarations by Eisenberg, ‘Cummings is a liar who invents things that make him look important’ (2002).11
Lorca’s letters written to his family and friends during his stay in New York refer several times to his friendship with Cummings, whom he had first met at the Residencia de Estudiantes in Madrid in 1928, and who accompanied him on the first part of his trip to New York in 1929, sharing the first part of the journey from Madrid to Paris, where they both went their separate ways with a promise to meet in America later that summer. Lorca mentioned in a letter from New York to his family on 28 June 1929 (i.e., before the Alhambra publication) that Cummings had indeed translated two of his ‘romances gitanos’ into English the previous summer as a class project. It is clear, therefore, that when he arrived in New York in the summer of 1929 Lorca knew about these translations. But did he have a copy? Did Cummings give them to him during their shared part of the journey or when he visited him in Granada in 1928? This link is now almost impossible to establish as their correspondence has largely been lost, like that of so many of the letters to and from Lorca’s compromising homosexual friendships. One of the obvious reasons which might cast some doubt on Cumming’s potential role as Lorca’s first published English translator is that it is not at all clear what relation he could have had with the editorial team of Alhambra (Federico de Onís and Ángel del Río claimed then and for many years afterwards not to know who Cummings was when Lorca went to visit him that summer).12 It is of course possible that if Cummings did indeed give these translations to Lorca, the latter could have easily passed them on to the editorial team of Alhambra, with whom he met regularly in New York.
However, in the 1976 publication of Cummings translation of Canciones, whose main characteristics I examine below in greater detail, Daniel Eisenberg refers in his Introduction to the first two translations published in Alhambra in 1929, but makes no claims for Cummings’s authorship. It seems extremely unlikely that Cummings, who it seems was rather hurt by what he felt had been his exclusion from the official version of Lorca’s American experience, would neglect to mention to Eisenberg that those were his own translations if he knew they had been published. In fact, until his death in 1991, Cummings never claimed to have made these first published translations, despite his desire for protagonism in the Lorca American narrative, his tendency to embellish, exaggerate and even invent Lorcan anecdotes in order to achieve this goal, and the hurtful feeling that he had been written out of the story for spurious reasons.13 If the translations published in Alhambra were indeed those made the year before by Cummings, I believe that the only plausible explanation for his uncharacteristic reticence to stake his claim to fame in the Lorca US narrative is that he simply he did not know anything about this publication and that it was made without his consent.
Eisenberg (1976: 12-13) states that ‘in 1929 Federico’s poetry was all but unknown in English’, and that Cummings had initially suggested to Federico an English version of the Romancero Gitano, some of which had allegedly already been translated into French.14 According to the version which Cummings narrated to Eisenberg, ‘Federico challenged him instead to translate Canciones, the work which the French translators did not want, the work which, Lorca said, had much more of what was dear to him’ (1976: 13). At this point, it would be pertinent to ask ourselves why Lorca wanted this book to be translated into English and not the much more successful and celebrated Romancero Gitano. Perhaps the answer lies in the fact that, as is well known, Lorca was heartily sick of what he saw as the overly literal readings of the Gypsy Ballads and the predominant tendency to associate him with gypsy folklore and poetic traditionalism. Indeed, writing in Mexico in his 1957 series of articles on Lorca,15 his close friend, the theatre director Cipriano Rivas Cherif, recalled a conversation with Federico in which the latter confided in him that ‘Al Romancero Gitano y particularmente a “La Casada Infiel” les tenía ya verdadero asco. Por culpa de sus peones (sic) divulgadores’ (Inglada 2017: 580). Moreover, Rafael Martínez Nadal even suggested in his 1939 Introduction to Spender and Gili’s first English anthology of the poems that Lorca was ‘depressed’ by the success of his ‘Gypsy’ Book. Nevertheless, what is clear is that by the time he arrived in New York in June 1929 Lorca was keen to see his poetry translated into English and published in the USA. Indeed, according to Mildred Adams, ‘several times that year Federico, who had no skill in any tongue but his own, showed an insistent desire to have his work put into English’ (1977: 106) and she further opined that:
To a man as gregarious and vocal as he was, the frustration of being surrounded by people who could neither understand nor respond to his poems […] was repeated torture. What he wanted most of all was to be understood, admired, loved. These gifts his poems won for him in Spain. If they were to be withheld in English-speaking America for lack of understanding, then let someone turn his poems into English. That this was a task of great difficulty, requiring a poetic talent almost as great as his own, he hardly recognized. (Adams 1977: 106)
The 1929 Alhambra translations
Having examined the biographical and circumstantial evidence that could point to either Flores or Cummings being the first English translators of Lorca’s poetry, I will now present some of the most salient characteristics of these two versions with textual examples which I will then compare with the translational style evinced by Cummings in his 1929 version of Canciones and that of Flores in his 1946 translation of Pablo Neruda’s Residencia en la Tierra. Evelyn Scaramella, who has looked in some detail at the history of the journal Alhambra and its role in the early reception16 of Lorca in the USA, including these anonymous first English translations of the poems and the two candidates that have been proposed as their author, has stated that ‘a study of the translation styles of each of these intellectuals would be necessary to make a more conclusive determination about the identity of the translator’ (Scaramella 2017: 431). In order to try to address this request and establish who is more likely to have been the first translator, I will now examine some notable examples from these two versions of Lorca’s ballads in search of any significant lexical preferences or syntactic tendencies which might characterize the translator’s own idiosyncratic poetic register.
The initial and lasting impression created by a reading of the 1929 Alhambra translations is the powerful sensation that the translator did not entirely understand the original texts, which is simply not plausible in the case of the eminently bilingual Flores.17 Moreover, the article about Lorca signed by ‘Daniel Solana’ shows no signs whatsoever of having been translated by the same person. Instead, the style of this ironic biographical text is redolent of the prose of a native Spanish speaker with a strong command of English and a notable tendency to translate somewhat literally but not archaically, as is the dominant tendency in these translations of two of the Gypsy Ballads. Thus, in the aforementioned article we find references to how Lorca manages to ‘sow and harvest grace’, which is clearly a calque of a rather more natural though cultured original expression in Spanish (cosechar y recoger gracias). Moreover, it also seems extremely unlikely that this prose text would have been translated by anyone other than Ángel Flores, the editor of the journal who had been appointed to this position due in no small extent to his bilingual translational skills between Spanish and English.
In these 1929 translations, there are frequent bold departures from the Spanish source text in order to create a plausible metre in English, and a constant tendency towards archaic use.18 Whereas the aforementioned article by ‘Daniel Solana’ calls ‘el viento hombrón’ a ‘big wind’, in this translation of ‘Preciosa y el aire’ it is now ‘the man wind’, a clumsy, non-native, direct translation which points much more towards the hand of Cummings than that of Flores, and would suggest that the article and the poems were not translated by the same person. A simple reference to ‘cristales’ in ‘Preciosa y el aire’ is overtranslated as ‘crystals’ instead of ‘glass’, suggesting a poor acquaintance with the Spanish language and resort to what is essentially a false friend. The general impression created by a close reading of the two versions is that the translator shows a strong tendency towards archaic and unnatural syntax in English, presumably in an attempt to achieve a hyperpoetic register. Thus, the very first lines of the 1929 translation of ‘Preciosa y el aire’ offer us a very curious and unnatural overtranslation:
|‘Preciosa y el Aire’ (1928)||Alhambra (1929)|
|‘Su luna de pergamino /
||‘Playing her parchment moon /
It seems extremely hard to justify this archaizing addition of ‘the maid’ to the simple reference to ‘Preciosa’ in the eponymous ballad, unless the translator believed that syllabically this scanned better in English. The translator’s seemingly irresistible urge towards archaism is again evident in the translation of the following verses:
|‘Preciosa y el Aire’ (1928)||Alhambra (1929)|
|‘Niña, deja que levante /
||‘Maiden, let me discover /
Apart from the archaic overtranslation found once again in the use of ‘maiden’, it is worth asking ourselves why the translator does not opt for the direct equivalence of ‘vientre’ (which would surely be ‘womb’ as we are talking about a young girl) and instead renders this as ‘loins’, which is an overtly homoerotic and deliberate overtranslation. In this same poem, the drink offered to Preciosa by the English Consul is ‘una copa de ginebra’, appropriately enough for a British diplomat, but the 1929 translator transforms this offering rather bizarrely into ‘a wine-glass of absinthe’, and also fails quite spectacularly to achieve a dynamic equivalence for simple expressions:
|‘Preciosa y el Aire’ (1928)||Alhambra (1929)|
|‘el viento que nunca duerme’||‘the wind that sleeps not’|
In the translation of the ‘Romance de la pena negra’, we find that a simple exhortation to Soledad Montoya is also overtranslated in an extremely archaic manner, thus losing any kind of dynamic equivalence with the straightforward, demotic register of Lorca’s original Spanish verse:
|‘Romance de la pena negra’ (1928)||Alhambra (1929)|
|‘Soledad: lava tu cuerpo /
||‘Soledad, wash thy body
This tendency towards archaism and a lack of sensitivity to what are relatively simple expressions in an everyday register in the Spanish original are a constant presence in the 1929 Alhambra translation of these two Gypsy Ballads. And even for 1929, these two versions contain an abundance of frankly peculiar and utterly archaic syntax to translate simple and straightforward Spanish expressions:
|‘Romance de la pena negra’ (1928)||Alhambra (1929)|
|‘no me recuerdes el mar’||‘recall not the sea to me.’|
Another curious and unjustifiable modification is to be found when the poet addresses Soledad and says ‘Lloras zumo de limón’ and this becomes ‘We weep tears of lemon’, which would be a highly unlikely mistake for a bilingual translator to make and suggests a deficient knowledge of Spanish, a defect which is frequently evident in the two translations. Another basic lexical misreading can be seen in the reference to Soledad running from ‘la cocina a la alcoba’, which is translated as from ‘the kitchen to the chamber’, suggesting recourse to a limited bilingual dictionary rather than any real understanding of the original Spanish term ‘alcoba’. Nor does the translator appear to have understood Soledad’s lament that ‘Me estoy poniendo de azabache, carne y ropa’ as this is rendered quite literally and confusingly as ‘I am turning to jet, flesh and clothes’.19
Before venturing my opinion as to who I believe was the first English translator of Lorca, I will now offer some salient examples of the translational style of both candidates as evinced by their other Spanish-English translations of poetry, specifically Cummings’s 1929 translation of Lorca’s Canciones (Songs) and Flores’s 1946 translation of Pablo Neruda’s Residencia en la Tierra (Residence on Earth).20
The first and immediately apparent characteristic of Cummings’s translational style is a very unusual and even tortuous English syntax to render quite straightforward phrases from the Spanish source text. As a Spanish-English translator, Cummings tends notably towards unnatural and frankly unconvincing English expressions, even starting with the titles of the poems:
|‘Canción del día que se va’ (1927)||‘Song of the day which wanes’ (1929)|
|‘Qué trabajo me cuesta /
||‘What toil it costs me /
Moreover, he constantly resorts to rather overwrought, archaic language, even for 1929, such as the following extraordinarily antiquated English verse to translate Lorca’s straightforward, direct question:
|‘Zarzamora con el tronco gris’ (1927)||‘Blackberry with the grey trunk’ (1929)|
|‘Zarzamora, ¿dónde vas?’,||‘Blackberry where dost thou go?’|
Cummings’ seemingly irresistible resort to overtranslation means that Lorca’s refrain, which in the original composition is a fairly straightforward reference to such a mundane element of Andalusian reality at the time as a ‘jinete’ becomes an almost Quixotic allusion to a figure from a world of knights and legends:
|‘Canción de jinete’ (1927)||‘The Song of the Cavalier’ (1929)|
|‘caballito negro, ¿dónde llevas tu jinete muerto?’||‘Black little nag, where do you bear your dead cavalier?’|
Curiously, and for no apparent reason, a few pages later Cummings translates exactly the same title (‘Canción de jinete’) more simply as ‘The Song of the Horseman’. Indeed, Cummings was a very inconsistent translator and recurrent terms such as ‘jinete’ and ‘muchacha’ are translated in different ways throughout the same volume of poems. In the case of the verses of a poet such as Lorca, who made such extraordinarily productive use of rhythmic and lexical repetition, these translational changes are extremely hard to justify. For instance, in ‘Primer aniversario’, (‘The first anniversary’) we find the first of many examples of the overtranslation of ‘niña’ as ‘maiden’. Indeed, Cummings often seems to have no alternative to this old-fashioned term, and thus we see how Lorca’s ‘Soltera en misa’ also rather inevitably becomes ‘The maiden lady at mass’. Another example of this lexical limitation is found in his solution for the down-to-earth and quintessentially Andalusian reference in the title of the poem ‘Mi niña se fue a la mar’, which with grim inevitability becomes ‘My maiden has gone to sea’.
In reference to Cummings’s diary of Lorca’s visit to Vermont (‘August in Eden’) which was included in the translation of Canciones, Ian Gibson states that it is ‘redactado en un inglés pintoresco’ (Gibson 2016: 404) and, indeed, a close analysis of Cummings’s translational style in Canciones reveals a very notable tendency towards quaint, archaic usage such as a fondness for the form ‘’Tis’, which is used in ‘Little Song of Seville’ (‘Cancioncilla sevillana’) to render ‘Está en la flor azul’ as ‘’Tis in the blue flower’. Cummings evidently had an insuperable predilection for old-fashioned, hyper-literary terms such as ‘the morrow’, and a strong urge to archaically overtranslate quite simple references:
|‘Canción cantada’ (1927)||‘A chanted song’ (1929)|
|‘para entrar en el gris /
||‘to enter into the gloaming /
Cummings constantly transfers the source texts into a much more formal register in English, and like the translator of the 1929 Alhambra translations, the American Cummings also uses several curiously archaic British terms in renderings such as ‘French lassie’ for ‘francesa’. Further evidence of his archaizing translational style can be found in a decidedly old-fashioned and almost Shakespearean rendering of an altogether more prosaic question posed by Lorca in his original poem:
|‘Cancion con movimiento’ (1927)||‘Song with movement’ (1929)|
|‘¿Me marearé quizá /
||‘Perchance I shall be seasick /
Throughout his translation, there are abundant examples of similarly unnatural and foreignizing calques:
|‘Es verdad’ (1927)||‘It is true’ (1929)|
|‘¡Qué trabajo me cuesta quererte como te quiero’||‘Oh, what pain it costs me to love you as I love you!’|
Finally, in a poem which is a reminiscence of Lorca’s childhood in Granada, it is perhaps understandable that Cummings may have misunderstood the reference to the city’s main festivity of Corpus Christi (which in the city is more usually referred to simply as ‘el Corpus’) but it is certainly surprising that anybody who supposedly had a strong knowledge of Spanish language and culture would mistake ‘Nochebuena’ for ‘New Year’, even though the Spanish term for New Year’s Eve is at least morphologically somewhat similar. One can only conclude that during their ‘line by line’ discussions of the poems,21 either Lorca could not explain this to Cummings or the latter simply did not understand his friend’s explanation:
|‘Tíovivo’ (1927)||‘Merry-Go-Round’ (1929)|
|‘Corpus azul/ Blanca Nochebuena’||‘The Blue body/A White New Year’|
Residence on Earth (1946)
Ángel Flores achieved the distinction of being the first English translator of a whole volume of poetry by Pablo Neruda when in 1944 he published a volume of English versions entitled Selected Poems, a text of which only 100 copies were ever printed by the Peter Pauper Press. In 1946, this second, and more commercially ambitious, English translation of Neruda’s work by Flores was published by that most Lorcan of American literary presses, New Directions.22 Its full title is Residence on Earth and Other Poems, and the published text contains no introduction or biographical notes. The book is divided into Part 1 (1925-1931) and Part 2 (1931-1935), and also includes a section of Neruda’s Spanish Civil War poetry entitled Spain in the Heart (España en el corazón) from 1936-1937. In Flores’s version of Neruda, one can instantly detect a much more demotic use of English in comparison to Cummings’s 1929 translation of Canciones. Unlike the latter text, in Residence on Earth there is absolutely no tendency to archaism or to overtranslation, and this is abundantly clear from the first verse of the first poem:
|‘Alianza’ (1933)||‘Alliance’ (1946)|
|‘De miradas polvorientas caídas al suelo’||‘From dusty glances fallen to the ground’|
Flores consistently opts for simple, straightforward translations and decidedly non-archaic language to render Neruda’s equally transparent original, achieving a notable dynamic equivalence:
|‘Sólo la muerte’ (1933)||‘Death alone’ (1946)|
|‘Hay cementerios solos, /
||‘There are lonely cemeteries, /
Unlike so many other anglophone translators of Spanish poetry and particularly the legion of those who have tried their luck with the work of García Lorca (whose Spanish, according to Ted Hughes, ‘cannot be Englished’) Flores does not overtranslate what are frequently quite simple, everyday expressions, which other translators like Cummings have turned into either archaic and unconvincing English, or excessively cryptic and impenetrable, surrealistic English, as in the case of Ben Belitt’s 1955 translation of Poet in New York, a text which subsequently exerted considerable influence on the US Beat Generation. For Flores, ‘caballo’ is just ‘horse’, not ‘nag’ or ‘steed’ or some other such contrived overtranslation favoured by Cummings and the translator of the 1929 English versions of the two Gypsy Ballads. Indeed, one shudders to think what this first published translator of Lorca would have done with Neruda’s title ‘Caballo de los sueños’. Flores just called it ‘Dream Horse’.
|‘Angela Adónica’ (1933)||‘Death alone’ (1946)|
|‘Hoy me he tendido junto a una joven pura’||‘Today I lay down beside a pure girl’|
There are no ‘maids’ or ‘maidens’ to be found anywhere in his English text, and for the bilingual Ángel Flores, ‘niñas’ are ‘girls’. Nor does Flores use ‘thees’, ‘thys’ or ‘thines’, or any such archaisms, and instead he consistently opts for a simple, demotic rendering of poetic Spanish into everyday English:
|‘Barcarola’ (1933)||‘Barcarolle’ (1946)|
|‘Si solamente me tocaras el corazón, /
||‘If you were to merely touch my heart, /
In fact, in all of the translations in Residence on Earth, Flores uses ‘you/ your’ for the Spanish second person forms ‘tú/tu’, never ‘thou’ or ‘thy’. Essentially, the English used by Ángel Flores in his poetic translations from Spanish is much more natural and demotic than the frequently overtranslated and archaizing verses of the aspiring poet Cummings, whose lack of dynamic equivalence and frequent mistranslations betray a deficient knowledge of Spanish and a poor comprehension of the source text.
Bassnett and Lefevere consider translation to be ‘a shaping force in the construction of the “image” of a writer and/or a work of literature’ (1990: 10), and the construction of Lorca’s image in the English-speaking world began with these two first translations. With regard to the importance of the first translation of a text into a foreign language, the Retranslation Hypothesis23 posits the theory that it will necessarily tend to domesticate to a certain extent and thereby adapt to the cultural norms and aesthetic conventions of the target language, whereas subsequent retranslations will tend to foreignize more and stay closer to the original text. This hypothesis essentially presupposes that one of the basic functions of the first translation is to introduce the work into the target culture and, consequently, make it as comprehensible and accessible as possible for readers who are not necessarily familiar with the culture of the source text (Bensimon 1990: 9). These first anonymous translations certainly tend towards domestication (albeit with the occasional foreignizing calque due to direct mistranslation) and to render Lorca’s ballads through the use of a hyperpoetic and archaic register in English, which was frustratingly quite the opposite to how he wished to project himself through translation, that is, to show that universal themes could be represented in a modern way using a popular idiom. Given the immense popular and critical success of the Romancero Gitano in the Spanish-speaking world, these first English versions were certain to create a considerable ‘horizon of expectations’ (Jauss 1982) in terms of Lorca’s reception in the USA, and whoever translated them so archaically and incongruously not only did Lorca a disservice in the summer of 1929, but also sowed the first, and sadly misleading, seed in his extraordinarily productive reception through the subsequent deluge of translations which currently shows no signs of abating.
In my opinion, having compared and contrasted some salient examples of the translational style of both Cummings and Flores with these first 1929 English versions of Lorca, the overwrought, syntactically unnatural and doggedly archaic style of the two translations of Lorca’s Gypsy Ballads published in Alhambra points unequivocally to Philip Cummings as the author. Although there is a good deal of circumstantial evidence and biographical coincidence that would support the claims of Flores to be the first translator, in my view there is no textual or translational evidence to support this. On the contrary, the translational style of Ángel Flores observed in his versions of Pablo Neruda’s Residence on Earth reveals an unequivocal tendency towards a much simpler, more demotic and more dynamically equivalent approach to Spanish-English poetry translation, which eschews unnecessary complications and recourse to old-fashioned and frankly erroneous equivalences between the two languages and poetic discourses. Instead, these translational defects are entirely characteristic of the style employed by Cummings in his 1929 version of Canciones. And there is also some notable circumstantial evidence that would support the hypothesis of Cummings being the first translator, such as the fact that Lorca specifically mentioned in the aforementioned letter to his family on 28 June 1929 that Cummings had translated two of his Gypsy Ballads.
What remains to be established is how these translations might have reached Lorca and the editorial team of Alhambra, and in the absence of any further correspondence between Lorca and Cummings finally seeing the light of day, that specific detail seems unlikely to be satisfactorily resolved. One major objection that might justifiably be raised concerns Cummings’s silence on the matter. If he was indeed the author of the first published English translations of Lorca’s poetry, and bearing in mind his desire for protagonism in the narrative of Lorca’s USA stay, as well as a sense of being unjustly written out of this narrative, why did he never stake his claim to being the first translator? I believe that this was simply due to the fact that he was quite deliberately ignored by Lorca’s Hispanic circle in New York and was therefore entirely unaware that his translations had been used. And he remained so for the rest of his life, thus depriving him of some much desired protagonism in the Lorca New York story by passing into posterity as the poet’s first translator into English.