Translating Polish Poetry into Scots: An Ethical Question

Though ideally a translator should have a sound knowledge not only of the language of the source text but of the literary culture from which it has arisen, examples can readily be found of satisfactory poetic translations made by translators with little or no knowledge of the original language. Examples also abound of cases where an inadequate knowledge of the source language has led a translator into errors of interpretation, which may or may not be counterbalanced by felicities of expression in the target-language text. The author’s Scots translations of poems in Polish, a language of which he has only a rudimentary knowledge, are presented and examined as case-studies of the practical and ethical problems of translating from an imperfectly-known language.


Translation: Quality of Writing in the Target Language
This paper is based mainly on my own experiences of translating poetry, but I of course hope the issues raised will be of general interest: principally, the question of how well a translator needs to "know" the source language in order to produce valid poetic translations. Prima facie, a natural assumption is that the "ideal" translator of a poem is a reader who has, fi rst, a thorough knowledge of the source language and of the literary culture of which it is the medium; and second, a full familiarity with the target language and a degree of technical skill in handling it suffi cient to produce a text comparable to the original in respect of literary merit. The second of these is surely indispensable: if a poetic translation, or what is off ered as one, is markedly inferior as a work of art to its original, the translator is culpable not only from a literary but from an ethical point of view, since he will be giving readers of the translation the impression that the original is less worthy of respect than in fact it is. As an example, consider the following lines from Leopardi's A Silvia, and Christopher Whyte's translation: Tu pria che l'erbe inaridisse il verno, Da chiuso morbo combattuta e vinta, Perivi, o tenerella.
Before winter could turn the meadows wan an insidious disease eroded your fragility, and overcame you, and you died. (in Jack et al. 35) The translation of the whole poem is certainly not bad (I have, admittedly, picked some of its least satisfactory lines to illustrate my point), but it has undeniable faults and infelicities, shown unmistakeably in this tiny extract. Leopardi's rhymes and alliterations are gone without trace, his graceful hendecasyllabic and heptasyllabic lines are replaced by lines with no consistent metrical pattern at all, and some of the line divisions are simply perverse. The word order of Leopardi's fi rst line is unusual, but the words themselves are not (except for the aphoeresis of inverno to verno): Whyte's wan, by contrast, whether used in its original literal sense or (as here) metaphorically, is not a familiar word in English; and the emotional charge of the adjective is un-called for, to say the least, in a poem whose emotional force assuredly needs no augmentation. "[A]n insidious disease eroded your fragility" not only is tongue-trippingly awkward with its cataracts of unstressed syllables: eroded suggests an action diff erent from and much less forceful than what is conveyed by combattuta, and by rendering part, and only part, of the meaning of Leopardi's vocative o tenerella by the abstract noun fragility, Whyte has completely lost the sense of intimacy which is a key feature of the original poem. Tenerella is quite impossible to translate exactly in a single English word; but on no showing has Whyte made even a passable attempt: even Alasdair Mackie's simple "sweet lass" in his Scots translation of the same poem (in ; also in Mackie 366-367) is more satisfactory. Anyone who imagined that by reading this version he was acquiring a knowledge of Leopardi would be under a complete misapprehension. Whyte is a poet of high reputation and proven ability; but misapplied skill can result in an unsatisfactory translation as surely as can a simple lack of skill.

Translation: Understanding of the Source Language
As there is no diffi culty in demonstrating, then, the ability to write as well, or nearly as well, in the target language as the original poet has done in the source is an absolute necessity in poetic translation. The other condition, though, is more debatable. That good poems which are literary re-statements of ideas expressed poetically in another language can be made by writers ignorant of the original language is hardly questionable. As an example, the companion anthologies by France and Glen (1989, translations into Scots and English) and Thomson (1990, translations into Gaelic) both contain works by George Campbell Hay stated to be from Croatian originals. But they are not from the Croatian originals. They are from Italian versions in Salvini 1942. This fact is not mentioned in either anthology, though in Hay 1948 andHay 1952, where the translations were fi rst published, it is stated specifi cally. Salvini's anthology does not contain the Croatian poems, and Hay, polyglot though he was, could not have read them in any case. The Italian renderings are without distinction of any kind: they are printed in lines and stanzas and to that extent refl ect the fact that they are translations of poetry, but make no pretensions to being poetry in their own right. Hay's renderings are poetry of merit, achieved by freely elaborating on the plain prose of the Italian (for discussion see McClure 2007). It is permissible to imagine (and indeed, to expect) that the Croatian poets, had they seen and been able to read Hay's versions, would have been gratifi ed at having inspired poetry of such quality, but since Hay's poetic embellishments to the Italian texts were of course done with no reference whatever to the originals, they would certainly see substantial diff erences between their work and his. In any event, to describe Hay's poems as being "from the Croatian of …" is wholly misleading. Hay's Poeti Croati Moderni translations should not be measured directly against the Croatian originals, for the simple reason that they were not off ered as translations of these and were not intended to be compared to them. The case of versions which are off ered as translations of specifi c foreign originals is somewhat diff erent. A translator's moral responsibility to produce something closely and recognisably related to the original is much more strongly present in such a work than in one where the translator has simply used the original as an inspiration to his own creative skill. And an extra dimension to the diffi culty of meeting this obligation is present in the case of translations by writers without the requisite knowledge to work directly from their originals and obliged to use an intermediary translation; for the obvious reason that the second translator must place an unconditional faith, which may be misplaced, in the accuracy of the translation from which he is working. As an example which I have previously examined more fully (McClure 2018), to an Italian translator with no knowledge of Gaelic, the process of putting an English translation of a Gaelic poem into Italian is from one point of view exactly the same process as doing so with an original poem in English. But in addition to the possibility of inadequacies or imperfections in the Italian translation of the English, there is the certainty that any inadequacies or imperfections in the English translation of the Gaelic will leave their mark in the Italian.

Translation: The "Either True or Fair" Fallacy
It goes without saying that any poetic translation will diff er from its original in some respects. I have expounded before now on what I call "the 'either true or fair' fallacy," that is, the naive and misguided notion that the process of poetic translation is a matter of striking some kind of balance between literal accuracy and poetic merit, as if the two existed in a fi xed relationship: the resources of the target language will always be diff erent, sometimes vastly diff erent, from those of the source; and the right to manipulate the target language so as to produce the best and most fi tting eff ects which it can aff ord, regardless of how closely they resemble those achieved by the original poet, is not a licence granted to a translator but a fundamental sine qua non of the practice. But there is, clearly, an enormous diff erence in principle between a non-correspondence from the original to the translation resulting from a simple failure on the translator's part to recognise, understand or appreciate what the original poet has done, and one which results from a deliberate artistic decision made by a translator who has understood the original perfectly well. If taxed with the fi rst, the translator can only resolve to do his homework better in future; if with the second, he can retort with the fi rst and greatest practitioner, not only in the Scottish but arguably in the entire European tradition of poetic translation, Gavin Douglas: "Quha can do better, sa furth, in goddis name!". 1 Yet, a distinction must be made between the ethical and the literary aspects of such cases. Clearly a translator should not make simple mistakes; yet even a mistake need not reveal itself in the poetic quality of the translation. Leopardi's "per poco / il cor non si spaura" does not mean "for a wee the hert faas lown" (Mackie 368), but the segmental and rhythmic pattern of Alasdair Mackie's cadence is masterly nonetheless. The diff erence in principle need not be a diff erence in practice, at least if judged solely by the results.
The conclusion to which this line of argument points is that the task of assessing a translated work as a poem and that of assessing it as a translation, though clearly linked, are not directly related; and an overall assessment must balance the two, taking account of the poet-translator's intention. (A key question here is whether the translation is or is not designed to assist readers who are actually studying the works of the original poet.) Poet-translators in their works may run the whole gamut of what Dryden called metaphrase, paraphrase and imitation: two outstanding Scottish examples from diff erent periods are William Drummond of Hawthornden and William Soutar (for discussion see McClure 2001 and2000). And a poet-translator may work from originals in a language he knows well or one which he knows imperfectly, or from intermediary versions made out of one he does not know at all: Drummond was perfectly at home in the French, Italian and Spanish of his models, whereas Soutar's renderings of Pushkin, Pasternak, Ady and Bloomgarten were made from English translationson which he immeasurably improved. The simplistic (and dangerous) conclusion that a work's status as a good poem may excuse its being a bad translation is no more warranted than the converse proposition would be: overall judgement is a case of balancing the two aspects of the poet-translator's achievement in the light of his intention; and an infelicity in either aspect is not nullifi ed, though it may be counterbalanced, by success in the other.
And if the foregoing is seen as a device for covering my back in the discussion of my own work which is to follow, this may be entirely correct.

Translation: The Author's Experience and Practice
I can, by now, claim a fair degree of experience in translating poetry into Scots. In this I am deliberately making my own contribution, whatever its value may be, to the imposing corpus of translations which form an integral part of Scots poetry in the 20 th and 21 st centuries, a corpus to which almost every major poet in the fi eld, and many minor ones, have contributed. Many, perhaps most, of these translations were made directly from the originals by poets who could claim, if not native-speaker fl uency, at least a sound knowledge of their source languages: Alasdair Mackie's renderings from French, Italian and Russian, or Douglas Young's from Gaelic and the classical languages, are cases in point. For my own part, the languages from which I have translated most extensively are Italian, in which I am tolerably competent, and Gaelic, of which I have a knowledge of the grammar and vocabulary but a much lesser degree of conversational fl uency. Other languages from which I have translated are French, German and Spanish. In all these cases, and in that of Old English, I have at least a working knowledge of the languages. Occasional translations from Provençal, Sicilian and Milanese (Frédéric Mistral, Marco Scalabrino, Carlo Porta) have been made with reference to the facing-page translations into French and Italian with which the poems are published: their relationships to the national languages are usually clear, and the points of comparison a source of great interest. The only time I have translated from an English rendering of a work in a language of which I have no knowledge at all, namely Flemish, was when invited by Frank Adam to translate his De komiekenkonferentie van Rochefort, which I rendered as The Comic Confeirin o Creiff ; but even there the relationship of the language to German ensured that it was not entirely opaque. Polish, however, is more dubious. The main result of two years and more of trying to learn it is an impression of almost fi endish difficulty, not mitigated by assurances from kind informants that even native speakers sometimes make mistakes in the intricate infl ectional systems of nouns, pronouns and adjectives. However, the Polish poems to which I have been introduced (principally by Agnieszka Skrzypkowska, postgraduate law student at Aberdeen University, to whom I record my gratitude) made an instant appeal, and tempted me at once into the venture of producing Scots renderings. I did so using English translations, both poetic and literal, but at all stages making sure, by consulting a printed dictionary and online translation tools, that I at least knew what each word meant, and how each sentence was constructed; that is, my translations are made from the originals, but with considerably more help in understanding them than I require when translating from Italian or French.
My practice as always has been to produce translations in a comparable poetic form to their originals. If a model poem rhymes and scans, I incorporate rhyme and some regular metre into my translation. If the metre is impossible to reproduce exactly, for example if the translation is fro m a syllable-timed language like French, I replace it with one which reads well in Scots; and I have at times allowed myself some laxity in rhyming: for example, since rhymes are much easier to fi nd in Italian than in Scots, when translating an Italian sonnet with only two rhymes in its octave I have often used four. My medium is most often the general (i.e. non-regional) literary Scots as it has been developed in the modern period; but since Scots is an extremely fl exible and multi-faceted language with an abundance of dialects and registers I at times use a regionally or socially marked form, such as North-East Doric or Clydeside demotic.

Examples and Discussion
Free verse poetry is, in an obvious sense, the easiest kind to translate. My example here is Tadeusz Różewicz's "Ocalony" ('Survivor'). (The original poems and the translations are printed after the main body of the article.) Though not organised into regular lines, the poem certainly exploits the rhythm of the language (Polish is a stressed-timed language like Scots) in a more than random manner; and I have attempted to do likewise in the translation. The opening line is in Polish a trochaic tetrameter, and rendered in the Scots as an iambic trimeter "I'm twinty-fower year auld" (not just "I'm twinty-fower," though the meaning would be the same without the last two words). In "Wyc't tae slauchter / I wan throu" I have used an idiom implying "come through [a task, diffi culty or ordeal] successfully" for ocalałem, in order to end the phrase on two heavy monosyllables. I have slightly paraphrased "To są nazwy puste i jednoznaczne" to "Thir wirds is tuim an diff ers nane" to bring another forceful monosyllable to the end; and in the following sequence of antonymic pairs I have reversed the order of "wróg i przyjaciel" for a rhythmic symmetry with the preceding line, and chosen translation equivalents which fortuitously give patterns of alliteration and assonance: a slight embellishment of the original which I trust is not presumptuous. I have simplifi ed the grammar of "Człowieka tak się zabija jak zwierzę," again partly for the rhythm; have made explicit the fact that the ludzi are dead to obtain an alliteration, and invited an emphatic stress on thaim to underline the bitter nihilism of the statement. Alliteration has been my guide in the translation of pojęcia as wittins ['information, knowledge imparted']: later I translate it as thochts for the same reason. The obvious translation for "cnota i występek" would be "virtue and vice," and though neither word is specifi cally Scots both would be perfectly understandable in a Scots poem 2 ; I have, however, chosen to render it as "guid natuir an ill": in the recurrence of the idea a few lines later, this wording underlines the stated falsity of the opposition.
My Scots version, overall, is in slightly less free verse than its model and is not an entirely literal translation; however, it is for those with a more intimate understanding of the language and style of t he original than I have to judge whether my translation fulfi ls its responsibility to its model.
When translating a poem in a defi nite and well-established metrical format, a corresponding format in the target language is required. The sonnet form is a classic example. Polish sonnets are written in alexandrine (thirteen-syllable) lines; Scots ones in iambic pentameter. In translating sonnets from Italian, the problem which invariably arises is that since Italian words are often longer than their Scots (or English) translation equivalents a translator is liable to fi nd himself faced with the problem of fi lling up a line: I am familiar with this, since one of my most prolifi c translation sources has been Cecco Angiolieri. The same is true of Polish, but here the eff ect is countered at least to some extent by the longer line. In translating Adam Mickiewicz's Crimean sonnets -poems of which the pervading theme of exile and longing for the calf-ground is certain to appeal to a Scottish reader or translator -I have found, as always, that the main diffi culty is in maintaining rhymes and a strictly regular metre: most often, I have not been obliged either to pad the lines out or to omit material in the original; but not always.
The phonaesthetic power of Scots words is, as is well known, one of the greatest assets of the language as a medium for poetry. Hirstie, used of land and meaning 'barren, unproductive' as well as simply 'dry,' could perhaps be criticised as saying more than suchy, and scaps 'extents of barren ground' reinforces the same idea. An issue which frequently arises in Scots translations is that the words chosen are liable to have greater semantic force than those to which they correspond in the original poems; but here I simply follow other translators in availing myself of the resources at my disposal. Kelter, suggesting rapid, irregular, tumbling motion, will surely do for nurza (my dictionary translates this verb as plunge); and also provides some alliteration. The same ornamentation literally fell into my lap from the fact that the Polish word fali translates accurately as swaws 'waves' and "kwiatów powodzi" requires only a grammatical change to become "fl ouers at sweel" ['surge']. Leck, meaning 'a fl at stone or rocky islet in the sea,' was chosen to underline the tactile contrast with thrissle-taps. A cairn is 'a pile of stones set up as a landmark'; not the same thing as a kurhan, which is 'a burial-mound': if charged with avoiding the dark overtones of Mickiewicz's word, my reply is that for many readers cairn will instantly recall Burns's rhyme " […] where hunters fand the murdered bairn"! "Thon skyrie clift" ['shining fi ssure'] is admittedly one point where I have departed from the original; but besides the necessity of fi nding a rhyme, the idea of a river seen in the distance, in semi-darkness, as a clift in the landscape is not impossible; and skyrie is in keeping with the imagery of this quatrain. My alteration of Akerman to Moldavia is another liberty; but it is partly for the metre and partly because the substituted name is at least slightly more likely to be recognised by Scottish readers; and is not a serious geographical inaccuracy. (Given the transformation of the geopolitical map of South-Eastern Europe between Mickiewicz's time and the present, I have chosen the historical form Moldavia rather than the name of the contemporary independent state Moldova. I am open to persuasion, however, on whether this is indeed the best procedure.) Another unavoidable loss is Mickiewicz's dexterous use of the place-name as a rhyme word, coming as the climactic point of the entire octave. A sokół 'falcon' is probably not the same kind of bird as a gled 'kite,' but since the essential implication is of a predator the substitution may stand; the wąż is not specifi cally identifi ed as an ether 'adder,' but since a disyllabic word was needed, the same applies; making the snake's piersią 'breast' into its side preserves the rhyme and alliteration (and in any case, breast -breist in Scots -seems an odd word to use of a snake). Finally, an inescapable dilemma was presented by the fact that though Litwa has only two syllables in Polish, Lithuania has fi ve, that is half a line, in Scots. If my shift of rendering "głos z Litwy" as "Baltic vyce" is criticised, besides the metrical argument I off er the consideration that the poet's birth town of Zaosie or Zavosse is not in fact in what a modern reader understands by Lithuania but in Belarus. Incidentally, in translating another of the Crimean sonnets, "Pielgrzym" ['The Exile'], I did retain the name Lithuania, rendering the lines Litwo! piały mi wdzięczniej twe szumiące lasy Niż słowiki Bajdaru, Salhiry dziewice, as Och, Lithuania! Your reishlin wuids Mair sweetly sang tae me nor bird or lass O Tatar steppes… but at the cost of submerging the two specifi c geographical allusions under one heading. Here as always in such cases, a translator's only possible retort to criticism is: well, how would you do it?
Polish literature has not been a prolifi c source for Scottish translators. Since Polish is not a language which many foreigners learn, this is to that extent understandable, but the close, long-established and enduring links between the two countries make it also a matter for regret. Bardachd na Roinn-Eòrpa an Gàidhlig (Thomson) contains one poem by Zbigniew Herbert translated by Iain Crichton Smith and three by Tadeusz Różewicz translated by Christopher Whyte; European Poetry in Scotland (France and Glen) one by Józef Czechowicz translated by Burns Singer and two by Piotr Sommer translated by Douglas Dunn. His translations are into English: I have put one of the Sommer poems into a demotic Scots, refl ecting the dreary urban setting. This form is unmistakably (I hope) diff erent from the literary Scots which I have used for the Różewicz and Mickiewicz translations: distinctively Scots lexical items are almost absent, but colloquialisms like yir missis, fags and blethers, the common interjection ken (like "you know"), the hesitation-fi ller "sort o like" and the contemporary slang coupon for 'face' demonstrate that a spoken register is being used. The phonetically spelt yir instead of your also emphasises, in intention at least, the sound of a speaking voice. 3 My translation method is diff erent too: I have virtually ignored the syntactic structures, though not the lineation, of the original, and written in the loosely-constructed sentences of an untutored speaker's monologue, a format which certainly suits the disorganised sequence of thoughts expressed. Neither rhyme, alliteration nor any consistent rhythmical pattern is used. My intention has been to suggest the imaginative world of, say, James Kelman, renowned for his fi ctional evocations of the gloomy and circumscribed lives of the urban working class. I am not suffi ciently well acquainted with Sommer and his work to know whether this would meet with his approval; but hopefully it may stand.
Finally, I off er an attempt, which may be considered presumptuous, at translating an inspired Polish war song, "Czerwone maki na Monte Casino," into singable Scots verse. The stirring paean to the heroism of Poland's troops, like all things of its kind, may have lost some appeal in the present age, but to another nation with a long history of military prowess it cannot fail to strike a sympathetic chord, and in any event the respectful commemoration of a victory won by the self-sacrifi cing heroism of the combatants is morally unimpeachable. The essential thing which I have endeavoured to preserve in my translation is the insistent "rum-pa-PUM" rhythm: not a diffi cult task, in fact. The rare Gaelic-derived word scarnach 'scree' in the fi rst line was chosen not only for its phonaesthetic vividness but because it sounds well in the musical context. Ideally, a syllable sung on the upbeat in a setting with this rhythmic structure should be long (that is, should contain a long vowel, a diphthong or a consonant cluster, and/or should precede a word boundary) and a high-sonority vowel (i.e. one like [a] or [ɔ] rather than one like [i] or [ɪ]). Scarnach fulfi ls these conditions: larachs 'ruins' would have been a closer translation, but lacking the internal cluster of resonant voiced consonants -rn-would not have been so satisfying to sing. Other cases where the position of a syllable on an upbeat in the tune has guided my choice in the translation are spang'd, mynins, gyties. Cases where I have unfortunately had to settle for a short syllable in a position where a long one would have been better are scuggit and thonner. In the former case the word would be more eff ectively sung with a "Scotch snap" rhythm than by prolonging the vowel.

Conclusion
All these examples, no doubt, are open to criticism on some grounds or others, but in the last analysis, poetic translation is a purely pragmatic activity. There are no rules more precise than the obvious ones that a translator must write as well as he can, in the sense of producing the most literarily satisfactory a poem as he can in the target language, and must avoid obvious mis-translations unless (as may happen) there is simply no feasible alternative. And since the venture of translating poems which accord well with the literary tradition of the target language and make a strong appeal to the translator is very enticing, a translator may yield to the temptation of exercising his skill even in a language of which his knowledge is very limited. Whether this particular translator would have done better to resist the temptation is for his readers to judge. Rist nou -whit lown! I hear the cranes at fl ee: The gled's ee follas thaim, but sauf thay'll bide; I hear the gress showd wi the butterfl ee, The sprats skiff 't wi the ether's slidderie side. Quaet in the saucht, I harken aiverie For Baltic vyce -Nane comes. Haud furth; we'll ride. JDMcC, efter Adam Mickiewicz.

Atweesh the Bus-stop an the Hous
Efter the picturs ye're awa tae see yir mate, yir missis is at hame by hersel, yir maw (ye jist stertit thinkin about her when ye got aff the bus) is in some ither toun no weel, ye got a telegram fae her yesterday. Atweesh the 140A bus stop an yir mate's hous -ken, whaur ye pass a shop at's shut an gae tae the kiosk tae get yir fags, an when ye're jist in front o the hous, an when ye're still in the liftafore ye gae intae his hous an sit doun tae yir nicht o blethers wi him an his missis ye're aa by yirsel. Yir laddie gaed aff yesterday tae see his gran an granda. He's by his sel, he hisnae got you nor his maw.
Ye're thinkin about aa this afore the door opens, wi the snaw spatterin yir coupon (altho it's near the enn o March) an ye're hurryin the couple o steps atweesh the bus-stop an the hous. An suddenly ye think tae yirsel: ilka day, this lanesomeness, sort o like agin aabody, an agin yirsel, an agin the fowk ye're thinkin o. JDMcC, efter Piotr Sommer