Axon: Creative Explorations

A few months ago, I received an email from an old friend who lives in Pakistan and who had stumbled upon English translations of mid-twentieth-century Sindhi poems, which he then decided to translate into French. In asking me for feedback on his translations out of the blue, he couldn’t have known how he catapulted me into a realm I usually only occupy in theory here in Australia. Despite being a simultaneous trilingual, I am rarely required to deploy all of my primary languages at once, and certainly never in a creative endeavour of this calibre. Most of the time, a limping bilingualism will do.

At first, I was unsure of how well my mostly vernacular Urdu would map onto the transliterated Sindhi originals he had included or my somnolent French grapple with questions of prosody. As I read his translations, however, I found myself immersed in the exercise of not just rousing but keeping awake all three (and a half?) languages long enough to move between them and provide feedback. Soon, in the process of finding the ‘right’ word, we started taking the few Sindhi poems we had apart, slipping in and out of all three of our languages to explain particular nuances, echoes, implications that we each thought ought to be preserved in translation1.

‘Come play bilingual games with me,’ urges Doris Sommer in Bilingual Aesthetics, for in the crossing and twining of languages lie opportunities for happy accidents and the joy of serendipitous discover (Sommer 2004). In the exchange with my friend, we not only take up this invitation but add several languages to the mix as well and, in doing so, are watching and listening as the sounds of one language have taken up residence in another.

The exchange is ongoing but has had a few results already. The first is that it has reinforced my suspicion that it is far easier to transfer the music of one language into another if both lack lexical stress (Niaz 2007). The translations into English (a language in which stress placement can alter the meaning of a word2), although accomplished, are a speedbump on the road between Sindhi and French to my/our ears. The second is that it has made me wonder at how poetry can find itself echoed across vast distances, the dance of word and sound and meaning that must all fall into step for the ‘poetry’ in poetry to work. The third and final outcome has been the sheer joy of cross-linguistic play that this exercise has reignited for me.

This paper primarily takes sound as the base element of poetry (Perloff 2009) and looks at how this might be used to consider poetry itself as essential to language rather than a product of it. First, I explore the relationship between sound and sense and then then move into a discussion of the use of sound as a way into poetry that weaves together multiple languages. From there I examine the effectiveness of translations as a further example of how sound and poetry and language all depend upon each other.

I should note that when I say ‘language’, I am necessarily restricted to languages to which I have direct native-speaker access – English, French and Urdu – languages to which I have direct second-language or learner access – Punjabi, Spanish, Italian – and languages to which I have cultural or social access (by which I mean their rhythms and conventions are familiar and usually intelligible to me) – Persian, Turkish, Pashto, Sindhi, German, Portuguese, Gujarati, Arabic, Saraiki, Hindko, Nepali, and a few other associated languages.

So, while I do speak about ‘language’ with a certain degree of confidence, I am aware that this is not an exhaustive list by any means and that it is not possible for a single person to have access to all human languages. However, I hope that if I set up my argument well enough, it can then be tested by speakers of multiple languages.

The idea of sound as basic to poetry is not new and many poets and theorists have spoken at length about it. For Robert Pinsky, as just one example, poetry is ‘above all a vocal art, an art that speaks to us in part because we take such delight in speaking it.’ (Rodgers 2017).

‘Delight’ is a useful word here. Young hearing children who are learning to speak move freely between sense and non-sense sounds. And even the earliest sounds humans make can be said to have some sense to them – babies gurgles are quite different from their various cries, all of which communicate much to the people around them. As humans develop the ability to speak, they start dropping the non-sense sounds and stick more and more carefully to the sense sounds of language. This rigidity about the sense of uttered sounds is partly to blame for the widespread belief that people cannot learn a new language as adults. Although there is a lot to be said for the ability of children to assimilate new information, adults are generally more capable than they let themselves be. Adhering so strongly to the idea that adults must make sense at all times can inhibit the linguistic mistake-making that is essential to developing facility a new language.

Poetry is perhaps the only space or medium where sound is liberated from the bonds of sense just enough to evoke in adults that early delight they took in sound. Much of that delight arises from rhyme in which there is indeed ‘some sense of fun that we never want to entirely refuse’ (Haskell 2018), but it should be noted that rhyme is only one kind of sound or aural effect considered here.

The cultural assumption, at least in English, that poetry that is to be read beyond early childhood is deliberately obscure or obtuse, acts as a kind of safety blanket too in that we are ‘allowed’ to not understand poetry as soon as we hear it. Indeed, as we are too often taught in school, and as Billy Collins laments in ‘Introduction to Poetry’(Collins 1988), poems are recalcitrant artefacts and must be tortured at length before they give up their meaning. Although this preoccupation with ‘meaning’ and the allied assumption that poems are intentionally hiding something can be frustrating to poets themselves, the benefit of this cultural assumption is that adults are absolved of the responsibility of grasping a poem’s meaning upon first encountering it the way that they would be expected to do with prose.

If ‘not understanding’ poetry is the general, acceptable default, encountering poetry as an adult can be said to be akin to encountering language as a child. It is something whose meaning is not wholly grasped but whose sound, if we are so inclined, we can delight in nonetheless.

Sound taken apart from language, or rather, sound standing in for language, is something that can be observed in the way adults communicate with young children, particularly pre-verbal ones. In research conducted around the world, neuropsychologist Anne Fernald noted that there are ‘cross-language similarities in the prosodic contours used to express approval or praise to infants, in contrast to those used to express disapproval or prohibition’ (Fernald 1993). Regardless of actual language spoken, people around the world use a higher pitched rise-fall or trochaic pattern to express praise and a short, sharp and lower pitched spondaic or staccato pattern to prohibit or correct. As an example, think of how an adult saying the words ‘well done!’ to a child versus how they would say it to a colleague. Similarly, consider how an adult would check a child who is about to do something dangerous versus how they would say the same thing to an adult. It may initially appear that these expressions are different, but on closer analysis, the difference lies only in the fact that we exaggerate and lengthen our tones when addressing children in order to make ourselves clearer, and do away with the exaggeration when addressing adults. The sonic patterns used to express approval and disapproval nonetheless remain the same (Fernald 1993).

Adults using sound patterns to communicate with children tells us that, even after language has taken root, specific sound patterns continue to be meaningful to humans. But to know whether the meaningfulness of these patterns resides in their association with language or in something essential to the sounds themselves, it is important to see whether the intended recipients, i.e. pre-verbal hearing children, also read them in the same manner as do adults.

After the first observation, Fernald devised an experiment to see if infants actually responded to these variations in pitch and cadence such that they could be determined to be meaningful to the children themselves and not just the adults uttering them. The children selected for the experiment were North American six-month-olds whose only exposure was to North American English. The children were given no visual cues, only praise and prohibition sounds in multiple languages as well as in a nonsense language. The experiment found that infants did respond positively and negatively to positive and negative cues, respectively, and did so most strongly when the language and non-language sounds were closest to the language they most often heard at home (Fernald 1993).

This experiment shows us that hearing humans are receptive to meaningful sound long before we are able to understand language itself, that cadence and pitch and melody all have meaning for us even before we can articulate anything (but also that our ears start attuning themselves to the sounds of our particular languages very early too).

Poetry, I argue, can allow a brief return to this aural state in which the sound of words arranged in a particular way evokes a response that goes deeper than just what a poem ‘says’.

This is also how multilingual poetry can work for and reach a broader audience.

I define multilingual poetry as individual poems that incorporate two or more languages. As in the examples below, this cannot be straight translation or even purely macaronic verse but rather two or more languages braided together to form a single unit, with each language contributing something distinct to the poem without which it would not work aesthetically or semantically. Plurilingual poetry is another term that is used for this kind of verse.

Take, for example, the opening lines of Lorna Dee Cervantes’s poem, ‘Freeway 280’:

Las casitas near the gray cannery
nestled amid wild abrazos of climbing roses (1981)

The long ‘a’ of ‘las casitas’, the echo between ‘gray’ and ‘cannery’, and the ‘r’, ‘o’ and ‘z’ sounds that ‘abrazos’ and ‘roses’ have in common are obvious chimes, but so too are the ‘s’ sound in ‘casitas’ and ‘nestled’ and the ‘i’ in ‘wild’ and ‘climbing’. ‘Wild abrazos’ and ‘climbing roses’ are also metrically matched. ‘Abrazos’ instead of its English equivalent ‘embrace’ is a particularly inspired choice because, where ‘embrace’ has a softness to it, ‘abrazos’ sounds enough like ‘abrasive’ to remind the reader of the thorns concealed in the idyllic picture painted here. Spanish and English are mixed to great effect in this poem, creating a unique rhythm and soundscape and playing off each other to expand the poetic resonance of the poem in terms of meaning and aural impact.

Multilingual pragmatics – the whens and hows of using more than one language in a single conversation or speech act – are dependent on the audience. When speaking to people with a shared language mix, multilingual people will often switch or mix codes, that is, switch back and forth between two or more languages during the conversation, often using the languages in the same sentence. Penelope Gardner-Chloros discusses this at length in her book, Code-switching, finding that conditions for switching include not just having compatible language mixes but also speakers’ attitudes towards the languages and the individuals in the conversation (Gardner-Chloros 2009).

For the purposes of this paper, let us imagine a conversation between multilinguals who share their languages and are comfortable mixing them on the fly (myself and my translator friend, for example). What tends to happen is what I liken to grabbing the bag or umbrella that is closest to the door on one’s way out of the house – the choice depends largely on which word is physically easiest to articulate given the words preceding it. This means that when concepts are repeated in a conversation, it is likely they will be articulated in whatever language is most immediate to hand in that particular place in the conversation.

This is largely automatic; people don’t labour over word choice in most casual conversations, and multilingual conversations are no different. Sometimes the word ‘dhania’ is what works, and sometimes ‘coriander’. When you have an audience who understands either readily, with all other considerations being equal (Gardner-Chloros 2009: 65-88), the automatic choice depends almost entirely on the way the sound feels to say for the speaker. Sounds that are comfortable to articulate one after another generally work well together to produce an aesthetically neutral-to-pleasant effect for the listener.

When this turns to poetry, there are some immediate issues. Because poetry is currently so mired in the visual and linguistic, one problem is whether the multilingual poem will be readable. The mixing of scripts is a related issue. If a reader does not speak a particular language but the poem itself uses only one alphabet, it is possible that the reader will still manage to ‘hear’ the poem or work out what the ‘other’ words mean, as in the example above. If a poem mixes scripts, on the other hand, it is almost guaranteed that the majority of readers will lose all access to sound and meaning, meaning that part of the poem will be completely closed off.

The way I choose to address this problem of access in the first issue of the Australian Multilingual Writing Project is to publish audio files alongside the texts of poems. In creating space for sound alongside – and perhaps instead of – sight in this manner, the project gives readers or listeners access to the soundscape of a particular poem. The poem retains semantic and syntactic value through the language or languages that the reader-listener can access, but it also gains an overt aural dimension (2018).

This idea comes from Antoine Cassar’s practice of making recordings of his multilingual poems available, starting with his collection Muzajk/Mosaics. In the first section of the book, he writes poetry that mixes the five languages he speaks – English, French, Spanish, Italian and Maltese, and the recordings go a long way towards making the poems accessible, particularly for people who have not encountered Maltese before (as the other languages use a fairly unaltered Roman script). The second section, in which he uses ‘guest’ languages and the third, in which he uses ancient languages that few people can read and for which we have no known pronunciations, on the other hand, are curiosities and interesting experiments, but largely inaccessible.

Still, there is an exuberance to the mixing of languages in Cassar’s work, an easy sliding between one and the other that shows a multilingual writer allowed free rein. Instead of just ‘the pleasure of the text enraptured with language’ (Bacon 2018), we see the pleasure of the poet enraptured with sound.

Fuq blata, ħadt xemxata, parlant avec les muses,
le onde fino alle ossa del petto, e più in su,
dawra mejt, jien sturdejt, ma tête, ma tête elle tourne,
high and low, to and fro, no side, no tide, no moon, (Cassar 2008)

‘Bateau Ivre’ is a giddy poem, as the title suggests, and this joy is audible when the poet reads it aloud. ‘Samota’ is similar in content, but a bit more grounded:

Vasto texto del mundo, themes many and themes none,
insondable dimension, pas d’espace, pas de temps,
insir l-art, l-arja u n-nar, nikħâl, nibħâr, ninħall;

c’est la vie autrement, la paix trouvée dedans,
the soul a dream blue sea, life should not mean, but be,
en transe, la fête des sens, je pense, je danse, je suis. (Cassar 2008)

This is a poem in which sound serves as the connecting force between the languages – the rhymes and chimes Cassar finds to fit the rhythm of the lines as he weaves in and out of five languages carry the poem forward and communicate something of the turmoil and wonder he expresses.

Although less enchanted by multilingualism for the sake of multilingualism, Tess Liem too is interested in the ways in which sounds drawn from different languages can interact. In her poem ‘Kitchen Linguistics’, she writes:

My own oral tradition began in the kitchen
            caught in
never learned how to spell bakwan/batwan
            but when
a corn, carrot, onion fritter
            a flitter
that my father deep fried in a wok,
            woke
seasoned before I was even a thought. (Liem 2017)

This poem has an entirely different tone to Cassar’s and aligns more with Cervantes’s themes of loss and longing, of a heritage denied and but sought nonetheless. In this poem, the italicised English lines echo the words at the end of the preceding line, but they are an intrusion, an interruption both of the poem’s flow and the poet’s connection to her father’s language, which, to her is contained not just in vocabulary but also in accent. The poet is forced to ‘hear’ English and her father’s language in tension with each other and the poem is structured such that reading it out loud is deliberately uncomfortable. Even without knowing the language Liem is referencing, the reader/listener can recognise this discomfort in the halting music the poem creates.

There is, of course, never any guarantee that the multilingual poet wants to be understood by a general audience in the first place. In their study of the creation of in-groups and out-groups through code-switching, Eva Mendieta-Lombardo and Zaida Cintron found that there are multiple layers to the use of multiple languages in a poem, and that some code-switching is ‘marked’ while some is ‘unmarked’ (Mendieta-Lombardo 1995). Assuming English-Spanish bilingualism in the context of a monolingual English-speaking audience, for example, if a word used is technically Spanish but readily available to a monolingual English-speaker (taco, adobe, buenos dias, etc.) then the use of it signals the availability of the poem to a mixed audience. On the other hand, if the word is more obscure to a monolingual English speaker, then one can assume that the poem – or at least that part of it – is not ‘for’ the non-Spanish speakers in the audience and is creating a subset of connections within the audience with people who do speak Spanish.

This creation of in-groups and out-groups is exemplified by two very different poems by Sandra Maria Esteves and another by Lorna Dee Cervantes. In ‘Black Notes’ the Esteves is exuberant and immersed in her own experience, throwing out melodious lines with little care for anyone not versed in her musical taste. The out-group is not non-Spanish speakers here but rather people who do not recognise the names she lists and the musical style her poetry echoes. And yet, because of the tumbling, speeding rhythm of the lines and the short, sharp words she uses, the poem sweeps the reader along.

Mezclando manos in polyrhythm sync to fingers,
to keys, to valves, to strings, to sticks,
to bells, to skins, to YEAH black.
Bringin’ it home black.
The bad Fort Apache tan olive brown beat black.
Bringin’ it all the way up fast black.
Flyin’ across Miles ’n Sony,
across John, Rhasaan and Monk’s ’81;
across Dizzy blue conga Gerry horn,
’n básico Andy mo-jo black.
Across Nicky’s campana timbaleando tumbao black.
’n Dalto’s multi-octave chords with all those keys black. (Esteves 1990)

In ‘Puerto Rican Discovery #3 Not Neither’, on the other hand, Esteves moves much more slowly but the audience is much narrower.

Pero con what voice do my lips move?
Rhythms of rosa wood feet dancing bomba
Not even here. But here. Y conga
Yet not being. Pero soy
And not really. Y somos
Y como somos–bueno,
Eso sí es algo lindo. Algo muy lindo.

In contrast, the Spanish Lorna Dee Cervantes uses in part IV of ‘Coffee’ is much more accessible to non-speakers of Spanish:

No more Genocide
in my name. We shall not overcome. We shall fight
this way forever. Estas son mis armas:
la computadora, el video, la pluma.
La plumage de justicia hangs from the broken
arrows of palabras breaking the media block
of Truth and Consequences of Free Trade Agreements. (2005).

From the context of the first two lines, we can read ‘armas’ as arms meaning weapons. ‘Computadora’ and ‘video’ are both easy enough to map onto the English ‘computer’ and ‘video’ and ‘La plumage de justicia’ can be read as ‘the plumage of justice’ first and then, in the context of ‘arrows’ as fletching. ‘La pluma’ and ‘palabras’ are the only real puzzles here and require some grounding in Romance languages to figure out. Even so, the overall intent of the passage – certainly in the context of the entire poem – is evident enough that a few missed words do not really compromise one’s understanding or ability to appreciate the overall effect of the poem.

The idea of processing effort, that is, the amount of effort it takes to decode this particular form of poetry, is worth considering here. Relevance theory posits a cost-effort balance where, for a communication to be successful, ‘extra processing efforts will be accompanied by extra rewards’ (MacMahon 2007). I think it is safe to say that decoding multilingual poetry, particularly when part of it is in a language one does not speak, requires ‘extra processing effort’. What, then, is the reward?

Here we return to Pinsky: The reward is delight, surprise, wonder, and hopefully an expansion of one’s poetic and linguistic ambit.

The aim of multilingual poetry is not to teach general readers and listeners new languages, but rather to indulge the multilingual poet’s desire to mix languages in a way that is natural to them – to us. In doing this we expose the general reader to sounds that, while unfamiliar, can still contain glimmers of access and can potentially help them make connections across languages. Much depends on individual poetic intention and the reader’s own mix of languages, of course. But there is delight to be found in encountering a connection across languages that you had been unaware of, in discovering what you didn’t know you didn’t know.

And, always, back to sound and meaning-making out of it. I said earlier that there is a looser tether between sound and meaning in poetry. Meaning is a shifting goal – or not a goal at all – in line with Mirza Ghalib’s idea that poetry that means one thing isn’t poetry at all (Pritchett 1994). Sound is a launching point for readers and listeners into the ‘multiplicity of meaning’ inherent in poetry (Goodman 1984). The presence of this multiplicity is what makes poetry both possible in the first place and captivating – there is always something ‘else’ there that we can glimpse but not pin down completely.

This multiplicity is also what makes translating poetry so difficult. During my postgraduate studies, I had the opportunity to examine the translation of a free-verse Urdu poem into English and French. There were five English translations and two French ones. While English and Urdu have a few centuries of history, French and Urdu do not. Because of this, I assumed at the outset that English would naturally be more successful at conveying the meaning of the poem than French, which has fewer cultural referents in common with Urdu. What I was surprised by was the discovery that, while English successfully and accurately conveyed much of the meaning of the poems, often in exhaustive detail, the French translations recreated the original language’s sound and rhythm to an extent the English translations were unable to do. In the end, the echoed sound did far more to evoke the turmoil of the original poem than the explanatory poems (Niaz 2007).

Now, I note that judging the success of translations is a tricky, and some would say pointless business, and also that translations are not created for readers like myself in the first place. Over the years I have come to think that access to the original work is sometimes an impediment to the enjoyment of a translation for itself and that commentary like this is not very useful to translators. What it is useful for, however, is demonstrating again that sound in itself is an essential element of poetry and that it has meaning that is interpretable without recourse to language. While French might not have been equipped to fully render the intricacies of the cultural references in the original poem, it captured the way that most Urdu-speakers would have encountered the poem in the first place – not as an intellectual artefact but as a series of images presented in musical language, i.e. that sounded impassioned and moving and so moved them.

Since books became easy and relatively cheap to produce, poetry has gone from being a primarily oral occupation to a visual one. This has produced some wonderful innovations, particularly in the 20^th century, and I am not taking up against print or ‘page’ poetry in the least. However, I do find it telling that, despite having trapped poetry on the page for generations, much of the discourse we engage in around it is in the language of sound. We speak of assonance and consonance and metre and rhyme, of the musicality of a phrase and the chiming of syllables while looking at a static image as if we know it belongs in motion, in Yeats’s mouthful of air.

Robert Pinsky has said that ‘“write” is almost the wrong verb for what I do. I think “compose” is more accurate, because you’re trying to make the sounds in your mind and in your voice” (Rodgers, 2017).

I take heart in the growing culture of recitation, of open mics and slams and readings. Poetry has always belonged to ‘the people’, by which I mean an audience; it has a robust tradition of performance, even in English, and certainly in my other languages. Poets are not just simple music-makers; they – we – are innovators, revolutionaries and troublemakers of all kinds. And the reason we are all of these things is not just because of the power of our words and our rhetoric, but the power of our very voices, of the pre-verbal ghosts that haunt the sounds we make when we lift poetry off the page and let it sing in as many tones as we can.

 

• 1. Sindhi and Urdu both belong to the Indo-Aryan family of languages and so share a considerable amount of vocabulary and grammar. This, along with the fact that these poems are in modern Sindhi, which shares many cultural referents and preoccupations with Urdu and other Pakistani languages, means that the originals are intelligible to speakers of these related, cohabiting languages. This mutual intelligibility is a hallmark of the languages spoken in Pakistan and Northern India and is explored in depth in Tariq Rahman’s Language, ideology and power: Language-learning among the Muslims of Pakistan and North India (Rahman 2002).
• 2. For example, the English word ‘CON-tent’ is entirely different from the word ‘con-TENT’. Stress placement can alter the meaning or intelligibility of a word or even render it nonsensical in English. Many other languages, on the other hand, are ‘unstressed’ in that the placement of stress on one syllable or another does not interfere with the meaning or intelligibility of a word.