The task of the translator is to faciliate [the] love between the original and its shadow, a love that permits fraying. (Spivak 1993)

Having discussed possible meanings of translation Bell (1991) defines it as follows:

Translation is the replacement of a representation of a text in one language by a representation of an equivalent text in a second language. This process of ’replacement’ involves that the target text reads as close to the style and manner of the text in the source language (SL). (p. 6)

From the outset I also wish to reiterate that this is not a linguistic analysis of the translation under discussion but rather an attempt to focus on the decisions taken by Coetzee as translator to render this very complex and densely written prose text into English. My methodology is based on a close reading of the text in the source language (Afrikaans) and the target language (English) to subsequently point out the similarities and the differences between the two texts. Following Pinto (2000:147) an important question underpins my discussion, namely: What is the nature of equivalence between translations and their originals?

In order to answer this question, my approach is based on some of the tenets set out in Toury’s (1995) theory of descriptive translation studies and in particular in the following: ’Compare segments of the source text and target text to determine the linguistic relationship, by mapping the target text onto the source text to find “coupled pairs”’ (quoted in Hodges 2010).

In comparing segments and pointing out linguistic relationships one necessarily has to address the thorny issue of equivalence in translation as well. According to Lefevere (1992), any approach to translation dominated by equivalence:

is likely to focus on the word as a unit of translation, since words can be pronounced equivalent to other words more easily than sentences can be pronounced to other sentences, paragraphs to other paragraphs or texts to other texts. (p. 7)

Venuti (2000:5) on his part regards equivalence and function as the two main concepts in any translation process.

Finding equivalents is the most problematic stage of translation. However, it does not mean that the translator should always find one-to-one categorically or structurally equivalent units in the two languages, since two different linguistic units in different languages can have the same function. In post-1994 South Africa it has become increasingly important for Afrikaans authors to be read globally and as a result several of them publish their works in both Afrikaans and English. André Brink was one of the first Afrikaans authors to publish his work in both Afrikaans and English, particularly following the ban on his novel Kennis van die aand (Looking on darkness) in 1973 (see Brink 1973). More recent examples of authors whose works have been translated into English include Karel Schoeman, Marlene van Niekerk, Ingrid Winterbach, Etienne van Heerden, Antjie Krog and Eben Venter. The translators who are responsible for this include Elsa Silke, Michiel Heyns and Leon de Kock who usually work in close collaboration with the respective authors. David Medalie (2012:15, n. 1) describes Michiel Heyns’s award winning translation of Marlene van Niekerk’s (2006) Agaat, as such a ’creative translation’ of a text of such ’highly allusive quality’ that it could be deemed as ’an original work in its own right’.

For the purpose of this discussion I wish to focus on the Nobel laureate J.M. Coetzee’s translation and try to show to what extent the translated text opens new discursive possibilities for its reader.¹ I am interested in the way in which literary texts are translated from one language into another.

In an essay accompanying his translation of Achterberg’s poem, ’Ballade van de gasfitter’, Coetzee (in Attwell 1992) explains his views on translating a literary text as follows:

[…] my translation itself is part of the work of criticism. This is so because, in the first place, it is in the nature of the literary work to present its translator with problems for which the perfect solution is impossible and for which partial solutions constitute critical acts. A literary work is, among other things, a structure in which form has become meaning. When form is disrupted, meaning is also disrupted. Such disruption is inevitable, for there is never enough closeness of fit between languages for formal features of a work to be mapped across from one language to another without shifts of value. Thus, the work continually presents its translator with moments of choice. (p. 88)

Coetzee continues to emphasise the element of ’choice’ available to the translator reiterates that the translator ’chooses in accordance with his conception of the whole. ’For Coetzee as translator translation is not ’simply translating the words’ but also to take decisions based on ’preconception, prejudgment, prejudice’.

When the original Afrikaans text was published in 1981 it received critical acclaim. It was read, amongst others, as a metaphor for the South African racially divided political system but it was in particular the nuanced poetic style of the text that impressed the critics. One should, however, remember that Stockenström is an eminent poet in Afrikaans and her poems are now available in English, thus giving a global audience access to her work.² In the foreword to the recent 2004 reissue of Die kremetartekspedisie Antjie Krog comments on the ’audibility of the novel’:

One hears whilst reading. Somebody is addressing the reader in a lyrical and evocative manner. The speaking subject of the text does not allow convention to prescribe to her. She invents her own language. (n.p)

Tracing the history of the Coetzee translation, Human, the publisher of the novel in Afrikaans, points out that Coetzee, as the translator, was in constant contact with the Stockenström and they discussed ’the subject of plant names’ (2004:11).³ The translation of the cultural references in Die kremetartekspedisie (Stockenström 1981) must have indeed been a daunting task, bearing in mind, as Eco (2003:82) puts it, ’translation is always a shift, not between two languages but between two cultures – or two encyclopedias.’ Similarly, discussions on the theory of literary translation emphasise the ’communicative purpose’ of the translation:

The question is how far translators should prioritize loyalty to the source writer versus producing a text that works in receptor-genre terms. How far, for example, should they adapt or update? (Schäler 2008:157–161)

In her study of Coetzee’s work Clarkson (2009:2) shows how Coetzee’s ’explicit preoccupation with language from the perspective of the linguistic sciences’ has been neglected by his critics. Existing studies, according to her, focus on his novel as ’the embodiment of a given theoretical or philosophical position.’

The translation of Stockenström also predated Coetzee’s own Booker Prize winning novel, The life and times of Michael K, published in 1983. Incidentally, Coetzee’s acclaimed novel Waiting for the barbarians appeared in 1980 and there are clear intertextual links between the two novels. Whether the peripatetic slave woman’s narrative serves as an intertext or even as an Urtext for Michael K does not fall within the ambit of this article. In one of the interviews conducted by Attwell (1992) during the translation period, Coetzee acknowledged that what he liked about 18th century English prose was:

… its transparency, particularly the transparency of its syntax, even when the syntax is quite complex… Foe which I began to write in 1983, is a tribute of sorts to 18^th century English prose style. (p. 146)

Not only 18th century prose but also his translation of Die kremetartekspedisie (Stockenström 1981) has given Coetzee ample opportunity to study the intricacies of syntax, especially since the author uses a dense, poetic discourse in her text. The syntax of the source language demanded an analysis analogous to the one Coetzee did on the novels of Beckett as part of his doctoral thesis. Indeed, Zeiss (1991) studied The expedition to the baobab tree (Stockenström 1983) in conjunction with some of Beckett’s novels. Coetzee’s (1988) translation of the silenced slave narrative must have influenced his writing of Foe because it anticipates his preoccupation with the voiceless Friday. Furthermore, Coetzee’s heroine Susan Barton in Foe replaces the archetypal male colonial hero Robinson Crusoe, just as the slave woman of The expedition to the baobab tree (Stockenström 1983) manages to evade all her captors and suitors and end up being some sort of goddess whilst living in the baobab tree.⁵

[The translator] must solicit the text to show the limits of its language, because that rhetorical aspect will point at the silence of the absolute fraying of language that the text wards off, in its special manner. (Spivak 1993:205)

Similarly, Spivak (1993) comments on the task of the translator from a feminist’s point of view and suggests that the task of the feminist translator is to consider language as ’a clue to the workings of gendered agency’. Translation then becomes an act of subversion, especially if the feminist translator translates a text written from a male-dominated perspective.

when it manages not to remind us that it is a translation. A translation, in his words, most coincides with its original when it is most transparent, when it approximates pure resemblance. (n.p.)

Once the interlingual contract has been established between the original text and the new translated text it opens up new interpretative possibilities to the reader. But, despite the fact that the new translated text is associated with its original author, we should always remember, as Hermans (1996) describes it that the text has an ’other voice, the translator’s voice.’

With regard to the issue of gender and language, Baker (2009:91) is of the opinion that in languages that have a gender category, ’the masculine term is usually the dominant or unmarked term.’ To complicate matters even further, in some languages like Arabic, the gender distinctions are also reflected in the ’concord between the nouns and pronouns and their accompanying verbs and adjectives.’ In a phallocentric society based on binary oppositions the lexical item referring to the male is also the dominant in such an opposition, for example, male or female; active or passive – a tendency of which Cixous (see Ives 2010) in particular is very critical and and which she wants to subvert so that the masculine is not taken to be the dominant one in the opposition. Ideally, one should do away with binary oppositions.

When reading Coetzee, the translator, the question does come to mind: To what extent could we speak of a textual transvestite at work in this translation of Stockenström’s work? The translator becomes the voice of the pre-colonial subject and just like the narrating slave marks her passage with beads or other identifying symbols, he marks the text with his phallic markings, with his intrusive discourse. Michael du Plessis (1988:126–127) hints at this when he observes that although the text is by a woman in which a woman speaks, it is translated by a man – and he even refers to André Brink’s remarks about ’the female body’ and ’the female mind’ and on the excellence of Coetzee as translator. He goes on to call it ’gender ventriloquism’ which anticipates Lucy Graham’s (2006) ’textual transvestism’. Spivak’s remark that the ’rhetoricity’ (1993:209) of the text needs to be undermined is also appropriate in this instance. In the case of Coetzee’s translation, one has to ask whether the translator, in Spivak’s words, has ’internalize[d] sexism as normality’ and is acting out at feminism. I will return to this in my conclusion.

The narrator in The expedition to the baobab tree (Stockenström 1983) remarks: ’If I could write, I would take up a porcupine quill and scratch your enormous belly [i.e. that of the baobab tree] full from top to bottom …’[.] And further on: ’If I could write. Even then, melancholy would take possession of me.’ (1983:29).⁶ In discussing this Du Plessis (1988:119) points out that the novel should be seen as ’a textual enunciation of femininity, an emission of signs, to appropriate Cixous’s liquid metaphors from the place of the feminine.’ And: ’This writing is not the writing we are reading, because it is introduced and rounded off by a subjunctive modalisation’ (p. 122).

Following Kristeva du Plessis (1988) remarks that language:

consists of physical signifiers, graphic marks or phonemic articulations, and of the silences and blanks on which these signifiers are predicated. Now it is precisely these elements in the language, marginal to it, but ineluctably inscribed in it, that one can identify as feminine. (p. 121)

Language associated with the feminine includes metaphors of liquescence, metaphorical expressions relating to holes and round womb-like structures. For Cixous, the body is ’the source of all metaphors: Cixous writes with her body – in, with, under, over, around her body. The fluids of maternity and jouissance – milk, saliva, the fluids of sex – are sensual metaphors for the economies of writing. As milk flows, so words flow. The titles of her works suggest movement, flight, opening, transgression (Ives 2010:69). We cannot decipher the narrator’s writing nor is the writing subject of the narrative able to deciper the slave’s marking of signs onto the body of the uterine-like baobab tree. She describes her attempts at writing as ’scratch’, ’carve’ and mentions the ’notches’, ’lobes and ’curls’.

Spivak (1993:202) in discussing the so-called ’double bind of literary translation’ mentions two theories of translation, namely that you add yourself to the original, or you efface yourself and let the text shine. She also uses the word ’ventriloquist’ to describe the position of the translator. Examining the relationship Coetzee has with Afrikaans, Barnard (2009:87) concludes that the subjectivity of the artist is a performative matter. When translating a text, Coetzee ’performs’ the role of reader-as-translator and this notion of a split subject fits in perfectly with Coetzee’s notion of autre-biography and his scepticism to talk about the self. This calls to mind Coetzee’s own observation that translation is based on choice and on the three p’s cited above, namely ’preconception, prejudgment and prejudice’ (in Attwell 1992:88).

Coetzee as translator can thus possibly be described in terms such as ’gender ventriloquist’, ’textual transvestism’ and ’ventriloquist’. By mapping this onto the translator one deconstructs the traditional essentialist approach of free versus literal translation or even the notion of equivalence. The traditional binary opposition of work versus translation is also put into question. One could even go as far as to use Spivak’s (1993:221) ’larger feminist agenda’ as a reading strategy for Coetzee’s translation. For Spivak, literary translation is ’no mere quest for verbal equivalents but an act of understanding the other as well as the self.’ And as Sachin Ketkar (n.d.) points out, Spivak’s notion of translation is typified by using metaphors such as ’submission and intimacy’.

The translator has to submit himself to the language of the text. But in the case of The expedition to the baobab tree (Stockenström 1983) the translator is also forced to look voyeuristically at the intimate details of the feminine experience in the text. He is forced to read about the female body, the changes that occur in her body during pregnancy. He observes her bodily sensations, her intimacy with her partner but also the way in which she uses her beauty as a form of power over her master. An example of where the translator has almost desexualised the narrator’s criticism of her sexual abuse by her master occurs when the translator translates the phrase bene oopmaak (Stockenström 1981:22) – literally opening one’s legs for his sexual pleasure – in an almost neutral and distanced ’it was legs apart for the owner’ (1983:22). One senses here that the male translator is not sharing the subject’s horror when describing the sexual abuse by her owner.

For the purpose of my analysis I will focus on the following aspects: the lexical items, the syntax, semantics, the poetic devices and the translation of the poems included in the novel. The question arises: Does Coetzee select the more obvious equivalents or does his selection raise new discursive possibilities? Does his translation inscribe the binary opposition between male and female? Du Plessis (1988) compares Coetzee’s novel In the heart of the country (1977) with the translation of the Stockenström text and he makes some interesting points regarding masculine and feminine language against the background of, amongst others, Hélène Cixous’ theory of ‘écriture féminine’. Some of his arguments will be developed further in this discussion of Coetzee’s translation.

These are a function of the propositional meaning of a word. We expect a human subject for the adjective studious and an inanimate one for geometrical. Selectional restrictions are deliberately violated in the case of figurative language but are otherwise strictly observed. (p. 14)

Coetzee uses the English equivalents available in the lexicon and does not invent new equivalents. Whereas the names of the birds are very poetic in the source text, the equivalents in the target language sound more neutral and scientific.

The translator has opted for Standard English equivalents in the case of most of the archaic Afrikaans word in the source text. Examples from Afrikaans vernacular and words derived from Khoi or Old Germanic are rendered into Standard English. The poetic language of the mother text is negated in favour of the more formalised, patriarchal language of the Father and is an insightful example of the way in which the language of the chora is silenced in the symbolic. The speaking subject of the translated text, in this case, is definitely not the speaking subject of the source text. Her association with the language of the chora has been silenced and reinscribed in a new, formalised manner.

Coetzee opts for ’dynamic equivalence’ (Nida 1964:127) because often he has to create an equivalent effect in the target text for the words listed under earlier. This occurs where a direct translation of the word in the source text proves not to capture the essence of what the author is trying to convey in her poetic use of words. An interesting example from the list, however, is the translation of singende lyne as ’circumambient lines’. The female author describes her writing in terms of musicality and singing. We learn from the text that the narrator is illiterate, but this attempt at ’writing’ associates her with the chora, where she is able to express herself in a creative and free flowing manner. She is creating her own form of writing and her own sound system, which are not part of the predominantly androcentric discourse of the patriarchal society. The equivalent selected by the male translator does not convey the same symbolism.

’Circumambient’, according to The Merriam-Webster online dictionary (Merriam-Wester.com n.d.) is derived from the Late Latin circumambient-, circumambiens, and present participle of circumambire, to surround in a circle, from Latin circum- + ambire to go around. Whereas the singende lyne (literally ’singing lines’) of the source text suggests freedom and uninhibited expression, the connotations of ’surrounding’ and ’enclosing’ elicited by the equivalent in the target text suggest the opposite. The translator seems to act as representative of a masculine discourse when he describes her writing as consisting of geometrical designs, namely the circle.

On the whole, however, deviant grammatical configurations are simply not acceptable in most contexts. This means that, in translation, grammar often has the effect of a straitjacket, forcing the translator along a certain course which may or may not follow that of the source text as closely as the translator would like it to. (p. 85)

The following are some examples of how the sentence structure in the source text is translated in the target text:

In twee lae hutte met ingesakte dakke woon ons, die slawe, almal saam, nie volgens geslag geskei nie. Van ligdag tot laatnag, sloof ons ons vir hom, die speseryhandelaar, af. (Stockenström 1981:22)

In two low huts with collapsed roofs we lived, the slaves, all together, not separated by sex. From sunrise to late at night we toiled for him, the spice merchant. (Stockenström 1983:21)

In the Afrikaans sentence the placement of die speseryhandelaar, that is, the object of the sentence, is shifted towards the end of the sentence; almost as if in parenthesis. Coetzee also places the equivalent, ’the spice merchant’, at the end of the sentence. Prominent in the source language sentence is the way in which the author manipulates the syntax. The sentence starts with a clause and the subject of the sentence ons [we] referring to die slawe [the slaves] is only in the middle of the sentence. The second sentence also begins with an adverbial phrase, followed by the subject of the sentence. This pattern is repeated in the target language but because of the difference in the verb structure in the past tense and to retain the logic of the sentence, there is a difference, (woon ons [we lived]). The indication of time in both the source language and target language sentences are placed at the beginning of the sentence for emphasis.

Similarly, when faced with an expression such as kraai die dag rooi he translates it literally as ’crow the day red’ and retains the poetic allure of the expression:

Haan, haan, ons lus jou. Haan, haan, vlieg op die dak van ons hut se nok en kraai die dag rooi. (Stockenström 1981:43)

Cock, cock, we want to eat you. Cock, cock, fly onto the roof ridge of our hut and crow the day red. (Stockenström 1983:43)

In the following difficult passage from the novel, he adjusts the sentence structure slightly:

Klouterplante se loshangende stengels swaai hulpeloos rond. Hulle magenta trompetterblomme loer sidderend maar skalks van elke vlak van die gasheerboom. In hierdie oper vlaktes dryf die wolke onafhanklik deur die blou en kom hulle net soms, asof geroep, na mekaar toe om donder en blitse te vervaardig en in reënbuie op te los. (Stockenström 1981:81)

The loose hanging stems of creepers swung helplessly about. Their magenta trumpeter flowers peered tremulously yet archly from every level of the host tree. In these more open plains the clouds floated in the blue independent of each other and came together only at odd times, as if called, to manufacture thunder and lightning and dissolve in rain showers. (Stockenström 1983:82)

In the source text the sentence starts with the subject, klouterplante, in the focus position of the sentence, whereas in the target text the translated equivalent, ’creepers’ is shifted to the sixth position in the sentence but opening with the qualifying description of the creepers. The focus is now shifted to the ’stems’ and how they look. Other than that, the structure is strongly equivalent to the one in Afrikaans. Ponelis (1979:539) points out that the first and last positions in a sentence are important, especially with regard to emphasis. In the source text language (STL) the emphatic position is allocated to the plants in general, whereas in the target text language (TTL) the emphasis is on a part of them, namely the stems.⁷

Another aspect of the syntax in the source text is the use of repetition. Often a word is created analogous to the previous one, as is the case in the following excerpt:

Ja, beaam ’n ander een, eers die waters, die groot blink, die blomryke, skaduryke, wildryke, die lugspieëling wat ’n werklikheid sal blyk te wees [...] (Stockenström 1981:83, author’s own emphasis)

Yes, confirmed another, first the waters, the great shining, profuse in flowers, profuse in shadows, profuse in game, the reflection that would seem to be a reality [...] (Stockenström 1983:83, author’s own emphasis)

Since there is no direct equivalent for the three words in the source text, the translator has to use another construction in the target text and still convey the meaning of the original. He does, however, repeat the same construction as is used in the source text.

No radical deviation occurs in the translation of the sentence structure into English as is evident from the examples discussed above. No masculinist features of language are inserted into the text.

Die dag na my weldoener se dood toe ek, voos van verliefdheid en verward, die vreemdeling gaan soek het, toe ook was die angs met my en dit was angs en verlange wat my voortgestu het en onsekerheid, die enigste sekerheid waarop ek altyd kon staatmaak, het my na die strate gelei waar verskimmeling die mure in veelkleurige sere laat uitslaan en die poortdeure skeef en verrot hang en ek het `n gebou herken, ek het van die slawe herken [...] (Stockenström 1981:47, author’s own emphasis)

The day after my benefactor’s death, when I, soggy with love and confused had gone in search of the stranger, then too the fear was with me, and it was fear and longing that propelled me forward: and uncertainty, the only certainty I could always count on, led me to streets where mould make the walls break out in multicoloured sores and the gates hang askew and rotten and I recognized a building , I recognized some of the slaves [...] (Stockenström 1983:48, author’s own emphasis)

This is an example of a crafty piece of translation because although the equivalent meaning is retained in the target text, the translator’s selection of words starting with an /s/ -sound is often even more evocative and poetic. A change in the nuanced meaning occurs in the following case.

Voos van verliefdheid/soggy with love. The word voos in the source language has connotations of worn out, lacking vitality and without any value, whereas the equivalent soggy in the target language conveys the opposite: juicy, filled with sap and vitality and wet.

The more matter of fact sentence, Ek het van die slawe herken, is translated as ’recognized some of the slaves’, thus gaining in poetic expression through the use of the s-alliteration.

Cock, cock, we want to eat you. Cock, cock, fly onto the roof ridge of our hut and crow the day red. (Stockenström 1983:43)

Once again we find direct equivalence between the two passages but the expression ons lus jou in Afrikaans is more ambiguous than the English, ’we want to eat you.’ The Afrikaans lus has erotic connotations of lusting after, desire, yearning for, whereas the English equivalent is more neutral. This is an example of Lefevere’s (1993) interpretative poetic translation, because the translator has interpreted the poem and wants to alert the reader to the erotic associations with the word cock.

One could also suggest that this is partially a literal translation because the structure of the sentence in the source language is closely imitated in the target language. In the target language, the word ’cock’ is used in a vulgar manner to refer to the male sexual organ and by selecting this equivalent in the target language, the translator is inscribing a phallic equivalent into the text. One could also view his translation as an attempt to make up for the loss of the erotic connotations associated with lus.

The rendition in the target text reads as follows:

Water yes water
you live in the reed’s bed
and in the hollow of the baobab
water you come out of the air
water you well up out of the earth
you cover the earth
you live under it and above it
your spirit is as great in a drop
as in flood and storms
eagerly I collect you and drink you
water you are in me[.] (Stockenström 1983:64–65)

Once again it is directly equivalent to the original and only in the second last line one finds a change in sentence structure with the placement of the adverb ’eagerly’ at the beginning, unlike in Afrikaans where it is at the end of the line. In the source language the word graag refers to the act of drinking but the emphatic position of ’eagerly’ at the beginning of the sentence in the target text results in an ambiguity: Does it refer only to the collecting of water or does it also include the drinking of the water? Such an inference opens the text to a new reading of the poem.

Striking is also the choice of ’great’ for the translation of groot, because ’big’ or ’large’ would not have had the same effect, especially since it refers to the greatness of spirit.

This is both a poetic and interpretative translation of the original poem in the source language. Again the structure of the source language is imitated but the translator allows himself some freedom when conveying the intended meaning of the poem. Water is associated with femininity and liquescence in Cixous’ discussions of feminine writing (Du Plessis 1988:123–124).

For Cixous, as for countless mythologies, water is the feminine element par excellence: the closure of the mythical world contains and reflects the comforting security of the mother’s womb. The translation of the poem about water suggests that the translator has managed to imitate the feminine imagery and way of writing of the original source text. Both Stockenstrom and Coetzee use words associated with water as feminine element. There are also references to the hollow uterine qualities of the tree and the use of the particular prepositions reiterates the feminine elements in the texts.

Discussing his translation of Dutch literature in particular, Coetzee (in Kannemeyer 2012:383) observes that in theory, he prefers staying as close as possible to the source texts when translating and is even willing to sacrifice the fluency of the target language.⁸ From this we can deduce that he rejects the idea of textual ventriloquism and phallic penetration when it comes to the translation of a text. Yet, as I have shown in my analysis of this translation, there are attempts to undermine the overtly feminine writing of the source text by opting for a more neutral (androcentric?) equivalent. In his own later novels he acts as a textual transvestite, who assumes female subject positions in the text, whereas in the case of this translation, it was not necessary to do so. It is a female-authored text with a female main character.⁹

Comparing the two texts, one indeed experiences the contrary, namely that the target text does not read like a translation but is very close to the original. Coetzee’s translation of Stockenström’s poetic novel is indeed exceptionally well and shows to what extent Coetzee as translator was able to produce a fluent variant in English.

Returning to Spivak’s (1993) issue with appropriation and gendering in the translation process, I am of the opinion that Coetzee is not directly attempting to silence the subject of the text. He acts indeed as some sort of ventriloquist who is passing on the female writing subject’s discourse on the enslavement of women as closely to the original as possible.

Baker, M., 2009, In other words: A coursebook on translation, Routledge, Abington. http://dx.doi.org/10.4324/9780203327579

Barnard, R., 2009, ‘Coetzee in/and Afrikaans’, Journal of Language Studies 25(4), 84–105.

Bassnett-McGuire, S., 1980, Translation studies, Methuen, New York, NY. http://dx.doi.org/10.4324/9780203427460

Brink, A.P., 1973, Looking on darkness [Kennis van die aand], W.H. Allen, London.

Bell, R.T., 1991, Translation and translating: Theory and practice, Longman, London and New York.

Clarkson, C., 2009, Countervoices: J.M. Coetzee, Palgrave MacMillan, London. http://dx.doi.org/10.1057/9780230245440

Coetzee, J.M., 1977, In the heart of the country, Secker & Warburg, London.

Coetzee, J.M., 1980, Waiting for the Barbarians, Penguin Books, New York, NY.

Coetzee, J.M., 1983, The life and times of Michael K, Martin Secker and Warburg, London.

Coetzee, J.M., 1988, Foe, Penguin Books, New York, NY.

Coetzee, J.M. [1992] 2005, Landscape with rowers: Poetry from the Netherlands translated and introduced by J.M. Coetzee, Princeton University Press, Princeton, NJ.

Du Plessis, M., 1988, ‘Bodies and signs: Inscriptions of femininity in John Coetzee and Wilma Stockenström’, Journal of Language Studies 4(1), 118–128.

Eco, U., 2003, Mouse or rat? Translation as negotiation, Weidenfeld and Nicolson, London.

Emants, M. & Coetzee, J.M., 2011, A posthumous confession, New York Review Books, New York, NY.

Graham, L., 2006, ’Textual transvestism: The female voices of J.M. Coetzee’, in J. Poyner (ed.), J.M. Coetzee and the Idea of a Public Intellectual, pp. 217–236, Ohio University Press, Athens, Athens, OH.

Gray, S., 1991, ‘Some notes on further readings of Wilma Stockenström’s Slave Narrative, The expedition to the baobab tree’, Literator 12(1), 51–59. http://dx.doi.org/10.4102/lit.v12i1.745

Hering, J.T., 2011, ‘Mistress and mother and Goddess: Enough to make you laugh’, MA dissertation, Dept. of English, State University of California, viewed 03 June 2012, from http://csusdspace.calstate.edu/bitstream/handle/10211.9/1499/Thesis%20Format%20Final.pdf?sequence=1

Hermans, T., 1996, ‘Translation’s other’, Inaugural lecture, University College, London, 19 March.

Hodges, P., 2010, ‘Literary approach to translation theory’, TranslationDirectory.Com, viewed n.d., from http://www.translationdirectory.com/articles/article2085.php

Human, K., 2004, ‘Stockenström se “Die kremetartekspedisie” verdien herlewing’, Die Burger, 16 Augustus, bl. 11.

Ives, K., 2010, Cixous. Irigaray. Kristeva. The jouissance of French feminism, Crescent Moon, Maidstone.

Kannemeyer, J.C., 2012, J.M. Coetzee: ’n Geskryfde lewe [J.M. Coetzee: A written life], Jonathan Ball Publishers (Pty) Ltd., Cape Town.

Ketkar, S. n.d., ‘Literary translation: Recent theoretical developments’, in Translation Directory.com, viewed 06 April 2011, from2011, from http://www.translationdirectory.com

Kristeva, J., 1984, Revolution in poetic language, Columbia University Press, New York, NY.

Lefevere, A., 1992, Translation, rewriting & the manipulation of literary fame, Routledge and Kegan Paul, London.

Medalie, D., 2012, ‘’’To retrace your steps”: The power of the past in post-apartheid literature’, English Studies in Africa 55(1), 3–15. http://dx.doi.org/10.1080/00138398.2012.682460

Merriam-Wester.com n.d., ‘Circumambient’, in Merriam-Webster, viewed n.d., from http://www.merriam-webster.com/dictionary/circumambient

Nida, E., 1964, Towards a science of translation, Brill, Amsterdam.

Pinto, D., 2000, ‘Power relations in the Portuguese translation of Alan Paton’s “Life for a life”, Language Matters’, Studies in the Languages of Africa 31(1), 130–154.

Ponelis, F.A., 1979, Afrikaanse sintaksis, Van Schaik, Pretoria.

Schäler, R., 2008, ‘Localization’, in M. Baker & G. Saldanha (eds.), The Routledge encyclopedia of translation studies, Routledge, London, pp. 157–161.

Spivak, G.C., 1993, Outside in the teaching machine, Routledge, New York, NY/London.

Stockenström, W., 1981, Die kremetartekspedisie [The expedition to the baobab tree], Human & Rousseau, Cape Town.

Stockenström, W., 1983, The expedition to the baobab tree, transl. J.M. Coetzee, Human & Rousseau, Cape Town.

Stockenström, W., 2007, The wisdom of water: A selection, transl. J. de Lange, Human & Rousseau, Cape Town.

Toury, G., 1995, Descriptive translation studies – and beyond, John Benjamins, Philadelphia.

Van Niekerk, M., 2006, Agaat, transl. M. Heyns, Jonathan Ball Publishers (Pty) Ltd., Cape Town.

Venuti, L. (ed.), 2000, The translation studies reader, Routledge, New York, NY/London.

Viljoen, L., 2001, ’Revisiting the baobab: Transforming space into place in Wilma Stockenström’s The Expedition to the Baobab Tree’, in C. von Maltzan & R. Wilson (eds.), Spaces and crossings: Essays on literature and culture in Africa and beyond, pp. 161–174, Peter Lang, Frankfurt am Main.

Wicomb, Z., 2002, ’Translations in the yard of Africa’, Journal for Language Studies 18(3/4), 209–223.

Zeiss, C.S., 1991, ‘Myth and metamorphosis: Landscape as archetype in quest narratives by Samuel Beckett and Wilma Stockenström’, Irish University Review 21(1), 56–81.

2. Translated by Johann de Lange and published as The wisdom of water: A selection (Stockenström 2007).

3. I am indebted to Helize van Vuuren for alerting me to this article by Human. In his biography of Coetzee, Kannemeyer (2012:382) points out the resemblances between Stockenström’s novel and some of Coetzee’s earlier novels.

4. In the interview preceding these essays Coetzee refers to his attempt to ’evolve a linguistic stylistics with some kind of critical penetration’ (in Attwell 1992:197). From this we deduce his ongoing preoccupation with language, stylistics and the structure of grammar.

5. Viljoen (2001) discusses identity politics in the novel, whereas Hering (2011) reads the novel from a feminist perspective.

6. Page references to the two texts will be indicated as follows: Kremetart will refer to the original source text, whereas Baobab refers to the target text.

7. I am indebted to my linguist colleague Annél Otto for pointing this out.

8. Coetzee describes this as ‘making [his] English as invisible as [he] can to the reader’ (Kannemeyer 2012:383). Coetzee also refers to the difficulty in translating culture specific lexical items without having the ‘luxury of the footnote’.

9. See Graham (2006) for a stimulating discussion of this aspect of Coetzee’s work.