Writing the Transatlantic Imaginary in Agualusa’s Nação Crioula and Monénembo’s Pelourinho

: This article explores the complex relationship between African literatures in European languages and the construction of the historical record in Agualusa’s Nação Crioula and Monénembo’s Pelourinho. It problematizes the idealised notion present in lusofonia of the Atlantic triangle as characterised by fluid identities and hybrid spaces, to argue that in both novels boundaries are ruptured but also reinscribed. In exploring unofficial acts of writing which diverge from official discourse: unpublished texts, secret correspondence, and ways of inscribing memory that do not involve writing, these two novels call into question the relationship between voice and writing and the way the historical record is constructed. By exploring the work of Agualusa and Monénembo in comparison, we can demonstrate some of the points of crossover between these two texts, which transcend both intra- and inter-continental boundaries. Such a comparative analysis allows for new readings of the texts to emerge, which may not be apparent when viewed solely from a Lusophone or Francophone canonical perspective.


INTRODUCTION
Assia Djebar has pointed out that a certain category of writer is very frequently asked in interviews the "banal" question of 'why they write', followed up by the question of 'why they write in French' (or, by extension, any language that is in some way 'not theirs' from the perspective of the interviewer). Her analysis of this interview habit is "Si vous êtes ainsi interpellé, c'est, bien sûr, que vous venez d'ailleurs." (Djebar, 1999: 7). 1 The extent to which elements of the personal identity of the writer "from somewhere else" feature as part of debate in African literature is striking. The linguistic groupings represented by lusofonia and francophonie, and by the terms lusophone and francophone, are far from simple. For example, the relationship to the metropolitan canon for the writer "from somewhere else", is complex, should such a relationship even be desired. Elsewhere, a certain critical enthusiasm for grouping writers has led to a plethora of possible group identities which recast all such relationships, but which can be baffling: in the search for enabling readings, how to choose among the figures of the Afropean, Afropolitan, lusotropical, négro-africain, négropolitain, or tropicopolitan writer? Sabrina Brancato's essay on the term "Afro-European" may perhaps give one answer. Brancato shows how the term "Afro-European" might be used positively to allow some greater transnational and translinguistic discussion of the writers she refers to, but she also concedes that the term could also reinscribe an "ethnic principle [that] may lead to a revival of old racial categories that do not do justice to the complexities of identity" (Brancato, 2008: 2). The key pointrather underscored by the fact that her highly detailed and critical attention to the term "Afro-European" unfolds while she rather oddly allows the word "Afrosporic" to repeatedly slip by unquestionedis that writers rather than writing become the focus in such analyses.
Lilyan Kesteloot's approach, in both her Anthologie négro-africaine (1992) and Histoire de la littérature négro-africaine (2001), is a case in point. In setting out to give an overview of a global "Black African" literature, she posits an identity of texts by black writers from Africa, the Americas and the Caribbean as all being "manifestation et partie intégrante de la civilisation africaine" (Kesteloot, 1992: 7), both by "character" and themes. Such an approach has its critics: for example, Alioune Tine has criticised such work as giving in to a tendency towards "le domaine des généralités, du flou, du vague", and outdated ideas of "African sensibility" and/or soul (Tine, 1985: 100). Kesteloot's purpose is perhaps more about making the point that seeking to integrate black writers into dominant European canons would represent "Autant de vains efforts pour enfermer des écrivains dans une classification artificielle!" 2but in the very next sentence she tells us that the correct context is "les sociétés nègres colonisées" (Kesteloot, 2001: 7) [colonised black societies]. Her focus on race, and chapter titles such as "L'euphorie des indépendances" (1960-69) (ibidem: 232), sit rather awkwardly with the history of lusophone African literature. For example, it seems that Agualusa's work must be disconnected from this literary history, based on his personal biography, and also that lusophone countries can be marginalised, as their experience of independence was not that of the "euphoria" of the 1960s.
Kesteloot's work, as well as any text purporting to present a history of any literature, must be read carefully. It is important to question the patterns set out in such texts, and the assumptions that underpin these, but it is unhelpful just to point to writers and texts that do not fit the patterns suggested. For example, in the introduction to his edited book The Postcolonial Literature of Lusophone Africa, Patrick Chabal warns repeatedly of the danger of sweeping statements, before nonetheless going on to identify "the four phases which characterise the development of African literature so far" (Chabal, 1996: 10), the last of which is "consolidation", a phase characterised as "a period where writers feel secure […] in their position as writers" (ibidem: 11). Chabal's book was published in the mid-1990s, as were the two novels we examine here, so it would be tempting to ask if Nação Crioula and Pelourinho are part of a 'consolidation phase', or a later one.
The pattern in scholarship of making vast categories and then vast statements on 'lusophone literature', 'the African writer', and so on, draws upon a multiplicity of very important historical engagements on the part of writers. But it can also become a self--propagating kind of canon play, where writers and texts (possibly in that order) are co-opted into ever-changing and ever-competing grand narratives.
Having set out some of the problems with making vast categories, in this article we will comparatively analyse Nação Crioula and Pelourinho 3 with a focus on the writing rather than the writer. Both texts actively explore questions of writing and non-writing; of writing as part of the Historical Record versus writing that remains unofficial and non--canonised; and of writing which may not be formalised with pen and paper, but may rather be inscribed into objects and bodies. Such questions regarding the processes of writing, publication and canonisation call to mind Woods's assertion that "history and memory are not merely literary tropesthey are the crucial sites where postcolonial national and cultural identities are being formed and contested" (2007: 3). In their focus on transatlantic history and memory, and given the refusal, within the narratives, to canonise certain acts of writing, both Nação Crioula and Pelourinho transcend structural, literary boundaries. Neither of these texts can therefore be adequately contained by the kinds of canons or categories we have discussed thus far.
Where canons and categories suggest one set of boundaries, the second set of boundaries, which relate to the contestation and formation of identities Woods (2007) suggests, are the geographical and cultural boundaries, and the transcendence of them, within the Atlantic Triangle. Particularly within the lusofonia imaginary, influenced as it is by Lusotropical thought, this transatlantic space between Africa, the Americas and Europe is conceived of as one in which national identities can become fluid and creolised. This idea, in official discourse, can be traced back to the Brazilian sociologist Gilberto Freyre, who, in his justification of Portuguese colonialism in Africa, O luso e o trópico, put emphasis on the "formas híbridas de homem e de cultura, das quais vê, participando raças, ambientes e culturas tropicais" (1961: 88). 4 However, in both Nação Crioula and Pelourinho, the Middle Passage between Africa and Brazil represents not necessarily a hybrid space (though this is seemingly suggested by the title Nação Crioula, since it is the name of the slave ship which takes the characters from Luanda to the Recôncavo) but a point of rupture between identities. We will demonstrate how transnational boundaries are transcended within this transatlantic space, and how the act of crossing them provokes contestations and renegotiations of identity. Linking this with the ways in which each novel presents the canonisation and refusal of different historical discourses will show how the two spaces and two sets of boundaries (canonical and categorical in the literary systems; geographical and cultural in the transatlantic space) relate to one another, and how they are contested and problematised within these texts.

WRITING THE LUSOPHONE TRANSATLANTIC IN NAÇÃO CRIOULA
If there is one writer who is considered to best represent the transnational literary space of lusofonia, it is José Eduardo Agualusa. David Brookshaw, for example, points out Agualusa's "personal identification" with "transatlantic cultural hybridity" (2006: 135), given that he has spent most of his adult life outside of Angola in various parts of the Lusophone world, from Goa to Brazil. His own lived experience, therefore, is emblematic of lusofonia as a fluid, creolised space. His novels, and in particular Nação Crioula, reinforce such a representation, as is evident just from the title (it is interesting 4 It should be noted that O luso e o trópico (1961) tends to be outwardly discredited as colonial apologism in modern thought (see, for example Castelo, 2013;Corrado, 2008: 58-62) but the Lusotropical ideas which developed from it, including the creolised space of the Lusophone Atlantic Triangle and the unique nature of Portuguese colonialism, are tenacious, as will be explored below. "hybrid forms of man and culture, in which tropical races, environments and cultures participate". to note that the English translation of this novel is simply entitled Creole, thereby removing any trace of "nation"). As Malcolm K. McNee summarises: Agualusa's vision of lusofonia, as defined by creolity and fluid identity, often appears to be written palimpsestically upon a Luso-Tropicalist master-text […] highlighting a splendid chaos of shifting individual and national identities. (McNee, 2012: 3) It is this idea that will be questioned within this section, exploring how Nação ainsi l'Afrique de son passé colonial et de la France" (Diallo, 2008: 88). 5 In Nação Crioula, the characters themselves seem to embody a Lusotropical ideal of the creolised subject: Fradique, as the novel progresses, identifies more and more with Angola, then Brazil, and becomes a fervent abolitionist. Through his union with Ana Olímpia, an Angolan aristocrat and the daughter of a Congolese prince, he eventually has a daughter who epitomises the miscegenation with which Freyre used to justify Portuguese presence in Africa as a uniqueand supposedly idealform of colonialism. Even more so than Fradique, Ana is the epitome of romanticised Lusotropicalism: she is an African beauty who speaks European languages as easily as African ones; her own transatlantic movement takes her from Angola to Brazil and back again, via Europe; and she appears as comfortable in the circles of the Parisian bourgeoisie as she does among the slaves on the voyage to Brazil. Crucially, she is both slave-owner and slave, at different points in the novel.
Given the ambivalence of identity Agualusa presents, particularly of slave/ slave--owner, it is compelling, if not unavoidable, to analyse Nação Crioula with reference to Santos's "Between Prospero and Caliban" (2002), the paper in which he sets out his theory about the specificity of Portuguese colonialism and postcolonialism. Indeed, Santos's theory has been widely employed as a lens through which to analyse the hybrid identities and intermediary spaces of the novel (Martins, 2012;McNee, 2012;BeeBee, 2010). Indeed, Fradique's final letter, addressed to Eça de Queirós, explains why he cannot write an article about Portuguese colonialism in Africa, because to do so would highlight Portugal's failure as a coloniser. This letter bears some striking similarities to Santos's theory, which will be briefly outlined here. Fradique writes: Desgraçadamente Portugal espalha-se, não coloniza [...] Pior: uma estranha perversão faz com que os portugueses onde quer que cheguem, e temos chegado bastante longe, não só esqueçam a sua missão civilizadora, isto é, colonizadora, mas depressa se deixem eles próprios colonizar, isto é, descivilizar, pelos povos locais. (Agualusa, 1997: 134) 6 In his paper, Santos outlines "the practice of ambivalence, interdependence, and hybridity" (2002: 16) which, for him, characterises Portuguese colonialism, and says that "the identity of the Portuguese colonizer does not simply include the identity of the colonized other. It includes […] the identity of the colonizer as in turn himself colonized" (ibidem: 17). The parallels are clear to see, and both records of this Portuguese difference in colonialismone fictional, the other part of the official historical recordretrace and rework the Freyrean concept of Lusotropicalism. For Santos, this is the identity linked to wider theories of colonialism; he draws particularly upon the ideas of Frantz Fanon and Albert Memmi to demonstrate that "the link between colonizer and colonized is dialectically destructive and creative" (ibidem: 13). But where Memmi (1957) uses this to demonstrate how the identities of coloniser and colonised are inextricably linked, with the coloniser dependent on the colonised for existence, Santos develops this to assert that the Portuguese coloniser is ambivalentnot merely linked with the colonised, but functioning in a subaltern, not-quite-Prospero position that is itself vulnerable to colonisation. Both Martins (2012) and McNee (2012) link Santos's theory with this section of Nação Crioula, understanding the letterwhich represents a refusal to write -as part of Agualusa's satirical approach, in that Fradique pokes fun at Portugal's impotence as an imperial power, and suggests that the metropolitan centre was less fixed than imperial Portugal would have liked to believe. The problem with this, however, comes from the conclusion that Santos draws: The informal colonialism of an incompetent Prospero saved large sectors of the colonized peoples for a long period of time from living Caliban's experience daily […] They were often allowed to negotiate the administration of the territories and its rules with the European Prospero almost on an equal footing (our emphasis). (Santos, 2002: 36) Such a statement is perhaps the undoing of Santos's argument: whether or not Portugal was in a more ambivalent position vis-à-vis British colonialism, the idea that colonised peoples should feel grateful for being given even a modicum of power in the running of their own country and society demonstrates the breathtaking injustice and inequality at the heart of any colonial system. Rothwell astutely points out that "the perversity, in reality, is somehow to play the victim of one's own colonizing process as part of the official discourse" (2010: 318). Rothwell posits that Agualusa is satirising the very idea of Portugal as an ambivalent and somehow "victimised" coloniser, arguing instead that Agualusa provides a "double-edged critique of the meeting of colonial powers in Africa" (ibidem: 316). Where Santos draws distinctions between the British Prospero and Portugal as "between Prospero and Caliban", it could be argued that Agualusa brings these two colonial systems together not to show the relative weakness or victimisation of one but to demonstrate the perversity of both. 7 There is then debate about the level of satire on which Agualusa is operating (Martins, 2012;McNee, 2012; cf. Rothwell, 2010). With that in mind, the act of non-writing, or refusal to write, takes on particular significance. Rather than simply being a prelude to the opinions on 7 See Agualusa (1997: 132-133) for the narrative passage on the meeting of British and Portuguese colonials in Angola.
Portuguese colonialism Fradique presents, it perhaps acts as a marker to the reader to consider what comes next in a more critical light.
Having demonstrated some of the problems with the mediation of the colonial past within the sphere of lusofonia, and how this has been both represented and interpreted in Nação Crioula, it is pertinent at this stage to consider some of the limits to the "creolity and fluid identity" (McNee, 2012: 3) identified in this text. The characters in Nação Crioula do offer enticingly ambivalent and fluid creole identities, moving between continents, social hierarchies, and navigating racial difference with ease, representing the Lusotropical ideal. However, the fact that this novel is in epistolary form raises some interesting questions regarding the practice of writing and the official structuring of the historical record.
Thomas Beebee (2010) provides an informative reading of Nação Crioula in its intertextual relationship with Eça de Queirós's A correspondência de Fradique Mendes.
For the purposes of this article, and specifically the inter-relationality between acts of writing and the transatlantic space, Beebee's understanding of Nação Crioula as not just an act of writing by Agualusa, but also one of translation, provides a useful starting point. Setting out the book as a translation of "African historical reality […] into the Portuguese literary mainstream" and "Eça's text […] into the postcolonial present", Beebee positions Agualusa within a "triangulated situation" in which Agualusa, as writer-translator occupies one corner of the triangle, between the source and target texts and their respective linguistic and cultural specificities (Beebee, 2010: 202). This authorial triangle is just one of many triangular formations operating in Nação Crioula.
In particular, the writer-characters form a triangle of paradoxical connections and communications: Eça de Queirós's position as a silenced writer within the novel is clear, but Fradique is also ultimately silenced, paradoxically and unintentionally, by Ana Olímpia. It is Ana who writes the final letter of the novel, yet she too is silenced. These three characters inhabit the Lusophone Atlantic Triangle, corresponding with each otherand crucially, failing to correspond with each otheracross its triad of transatlantic points.
It is only when we reach the final letter of the novel that the significance of this being the 'secret' correspondence of Fradique Mendes is revealed. Ana Olímpia's letter, addressed to Eça de Queirós, is dated August 1900the year of publication of Eça's A correspondência de Fradique Mendes, which not incidentally for this narrative, was a posthumous publication: he died that very August. In the letter, Ana makes reference to the fact that she had refused permission for the letters to be published two years earlier, and now, ironically, gives Eça permission to publish them, when it is too late. Therefore, as Beebee points out, the preceding letters were never destined for publication or for canonisation within Portuguese literary history. The African and Brazilian histories of the Lusophone space are merely alluded to in Eça's original text, therefore these secret, 'unpublished' letters fill historical and textual gaps in a kind of "literary game" (Beebee, 2010: 196). This perspective focuses on Agualusa as the writer exploiting intertextual gaps in the Portuguese literary canon through the voice of Eça's character. However, if we look at the writing rather than the writer, as is our emphasis here, then it becomes a significant detail that it is Ana, not Eça, who prevents the publication of Fradique's letters, albeit unknowingly. It is Ana's voice, rather than Fradique's, which provides the final version of events; and while this serves to verify the historical record as presented by Fradique, her interpretation of those historical events diverges, thus allowing for the contestation and reshaping of memory.
This contestation of memory is made explicit early on in her letter. She declares: "não é a história da minha vida. É a história da minha vida contada por Fradique Mendes. Conseguirá V. compreender a diferença?" (Agualusa, 1997:138). 8 While Ana is ostensibly addressing Eça de Queirós, it seems that Agualusa is directing this statement to the reader, allowing for the questioning of the official discourse of Portuguese colonialism and Angolan-Brazilian history, as represented by Fradique's narrative. Ana's voice is marginalised within the text; she is given just 24 pages to Fradique's 123 in which to recount the same events, and her letter is silenced to an even greater extent than Fradique's (it is, after all, implied that his letters were read, whereas Ana's letter will never be read by its intended recipient). However, for the reader, her written testimony is just as accessible to the reader as Fradique's, and through her voice, questions are provoked about the fluid identities and hierarchies apparently present in the preceding account.
It is true that Ana moves between different social positions and countries, but her experience of events demonstrates that this is no easy, fluid process. In fact, she reveals that each movement provokes trauma and ruptures of identity. Regarding her ambivalent identity between slave and slave-owner, she says: "Eu só soube o que era não ser livre, quando, depois de ter sido senhora de escravos, regressei (da forma mais brutal) àquela condição" (1997: 152). 9 For her, there is no ambivalence, or a "between Prospero and Caliban" state of beinginstead, she has been free, then is not free, and she cannot understand the former until suffering the latter. Similarly, in terms of national/transnational identity, she does not occupy a fluid, transatlantic space between Angolan and Brazilian identity; rather, the journey to Brazil symbolises her death as Angolan and her rebirth as Brazilian. Once she becomes Brazilian, she cannot maintain a healthy Angolan identity, and Angola becomes, for her, "uma doença íntima, uma dor vaga, indefinida, latejando num canto remoto da minha alma" (1997:

158). 10
Nação Crioula, therefore, can serve to raise questions about the attainability of a romanticised creole identity and fluid movement between geographical and cultural spaces. These concepts are central to lusofonia, but Ana's letter, in its contemplation of slavery, freedom, identity, and the traumas and ruptures they can invoke, offers a divergent point of view from that presented within official history and memory, and opens up possibilities for the reclaiming and renegotiating of historical discourse within Angolan literature. Furthermore, through acts of non-writing and non-publication on a diegetic level, Agualusa prompts the reader to question the actual official discourse and historical record upon which lusofonia is based, and creates subversive spaces in which alternative discourses might begin to emerge.

THE TRANSATLANTIC IMAGINATION IN MONÉNEMBO'S PELOURINHO
Like Agualusa, Monénembo is a writer whose work reflects a global imagination. He was born in Guinea, but was forced into political exile in his early twenties, a traumatic experience that he has acknowledged as central to his early work (Pia-Célérier, 1996) and a theme that has inspired some excellent analysis of his thought (Gbanou, 2003(Gbanou, , 2007Teko-Agbo, 1996). He has described himself as an "écrivain en fugue" [runaway writer] (apud Célérier, 1996: 111), developing this image to cast himself as a mischievous child absconding from a boarding school, or a rebellious artist who cannot be contained in one literary school or movement. Monénembo's comment in this particular interview is meant quite light-heartedly, but his image of the 'fugue' also has associations with a specific kind of amnesia, with identity and memories lost and then returned. Motifs of lost and found memories, unwritten histories, and the instability and contingency of the historical record are very common in Monénembo's work, with voices that tend to escape the written record his main focus. In this section we will examine Monénembo's imagination of the transatlantic space, primarily in Pelourinho (1995), set in São Salvador da Bahia, with reference to the parallels in Les coqs cubains chantent à minuit (2015), set in Havana. As in Nação Crioula, these two novels both invoke and destabilise the 'Atlantic triangle' of Africa-Americas-Europe while raising questions about which voices are heard in the historical record of this highly charged space and its literature. The early chapters of the later novel in some ways read as a straightforward transposition of the entire narrative of Pelourinho from Brazil to Cuba, such are the similarities between the two novels. Both have a contemporary urban setting, with narrators directly addressing a mysterious 'African' visitor who has caused untold upset in the local community. The first of Pelourinho's two narrators is Innocencio; while the narrator of the later novel is Ignacio. Innocencio and Ignacio are similar in many ways: both make a precarious living by, amongst other things, hiring themselves out to tourists as general 'fixers'; and both attach themselves to the mysterious 'African' visitor, initially seeing him as simply a potentially lucrative mark.
In each novel, confusion over the identity of the visitor abounds, particularly over where he is really from. In Les coqs cubains chantent à minuit, the visitor is invited to a party and introduced as "Mon ami Tierno, il vient de Guinée… enfin, plutôt de Paris. Je veux dire, un Guinéen vivant à Paris." (Monénembo, 2015: 72). 11 It becomes clear immediately that nobody is altogether very sure of where Guinea is. In the ensuing conversation, the South American country of Guyana is confidently placed in Africa, and Guinea in Guadeloupe, "pas très loin des Caraïbes" [not far from the Caribbean] (ibidem: 73). In both novels, the visitor announces that he has come to reconnect with his roots, and this quest is greeted with some incredulity, particularly on the part of Ignacio and Innocencio. For Ignacio, Un Africain à Cuba à la recherche de ses racines! C'était bien la première fois que j'entendais ça. En temps normal, c'était l'inverse qui se produisait. (ibidem: 27) 12 He goes on to give a characteristically acerbic description of those who 'returned' to Africa in the 1960s, and the disappointments that awaited them: The African visitor in Pelourinho reverses the direction of the 1960s returns, but with the same intention of tracing family links that go back for generations and that 11 "My friend Tierno, he comes from Guinea... well, rather, from Paris. What I mean is, he's a Guinean living in Paris". 12 "An African in Cuba looking for his roots! First time I'd heard that one. Usually it happened the other way round". 13 "thousands of Blacks from Harlem, Louisiana and elsewhere [who] streamed into the ports of Guinea and Ghana, tears in their eyes and cameras round their necks, in a desperate search for their ancestors". survived the Middle Passage. He also announces that he intends to write a book on this lost connection (Monénembo, 1995: 149-150), which leads the locals in Pelourinho to nickname him 'Escritore'. Escritore's search for his 'cousins' involves a direct lineage, and the dramatic story of a shared ancestor who was sold into slavery at Ouidah (ibidem: 137-143), which Monénembo bases on a true story but which Innocencio initially sees as 'idiotic' (ibidem: 143), asking the Escritore to at least concede that "vue du barzinho de Preto Velho, ton histoire paraissait bien compliquée" (ibidem: 137). 14 Ignacio, Innocencio and the people in their immediate circle may have no sophisticated geographical knowledge or highly theorised responses to the idea of a shared transatlantic history, but they reflect the realities of the community the visitor has come to find. And it is via such apparently superficial conversations and sceptical throwaway comments that Monénembo signals the forgotten presence of a vast transatlantic history and memory. Whereas the quest in Les coqs cubains chantent à minuit is ultimatelyif uneasilyresolved when lost memories relating to two generations of a family are recovered, any resolution of the quest in Pelourinho requires a return to the time of Nação Crioula, and perhaps further. Christopher L.
Miller, discussing the relative rarity of treatments of the transatlantic slave trade in francophone African literature, singles out Pelourinho as an exception offering "a truly Atlantic perspective" (Miller, 2008: 95), and it is in Monénembo's discussion of slavery in Pelourinho that the full complexity of the transatlantic connection comes through, and all sense of tidy borderswhether linguistic, canonical, geographical or temporalis problematized.
Linguistically, Monénembo is far from unusual among 'francophone' African writers in his use of words from languages other than French in his novels, though Albert Gandonouafter counting the appearances of words from Portuguese, Yoruba and English in the textconcludes that the particular language mix in Pelourinho makes it nonetheless "un bien drôle de roman africain!" (Gandonou, 2002: 52) [a very funny kind of African novel]. These unexplained words might well be disorienting for the reader, as they include words and phrases that clearly refer to all manner of complex issues connected to history, physical space, race, and religion, but they may also reflect the notion that Escritore himself is "étranger aussi bien à la langue qu'à l'espace social" (Gbanou, 2003: 57). 15 Innocencio tries to get this point across, arguing that while Escritore might have all manner of historical knowledge of the slave trade, he has local knowledge which is equally important to the visitor's quest. This conversation takes place in the "barzinho de Preto Velho" mentioned above, a place which seems emblematic of much of the sense of lost memories in the novel. The bar, with all its day-to-day ordinariness for the residents of Pelourinho, becomes the stage for a struggle to connect contemporary life to the possibility of an older transatlantic identity and the scene of passionate arguments over imaginings of origins and 'home'. Preto Velho proudly claims his African heritage and a Mozambican great-grandmother, only to be told bluntly that "Nous sommes du Reconcavo" (1995: 63). In this instance, Innocencio agrees, pointing out that he too can be confident that his close family are from the Recôncavo region around Salvador, but that he can make no sense of ideas that he might or might not also somehow be 'from Africa'. The characters here insist on a Brazilian identity, setting aside Recôncavo's status as a key locus of the transatlantic slave trade in Brazil, and rejecting the notion of a transatlantic identity. In a sense, they refuse Escritore's invitation to see themselves as part of an African diaspora.
We can draw a clear comparative point with Nação Crioula here: as discussed above, Ana describes the loss of her Angolan identity and the gain of a Brazilian one.
She does eventually return to Angola, but with the passage of time, her Angolan identity has been erased from wider societal memory too, and she is known in Luanda as the 'Brazilian' (Agualusa, 1997: 159). If Ana's Angolan identity can be all but lost in just one lifetime, then it makes sense that there are barriers in place which prevent Innocencio having recourse to his distant African ancestry. Escritore, in a quest to find his own roots, attempts to break down these barriers, but he is met with resistance. As demonstrated earlier, in Nação Crioula, the Middle Passage symbolises deatha ruptureas well as a fluid space where identities can be hybridised. This is reinforced by a newspaper report claiming that Fradique and Ana Olímpia died on the voyage, and in a letter to his godmother not long after their arrival in Brazil Fradique writes: "trago-lhe uma funesta notícia: morri" (Agualusa, 1997: 77)  and 'History', and that recalls Monénembo's positioning of himself as a 'runaway writer' who draws his imaginative power more from the realities of life than from any school or theory. Innocencio does indeed find Escritore's 'cousins', and has to change his mind about Escritore's ideas of a possible transatlantic and transhistorical identity, but remains sceptical about whether the full violence of the slave trade, or the complexity of the lived transatlantic connections he has come to perceive in his daily life, could ever be contained in a book (Monénembo, 1995: 144-145). It is the only second narrator of the novel, Leda, who could reveal to Innocencio the truth of his own personal history, the insight he needs in order to understand his newfound sense of history and identity, and perhaps also the limitations to Escritore's project of writing a book. However, her life and her narrative remain isolated from Innocencio's, along with her access to Pelourinho's past.
Noémie Auzas has written of Escritore as a "sorte d'incarnation vivante de la transmission du passé" (Auzas, 2004: 117), 17 but this seems an equally apt description of Leda. Leda lives in present-day Pelourinho, but is physically closed off from the teeming day-to-day life of Innocencio's narrative. She is blind, and it seems that she never leaves her room, but she nonetheless 'sees' Escritore in strange "vertiges" (Monénembo, 1995: 33). 18 Leda calls upon the sensual and the supernatural throughout her narrative: the scent of jasmine signals the presence of the ghost of her childhood friend Lourdes, while mystical changes in her sense of light and colour announce visions of Pelourinho's past.
Leda uncovers the traumatic history of the slave trade in Pelourinho in two sequences (Monénembo, 1995: 81-86;125-127) where a slave is whipped to force him to take his master's name. When he resists, his torturers threaten to give the same treatment to the woman carrying his child. This torture takes place publically, in the picturesque streets that tourists now visit, and where Escritore and Innocencio meet.
The pillory the slave is tied to may be the pelourinho that gave the district its name. The reader's attention is called to a chair made of jacaranda wood that the slave owner sits in, an apparently tiny detail that recalls the motif of hidden histories: the same chair seems to be in Preto Velho's bar in the present-day narrative, an odd and ancient piece of furniture that Escritore chooses when he 'holds court' (ibidem: 25) there, and that Innocencio finds oddly troubling. Leda's detailed account of the past also reveals Innocencio's direct connectionsby blood and by nameto this violent scene.
Throughout her narrative, she draws attention to the myriad ways the past is linked to the presentin physical space, artefacts, songs, names, scents, coloursyet may not 17 "a kind of living incarnation of the transmission of the past". 18 "literally 'dizzy spells', but here a kind of visionary trance induced by spiritual forces". leave a written record. As she tells Escritore, "Tout cela est inscrit en moi comme en un tas de vieux livres impossibles à feuilleter". (ibidem: 127) 19 This sense of a physical 'inscription' of history in Leda links to the way Escritore finally finds his cousins: they all carry the same tattoos on their shoulders. Her narrative also makes the reader question to what extent Escritore's plan of writing a book could ever have succeeded, or indeed was ever real. Rereading what he says about his book in the light of Leda's narrative, it is striking how he talks of the book in very physical terms, as "un livre de chair et de moelle [...] un agneau à immoler" (ibidem: 154), 20 and how he sees the book and his life as the same thing. Here Daniel Delas' reading of Escritore's quest is instructive: in meeting his 'cousins', "La boucle était bouclée, la quête trouvait sa vérité, le roman était écrit d'un coup, l'auteur pouvait mourir" (Delas, 1996: 107). 21 Read this way, whether or not Escritore actually writes anything before his death is not important: the essential is that he succeeds in his quest in re--establishing the family link he came to Pelourinho to find.

CONCLUSION
The two texts discussed in this article explore the memory of the Atlantic triangle, foregrounding the many silences, assumptions and false trails that haunt the writing of such a history. Nação Crioula appears at first to be a one-sided epistolary novel that consists of letters from a nineteenth-century Portuguese aristocrat sent back to the imperial centre at Lisbon. Pelourinho opens on the model of the murder mystery, with the oddly-named 'Escritore' dead in a ditch and his projected book recreating the link between Africa and Brazil an impossibility. However, both texts go on to undermine the idea of the written (or published) record, and to point to its insufficiency. There are no responses to Fradique's letters in the text of Nação Crioula, and Ana's letterfor all that it is essential to the truth of the transatlantic journeys she made with Fradiquewill never be read or published. Escritore's book joins the silence, though Innocencio tells us how he feels part of a generalised local "hallucination" (Monénembo, 1995: 198) that this book will somehow still appear and resolve all of the problems Escritore raised. Agualusa and Monénembo thus write against the both the record and the silence of the Middle Passage, at once making and breaking links in the transatlantic space.
Revisto por Sofia Silva 19 "All this is inscribed in me as in a pile of old books that cannot be leafed through". 20 "a book of flesh and blood [...] a lamb to be sacrificed". 21 "Things come full circle, the quest was fulfilled, the novel was written instantly, the author could die".