[t]he Irish cultural obsession with the past” (2011: 16). Pine makes her point in the context of a discussion which is remarkably relevant to O’Connor’s novel: she notes the imperfect quality of cultural remembrance symbolized by the figure of the ghost which keeps haunting the present. A cultural memory which has been crowded by spectral shadows of departed souls is evidently traumatized and needs “to exorcise the ghost of the past” (Pine 2011: 17). In order to express their discontent with a haunted culture, according to Pine, the Irish should follow “the imperative of ethical remembering” (2011: 16): they should perform the remembrance of the past with a view to the future. Only “an ethical form of memory” (Pine 2011: 17) can possibly restore a healthy balance between the past and the future and make reality proof against any visitations from incorporeal intruders. Ghost Light, as an account of an individual trauma induced by a very persistent apparition, whose presence and permanence in Irish culture seem to be incontestable, conforms to this imperative perfectly well. Even though Molly, as a character, embraces her trauma and accepts the haunting of the past, the larger effect of the novel is to reinscribe her own memory in Irish culture and, symbolically, bring her back to life by saving her from ghostly oblivion.

What Molly’s intimate memories of Synge bring into sharp focus is an irreconcilable conflict between her own idiosyncratic, uncensored and possibly wayward version of the past on the one hand, and the received and publicly accepted account of their relationship that Dr. Marcia Vinson expects to learn more about. The American scholar, in Molly’s view, is driven by the idea of “a normative biography” (Berntsen and Bohn 2009: 63), unfolding according to certain social and cultural norms and expectations. Molly is convinced that Dr. Vinson wants to hear only a carefully edited version of the events from the first decade of the 20th century in Dublin. To put it in more technical terms, Molly is keenly aware of the discrepancy between her life story and the life script it is supposed to conform to. According to Berntsen and Bohn, a life script “represents a series of life events that take place in a specific order, and it represents a prototypical life course within a certain culture” (2009: 64). Clearly, Molly’s biography deviates from the cultural norms of her times; her apprehension about sharing her memories with strangers is a consequence of her realization that her actual life narrative may be too much of a challenge to society at large. Instinctively, in the early 1950s, she recoils from contributing her personal insights and most intimate recollections to Irish (and, more generally, Anglophone) remembrance culture because she intuits that those are incompatible with Synge’s public stature and the overall tenor of the cultural memory associated with the period of his artistic activity. Ironically, Synge may have viewed himself as an outsider in Irish culture, too, but after his death Irish culture itself quickly forgot about that and happily embraced him as one of its major figures.

In the 21st century, however, Molly may finally share her memories, dreams and reflections with whomever is willing to lend her their ear. Of course, it is too late for her to produce any form of autobiographical discourse; she has to speak posthumously, by proxy, through fiction. Granted, Ghost Light yields us an image of fictional history, but there is no other image of this particular aspect of the past which is readily available to scrutiny. Under such circumstances, fiction may usefully fill in the gaps which historical discourse is powerless to document. Hayden White insists that the two discourses – historical discourse and fictional discourse – work side by side to give us a unique insight into the true and the real. Following Michel de Certeau, White assumes that “historical discourse wages everything on the true, while fictional discourse is interested in the real – which it approaches by way of an effort to fill out the domain of the possible or imaginable” (2005: 147). The real, on this model, includes but also transcends the empirically verifiable, and gestures in the direction of purely cognitive categories. According to White, the real encompasses both actuality and possibility: “[t]he real would consist of everything that can be truthfully said about its actuality plus everything that can be truthfully said about what it could possibly be” (2005: 147). Historical fiction explores those possible pasts which seem to be of particular interest to contemporary writers and readers, a demand which historiography is evidently incapable of satisfying.

Fiction has a particularly useful potential for stimulating reflection on the past. According to Julie Hansen, it also reveals memory’s reliance on imagination. She claims that “[l]iterature resides in a grey zone between history and memory, lived and imagined experience. Although it reflects and refracts the real world, fiction frees us from definitive truth claims about reality” (Hansen 2015: 198). Hansen believes that the critical potential of fiction manifests itself also in its capacity to challenge history: “Imagined pasts allow us to contemplate alternative interpretations, to question accepted historical truths and to problematize the relationship of society and individuals to past events and even memory itself” (198). The current popularity of historical fiction (including historiographic metafiction – see Hutcheon 1988) seems to testify to the demand for a plurality of perspectives on the past.

This is not to say that fiction has only recently discovered a new preoccupation with history. Richard Terdiman, in his study of late 19th-century writings, puts considerable emphasis on the memory crisis which occurs in the literary texts of that period. Recalling the past has been a central topos in so much fiction ever since that Terdiman does not hesitate to conclude that “[n]ovels are exercises in the process of memory” (1993: 25). The earliest historical novels in English, however, much predate Terdiman’s memory crisis. Arguably, the harbinger of the genre comes from an Anglo-Irish writer, Maria Edgeworth, who published Castle Rackrent in 1800. Although Kersti Tarien Powell acknowledges that Edgeworth’s treatment of history and memory does not meet all the criteria of historical fiction and, accordingly, the honor of the first English writer of historical novels should go to Walter Scott (2004: 134-135),1 Edgeworth’s influence on both British and Irish authors (including Scott himself) is beyond dispute. Diana Wallace also juxtaposes Edgeworth with Scott to conclude that Castle Rackrent, unlike Waverley, focuses on “no great event” and that is the only reason why it is considered a regional novel, rather than the first historical novel (2005: 9).

Powell identifies two writers as the pioneers of historical fiction in Ireland: James McHenry and John Banim. McHenry’s novel of 1824, O’Halloran, or The Insurgent Chief: An Historical Tale of 1798, is concerned, as the title implies, with the 1798 rebellion, and, crucially, it seeks to represent history in an impersonal and unbiased manner (Powell 2004: 135). The Boyne Water (1826), jointly authored by John and Michael Banim, owes much to Walter Scott and yet, emblematically for the bulk of contemporary Irish historical writing, “the history that the Banim brothers were dealing with, the Jacobite-Williamite War of 1689-91, [is] problematically open – an unfinished tale of continuous violence between Protestants and Catholics” (Powell 2004: 136). In later periods of Irish literary history prominent writers of historical fiction would often appear in tandem, too. For example, in the 1930s there is an intense preoccupation with history in Sean O’Faolain’s novels (e.g. Bird Alone) and biographies alike. Simultaneously, Kate O’Brien’s works from the same decade evince a keen interest in Limerick’s past (disguised as a fictional city of Mellick).

A genuine interest in Irish history and the workings of both personal and collective memory is particularly evident in late 20th century and early 21st century Irish fiction. Powell (2004: 138) notes the growing popularity of historical novels in Ireland already in the 1970s while Eve Patten (2006: 259) associates the heightened status of this kind of writing with the rapid transformations in the Republic since the mid-1980s and the publishing boom of the 1990s. Significantly, both Patten (2006: 263) and Liam Harte (2014: 11), in his Reading the Contemporary Irish Novel 1987-2007, detect pronouncedly postmodern qualities in the treatment of history and memory in the recent writings: they point to the fiction’s self-consciousness and the use of irony. What emerges from Harte’s discussion of a considerable group of contemporary Irish writers (including Sebastian Barry, Seamus Deane, Anne Enright, Edna O’Brien, John McGahern, Patrick McCabe, Roddy Doyle, William Trevor and Colm Tóibín) is their common stance on recent Irish history: they “find deficiencies in totalizing narratives of the past, refuse to fix the nation in unambiguous paradigms, and pose awkward, complex questions about the adequacy of nationality as a foundational fiction for the self” (Harte 2014: 3). There are several plausible explanations for this concern with the past, memory and identity in Irish culture at the turn of the 21st century but it seems that it was mostly a series of political and social upheavals, including the eruption of the conflict in Northern Ireland, that galvanized public opinion and inspired the writers to raise historical themes in their novels.

What may also have prompted an open discussion of various aspects of the Irish past was a (re-)emergence of “revisionism” in historical studies in the final decades of the 20th century (Garratt 2011: 2)2 to the effect that even those versions of Irish history that used to be blotted out have been rehabilitated in some recent writings (e.g. by Sebastian Barry). At the same time, as George Boyce and Alan O’Day remind us, “contemporary historical heterodoxy [largely ushered in by revisionism] is comparable to religious heresy and inspires much the same suspicion and hatred from those who feel threatened by it” (1996: 2). What revisionists often question is the nationalist orthodoxy, which subordinates historiography to politics. It comes as no surprise, then, that much of this intellectual ferment is reflected in Irish novels, especially those that address recent historical as well as social and political issues. Symptomatically, in the section of his Luck and the Irish concerned with modern Irish culture, the historian R. F. Foster sums up his discussion of the latest developments in fiction by declaring: “history remains the preoccupation” (2007: 170, my emphasis).

What distinguishes fiction from historiography is that it is uniquely capable of animating the past and telescoping the temporal distance between the then and the now. In Ghost Light this is vividly illustrated when the past visits Molly in post-war London: Synge haunts her in person, an anthropomorphized memory of their relationship, which proves to be both irritatingly persistent and life-sustaining. He resembles an occasional visitor on a par with other inhabitants of the city who seem like shadows to the almost constantly intoxicated Molly:

But this morning someone else is come to you again, out of the same light, somehow, out of an unseen room, out of a city you have lived in the last thirteen years but have never found a reason to call your own. This has happened to all of us: a coasting across the mind by one we had thought forgotten or purposefully banished. But today will prove him a wanderer reluctant to be exiled, an emigrant still attempting to come home. (O’Connor 2011: 2)

On several occasions, Synge comes to Molly on his own, unbidden, and takes control over her mnemonic faculties which have been impaired by alcohol. Her memory becomes a peculiar locus, a host to an alien identity which can survive only in her recollections. She is overpowered by his presence; her present existence yields to the past even if some memories of their relationship turn out to be painful and depressing. Clearly, though Synge is dead and buried, he is still able – from beyond the grave – to hold Molly hostage through the workings of her own memory.

Sometimes it is only his voice – effortlessly formed epigrams, patronizing words of advice, a sporadic term of endearment – that accompanies Molly whether she likes it or not. These prosopopoeic visitations from the past seem to be perfectly natural to her. Synge’s phrases are in tune with her own thoughts and recollections, a smoothly integrated intertext in her personal account. The heteroglotic quality of her soliloquizing testifies to Max Saunders’s claim that “[l]ife-writing is fundamentally intertextual” (2008: 322). Thereby her individual story transcends her own memory and becomes part of a collective past. According to Saunders, “[i]f other genres or sub-genres or forms [apart from autobiography and biography] can be read as life-writing – such as novels, poems, short stories, travel writings, topographical books, historiography – they can all be used as routes into cultural memory” (2008: 322). Molly’s memory resonates with various voices from the past, most notably Synge’s, but also Yeats’s, Lady Gregory’s, Molly’s sister’s and those of several other people. At some points, her prosopopoeic memory can barely contain all this vocal excess, an unbearable noise that she seeks to quieten by means of alcohol. Still, most of the time, voices from the past are the only performance that she truly enjoys in the present.

Molly’s reliability as a chronicler of the past raises some serious doubts. She has problems recalling the names of several people and titles of certain movies. Irritated by those failures, she complains about the efficacy of her memory: “Oh so maddening, how memory works, or doesn’t work, when you are old. Every square in a counterpane you owned when you were five, you remember the sequence, the colours. But then people so important to you, and now even their names are melted away” (O’Connor 2011: 130). Nevertheless, Molly seems to recall perfectly well what she needs in order to sustain a particular vision of her youth that makes her present predicament more bearable. It is the memories which are concerned with Synge that are constitutive of her current sense of selfhood so it should not come as a surprise that her memory is highly selective (Neumann 2008: 333).3 According to Helen L. Williams and Martin A. Conway, to have a coherent view of the self, one tends to make the contents of one’s autobiographical memory congruent with what is perceived as a desirably coherent self. As a consequence, “memories can be altered, distorted, and fabricated in order that beliefs and knowledge about the self may be confirmed and supported by specific autobiographical memories” (Williams and Conway 2009: 43). In Ghost Light, it is tempting to attribute Molly’s license to invent or modify memories to her abuse of alcohol and yet both her drinking and her confabulations may be merely symptoms of a problem with identifying a coherent self in the present.

In Molly’s case, going back to the past involves not only reinventing actual circumstances and conversations; she is also capable of recalling what she used to imagine when she was engaged to Synge. Imagination and memory have in fact much in common (Lachmann 2008: 303). Both rely on a mental representation of something absent, a mere image, a shadow of the original. What makes Molly’s imaginings interesting is that they are quite vivid and detailed even though they come from a distant past: she can still remember clearly – or at least she claims so – what she dreamt about, e.g. the possible futures she might have had together with Synge. As an old woman, she retains a very lucid picture of her youth although her present is practically a blur. Intoxicated by gin, she dozes off in a cinema; her somnolent mind rehearses a conversation with Synge but when she wakes up, she is at a loss whether the conversation was a memory or a figment of her imagination.4 At another juncture, towards the end of the novel, she asks herself parenthetically: “Did any of this happen, Molly? Aren’t the dates incorrect? Sure it’s only a story. What matter?” (O’Connor 2011: 213). Apparently, she is not concerned about fidelity to actual facts; in the grand scheme of things, her autobiographical memories are not meant to clear her name and right the historical record. Instead, they are meant to keep her alive.

Molly’s mnemonic performance functions as a generic marker for the entire narrative, which may be identified as belonging to the category of “fictions of memory” (Neumann 2008: 334). In her discussion of memory in literature, Birgit Neumann points out that such works are not only non-referential stories but they are fictions, too, inasmuch as the protagonist’s or the writer’s present self needs consolidating even at the cost of imaginatively modifying the past:

These stories can also be called ‘fictions of memory’ because, more often than not, they turn out to be an imaginative (re)construction of the past in response to current needs. Such conceptual and ideological fictions of memory consist of predispositions, biases, and values, which provide agreed-upon codes for understanding the past and present and which find their most succinct expression in literary plot-lines and myths. (2008: 334)

Fictions of memory are thus not simply falsifications of the past; as artefacts they are adaptations which we have developed to cope with it. Whether we take autobiography – Rousseau’s blatant lies in his Confessions come readily to mind here – or novels like Ghost Light, the mechanism will be similar. As Renate Lachmann notes, from a cultural perspective, all writing is “both an act of memory and a new interpretation” (2008: 301). In fact, every act of memory is already an interpretation of the past in its own right. Literary representations of memory, in turn, are never meant to reconstruct the past; they are meant to explore the nature of memory and selfhood in the first place. By the same token, Ghost Light is not a novel about Molly Allgood’s youth; it is a novel about an old woman’s attitude to her past.

The very phrase “fictions of memory” sounds oxymoronic enough to generate epistemological skepticism in those who are not enthusiastic about the exuberance of postmodern rhetoric. It is true that, for example, what Hayden White articulates in his writings on history – i.e., his claims about its inalienable literariness and problematic representational status – may seem hyperbolic if we see it in isolation, and yet his position is endorsed by such prominent continental philosophers as Paul Ricoeur (2006: 251) who, in his magnum opus concerned with ars memoria, recognizes White’s contribution as especially valuable in the context of the relation of history to fiction. On the whole, contemporary discourses on cultural memory, often aided by recent breakthroughs in neurosciences (like the discovery of the role of the hippocampus in binding together memories of the past with the capability of envisioning the future – see Miller 2007: 312), tend to collapse oversimplified binary distinctions: past/present; history/fiction; memory/imagination; present/absent; genuine/spurious (rhetorical), etc. In this way they pave the way for such reassessments of the past that are not encumbered by traditional representationalist epistemologies.

This is aptly illustrated by what Ricoeur discusses in his Memory, History, Forgetting as the pragmatics of forgetting. Even though the bulk of the book is concerned with memory and history, Ricoeur realizes that the representation of the past will always involve a “dialectic of presence and absence” (2006: 414). Therefore he gives oblivion its due in a section which, strategically, precedes his discussion of forgiveness. In keeping with contemporary approaches to memory and its interrelatedness with other faculties,5 Ricoeur refuses to address forgetting as an isolated category:

[F]orgetting is bound up with memory …: its strategies and, under certain conditions, its cultivation worthy of a genuine ars oblivionis result in the fact that we cannot simply classify forgetting through the effacement of traces among the dysfunctions of memory alongside amnesia, nor among the distortions of memory affecting its reliability. Certain facts we will discuss later lend credit to the paradoxical idea that forgetting can be so closely tied to memory that it can be considered one of the conditions for it. (2006: 426)

Ars memoria and ars oblivionis are jointly responsible for our representations of the past. It is tempting to identify Mnemosyne as a charitable goddess and Lethe as a cruel one; in fact, however, the latter may often be a merciful spirit, while the former may prove to be an unwelcome visitor. Ricoeur implies that what we remember is an end product of complicated negotiations between several psychological faculties which interact within the same self. As a consequence, it is necessary to develop another art – that of bearing in mind what is better to be forgotten and what is truly worth remembering.

Molly Allgood is a skillful artist and as an actress she is very good at remembering selectively. Also, she knows how to use her props. Throughout Ghost Light her recollections are fortified with several letters, either quoted from preserved copies or recited verbatim from memory (she calls herself “a professional rememberer”). One of the letters came from Synge; it is the only personal communiqué from him that she decides to hold on to against the demands of his family, a precious memento of the very beginning of their relationship. Eventually, Molly is prepared to part with this letter when pressed for money and yet, on second thoughts, she hands it to an aspiring young actress as a gift. The letter from Synge is an invaluable keepsake but it does not carry the emotional load associated with their relationship which Molly has stored safely in her heart and memory. Written at a very early stage of their acquaintance, the letter is a little out of tune with the overall tenor of Molly’s recollections.

In contrast to that letter, there is another one from Synge, a letter of apology, which Molly has memorized word for word, which clearly indicates how emotionally invested she still is in their relationship. Although the original is beyond her reach, she can easily quote every sentence decades after receiving it. This is even more surprising given the fact that otherwise her memory is rather defective. She confesses to herself that, once she got it, the letter affected her so much that she slept with it under her pillow for a year and always kept it close to her heart during the day. Still, Molly knows that this is one of those letters whose contents she can never disclose: when she considers what she could possibly share with the American scholar she realizes that no passage from their intimate correspondence with Synge should become public. She does not realize that several decades after her death Synge’s letters will be published and everyone will have access to his part of their correspondence.

Finally, there is the most peculiar letter in the entire novel, the one that Molly claims to have destroyed and yet which is discovered among her papers after her death. Strategically, this letter is of paramount importance not just because it constitutes a coda to Molly’s confession but also because it is a synecdoche of the whole narrative. It is a love letter, pure and simple. Touchingly and beautifully written, full of yearning and passion, with some triviality and facetiousness too, but altogether endearing and heartfelt. This letter is like Molly’s soliloquy itself: never meant to reach any addressee, a figure of loneliness and longing, and yet one that discloses the very naked truth about her relationship with Synge. When she reminisces about the circumstances of its composition, Molly remembers being convinced that it had to be destroyed precisely because of its sincerity:

And then you simply let go, writing anything you felt, for you knew you would never send it, lacked the mettle to be revealed. Foolish phrases came crowding. It didn’t matter now. Quotations from love poems, from songs that had come to mean something. As you watched the letter burn, you felt the strange, bright hope that its destruction would somehow anatomise the realities it had tried to describe. (O’Connor 2011: 121)

Molly’s paradoxical assumption is that she can be honest and outspoken about their relationship only as long as her outpourings reach no other person but herself. Even Synge, when he is still alive, is not supposed to learn what she truly feels. Ironically, Molly seems more at ease with a ghost which inhabits her memory after her lover’s death. Therefore the conclusion of the novel is rather disconcerting: it turns out that despite her protestations to the contrary Molly preserved the letter until her own death.

What is even more disconcerting about Ghost Light as a whole is that Molly’s entire soliloquy is but a figment of Joseph O’Connor’s imagination. In the final section of the novel, he expressly acknowledges Molly’s love letter, too, as well as several other elements, to be fictional (O’Connor 2011: 244). The original would indeed have been destroyed – there is no inconsistency in her mnemonic performance here – but O’Connor insists on doing justice to Molly’s memory (that is, to our memory of Molly) by imaginatively recreating her memories and corroborating them with invented evidence. At the end of the day, what he delivers is a bold experiment in posthumous ventriloquism. Not a historical reconstruction, but a historical fiction. Of course, reading Molly’s soliloquy is not tantamount to learning the actual truth about the past, and yet having read O’Connor’s novel we realize that even Molly’s own, first-hand account would not have been an entirely reliable story of what really happened between her and John Millington Synge. Therefore, what matters as a contribution to Irish remembrance culture is that Ghost Light offers us an edifying narrative about a woman who has been shamefully neglected by mainstream historical discourses. Ultimately, let us remember, O’Connor’s literary project of rewriting the past should not be judged by epistemological criteria. It is an ethical project sensu stricto.