“It was all a plot.” “I thought you liked plots … In any case, it's the plot of history. It was simply inevitable.” “But you helped inevitability along a little. …” “There's a process … It charges everyone a price for the place they occupy, the stands they take.”

Malcolm Bradbury

1.

“Plenty of poets are poets only … I am a historical poet,” Cavafy is reported to have said toward the end of his life (Lechonitis, 1977: 19). Possibly “historian-poet” would be a more accurate translation, in the context of this remark, of poietés istorikós. George Seferis found this remark, “like much of casual conversation, lacking in clarity” (1974: 340); but its bald ambiguity neatly epitomizes the mingled puzzlement and fascination experienced by readers who in the fifty years since Cavafy's death have engaged themselves with the poems in which the poet and the historian appear to change mantles. This happens in approximately half of all the poems that Cavafy published in his lifetime, and in a rather smaller proportion of those which appeared posthumously. The most pressing problem which these poems have created for readers can be summed up as that of relevance. The academic historian may confine his interests to “what happened in history,” but history for the layman, which must normally include the poet, more often begins at home. Characteristic lay attitudes to history are, first, that the past teaches us about the present and, secondly, that history repeats itself. And the twentieth century has discovered that history, when it repeats itself, becomes myth.

Cavafy wrote poems involving the past which conform to each of these expectations. And, partly on account of this, his use of history in his poetry as a whole has been interpreted either in terms of didactic allegory (Malanos, 1957, 1963; Tsirkas, 1958, 1971; and Dallas, 1974) or in terms of myth (Seferis, 1974; and Keeley, 1976). None of the readings proposed by these writers is fully adequate: Malanos and Tsirkas admit that their respective decodings do not work for all the poems, and concede that there is an element of pure antiquarianism in Cavafy. Seferis's farreaching conclusions are substantially based on the reading of only one poem, while Dallas and Keeley attempt to accommodate Cavafy's poetry as a whole within a generalized model which fails to account fully for all its features. Correspondingly, the poems in which Cavafy either uses history explicitly as an allegory or “alibi” for the present, or enters the world of myth and legend in which history most readily repeats itself, are very few. And all of them were written before the “watershed” year of 1911.

There are altogether twelve poems in which Cavafy presents or refers to the past more or less explicitly as a moral allegory for the present.¹ The opening lines of “Thermopylae,” in the Keeley and Sherrard translation, aptly demonstrate this (1901/1903: A103, tr. 12)²:

Honor to those who in the life they lead
define and guard a Thermopylae.

(my emphasis)

Not the historical pass defended and lost in 480 B.C., but any crucial moral “pass.” A Thermopylae is a historical metaphor for a contemporary and generalized dilemma, the purpose of the appeal to history is to illustrate a perennial, and especially a present, moral truth. Similarly, in others of these poems: “our efforts … are like the Trojans'” (A26, tr. 17); the soul is to watch out for “some Artemidoros” giving warning (A18, tr. 24); Theodotos, bearing Pompey's head on a bloodstained platter, may even now be entering the house of some neighbor (A21, tr. 40); at the end of the Odyssean voyage, “you” will have learned “what Ithacas mean” (A23, tr. 29). The distinguishing features of these poems are the use of the present tense or imperative mood, and a first or second person where the speaker and/or addressee is not specifically included in the historical context of the poem. Thus in “The Satrapy” (A16, tr. 23), the absence of a consistent context for the person addressed forces it on the reader's attention that he is “not necessarily Themistocles or Demaratos or any other political figure” (Lechonitis, 1977: 23), and so allows the historical references that are in the poem (Sousa, the Demos, and the Sophists) to be taken metaphorically. “You” in the poem, in other words, is as likely to stand for Cavafy himself, or the reader, as for Themistocles or Demaratos, or even, pace Dallas (1974: 56-63), Alcibiades.

Two poems which could be included in this group, but which may be considered as borderline, are “The Sea Battle” and “Waiting for the Barbarians.” “The Sea Battle” (1899: Anekdota, 121) represents the chorus of defeated Persians at the battle of Salamis whose ululations are taken from Aeschylus' play on the same theme, The Persians. In appearance, the poem is the dramatic monologue of the Persians shortly after their historical defeat, but it contains the lines:

Why must it be: no sooner does one possess
renowned Ekbatana, Sousa
and Persepolis, than he calls up a fleet
and goes toward the Greeks to fight at sea.

(my emphases)

“One” here introduces a generalized moral which seems to lift the “we” of the poem out of a specific historical context, to be reconstituted as a general moral exemplar. And “Waiting for the Barbarians” (1898/1904: A107, tr. 14) seems to belong with this group because of the unusual absence of any indications of context in place or time. Since we are not told which particular incident in the decline of an ancient civilization is being depicted, it is the generality of the picture that strikes us most forcibly; it could be (almost) any time, any place, and so we are irresistibly drawn to read the poem as a metaphor for contemporary civilization. This impression is reinforced by the use of the present tense throughout, without any distancing devices such as Cavafy uses elsewhere to place his speakers in perspective.

All of these twelve poems were written before 1911, most of them considerably earlier. After that date Cavafy wrote no more poems (at least that he chose to preserve) in which the past serves as a moral allegory for the present. But he continued writing poems which deal with history, from which, however, explicit reference to the present is absent. The approaches to Cavafy's historical poems proposed by Malanos, Tsirkas, and, to some extent, Dallas, assume that the past continues to function as an allegory for the present even where this is not explicit, although no reading based on the allegorical interpretation of history has so far claimed to account satisfactorily for all of Cavafy's historical poems.

In terms of the perception of history as myth, we encounter a more complex situation. This is partly due to the ingenuity of the poet George Seferis, who first identified the “mythical method” as the mainspring of Cavafy's use of history, and partly to the dominance of the poetics of the Anglo-Irish-American writers Yeats, Joyce, Eliot, and Pound in both Greece and the English-speaking world for much of the last few decades. Before discussing the poems in which Cavafy makes use specifically of the mythical rather than the historical past, it will be necessary to summarize the debate initiated by Seferis and to comment on the position that Cavafy treated history as myth in order to “redeem the time.”

In a now historic lecture given at the British Council in Athens in 1947, Seferis gave striking evidence for parallels between what he called the “sense of time” revealed in Cavafy's poems, and that of T.S. Eliot (Seferis, 1974). Taking as his starting point the poem, “Those Who Fought for the Achaean League” (1922: B31, tr. 86), Seferis pointed out that this poem had been published, and probably written, in February 1922, shortly before the catastrophic defeat of the Greeks in Anatolia in that year. The poem is an epitaph for those soldiers of the allied Greek city states (the Achaean League) who died fighting against the Romans in 146 B.C., the year in which Greece finally became subjugated to Roman dominion. For Seferis, this epitaph for the victims of an ancient defeat for Hellenism was at the same time an epitaph for the coming disaster in Anatolia, while the “tailpiece” of the poem, ascribing the epitaph to an Achaean writing in Alexandria some thirty years after the disaster, presents Cavafy, a Greek in Alexandria, lamenting powerlessly from the wings.

From his reading of this one poem, Seferis went on to argue that Cavafy used what T.S. Eliot had termed the “mythical method,” and that he did so “systematically” in order to refer to the condition of the present by invoking the historical past. He concluded, “there exists [in Cavafy's poetry] a sense of temporal co-identity; the past is identified with the present and perhaps with the future,” exactly as Eliot had written in “Burnt Norton.”

Seferis's lecture provoked, within a matter of months, an intemperate but intelligent reply from Timos Malanos (1963: 121-61). Here it must be said that although Seferis is by far the more sensitive critic of poetry, all the evidence is on Malanos's side. The poem about the Achaean League, Malanos reminds us, must, at the latest, have been written some months before the final collapse in Anatolia, at a time when the outcome of the Greek campaign and the nature of the disaster that was to follow were becoming increasingly foreshadowed but not yet inevitable: in any case, it was hardly time to write the epitaph. More telling still, when the storm broke fully in August 1922, it produced not the slightest ripple in the poems that Cavafy published later that year. And Malanos had also read Eliot and been struck by how different was the “sense of time” of each of the two poets: “Cavafy refers to History in the manner of the reflective man, who experiences retrospectively its unique time: time past. (This is, besides, the tense of all of his poetry). On the other hand, every time that Eliot refers to History, he does so in the manner of the prophets of the Bible, his vision in parallel inspiredly equates three times: past, present and future” (1963: 150).

Malanos had already proposed that a good many of Cavafy's historical poems can be understood in terms of an alibi for personal obsessions. But he had also recognized that there were other poems which could only be termed “entirely historical,” that is, which had no reference either to the poet's personal life or to the present more generally. Malanos spiritedly defends his own reading of these latter poems and restates it as an effective counter to Seferis: “The way in which Cavafy perceived time … is none other than the way in which it has been perceived by all our Historians—of whom the foremost representative is Paparrigopoulos—who perceive, like an unbroken line from Homer to 1821, the History of the Greek Nation” (Malanos, 1963: 135).

Seferis's more general application of the “mythical method” to Cavafy's poetry is no more solidly founded than is his interpretation of the poem about the Achaean League. In his essay, he quotes part of Eliot's review of James Joyce's Ulysses, in which the term “mythical method” was first used and defined. The “mythical method” has often been discussed without reference to this context. Eliot, when he coined the term, was reviewing a novel which, instead of a plot, depended for its structure on following, in a modern context, the pattern of an ancient myth: Odysseus' homecoming to Ithaca, superimposed on the peregrinations of Leopold Bloom around Dublin on a June day in 1904: “In using the myth, in manipulating a continuous parallel between contemporaneity and antiquity, Mr. Joyce is pursuing a method which others must pursue after him. … It is simply a way of controlling, of ordering, of giving a shape and a significance to the immense panorma of futility and anarchy which is contemporary history” (Eliot, 1923).

When this definition is applied to Cavafy, we are bound to notice that myth, in the sense that Eliot uses the word to refer to the archetypal and infinitely repeatable story of Odysseus, is not prominent in Cavafy's poetry. That there is a “continuous parallel between contemporaneity and antiquity” in a body of historical poems which are, in the most literal sense, discontinuous seems on the face of it scarcely probable, and, in any case, one half of the equation is missing: there is no counterpart to Joyce's contemporary Dublin in poems which describe only the ancient world. Finally, even were myth more frequently present in Cavafy's historical poems, it is difficult to see how it might function in them to give shape and significance to contemporary history. The past history with which they deal is itself chaotic and futile enough.

The past in a poem by Eliot, or, for that matter, by Seferis, is most often the mythical or mythicized past. When these writers refer to history, they tend to treat it in terms of myth—to reduce the sense of the contingent, unique, and irrecoverable of the historical moment by moving it toward a system whose components are “meaningful,” symbolic, and cyclically repeated. Their procedure is to detach the contingent moment from its historical context by an (at least partial) integration into the cyclical, synchronic pattern of myth. In consequence, it acquires the status of a symbol. The progression is the same as that from the events of the Passion to the Easter liturgy. There is no clear sign of this type of movement in Cavafy's poetry, as there is no counterpart to the Grail myths, Tiresias, or the Tarot pack in “The Waste Land,” nor to the Odyssean quest that animates the poetry of Seferis.

A cogent attempt is made by Edmund Keeley (1976: passim) to uncover a counterpart for these uses of myth in Cavafy, through a determined and consistent reconstruction of “Cavafy's Alexandria” as a “myth in progress.” But although the “myth in progress” is, in Keeley's hands, a powerful metaphor for Cavafy's life work, Keeley is unable to demonstrate that the metaphor is also present in the poetry. This is in effect an imaginative treatment of Cavafy parallel to Eliot's of the Thames or Seferis's of the Aegean: namely, to integrate the poetry into a preexisting mythical pattern. The crucial difference remains that Eliot, Seferis, and Joyce themselves reveal this process as an organizing force within their texts, while there is no real evidence that Cavafy's texts, singly or as a whole, attempt this kind of integration. Myth in this reading of Cavafy becomes a “way of controlling and ordering” Cavafy's texts, rather than a device used within the texts for the controlling and ordering of history or personal experience.³

We must now turn to those poems of Cavafy which make explicit use of myth; and, like the “moral allegory” poems, these turn out to be far fewer in number than might have been supposed from the importance given to this element in Cavafy criticism. Allowing for a somewhat generous definition of myth, that includes stories of literary origin either based ultimately on mythical sources or having acquired a currency and symbolic status since first being written which goes far beyond their explicit historical context, or both, there are no more than twelve of these poems, of which two have already been encountered under the heading of “moral allegory.”⁴ All were written very early in Cavafy's career, in the decade between 1893 and 1903 (assuming the dating of 1894 for the first draft, now lost, of “Ithaca,” which would otherwise have been written much later than the others, in 1910). Several of the unpublished poems in the group use non-Greek material, but of the seven poems dealing with Greek mythology, it is interesting to note that five refer to characters and situations best known from the Iliad, while only one, “Ithaca,” refers to the Odyssey. The Iliad has always been felt to be the more historical of the two ancient epics. Of the remainder, “When the Watchman saw the Light” has its source in Aeschylus, while “Interruption” makes only passing reference to a number of ancient myths.

The last of these poems to be written, “Bad Faith,” clearly reveals the direction in which Cavafy's treatment of myth was tending, and perhaps also the reason why Cavafy never made use of myth again (1903/1904: A109, tr. “Unfaithfulness,” 13):

In the midst of Thetis' wedding to Peleus
Apollo stood up on the glittering marriage
board, and gave the newly-weds his blessing
for the future issue of their union.
He said: Never shall sickness touch him
and he shall live long.—His saying this
pleased Thetis greatly, because the words
of Apollo so well versed in prophecies
appeared to her a surety for her child.
And as Achilles grew to manhood, and as
his looks became the pride of Thessaly,
Thetis kept the god's words in her mind.
But one day came elders bearing tidings,
and told how Achilles had been killed at Troy.
And Thetis tore her purple robes in pieces,
flung from her and trampled
in the dust her bracelets and her rings.
And in her bitter grieving she remembered times gone by;
and asked them what was wise Apollo doing,
whereabouts was the poet who jumps on tables
to make fine speeches, whereabouts was the prophet
the day her son was slain in the bloom of his youth.
And the elders made answer that Apollo
himself, none other, had gone down to Troy
and with the Trojans killed Achilles.

The lengthy epigraph from The Republic which prefaces the poem does much more than refer us to the ancient source for Cavafy's treatment of the myth of Achilles (which is better known from the Iliad). It also refers to what is probably the first recorded rational critique of myth and its function in society, namely the part in The Republic where Socrates argues that only morally edifying forms or variants of myths should be allowed currency in the ideal state. The reference of this epigraph is precisely to the historical moment in Greek—and European—culture of the first rational attempt at curtailment of the mythic function.

The text of the poem that follows retells in narrative form the story contained in Thetis' words quoted in the epigraph from Aeschylus' lost drama. The subject matter belongs certainly to myth, but the even, reasonable tone of the narrative voice, with its controled detachment from either the joy or the grief of which it tells, resembles nothing so much as, in Malanos's words, “the linguistic idiom of the Historian” (1963: 153). This poem, ostensibly mythical, in fact probes the non-rational world of myth with the rational tools of the historian to expose the paradox of myth as a contradiction in reason and in rationally-grounded morality. Hence the poem's title, which alludes not only to Apollo's “bad faith,” but to the impossibility experienced by Socrates (in the epigraph) of any longer having faith or belief in such ambiguous material as inherited mythology. Cavafy “demythologizes” myth in this poem by rewriting it in terms of historical discourse. This is the exact opposite of the approaches to myth adopted by Yeats, Joyce, Eliot, and Seferis.

We are now left with the great bulk of Cavafy's historical poetry in which no reference is made either to the moral lesson to be drawn in the present or to the renewing power of myth. Allowing for some argument about borderline cases (is the speaker of “Ionic,” [1896/1905/1911: A53, tr. 28] placed in historical time or not?), there are, by my calculation, some seventy-six poems in the published canon and seventeen of the posthumous poems to which the following definition of the historical poem can be applied: a poem which narrates or mimetically represents actions and/or words (real or imaginary) which are more or less precisely fixed, historically, in place and time, and in which no apparent reference is made to the modern world. Apart from the “moral allegory” and mythical poems already discussed, this definition excludes only four poems in which the historical past is mentioned but plays a different role: “Caesarion,” which will be discussed below, and “Párthen” (1921: Anekdota, 183), which partly reduplicates it in technique; and “Days of 1909, '10, '11” (?/1928: B73, tr. 117) and “Following the Recipe of Ancient Greco-Syrian Magicians” (?/1931: B87, tr. 126), in which, in a contemporary context, a contrast is drawn between past and present.

How are all these historical poems, so defined, to be read? If history is not invoked in order to teach us in the present or to be “redeemed” in the guise of myth, there remains no other solution but that of Malanos after he had exhausted all of Cavafy's “alibis.” But if one is content to accept this and go no further, it is difficult not to share in Seferis's disappointment with a poet who merely “versified history.” We must go back to Cavafy's designation of himself as a “historian-poet” for the answer. And since the layman's or poet's attitude to history has proved inadequate to account for the role played by history in the majority of the poems, we must look now to the academic historian. Indeed, this was Malanos's solution, but Malanos seems to have had in view only one of the questions asked by the professional historian: “What happened in history?” Applied to poetry, this view of history could offer only the consolation of antiquarian dabbling, or the escapism of the historical novel, and Seferis was right to be discouraged at the prospect. But there is another kind of question implied, and today increasingly stated, in the work of the historian, namely: “What is history? How does the historical record enable us to understand the past? And what is the nature of the historical record?” These are questions that in all eras have a vital bearing on the present, and that determine the nature of our relation to the past and how we perceive it. What is represented, questioned, and constantly reexamined in Cavafy's historical poems are not the facts of history (“What happened?”) but history itself (“How is history made?”). Cavafy's historical poetry stands in the same relation to versified history as War and Peace does to the historical novel.

2.

Time is a ubiquitous preoccupation in Cavafy's poetry, and Malanos only exaggerated slightly in saying that time in Cavafy is always time past. It is this preoccupation with perceiving the past that unites Cavafy's historical poems with the scarcely smaller part of his oeuvre devoted to erotic experiences or imaginings set in the poet's own personal past. The treatment of time in the erotic poems is discussed elsewhere in the present volume; here it need only be said that a constant and unifying theme in these poems is the perception of time-as-experienced. In the process of making poetry, the transcendent present can only ever be the moment of writing; the present of writing is made transcendent by the permanence of the written text, while all lived experience, no matter how burningly recent, belongs at that moment to the past, and is therefore excluded from transcendence. But by a complementary process, experiences which the poet himself seems to regard as sordid and trivial in fact ironically become the raw material of “high art” because the true subject matter of the poems is not what was experienced but the artistic endeavor in the present, in which the awareness of loss to the past is set against an act of memory and imaginative re-creation.

The best way of tackling the problem of Cavafy's historical poems is to start from the treatment of time in the erotic poems. It is indicative that the two recorded fragments of conversation in which Cavafy spoke of himself as a historian manqué begin with remarks about time in his personal poetry. The first, dated merely to the last ten years of his life, states:

“… I am a poet of old age. The most lively events do not inspire me at once. First, time must pass. Then later I remember them and am inspired.

“Plenty of poets are poets only. Porfyras, for example, is a poet only. But not Palamas. He has written stories. I am a historical poet [or a historian-poet]. I could never have written novels or stage plays: but I sense within me 125 voices telling me that I could have written history. But now it's too late.”

The second is dated April 8, 1929:

“… With me the immediate impression does not provide the impulse for work. The impression must become part of the past, must be falsified of itself, by time, without my having to falsify it.

“I have two aptitudes. To make poems and to write history. I have not written history and now it is too late …”

(Lechonitis, 1977: 19-20; my emphases).

In neither case does the context fully account for the jump from Cavafy's method of distilling personal experience into poetry to his fascination with the historian's approach to his very different task. (The starting point for both conversations is Cavafy's extreme touchiness at any suggestion that, in his sixties, he had passed his poetic prime.) That the jump is made in two separate conversations that happened to be recorded clearly shows that the two things were closely linked in Cavafy's mind. If we say that the unifying theme of his erotic poetry is time as experienced loss, balanced or opposed by an act of memory and imaginative creativity, then the jump from the personal to the historical sphere would seem to be an extension of the same theme beyond the limits of what can be experienced as memory. Taking this starting point in the personal experience of time, the conundrum to which it leads would then be: how does one perceive historical time? That is: the idea of time beyond one's natural lifespan? Seen in this perspective, Cavafy's remark that other poets are “poets only” takes on an added hauteur.

This extension of the perception of time from personal experience to the historical plane is dramatized in a single poem, “Caesarion,” a poem which eludes my definition of the historical poems, and is like no other “erotic” poem of Cavafy's, but shares features typical of both (1914/1918: A69, tr. 58):

Partly to verify a date,
partly, too, to pass the time,
last night I took in hand a corpus
of Ptolemaic epigraphs to read.
The boundless praises and the flatteries
of all were much the same. All were glorious,
victorious, mighty, gracious in deed;
their every enterprise most wise.
As for the women of that ilk, they too,
Berenices and Cleopatras, all were wondrous.
When I had succeeded in verifying the date
I should have put down the book had a tiny mention,
quite insignificant, of King Caesarion
not drawn my attention suddenly …
Ah, there, you came with your indefinable
air of charm. In history a few
lines are all that speak of you,
that way I shaped you the more freely in my mind.
I shaped you beautiful and finely sensitive.
My art infuses your face with
dreamlike, congenial beauty.
And so fully did I imagine you,
that late last night, as the lamp
burned down—I left it on purpose to burn down—
I thought you came inside my room,
it seemed to me you stood before me; as you would have been
in conquered Alexandria,
wan and weary, ideal in your sadness,
hoping still that they might pity you
who basely were whispering of “too many Caesars.”

The poem's title refers to a historical character and situation, its beginning refers to historical source material, and its end to a specific historical situation which obtained in Alexandria in 31-30 B.C.; but although it refers to history, “Caesarion” is not a historical poem. Despite its title, it refers to the personal experience, with restrained erotic or at least hedonistic overtones, of the poet. The meeting point of the personal/erotic and the historical lies in the fact that the personal experience is the reading of a historical source, and its aftermath.

From the opening lines, two kinds of time, or two perceptions of time, are set up simultaneously. Both belong to the past, and both are precisely defined, but with different frames of reference. The opening, “Partly … partly,” is a favorite device of Cavafy for keeping distinct or contradictory ideas in play.⁵ The first “partly” introduces (impeccably) the language of the historian, and a perception of time as a fixed, and conceptual, series, in which the interest lies in establishing the correct relations between facts which are complete and unalterable: the series is closed. The activity described in this first line belongs to the academic historian; it is not part of everyday experience for most of us. The second “partly” introduces perhaps the lowest common denominator of all living experience, “to pass the time.” But in juxtaposition to “verifying a date,” the commonplace has a special significance. Over and above their most common, banal usage, the words mean literally to experience one's life as a temporal series. Time in the second line is an open series, the succession of one moment by another that puts all experience beyond our grasp as soon as it is registered as experience, and it ends only with the death of the experiencing “I” (one is reminded here of the early poem “Candles”). Two kinds of time have been established: recorded time, which is perceived as a closed series; and experienced time, which remains an open series.

If the first two lines establish, respectively, historical time and time-as-experienced, the two lines which follow repeat the parallel, but in reverse order. The third line refers to experienced time, the fourth to historical. The third line establishes the poet's experience as having happened “last night”; that is, it is perceived only in relation to the unspecified present which produced the text. (To have given a date for his personal experience, as Cavafy does in many of his erotic poems, would have been to reach out from time as experienced to time as recorded, that is, to a wider system of historical interrelations independent of the experiencing “I.” On the other hand, the life and early death of Caesarion cannot be detached from the epoch of the Ptolemies. Caesarion exists in the present only through the historical record. His unfortunate role in history, which sealed his fate, is all, ironically, that survives of the living youth.

The achievement of the rest of the poem is to engineer a convergence between the two sorts of time which had first been set up as parallel. Among the historical epigraphs, the mention of Caesarion stands out, paradoxically, for the king's historical insignificance. Because he was only a victim and not a maker of history, he is denied the platitudinous epithets that deprive the other Ptolemies of any possibility of human individuality. This silence on the part of the historical record provokes the leap whereby the poet experiences the imaginative re-creation of a historical personage, as if from something out of his own lived past. The convergence of the two kinds of time in the poem is achieved in the second section with its juxtaposition of a “mention” in a historical text and the unexpected adverb “suddenly,” located in the immediate past. The poem imaginatively reconstructs two pasts—the poet's own (“last night”), and the historical past of Caesarion, which, “last night,” the poet had imaginatively reconstructed as if it had belonged to remembered experience. The poem proposes an analogy between the perception and re-creation of personal memory (located in the recent past, “last night”) and the perception and reconstruction of a past foreclosed as part of history and remote beyond the range of personal memory.

It is not that history becomes present: Cavafy's imagination re-creates Caesarion in the poet's own room and time, but it does not in the process detach the king from the remote historical context in which he lived and died. “And so fully did I imagine you” (experienced time) is completed not by “you stood before me,” but by the lines following “as you would have been.” The imaginative achievement consists not in conjuring up a beautiful youth from the recesses of history (a poet who was “a poet only” could have done that), but in juxtaposing experienced and historical time in such a way that each comes to be seen in terms more usually appropriate to the other. In the last nine lines, it is the poet's experience of only “late last night” that is depicted with the finality of the past tense appropriate to the closed series of distant historical events; while the verbs relating to Caesarion “in conquered Alexandria” are deprived of time-reference in usages more appropriate to the experienced present, looking uncertainly into a future that has not yet been foreclosed: “you would have been,” “hoping still,” “they might …,” “who were whispering.” Experienced time becomes historically fixed, remote and determined, while historical time is experienced, not directly in the present of writing (as happens, arguably, with Eliot or Seferis), but in the immediate past in which ordinary, personal experiences are also perceived.

Time, which is perceived in irreconcilable ways in the words “to verify a date” and “to pass the time,” is finally seen as indivisible: history is experience and experience is history. What this means for the “historical poet” is that his exploration into the workings of time, although confined to the past, is not limited to the operation and span of personal memory, but may range freely through the millennia of recorded history. It may be argued that the achievement of the historical poems proper (of which “Caesarion” is not one) is to scrutinize the historical record through the perspective of time-as-experienced, and, in so doing, to raise searching and perhaps unanswerable questions about the nature of the historical process.

3.

Cavafy's historical poems range across a well-defined space, both geographically and in historical time. Geographically, this space is bounded by Rome in the west and by the limits of Hellenistic penetration in the east,⁶ but its capital city is Cavafy's own city of Alexandria. In time, once the poems dealing with myth and those treating history in terms of moral allegory are set aside, it becomes evident that none of the historical poems proper deals with a period earlier than the conquests of Alexander the Great and the consequent expansion of Hellenism, while an appropriate terminus in the other direction is set by two poems relating to the fall of Constantinople in 1453 that brought the Byzantine Empire to an end.⁷

The historical record for this long period (the “decline and fall” of classical Hellenism?) exists in two forms: in surviving inscriptions and papyri, and in the synthetic work of historians from Arrian to Sphrantzes, whose writings have been copied and preserved through the centuries. In order to see how Cavafy's historical poems probe and even subvert these forms of historical testimony, we must now look first at those poems dealing with ancient inscriptions, and then at others which focus more or less directly on the writing of history.

The experiment of writing poems in the form of mock-ancient epigrams is taken to its farthest extreme in the poem “In the Month of Athyr” (1917: A79, tr. 55):

With difficulty do I read                    on the ancient stone.
“Lo[r]d Jesus Christ.”                    A “So[u]l” I can distinguish.
“In the mo[nth] of Athyr”                    “Leukio[s] wa[s lai]d to rest.”
The evidence for his age                    “Years he lived,”
the letters XXVII reveal                    that as a young man was he laid to rest.
Among the lacunae I see                    “[H]im … Alexandrian.”
Then come three lines                    badly mutilated;
but some words I can make out—                    such as “our [t]ears,” “anguish,”
then “tears” again,                    and “to u[s his f]riends who [m]ourn.”
It seems to me that Leukios                    must greatly have been loved.
In the month of Athyr                    was Leukios laid to rest.

Here no difference is made between the time of reading the inscription and of writing the poem: the poet deciphering the ancient stone belongs to the present of the text (“now” to both poet and reader), and the experienced past is not brought into play as it was in “Caesarion.” The present tense, however, encompasses two activities, necessarily simultaneous, which are normally thought of as mutually exclusive. The “I” in the poem is at once deciphering an inscription, one of the drier aspects of the professional historian's task, and writing a poem, the province exclusively of the “inspired” creator. There can be no more striking demonstration of what Cavafy meant by calling himself a “historical poet.” The present tense of “I read” in the first line must logically imply another present tense: “I write”; and the existence of the fragmentary ancient text in the lines which follow implies yet a further present tense, which teasingly has been obliterated from the worn inscription, governing the words “tears, anguish, us who mourn.” The ancient text is the testimony to a moment in which the grief of the dead man's friends was a present experience, when his life, death, and their love for him, were part of the experienced past of the writer of the inscription.

The poem's impact lies in the incompleteness of the inscription, which is the more forcefully presented by the use of square brackets with which the epigraphist indicates those parts of the text which have had to be supplied by conjecture. Between the two presents, of the friend of Leukios composing his epitaph, and the “I” of the poem reading that epitaph and simultaneously composing the text before us, lies the gulf of time which has so ravaged the ancient stone. The present of the inscription is so remote that the name of the month (which places it in the familiar yearly cycle of time) is unfamiliar, while the fragmentary state of the inscription belies the permanence of the written word, which is an article of faith for historian and poet alike.

A more complex poem is “Kimon Learchou, Aged 22, Student of Greek Letters (at Cyrene)” (1913?/1928: B69, tr. 115). Here, the title provides the kind of red herring of which Cavafy was sometimes fond: its apparent fullness and precision in fact tell us little about the speaker in the second part of the poem, to whom it refers, and mention of the city of Cyrene does no more than reinforce the impression of a historical epoch which is not closely specified: the urban world of the Mediterranean in the late Hellenistic or Roman period.

“I met my end                    in my hour of happiness.
Hermotéles had me                    as his inseparable friend.
The last days of my life,                    for all that he affected
not to be concerned,                    I often sensed
his eyes full of tears.                    Whenever he thought
I had drifted asleep                    he would fall frenzied
on the edge of my bed.                    But we two, we were
boys of an age,                    we were twenty-three.
Fate is the traitress.                    Perhaps some passion
for another would have                    taken Hermotéles from me.
My ending was well;                    in love indissoluble.”
This epitaph                    for Marullus Aristodemus
deceased a month                    ago in Alexandria,
I mourning received,                    I, his cousin Kimon.
It was sent me by the writer,                    a poet of my acquaintance.
He sent it me because                    he knew I was
a relative of Marullus:                    that was all he knew.
My soul is full                    with sorrow for Marullus.
We grew up                    together, we were like brothers.
I am profoundly sad.                    This untimely death of his
every thought of malice                    has quenched within me …
every thought of malice                    against Marullus—for all that
he had stolen from me the                    love of Hermotéles,
and now should Hermotéles                    desire me once again
it will not be at all the same.                    I know my nature,
how susceptible I am.                    Marullus' ideal form
will come between us,                    and I shall think he
says to me, See here, now                    you should be satisfied;
See here, you've got him back                    as you wanted, Kimon;
See here, you've no more                    reason now to slander me.

Kimon does not in the end tell us whether the epitaph with which the poem opens was ever inscribed on his cousin Marullus' tomb. But it was written with that purpose, and so to immortalize Kimon's presumed grief over his cousin's early death. The touchingly sentimental epitaph can be read in the same spirit as the poem “In the Month of Athyr,” as a human testament at once present and infinitely remote. But the poem becomes more interesting when the speaker tells us that the author of the epitaph was a poet who knew almost nothing about the dead man. The purpose of the epitaph, potentially a tiny piece of the source material of history, turns out to have been not so much to record the facts of the young man's death, as to console or possibly to flatter a bereaved relative. And the irony is that the poet's well-meant shot in the dark has misfired badly: Kimon's true feelings are dominated by bitterness toward his dead cousin over the very relationship so idyllically extoled in the poet's epitaph. And the epitaph, immortalizing that relationship in verse, puts Kimon's lover Hermotéles forever out of his reach.

One effect of this poem is to cast doubt on the truth of the historical record as a transcription of actual experience. But it also introduces a new figure into the business of recording history: the poet. Since he knows almost nothing, and in any case could never know the literal truth of what he has written in the epitaph, the poet can only have created the whole text out of his imagination. Truth to fact (historical truth) is blatantly subverted by the interfering well-wisher, the poet. But what the poet has written has a power and permanence that have nothing to do with its truth: thanks to the poet's work, the dead Marullus will always come between Kimon and the object of his desires, Hermotéles.

Two poems which deal not with inscriptions but with the business of writing history are also important here. The first is “King Demetrios” (1900/1906: A27, tr. 18):

Not like a king, but as an actor, he disguised himself in a dark cloak instead of that tragic costume, and slipped way unnoticed.

Plutarch, Life of Demetrios

When the Macedonians deserted him
and proved they would rather follow Pyrrhos
King Demetrios (great
was his soul) did not at all—so it was said—
behave like a king. He went
and doffed his golden garments,
kicked off his royal sandals
all of purple. In simple clothes
he quickly dressed and fled.
Doing as an actor does
who when the show is over,
changes his clothes and comes away.

The essence of this seemingly simple poem is that Cavafy's text retells precisely the same facts—no more, no less—as the historian Plutarch in the “Life” from which the epigraph is taken, but completely changes their meaning. The historical record (unimaginatively) condemns the loser who, in defeat, behaves not like a king but with the shallow art of the actor, and for no better reason than to save his skin. Without changing any of the facts, or even greatly departing from the historian's dry manner of narration, Cavafy presents a very different interpretation. The difference lies in the interpolation (“great was his soul”) and in the last two lines. Where Plutarch's king adopted a disguise in order to conceal his true identity and “slip away unnoticed,” Cavafy's king behaves like an actor in that, at the end of the show, he takes off his disguise. In Cavafy's poem, it is the trappings of kinship that constitute the disguise,⁸ and the man within is therefore entirely honorable in modestly doffing them and leaving the stage of history when he has played his role to the end. Cavafy's poem aptly illustrates how even the baldest narrative of historical fact involves dramatic re-creation and moral evaluation. The juxtaposition of the two historical narratives, Plutarch's and Cavafy's, is a sly hint (more fully developed elsewhere) that it takes a poet to write history properly.

Of all Cavafy's historical poems, “Anna Comnene” refers most directly to the process by which history is written (1917/1920: B20, tr. 79):

In the prologue to her Alexiad, Anna Comnene
her widowhood bewails.
Her soul is in turmoil. “And
with copious tears” she tells us “do I bathe
mine eyes … Alas the stormy billows” of her life,
“alas for its reversals.” She is seared by anguish
“unto the bones and marrow and her soul's wrack.”
The truth however seems to be that one grief only
and that a mortal one was known to this power-avid woman:
one deep sadness and one only weighed
(though unconfessed) on this indomitable Greekling,
that she never managed, for all her dextrous grip,
to gain the Crown; it was snatched from her
almost from within her grasp by the upstart John.

Again the poem is relatively simple, and, on the face of it, all it does is repeat the judgment of one historian (Gibbon) on another. But in the context of Cavafy's historical poems and historical interests, the poem has more to say: it shows that the writing of history is itself part of history. It is the same human impulse that creates an empress and a historian. Not only that, but it was the historical accident of Anna's failure to become empress that gave us one of the richest historical sources for an important period of medieval history. Inevitably, later historians have asked themselves: how reliable is this embittered and self-conscious mourner for the past glories of her father's reign and her (then) hopes of future power? Cavafy's poem epigrammatically goes to the heart of the personal experience which is not only distorted by the historical record (the real reason for Anna's grief), but which actually calls the historical record into being. The experiences of Anna's early life are transformed into written history, and, it is implied, imaginatively re-created for the purposes of self-justification as a response to personal loss. Seen in these terms, the poem establishes a close parallel between Anna's activity in writing history and that of the poet, as it appears for instance in poems such as “Understanding,” “Their Beginning,” and “I Brought To Art.”

In probing, in these poems, the nature of the historical record and, so, of our perception of the past, Cavafy frequently alludes indirectly to the activity of the poet or creative artist. Many of his poems set in a historical context have poets or artists as their main figures, and, in doing so, bring into play the ideas of art, time, and the perception of time which are more explicitly mentioned in the personal poems, whose sphere is memory or time-as-experienced. One of the fullest explorations of these themes in the context of historical time is the poem “Dareios” (1917/1920: B18, tr. 78):

The poet Phernazes is working on
the important part of his epic poem.
How the royal throne of Persia
fell prize to Dareios son of Hystaspes. (From him
is descended our glorious king,
Mithridates, Dionysos and Eupator). But here
philosophy is needed; he must analyze
the emotions that Dareios would have had:
perhaps intoxicated pride; but no—more likely
recognition, in some sort, of the vanity of greatness.
Deeply the poet ponders the matter.
But he is interrupted by his servant entering
at a run, delivering news of utmost seriousness.
War has broken out against the Romans.
The most part of our army has crossed the frontiers.
The poet is aghast. Disaster!
How now can our glorious king,
Mithridates, Dionysos and Eupator,
concern himself with poems in Greek.
In wartime—imagine, poems in Greek.
Phernazes is beside himself. What a blow!
Now that he was certain with his “Dareios”
to make an impact, to dumbfound
for good his cavilling detractors.
What a postponement, what a postponement of his plans.
And were it but a postponement, no harm done.
But let's see if we have secure defenses
at Amisós. It isn't a city remarkably well fortified.
They are the most dreadful enemies, the Romans.
Can we hold our own against them,
we Cappadocians? Can such a thing be?
Are we to stand now against the legions?
Great gods, protectors of Asia, grant us aid.—
Yet in the midst of all his confusion and the uproar,
insistently the poetic idea goes to and fro—
most likely is, of course, intoxicated pride;
intoxicated pride would have been the mood of Dareios.

Characteristically, the poem is not about Dareios, as one might have expected from the title, but about an imaginary poet, Phernazes, writing a poem about Dareios, who had lived some four centuries before him. “Dareios” is a historical poem about the writing of a historical poem. The historical allusions place the main character, Phernazes, in a context of Pontos in Asia Minor shortly before the defeat of its last independent king, Mithridates VI, by the Romans in 71 B.C. The fictitious poet is writing an epic (the closest of the traditional poetic genres to history) about his ruler's ancestor, Dareios the Great, who ascended the throne of Persia in 521 B.C. But he is interrupted at the point where his historical narrative must be supplemented by gifts peculiar to the poet: “philosophy” is required, in order to analyze the “emotions” of his historical character: this, and not merely the crucial moment of Dareios' accession to power, is the “important part” of Phernazes' work.

In “Dareios,” a poet is interrupted in the composition of a historical poem by the intervention of history. Part of the poem's effect lies in the present tenses which depict Phernazes' predicament in terms of time-as-experienced, and which, in the same manner as in the poem “In the Month of Athyr,” link Phernazes' present with the present of the writing of the poem. The present tense here is not a “historic present”; it is closer to the present tense of interior monologue, where it is to be supposed that the events themselves and the recording of them are simultaneous. The narrative voice, although in the third person, depicts a series of present moments as they succeed one another in Phernazes' consciousness. In the act of writing, which, as we have seen, constitutes the only true present tense for the poet, the fictitious Phernazes experiences time which to Cavafy, and us, is historical: the same epoch is presented as at once historical (fixed in a pattern already fully determined) and experienced (in terms of an open future—Phernazes does not know what is going to happen).

The present of the poem (Cavafy's) includes the awareness of Phernazes' present as fixed in a pattern which could only be perceived long after the event. This dual perception of a historical moment is poignantly reinforced by the nature of the dilemma facing Phernazes in Cavafy's poem. He, Phernazes, faces the same problem that Cavafy, in bringing Phernazes' consciousness to us, simultaneously surmounts: how to bridge the gulf between the historical and the experienced past, how to depict the “emotions” (the nature of the experience) of a historical character at a historic moment. It is perhaps for this reason that Cavafy's poem bears a title apparently more appropriate to Phernazes' work-in-progress than to his own completed poem.

The poetic function fulfilled or attempted by Cavafy and his creation, Phernazes, seems then to be to recuperate historical time as experienced time, and bring it within the ambit of personal memory and the individual imagination with which the poet assimilates and refurbishes his own experience. History interrupts the poet's act of creativity, but that act was itself partly dedicated to the recuperation of the historical past, and is in turn recuperated for us as a part of history. Even the eruption of history in the form of real and decisive events is not unambiguously treated. D.N. Maronitis, in a perceptive essay on this poem (1970), goes beyond what is explicitly stated in the text and interprets Phernazes' dilemma as, in part, a pragmatic one: will the attribution of “intoxicated pride” to his illustrious ancestor be flattering to the king who is Phernazes' patron? (According to this reading, “philosophy” in the seventh line also means, ironically, something like “circumspection.”) Maronitis further believes that the last four lines of the poem are causally connected to what has gone before: because of what has happened, Phernazes makes up his mind to write, after all, that it was “intoxicated pride” that Dareios experienced. According to this reading, the historical tide that in the poem begins to sweep away Phernazes and “Greek poems” together with the kingdom of Pontos, in fact liberates the poet. It inspires him with the correct and satisfying solution to his artistic dilemma, but it simultaneously removes altogether the pragmatic dilemma that had faced him. If Mithridates is never going to read the poem and reward the poet, if, most probably, the poem is never even going to be finished, then what need is there to flatter anybody? What need, either, to worry about dumbfounding his critics?

Whether Maronitis's additional subtlety is accepted or not, what is ultimately validated in this poem seems to be the “poetic idea” of the third from last line. This is what matters: the act of creativity, whether the writing of verses, the making of wine bowls, or the fashioning of statues. The creative act is more important than the artifact in which it results: the insistence of the poetic idea that haunts Phernazes is more important than the poem he would in other circumstances have completed, and which would most probably have suffered, in due course, the fate of the fragmentary inscription of “In the Month of Athyr.” Poems and artifacts are not proof against time, as that poem movingly demonstrated. But the capacity to create anew, with its fickle relation to the contingent external forces that comprise historical fact, is ever-present, and in fact is no less necessary for the production of the historical record than for that of poems and fiction. Phernazes' poem, presumed unfinished, stands beside the palaeolithic cave paintings inscribed in inaccessible niches as a monument to a human impulse and power, whose value resides less in the created thing than in its creation, which is perhaps the greatest human defense against time and oblivion.

I have tried to show that Cavafy's historical poems are neither the mere dabblings of the antiquarian nor necessarily the “alibis” for private obsessions. Neither do they seek to be read as allegory, whether moral or political, nor as myth-making. They are rather the testimony to a rational confrontation between the creative imagination, whose sphere is the present moment, and the consciousness of the past as always beyond the reach of that moment. Cavafy's “sense of time” is more complex than either Seferis or Malanos acknowledged: it includes the awareness of two quite different perceptions of time, the historical and that of experience or memory. And his historical and erotic poems together dramatize the interplay between these two perceptions of time, and between either or both of them and the present moment of creativity. Cavafy's poetry as a whole can be seen as the attempt of a modern poet in the European tradition to transcend the idea of time without recourse to myth. History and memory occupy the space in Cavafy's work that in other contemporary writings is taken up by myth. If the result is an uncanny likeness between Cavafy's Alexandria and Joyce's Dublin, a “myth in progress,” it has been achieved in Cavafy's case by a method which is the opposite of that employed by his Irish and Anglo-American contemporaries.

Cavafy's method is not to seek to override or to “redeem” time by appeal to the timeless order of myth. But then he did not seek, as Yeats, Joyce, Eliot, and Pound all did, to discover or establish order in the present. The present in a poem by Cavafy has no meaning outside of the creative act. At the heart of his poetry lies the attempt of the experiencing and imaginative consciousness, present in the act of creativity, to comprehend and relate to all that is not present. Poetry provides the medium in which the present tense can be perpetuated as the act of writing, and in poetry the present tense can refer to nothing but the time of writing. In consequence, all personal experience and all of recorded history are banished from the creative present, but are at the same time inseparable from it because the past is intelligible only as the sum of an infinity of such momentary presents.

Cavafy's poetic method is to subvert history from within. Often echoing the voice of the historian and with copious and respectful references to the historical record, Cavafy breaks up the conceptual continuum of recorded history into its constituent moments of individual experience. And the constituent moments of history are frequently chosen and dramatized in order to contradict the perspective of the completed historical record. Not only is the continuous thread of causally related events broken up into its constituent elements, but the transference from history-as-event to history-as-text, on which all our perception of the historical past depends, is placed under the microscope. History-as-text requires the verisimilitude of being “true to experience,” and here it turns out that the historian needs also to be a poet. The resounding, and often subtly ironic, inscribed epigrams which appear in Cavafy's poetry are often the work of professional poets, whose truth to fact is less evident than is their desire to please. And the writing of history, as in “Anna Comnene” and “Dareios” is an activity closely allied both in motive and in method to the writing of poetry.

If time in Cavafy's poetry is the archetypal enemy, it is also a necessary accomplice. Only by the appeal to history can the poet of the imaginative present project his consciousness of the passing moment beyond the bounds of the present to illuminate, and subvert, our received conception of the past. The moment of creativity is validated by the appeal to history: like Phernazes' dilemma, it can be documented or placed in a historical context, but at the same time the recording of history is an impossibility without the imaginative—and often subversive—impulse of the poet. Cavafy's historical poems validate the poetic impulse as a constant factor throughout two thousand years of history, while simultaneously erasing the distinction between what he himself called “writing history” and “making poetry.”

Notes

1. “Ithaca” (1894?/1910/1911: A23, tr. 29); “Waiting for the Barbarians” (1898/1904: A107, tr. 14); “The Intervention of the Gods” (1899: Anekdota, 111); “The Sea Battle” (1899: Anekdota, 121); “Trojans” (1900/1905: A26, tr. 17); “Interruption” (1900/1901: A102, tr. 11); “Thermopylae” (1901/1903: A103, tr. 12); “Manuel Comnenos” (1905/1911: A47, tr. 46); “The Satrapy” (1905/1910: A16, tr. 23); “The Ides Of March” (1906/1910: A18, tr. 24); “The God Abandons Antony” (1910/1911: A20, tr. 27); “Theodotos” (?/1911: A21, tr. 40).

2. All references to Cavafy's poems in this article give the date of composition first (where known) and then the date of publication. Thus, “Thermopylae” was written in 1901 and published in 1903. Where three dates are given, the middle date indicates a revision. Unless otherwise indicated, all translations in this article are my own.

3. Both Keeley and Dallas bracket the name of Eliot with that of Ezra Pound in their comparisons with Cavafy. It does not seem to have been noticed that Pound comes significantly closer to Cavafy's perception of history than any of his English-language contemporaries. The Cantos stand as a monument to the attempt to treat history, in all its random diversity, as the raw material for myth; an attempt which is finally seen in the text as a qualified failure (“does it cohere?”). And there is probably nothing in English poetry that so closely approximates Cavafy's intense feeling for the past as past, as do such early poems of Pound as “Papyrus,” “Provincia Deserta,” and “Near Perigord.” The final lines of “Provincia Deserta,” in which the poet traverses the landscape in which the Provençal troubadours once sung and which lives in their songs, could serve as an epigraph for all of Cavafy's historical poetry:

   So ends that story.
   That age is gone;
   Pieire de Maensac is gone.
   I have walked over these roads;
   I have thought of them living.

4. “Priam's Night Journey” (1893: Anekdota, 51); “Salome” (1896: Anekdota, 87); “Chaldaean Picture” (1896: Anekdota, 89); “The Horses of Achilles” (1896/1897: A113, tr. 5); “Lohengrin” (1898: Anekdota, 103); “Suspicion” (1898: Anekdota, 107); “The Funeral of Sarpedon” (1898?/1908: A111, tr. 7); “King Claudius” (1899: Anekdota, 113, tr. 134); “When the Watchman Saw the Light” (1900: Anekdota, 123, tr. 137); “Bad Faith” (1903/1904: A109, tr. “Unfaithfulness,” 13). It should be noted that only three poems out of this list were ever published by Cavafy. In addition to the above, the following two poems from the previous list refer to the mythical rather than the historical past, as the basis for moral allegory: “Ithaca” (1894?/1910/1911) and “Interruption” (1900/1901).

5. Compare “Dangerous Things” (?/1911: A46, tr. “Dangerous Thoughts,” 30); and see Beaton, 1981: 520-1.

6. See “Epitaph” (1893: Anekdota, 57), “Philhellene” (1906/1912: A37, tr. 31), and “Coins” (1920: Anekdota, 181).

7. “Theophilos Palaiologos” (?/1914: Anekdota, 131, tr. 147) and “Párthen” (1921: Anekdota, 183). Interestingly, neither was published by Cavafy.

8. Compare “The Ides of March” (1906/1910: A18, tr. 24): “when you assume the role of someone that famous” (trans. Keeley and Sherrard), and “Alexandrian Kings” (1912: A35, tr. 33).