At one point in Claudius the God Claudius quotes his open letter to the Senate and People concerning his own shortcomings. Towards the end of this document he writes:

You all know the old Patrician saying: Aquila non captat muscas. The eagle... does not hawk for flies.... Let me quote an enlargement made many years ago by my noble brother, Germanicus Caesar:

Captat non muscas aquila; at quaeque advolat ultro

Faucibus augustis, musca proterva perit.

. (The couplet, translated, means: 'The eagle does not hawk for flies, but if an impudent fly comes buzzing...into its au st throat, that's the end of the creature'). (607)

This is an entirely typical passage from the Claudius novels in which Graves imaginatively extrapolates a passing reference in Suetonius, and uses his knowledge of Latin and the Roman mindset to build an amusing, apposite, and believable reference. But what few readers of 1934, and probably few since, have realised is that although aquila non captat muscas is a genuine Latin proverb it was also the Graves family motto.

Of course even if the fact had been generally known it is not likely that it would have affected many readers' interpretations of the Claudius novels. It is quite likely that the extension to the proverb was a school exercise of the young Robert Graves, or one or other of his siblings; or it may have been something written by Graves himself at the time of writing Claudius the God in 1934. And rather than go off into a wilderness of speculation as to why Graves should have inserted the reference here, and should have associated the extension to the phrase with Claudius's brother Germanicus, I would prefer to think of it as a mere joke, or possible even less than a joke, perhaps something that Graves gave no thought to at all as he penned the passage, and intended nothing by it.

However what I would like to note about this instance is how typical it is of the method that lies behind the reconstruction of the Roman world attempted in the two Claudius novels. For ever since the publication of these two novels they have been read in a very straightforward way, and although classicists and ancient historians will do doubt prefer the thin stream of academic scholarship surrounding the Emperor Claudius², it is probably fair to say that, to the extent which the Emperor Claudius is present to any general historical consciousness, it is in the version presented by Graves in these novels, or in the version presented by the BBC series of the 1970s, itself deriving from Graves' work. Not that simple, unitary readings of Graves' work are unrewarding, and Graves' skill as a writer enables these readings to be made and enjoyed with little of a guilty conscience. But recent critical work has warned us to be careful of taking for granted Graves' methods and intentions. Steven Trout, for example, has recently compared Goodbye to All That with Defoe's A Journal of the Plague Year; in Trout's view both narratives were received at the time of their publication as factual reportage, and both use purported original documents and eye-witness accounts to support their veracity. But both on closer reading mock the simplistic assumptions of literal truth, and both are concerned not to report on exactly what happened, but to recreate what it felt like for the participants to be involved in tumult of events and to be part of process of the frantic dissemination and invention of information, misinformation and mythology that people engage in under the stress of great events (179—87).

What I am arguing is that the recreation of the Roman world in the Claudius novels is not a simple as most readers have taken it to be. A closer examination of the methods that Graves uses would find many discordant and disharmonic features in the writing, and yet my purpose in uncovering some of these elements is not to suggest that the recreation of the Roman world in these novels is unsatisfactory or flawed. On the contrary it is to suggest that Graves in these works engaged in what might be termed a covert exercise in modernistic fiction, and the end result of his work is that the recreation of the past is in fact more satisfactory, and more satisfying, that if he had written a simple, straightforward text. The authenticity that Graves constructs in these novels, is laboriously contrived, but the more lasting because of this.

In writing a historical novel set in Roman times Graves had to face the problem of what has subsequently been termed 'incommensurability'. This the consideration that social institutions, practices and so forth from different societies cannot properly be compared with each other, for even similar institutions or practices function in the social fabric in different ways at different times and in different societies. An example from the Claudius novels is that of the Roman army. The Roman army of the early imperial period was organised into legions, each of about 5,000 men; Graves refers to these in the novels as 'regiments', yet regiments in the British army contain less than 1,000 men. On one level this equivalence is a mistake, yet on another it is correct, as the Roman soldier identified with his legion in the way that a British soldier identifies with his regiment. Another 'mistake' on Graves' part is the way in which, in the Claudius novels, he uses the term 'captain' to describe centurions—in British army captains are the third rank of commissioned officers, but in the Roman army centurions had all risen through the ranks, in one way they were the 'equivalent' of sergeants in the British army (this points to the fact that, compared to twentieth century Britain, the Roman Empire had a very much smaller upper class from which to draw its officers).

It is an interesting question as to why Graves should have used these terms when the words 'legion' and 'centurion' are so well naturalised in English, and surely would not have caused a problem to any reader prepared to undertake the reading of the novels in the first place. And another case in point is Graves' celebrated use of the word 'assegai' for the German warriors' framea. He explains this in his Author's Note:

Aircraftman T.E. Shaw... questions my use of 'assegai'.... He suggests 'javelin'. But... I needed 'javelin' for pilum, the regular missile weapon of the disciplined Roman infantryman; and 'assegai' is more savage-sounding. 'Assegai' has had a 300year currency in English, and acquired new vigour in the nineteenth century because of the Zulu wars. The long-shafted iron-headed framea was used, according to Tacitus, both as a missile weapon and as a stabbing weapon. So was the assegai of the Ama-Zulu warriors, with whom the Germans of Claudius's day had culturally much in common. (7)

And although this note, and his usage, is entirely reasonable, both the last sentence quoted above, and the extended mention of such a small matter at the beginning of the first novel, underline the principle of equivalence, as opposed to strict identity, both in this case, and in the construction of the Claudius novels in their entirety.

Chris Hopkins has pointed out that in the 1930s many historical novels were being written, and some at least were set in Rome (128—35). But many of these, especially those written from a left-wing perspective, were obviously, and in many cases self-consciously, written in what Graves was later to describe as 'an unsuitably modern style' (Introduction 26). Clearly the classically-trained Graves would not have been happy to write a Claudius novel which alluded in a too obvious, or partisan, way to modern world-politics. However it is also clear that he could easily have written an unreadably authentic Claudius autobiography, an amalgam of the styles of Caesar, Cicero, Pliny, Marcus Aurelius and any other autobiographically-inclined, or apparently autobiographically-inclined, Romans. Such a work would be unreadable because of the amount of knowledge that the reader would have had to have brought to it, and because it would have been difficult for anyone to read any modern relevance into it.

Instead of either of these alternatives Graves chose a middle course, and wrote the Claudius novels from a perspective which Chris Hopkins explains as a long one, not immediately politically-engaged (134—5). However it is apparent that despite this a great deal of contemporary political relevance creeps in, for example the sustained joke of the equation first century Germans = twentieth century Germans = dangerous, warlike barbarians, which begins in the Author's Note passage quoted above and carries on through both novels, or the passage in which Messalina seems to be espousing state-planning (in the 1930s as much the hall-mark of the fascist right as the socialist left) (601), or even Claudius' straight-faced description of his early rule as 'a gradual reconstructive programme' (478). Indeed I suspect that few readers note that the success of the Claudius novels is largely due to one unelaborated little fiction that Graves inserts right at the beginning: this is that as Claudius is writing in Greek he has to explain a great deal about Roman customs and religion, and translate a number of Latin quotations and puns, for his potential readers of the future (14), and if his Greek text has been conveniently translated into English on our behalf by Graves why then these explanations will do just as well for us, Claudius's English-speaking readers of the future. Thus it is that the Claudius novels move very smoothly when compared with some of Graves' later fiction, such as King Jesus, which at times seem to labour under the amount of historical information that must be communicated.

The middle course which Graves steers in his Claudius novels is not therefore a simple matter of Graves thinking analeptically, and letting Claudius speak, as many of Graves' readers, critics, and perhaps even Graves himself, might have wanted to think. Instead the effects that create the illusion of Claudius speaking are intricate and contrived. At times Graves did recognise the literary strategies that had to be employed when writing historical fiction. In the

Introduction to The Golden Fleece he wrote of the necessity to indicate 'at what vantage-point in time' the narrator was speaking from, otherwise it would not be clear how to understand the story (26). In a passage from the 'Historical Commentary' to King Jesus Graves first writes in 'fundamentalist' way about his analeptic method, but then, in the second part of this quotation turns to a metaphor that suggests the artificial, and conventional tactics that have to be employed by a historical novelist:

To write a historical novel by the analeptic method... one must train oneself to think wholly in contemporary terms. This is most easily done by impersonating the supposed author of the story, who has much the same function as the carefully costumed figure placed in the foreground of an architectural drawing to correct misapprehensions about its size, date and geographical position. (353—4)

In the aforementioned two passages Graves was writing as a wellknown and respected historical novelist. In the opening of I, Claudius, Graves first full-length historical fiction, the Author's Note and the other material at the beginning of the book, are not so much a commentary or introduction, but constitute the opening moves in the semiotic game of constructing the world of the Claudius novels. They establish the fiction of an autobiography lost for 1900 years, the genuine autobiography of a Republican Roman emperor, but at the same time they undermine it. The title page has first the title of the work, 'I, Claudius', and then the phrase 'From the Autobiography of Tiberius Claudius' ('from'?), and then this phrase is followed by a difficult to read piece of handwriting, close inspection revels it to be the two words 'Tiberius Claudius' written in Greek (Tau-iota-beta... &c). Presumably the handwriting is according to the model of 1ST CENTURY Greek script, but does a genuine signature of Claudius exist? Probably not, so this signature is like Claudius's official biography, in part written by his freedman Polybius, modelled on his own style (10), or like the various signatures improperly obtained from Claudius, or forged, by his freedmen for the benefit of Messalina in several places in Claudius the God.

On the next page is a piece of opening material added for the first Penguin edition in 1941. It is 'the Latin version of the Sibylline verses mentioned in the first chapter', by one 'Mr A.K. Smith I.C.S' (Indian Civil Service?). These verses are the ones quoted by Claudius from a manuscript given to him by Livia, about the succession of the six hairy ones. But no such set of verses exists amongst the extant Sibylline verses, and presumably the story is that Mr Smith was a Claudius enthusiast who took the trouble to translate Graves' verses into Latin and sent them to him, and Graves, with as much cunning as he showed in his self-conscious inaccuracies in Goodbye to All That, printed them in the Penguin edition with a note which besides mention of Mr Smith's irrelevant qualifications draws attention to the fictitious status of the verses quoted later (if they were a genuine Sibylline text why did Mr Smith need to supply the Latin version?).

One could continue with a similar commentary for every page of these two novels, but the point of all this prefatory material is at once to establish the principles behind the construction of the work— as with the 'assegai' passage already quoted—and to undermine it. The key sentence in all of it, quoted, probably out of context (!), from Tacitus, is:

so true it is that all transactions of pre-eminent importance are wrapt in doubt and obscurity; some hold for certain facts the most precarious hearsays, others turn facts into falsehood; and both are exaggerated by posterity.

The joke of all these pages, and of those that follow is that most readers at the time, and subsequently, and Graves himself at times, have come away from these novels thinking that the author had some privileged window on the truth—and indeed the pages that follow do give a good impression of this—but what they are also 'saying', paraphrased crudely, is: 'Every kind of historical representation breaks down because no certain meaning can attach to events, as the meaning itself becomes an object of contestation, and yet to write history is traditionally thought of as a process which establishes authority, and banishes doubt—what could be more absurd than to imagine a situation in which a Roman emperor is secretly a Republican? —and yet this the fiction that you, the readers, are swallowing hook, line and sinker.'

Readings of I, Claudius and its sequel which centre on the 'truth' of Graves' reconstructions of history usually put great emphasis on the Livy—Pollio episode of I, Claudius (102ff). Here the young Claudius runs into the famous historians Livy and Pollio, and in the course of their conversation a comparison is drawn between Livy the historian as the inspired propagandist, and Pollio as the recorder of what actually happened (in a von Rankean sense). It is sufficiently obvious from this episode that Graves has very little time for Livy, or any obviously propagandist historian, and yet it would a mistake to assume that the message to take away is one of unqualified support for Pollio. Pollio, who interestingly enough has, just like Claudius is later to do, written both an 'official' history and a private 'Supplement' which tells the unvarnished truth (105), is favoured, but this preference is rendered ambiguous in a couple of places. When Claudius's assistant Sulpicius, sums up the argument, for example, he describes Pollio in no better terms than:

. mere truth-tellers—"undertakers who lay out the corpse of history" (to quote poor Catullus's epigram on the noble Pollio)—people who record no more than actually occurred— such men can only hold an audience while they have a good cook and a cellar of Cyprian wine. (110)

Which is not exactly complimentary, and also rather puzzling as the only reference to Pollio in the text of Catullus—an author who has traditionally been credited as a Latin author uncontaminated by literary convention, an absolute truth-teller—is actually much more positive than Graves' invented reference (cf Catullus XII). Secondly after Claudius's meeting with Pollio, we learn:

His stipulation about my having the authority to make corrections [to his historical works] everyone treated as a joke; but I kept my promise to Pollio some twenty years later. I found that he had written very severely on the character of Cicero—a vain, vacillating, timorous fellow— and while not disagreeing with this verdict I felt it necessary to point out that he was not a traitor too, as Pollio had made him out. Pollio was relying on some correspondence of Cicero's which I was able to prove a forgery by Clodius Pulcher. (113)

Cicero may not have been, for the real Asinius Pollio and Tiberius

Claudius, quite the towering moral and political figure that postRenaissance historians have made him, but he was an important figure, and for Pollio to be reported as having made such a mistake on the basis of only one piece of evidence hardly inspires confidence in his historical judgement, especially as he would have been a contemporary of Cicero's in his youth. Graves seems here to be saying that even a historian of von Rankean integrity can slip up as badly as any other.³ Finally we should remember Pollio's own words in giving Claudius authority to revise his works—one of those stark, memorable phrases, unconnected with any of his other beliefs, that Graves is so good at supplying: 'Keep them up to date. Books when they grow old serve only as wrappings for fish' (112).

Thus far we have been concentrating on a narrow range of thematic references, references to historical method, to the question of commensurability between different epochs in history, and have found in the two Claudius novels a persistent feature of the writing in connection with these two areas is that it undermines the certainties it sets out. Of course in much more obvious way the Claudius novels undermine themselves: Graves is very careful to allow to emerge gradually the possibility that Claudius may not be a wholly a reliable character as his privileged status as narrator of his own story might suggest he is. To a much greater extent than any of Graves' other novels (almost all of them, by the way, first person narratives), we are allowed to see the other side of the story. Particularly in Claudius the God, and long before Claudius makes an explicit decision to act the part of King Log, plenty of details are supplied which indicate that not only has the monarchy had a deleterious effect on the morals and conduct of the Senatorial order—as Claudius is ever-eager to allege—but it has had a deleterious effect on Claudius's own morals too. The persistently theatrical, and at times farcical, nature of the presentation of most of the events of the story—a technique we are familiar with from Goodbye to All That—carries on into the events of Claudius's reign, particularly in the parodic activities of the clown Baba, which Claudius relates, even as he reports the words of his enemies in the Senate.

In the same way the construction of the novels themselves is a mixed, hybrid one. The dialogue, for example, whilst not 'unsuitably' modern, is none-the-less modern, and many references in the text are, like 'assegai', made by the principle of equivalence, for example the glass of 'vermouth' that Claudius's physician Xenophon prescribes for him as a sedative (780). It is also, as mentioned before, full of contemporary references, and can hardly fail to be. Finally it also contains many personal references, such as the aquila non captat muscas passage quoted earlier, or the passage, which may not have gone down very well with Laura Riding, where Claudius/Graves lays down that: 'I do not think it is natural for a normal man to live long without a woman' [ie without sexual relations with a woman] (202).

Graves' references to Roman society and history are also necessarily hybrid. Although is amazing, reading through Tacitus and Suetonius, how many of the wildly improbable episodes of the plot come almost verbatim from their pages, Graves necessarily had to invent and reinterpret a great deal. And yet such is the authority exuded by the other elements of the recreation it is often difficult to say whether or not a detail really has been invented by Graves. For example Sulpicius's quotation about Pollio from Catullus is certainly not from the text of Catullus, but it would be a knowledgeable classicist who could say definitely that no ancient source attributes that view of Pollio to Catullus, that it is pure invention. And often Graves seems to be playing with the notion of authenticity in his passing references. When Messalina, for example, is trying to trick Claudius into giving her a divorce, Claudius reports a saying of 'the philosopher Mnasalcus' about true love (571). In fact there was no such philosopher, or at least, the standard reference works to the Classics do not mention him—again it would be hazardous to say that there is not somewhere in the corpus of classical texts some reference to this person. But a later passage Graves goes on to insert his creation again in a letter from Herod Agrippa, as though underlining the invention, by mentioning him in the company of three other men who certainly did

exist:

Gadara was a rather run-down place by then, though in its prime it had produced four great men: Meleager the poet, Mnasalcus the philosopher, Theodorus the rhetorician... and Philo the mathematician.... (669)

Earlier I described Graves' purpose in the Claudius novels as 'a covert exercise in modernistic fiction'; it is clear that Graves himself detested the term modernism, writing many times against it, and once—in A Survey of Modernist Poetry—hijacking it for his own purposes. It is also clear that the Claudius novels are on first inspection nothing like the 'classic' texts of modernist fiction in English from the first half of the twentieth century. But on closer inspection, behind the appearance of traditional technique, many features of similarity can be discerned, most notably a persistent tendency for the narrative to undermine its own certainties on any number of levels. This is a narrative, after all, which after over 600 pages of first-person narrative breaks off before its conclusion and moves outwards into four, independent, incomplete and mutually contradictory accounts of the circumstances surrounding the narrator's death.⁴

As we have seen this convert modernism was a very necessary step for Graves, and one which has guaranteed the success of the Claudius novels down to the present day, for a truly 'authentic' narrative would be impossible, and one near-authentic would be almost unreadable except by dedicated classicists. On the other hand a novel set in Rome of the first century but concerned solely with modern concerns would have been a vulgarity, and one which could have had only a very short-lived success. The middle-course which Graves steers allows him to construct a greater authenticity in his narrative than either of the alternatives would have allowed him; it has allowed him to define for the twentieth century the historical figure of

Claudius, and has entered into the lists of twentieth century fiction a very powerful and lasting tale.

Lastly, it allows him to do the impossible, paradoxically the

awareness of the incommensurability that the texts demonstrate allow Graves certain moments in which the similarities between Graves the author, and his persona of the Emperor Claudius can overcome the historical and cultural distance between them, as for example in the moving stoic creed of Claudius/Graves in the meeting of Claudius with Livia:

I never take vengeance unless I am forced to by an oath or in self-protection. I believe that evil is its own punishment. All I want now is just to know the truth. I am a professional historian and the one thing that really interests me is to find out how things happen and why. For instance, I write histories more to inform myself than to inform my readers. (289)⁵

It allows, in other words, the most authentic account that is possible in twentieth century fiction of a first century Roman character. More than sixty years after their publication it is no disrespect to them or their author, to say that what the Claudius novels demonstrate is that it is indeed 'no great magic' 'to bring the dead to life', it is, instead a matter of highly accomplished literary technique.

Notes

¹ • | quote from the Penguin omnibus edition, which reprints I, Claudius and Claudius the God in one volume.

²• The most recent work in this area is Barbara Levick's Claudius, which presents a much more credible picture of the aged Claudius, afflicted by ill-health, uxoriousness and intemperance, and circumscribed in all his actions by the need to balance the powerful forces of the army, the Senate and the corps of imperial freedmen.

³ • | have often thought in connection with this passage whether we have not been misreading Graves' famous satire 'The Persian Version'. Granted it is a devastating parody of official war-communiqués, but does it not also cast doubt on the 'Greek theatrical tradition', and by implication on its historical tradition—a tradition that is the origin of all western historical-writing—as well?

⁴• It is also a narrative which Graves himself had occasion to reinterpret: in his article of the late 40s, 'Caenis on Incest', he gave a completely different account of the motivation of the leading characters in the story from that provided by Claudius. In the 60s he also came to provide an elaborate mycological and toxicological account of

Claudius's poisoning for Gordon Wasson—as it were a fifth account of Claudius's death to be appended to Claudius the God.

⁵ • It is interesting to note that Graves' literary techniques allow him in the Claudius novels to introduce many examples of the very different mentality of the first century Roman mind in his character Claudius, without any adverse critical comment. When, in the Introduction to his translation of Apuleius's The Golden Ass, he spelled these out in a non-fictional context with apparent approbation he received a great deal of adverse comment.

Works Cited

Graves, Robert. 'Caenis on Incest'. Occupation Writer. New York: Universal Library, 1950: 244—56.

I, Claudius (1934) and Claudius the God (1934). Reprinted in one volume. London: Penguin, 1986.

Historical Commentary to King Jesus, by Robert Graves. London: Cassell, 1946: 352-56.

Introduction to The Golden Fleece, by Robert Graves. London: Cassell, 1944: 9-27.

Introduction to The Golden Ass, trans Robert Graves. London: Penguin, 1950: 9—21.

Hopkins, Chris. 'Robert Graves and the Historical Novel of the 1930s'.

New Perspectives on Robert Graves. Ed Patrick Quinn. London: Associated Universities P, 1999: 128-35.

Levick, Barbara. Claudius. London: Duckworth, 1990.

Trout, Steven. 'Telling the Truth—Nearly: Robert Graves, Daniel Defoe and Goodbye to All That'. New Perspectives on Robert Graves. Ed Patrick Quinn. London: Associated Universities P, 1999: 175-87.