Sophia McDougall's Romanitas trilogy, begun in Romanitas itself (2005), and now continued in Rome Burning, demonstrates clearly the mutually exclusive ghettoization of SF and what is variously described as "literary" or "mimetic" fiction. One presumes that McDougall's comfort zone is in the latter category, and, to judge from interviews, she began writing under the impression that a Roman empire surviving to the present day was a wholly novel idea. Presumably the many times that SF has toyed with the notion, going back at least to L. Sprague de Camp's Lest Darkness Fall (1941), have simply never crossed her radar (and she is too young to remember the 1974 Tomorrow People story "A Rift in Time," where this was the central idea). But conversely, any SF critic wishing to make a point about alternate histories of Rome will almost certainly cite Robert Silverberg's Roma Eterna (2003 fix-up of stories published between 1989 and 2003) as the most recent example, and show no sign that they are aware of McDougall's books.

This may seem a bit odd given that McDougall's publisher, Orion, also publishes, through a different imprint, the SF Grand Master's work. But plainly SF is not where the particular branch of Orion that McDougall is signed up to wishes to place the Romanitas series. From the packaging and advertising, it would appear that the lengthy works are aimed at the airport novel market, and the sort of people who buy Tom Clancy and Clive Cussler novels. One even wonders if McDougall or her editor will be pleased to be reviewed in an SF context such as Strange Horizons.

Nevertheless, SF is plainly what Romanitas and Rome Burning are. Whether alternate history or counterfactuals belong in SF at all is an issue much debated, and varying views on that can be seen in Vector 254 (November/December 2007). But even if I didn't agree with Graham Sleight that, in the end, alternate history can and should be claimed for SF (especially if one is expanding the abbreviation into "speculative fiction," the literature of ideas, for all alternate history is driven by the idea of the change or "Jonbar" point), I would still claim Romanitas for SF. Because there's a much less arguable indicator—it's got people in it with paranormal powers.

The Greeks and Romans lived in a pre-Enlightenment age, where the supernatural, though rare, was nevertheless considered part of life. So it is not surprising that works like Homer's Odyssey are at the start of the European tradition of fantastic literature. When what is now described as the "ancient novel" emerged in the Hellenistic and Roman periods (second century BC to second century AD), it was only natural that fantastical elements featured, perhaps most obviously in the second century AD Metamorphoses or Golden Ass of Apuleius. This tradition spilled over into post-Enlightenment historical novels, bringing small moments of fantasy into works such as Robert Graves' I, Claudius; it's one thing to have a Sibyl that the characters believe in, but another to have that Sibyl accurately predict events that are known to the author, but could not possibly be known to the alleged autobiographer. (Is it worth recalling that Graves translated Apuleius, and indeed wrote an SF novel, Seven Days in New Crete?)

Romanitas sits within such a tradition. and there is indeed a Sibyl whose predictions seem accurate. But the healing powers (played down in this volume) of Sulien, one of the two escaped slaves who become the principal characters of the novels, and the mind-reading (not overplayed, but crucial to the plot at times) of Una, Sulien's sister, seem much more science-fictional, or at least are described using SF's vocabulary. McDougall may not see herself as an SF writer, but she has absorbed many of SF's tropes. This is hardly surprising though, in a writer born into a world where many of these tropes have been fed into mainstream popular culture by Doctor Who, Star Trek, and Star Wars.

Okay, let's get the history out of the way. It may seem a bit of an irrelevance, but this sort of book attracts professional historians as readers (and sometimes writers), so it's a legitimate issue. The trouble, as Edward James, in the selfsame issue of Vector to which I've already referred, notes in relation to Silverberg's novel, is that the basic concept is implausible. An eternal Roman empire is one of those notions, like a German invasion of Britain in 1940, which excites the imagination; but once one looks into either, it's hard to see how they can work. All empires have their dynamic, and rise and fall according to that. An awful lot would have to happen to give the Roman empire an extra 1,500 years (or 500, depending on whether you count the Byzantines, as I think you should), more than can be explained through one emperor not getting assassinated when historically he was. As a Jonbar point, McDougall's choice, the death of the emperor Pertinax in AD 193 (though interestingly just less than a quarter of a century away from Silverberg's Jonbar point), isn't that significant. Pertinax was an emperor elevated to the throne without direct military support. Almost without exception, such emperors led short lives which ended violently (the one exception survived by selling himself out to the army). Pertinax himself had already survived two assassination attempts, and it can be argued that surviving the third would merely delay the inevitable.

However, McDougall does some interesting things against this background. It is an easy (and lazy) assumption to fall into that technological development in alternate history progresses at the same rate as it did in what we recognize as "reality." But anyone who understands that a driving factor in the Industrial Revolution was a shortage of labour will perceive that a slave-owning society would remove one of the major forces for change. McDougall does not fall wholly for the easy assumption (though clearly something akin to the Industrial Revolution has happened at some point), and her technology is familiar but different. There is heavier-than-air flight, but only in the form of short range helicopter-like "spiralwings"; long distance travel relies on continent-spanning monorails. On the other hand, just about every telephone (or "longdictor") is equipped with a viewscreen, technology we were all promised long ago, but which telecommunications services have singularly failed to provide. (I wonder if McDougall saw repeats of Gerry Anderson's 1960s sf series Thunderbirds, as the technology is very similar.)

And McDougall never shies away from slavery. Where Silverberg and others can sometimes overlook the slave basis for Roman society, McDougall makes it a driving point of her plot, with conflict between abolitionists and non-abolitionists (and between those abolitionists who want to work within the legal framework of the empire and those who are willing to tear down the whole of society to eliminate the institution) looming large in the motivations of characters. There is no implication here that the world would be better had only the Roman empire survived—for McDougall, Rome is a state which inherently promotes suffering through its structures, even if those at the top don't mean to. Though the city of Rome is portrayed as not that much different from any modern European city in our reality, there is enough information slipped through to convince us that we wouldn't want to live there.

These are interesting issues, and place McDougall's novel in its generic and historical setting. But what is Rome Burning actually like? Is it, in fact, any good? Rome Burning picks up two years after the end of Romanitas. Marcus Novius is more securely the indicated heir to the empire. Sulien and Una, the escaped slaves who shared his adventures when a pro-slavery conspiracy tried to have him killed, are freed and his confidants, but resented for having risen too far above their station. Marcus' confidant Varius has devoted himself to legal forms of slave relief. On the world stage, the empire is entering a crisis, as multiplying border incidents, accusations and recriminations are leading towards war with the Nionian (i.e. Japanese) empire. Of course, anyone who's seen the James Bond films You Only Live Twice or The Spy Who Loved Me, or the Doctor Who story "Frontier in Space," will quickly realize what's really going on (a third party is trying to manipulate the two superpowers into war), but they may take considerably longer to deduce who is behind it, and why. Meanwhile, the plot is full of the sort of devious machinations, manipulations and double-crosses that typify the airport novel, and indeed are to be found in I, Claudius. Whether that's what the Roman imperial household was actually like is beside the point. It is how everybody expects it to have been, and therefore how everybody imagines a twenty-first century Roman state would work. It is to McDougall's credit that only one character is truly disreputable, and even with them McDougall makes a good stab at explaining their motives, and why they think they're doing the right thing. And that's not the main antagonist. That character is shown to be acting out of impeccable purposes—they have merely taken things too far.

It goes almost without saying that Rome Burning is, like almost any modern novel, too long, and could cheerfully lose about a hundred pages without doing any serious damage to the story. But in terms of literary technique, it marks a definite step up from Romanitas. In the first novel, McDougall had an annoying habit of changing point-of-view characters from one paragraph to another. This sort of narrative omniscience makes it easier to tell an "objective" story from all sides, but as used by MacDougall it confused the reader, who never quite knew whose head they were supposed to be inside. The device is largely absent from Rome Burning, and the narrative is all the better for it. Also, in the first volume the central characters were fifteen or sixteen. The result was a novel that often read as if it were intended as a young adult book, had it not been for the inordinate length (generally, J.K. Rowling aside, avoided in YA). There is nothing, of course, wrong with young adult fiction, but Romanitas seemed a little schizophrenic, as if McDougall couldn't quite tell if it should be an adventure novel for teens or a Tom Clancy-style political thriller. The result was a novel that, though not being sure which target to pick, potentially risked missing both of them. The teenage characters have now all aged, and crossed over into adulthood, thus removing any questions of inconsistency of tone.

And I have to concede that, despite prose that is never really more than competent, McDougall does make the reader concerned about the fate of her characters. One may feel, with some justification, that Marcus gets even more insipid than he already was in the course of this novel—but still one feels for him as he is forced to give up his unsuitable true love for a political marriage. And one cares about whether he and the other main characters will survive the novel.

Which is a question that is not quite answered. Romanitas wound itself up reasonably neatly, whilst still leaving threads to be picked up in a sequel. Rome Burning, on the other hand, ends (almost literally) with a cliff-hanger, the fates of several key characters unclear, much like the Moldavian massacre at the end of the fifth season of Dynasty.

But that's an unfair comparison. Rome Burning is better than just about any television soap you may care to name, even if the general ethos of the soap flows through its pages (as it does through most novels of this ilk). This is certainly not a bad read, and on some levels, an interesting example of someone writing science fiction whilst neither intending to, nor being aware that they are.

Tony Keen is professionally interested in alternate histories of the Roman empire. A more in-depth article comparing McDougall, Silverberg, and Stephen Baxter will appear in a forthcoming issue of Foundation.