Roz Kaveney has established herself as a major critic of popular culture, through works such as From Alien to The Matrix (2005), Teen Dreams (2006), and the edited volume Reading the Vampire Slayer (first edition 2001; second edition 2003). Now she has produced Superheroes!

The superhero genre is a funny one. Emerging as a distinct genre in the 1930s—out of the pulp magazines and characters such as the Shadow, Zorro, and the Lone Ranger—it almost disappeared at the start of the 1950s, but returned at the end of that decade. Superheroes then gradually eased many other genres out of comics (in the 1950s the major companies had a range of genres westerns, romances, crime stories, etc.), at least until the emergence of independents and graphic novels in the 1980s. And though many companies published them in the 1940s, just two now dominate the superhero market: Marvel and DC. Other companies do still have superheroes, but their characters lack the iconic status of those in the Big Two stables (and Kaveney writes little about them).

But for all that comics can do many things, superheroes are one thing comics have traditionally done well, and probably better than any other medium. Some characters (Superman, Batman, Spider-Man) have crossed over into popular culture as a whole, and are iconic figures, as recognizable as James Bond or Odysseus. At the time of writing, the big movie blockbuster of the summer is looking likely to be Iron Man.

At heart, superheroes are wish-fulfilment fantasies—that, for instance, under the geeky exterior the world sees, there's really someone worth paying attention to (Superman); or that someone will administer justice to wrongdoers (Batman). And so they can often be seen as adolescent. Someone dressing up in a colourful costume and fighting crime can seem rather odd from an adult perspective. Yet the best superhero comics are texts worthy of a great deal of study. And they're fun.

Kaveney aims, as she's stated in a recent interview, to write about superheroes seriously. Too often, she feels, they are left in a ghetto, and treated as childish. There's certainly a need for serious criticism on the genre; there is comparatively little, given how much superheroes have dominated comics since their invention in 1938. Paul Gravett's excellent Graphic Novels (2005) devotes only one of ten thematic chapters to superheroes, and that focuses on only two main examples, as opposed to the three or four found in other chapters. There are plenty of nonfiction works out there, but they often focus on individual characters (such as Les Daniels on Superman, Batman, and Wonder Woman, principally historical rather than critical works); or are produced by comics companies as celebrations of their work, thus blunting any critical edge (for example, Daniels again, with his books on the histories of DC and Marvel); or focus on some particular aspect of the genre (such as Trina Robbins's excellent The Great Women Superheroes); or are encyclopaedic in form (such as The Rough Guide to Superheroes). Quite a bit of critical work, as I rather grumpily complained in a letter to Foundation, shows a knowledge of three or four star creators, but displays rather less familiarity with the genre as a whole. There is definitely a need for a book that can function as a historical and critical introduction to the whole genre.

Superheroes! is not quite that book. It is an excellent work, which anyone with any interest in the superhero genre needs to own. Unfortunately, it can't be the only book you'll need, for two reasons. First, the work has a few too many errors (some are detailed later), so corrective works need to be kept around.

The second reason is that the book doesn't provide the broad historical sweep that I was hoping for. Kaveney certainly knows her superhero history, as is shown throughout. But almost all the runs she chooses to focus upon as examples come from the last twenty-five years. Many of these do deserve attention, but I was sorry that no time could be found for, say, Joe Simon and Jack Kirby's original Captain America run, or John Broome's 1950s revivals of the Flash and Green Lantern, or Stan Lee and Kirby on Fantastic Four, or Steve Englehart and Marshall Rogers's Batman stories in Detective at the end of the seventies. It's almost as if Kaveney thinks that superhero comics reached a level of artistic maturity with the arrival of Alan Moore in the mid-1980s (incidentally, the time when I started seriously reading the genre), and anything before that is only of historical interest. If this is the case, she may have a point, though I'd argue that the Englehart/Rogers Batman, or Denny O'Neil's Green Lantern/Green Arrow stories, are as sophisticated as some of the runs that she does discuss. And the chronological bias doesn't help people who want to write about Alan Moore or Frank Miller, but don't really know about the comics that Moore and Miller are responding to. (Kevin Jackson argues an alternate theory in New Statesman: that the bias is connected with the destruction of Kaveney's late 1960s and early 1970s comics collection, the fate of which Kaveney recounts early in the book.)

There comes a point, however, when complaining that a book is not the book one would have written yourself is a fruitless exercise; a reviewer has to address the book that actually has been written. The first thing that strikes me about Superheroes! is that the only illustration is on the cover. This is initially surprising for a book about a medium that depends upon the visual image, as comics do. But, as Kaveney explains, artists have been written about elsewhere, and she is interested in writers and the stories. And besides, as she further notes, illustrating a text like this adequately would be extremely expensive in terms of rights.

The book's format is worth noting. The main text proceeds as you'd expect. But from time to time, there are sidebars inserted. This is an effective way of dealing with such a wide-ranging subject as the superhero genre. One of Kaveney's main thrusts is that the Marvel and DC universes constitute the largest storytelling continuities (or Big Dumb Narrative Objects, to use the term she employs in From Alien to The Matrix) in existence. This means that a critical work on the genre as a whole wants to go off in all directions at once. The sidebars allow Kaveney to do this, and insert short, or sometimes not so short, digressions (one of which is a development of her Strange Horizons review of Heroes).

The first chapter sets the scene for discussion of superhero comics, and Kaveney's own relationship with the genre. There follow chapters on: Brian Michael Bendis's Alias, a series of which Kaveney is evidently particularly fond; Watchmen; some other notable runs from the past quarter century; "event comics" (i.e. Crisis on Infinite Earths and its successors); Joss Whedon; and superheroes in films. The best of these is the one on Whedon, and the way comics have influenced his work on Buffy and Angel. I hadn't quite understood how important the superhero aesthetic was to what Whedon does, but once Kaveney points it out, it's obvious. It's not a point Kaveney makes explicitly, but it seems to me that the doomed romances that end in death (Giles/Jenny, Buffy/Angel, Willow/Tara, Wesley/Fred) can be understood as Whedon trying to recapture the sense of tragedy of the Scott Summers/Jean Grey relationship, as portrayed in early 1980s X-Men.

But all the chapters are good. I don't rate Watchmen as highly as Kaveney (and almost everyone else) does (as I have discussed elsewhere), but her chapter does a very good job of explaining why it's important anyway. (I also disagree on the relative merits of Batman Returns and Batman Forever, discussed in the final chapter, but again I'm aware I hold a minority viewpoint here.) The reader certainly comes away with a good idea of the current state of the superhero genre. The big star writers, Moore, Miller, Grant Morrison, and Neil Gaiman, are of course covered. (Gaiman less than the others; as Kaveney sees it, while Gaiman's major comics work, Sandman, takes places i the DC universe and sometimes features superheroes , it is not itself a superhero comic, and has almost none of the semiotics of a superhero comic. Gaiman's more traditional superhero work, such as Black Orchid and his run on Miracleman, has had less impact.) One of Kaveney's strengths, however, is that she sees beyond those to the other writers, whose names you might not know—Mark Millar, Ed Brubaker, Kurt Busiek, and the like.

Kaveney says that she's a bit "theory-lite" (p. 4). But there's actually a lot of theory in Superheroes! The notion of the "thick text," the artistic work considered as a palimpsest of compromises between different individuals, read in the context of the cultural background and expanded continuities (which Kaveney more clearly explains in From Alien to The Matrix), is important. The theory just isn't obvious as such, because Kaveney sees no reason to dress it up in obfuscating jargon. So the book remains accessible to even the most pomophobic reader, and is packed full of useful insights (I'm grateful for the notion that Walt Simonson's Thor-as-Frog stories reference Homer's lost Batrachomyomachia).

Particularly usefully, she explains how the continuities grow, and how they can occasionally be marred by what she calls "strip-mining the continuity," creative decisions (usually made by editors) that invalidate fondly remembered stories. Her argument is that, sooner or later, someone else will come and put things right with a new story (as Chris Claremont put right the dreadful treatment meted out to Ms Marvel in the 1980s). I'm less optimistic about this than Kaveney is. My feeling is that the continuities have been developing for so long, and editors seem so interested in shock, that almost all the good story ideas have now been done. A few bright sparks aside, there are only stupid ideas left—the continuities seem generally to have stagnated, which is one reason why I have not been generally tempted back to the same level of interest I had in superhero comics in the 1980s, except for a few particular examples (the other being that the sheer scale of the Big Dumb Narrative Objects has become daunting, for me at least). Kaveney clearly disagrees (and in fairness, is much more familiar with superhero comics over the past ten years than I am).

I did promise you some nitpicking. (I'll leave aside the points where the grammar suggests that Kaveney began writing a sentence one way, and then changed without fully revising what was already written, which from time to time require a repeated reading of a passage to elicit the meaning.) At a couple of points, Kaveney creates a slightly misleading impression. So, for instance, on page 19 she says, "because the comics continuity ... moved on to an equal partnership between [Lois Lane and Clark Kent/Superman] ... it is sad that ... the 1970s Richard Donner films ... went back to the old assumption that Lois does not know [that Superman is Clark Kent]"; but Lois did not find that out in the comics until the 1990s. That isn't hugely important, and neither, I suppose, are the mistakes with names: for example, Frank Miller's The Dark Knight Strikes Again is consistently referred to as The Dark Knight Strikes Back (p. 145); Spider-Man's Uncle Ben is called "Uncle Henry" (p. 255). It doesn't take much to track down the correct facts if you want to.

There’s only one serious error that leapt out at me: the garbling of the history and relationship of Captain Marvel and Marvelman/Miracleman (p. 36). Contrary to what Kaveney writes, the name Marvelman was not coined in 1982, when the character was revived for Warrior, but in 1954, when Len Miller asked Mick Anglo to create new material to replace the American Captain Marvel stories he had been reprinting, which dried up when Fawcett agreed to cease publication of Captain Marvel as the result of a lawsuit brought by Superman's publishers. Marvel Comics's lawyers did not force the change to Marvelman (hardly the sort of name they'd be comfortable with), but forced the change to Miracleman when the Warrior material was reprinted by Eclipse in the US. And Neil Gaiman never wrote for the feature while it was still Marvelman.

Now, one might say that these sorts of errors aren't really important. But there are quite a lot (and these are only a selection of the ones that I noticed as I read). And I think they do matter, for two reasons. First, clearly a nonfiction book ought to get its facts right, especially one directed at a nonspecialist audience; otherwise, the wrong information will be propagated. Second, for anyone who does spot the errors, the number undermines the authority of Kaveney's opinions, and I think that does this important work a disservice. So, I hope it sells well enough to justify a second printing that will correct these errors.

It certainly deserves to. Any future scholarly article on superheroes that does not reference Kaveney just won’t have been done properly.

Tony Keen likes superheroes, and is too grown up to feel guilty about it.