The first eight of IW's chapters present a history of fantasy literature from the Victorians down through Ursula K. Le Guin. It's not a bad read, though Carter's scholarship is not to be relied upon, and his style is and will forever remain insufferably chatty to me. The last three chapters are a kind of manual for fantasy-writing, many of whose maxims are repeated in Patricia Wrede's excellent guide available at SFWA.
Let's begin with Chapter 9: "Of World-Making: Some Problems of the Invented Milieu."
Lin – I find somehow that I must call him Lin, and not Carter or Mr. Carter – Lin, I say, begins with the common observation that fantasy-writing presents unique technical challenges.
If you stop and think about it, you will realize that fantasy writers face a variety of technical problems that authors working in other genres seldom have to worry about. The problem of creating an imaginary world on paper is the largest and most serious of these, and it is a complex problem involving many different factors.He goes on to explain that getting James Bond from London to Lisbon is a relatively simple matter, since the readers are presumably familiar with telephones, taxi cabs, airplanes, Portugal, and the like, and no exposition need be expended on these things.
Here I have a quibble. Modern-type books that don't bother to build mental pictures are unspeakably boring to me. The action seems to take place in drab gray rooms and corridors. Perhaps that's a fault of my imagination. But not all modern-type books suffer from this. I enjoy Ian Fleming and Raymond Chandler and Dashiell Hammett considerably – sometimes I'd actually rather read them than fantasy! – one reason being their ability to paint a picture. Chandler is particularly good at this. Though terse, his descriptions are almost painfully vivid and precise, and rely almost exclusively on minor details. On the other hand, quite a few fantasies nowadays read more like the drab modern thriller. There's a broad enough base of reader expectation that they don't have to describe a throne room any more than they would a motel room. How unutterably depressing!
On the other hand, in my opinion, a good fantasy contains a preponderance of what is familiar to the author. A fantasy chock-full of marvels is very dull indeed. It remains a vague, silly dream, failing to take on a life of its own, like the Amadis de Gaul that Don Quixote was so fond of.
But back to Lin Carter. For the most part the chapter confines itself to discussing the construction of a believable milieu. Here I opine that the best of fantasy milieus weren't constructed from scratch as settings for stories. Either the construction was pursued for its own sake and eventually flowered into a story, or the story acquired its milieu in the telling. The most memorable fantasy worlds came into being over the course of decades, and produced stories only late in their authors' lives. Eddison's Mercury, for instance, had its genesis in childhood make-believe, while Tolkien's Middle-Earth was born in the trenches and mess halls of the Great War. It's pretty obvious which fantasies are the result of an author who sat down one day and decided to cook up an imaginary world. (Cf. Diana Wynne Jones' comments on MAPS.)
Lin categorizes the types of milieu most commonly encountered. Many fantasies are set in a known world of myth. Others are set in a purely invented world, the four varieties being: (1) remote antiquity (Hyboria, Poseidonis, Middle-Earth), (2) remote posterity (Zothique, the Dying Earth), (3) an alternate dimension (Narnia, Witch World), and (4) another planet (Barsoom, Perelandra). It's interesting that all of these are linked to our own space and time somehow. Most fantasies set in invented milieus nowadays are content to just take the independent world as a given. Another result of the establishment of reader expectations. Even the novels that came out in the seventies and early eighties still tended to connect their worlds with ours, e.g., The Sword of Shannara, Lord Foul's Bane, Mythago Wood. Earlier examples of complete autonomy include Earthsea and Gormenghast Castle, but these are fairly rare.
A discussion of techniques in world-building follows. Lin recommends familiarity with geology, climatology, and the like. Deserts don't just taper off into jungles. Great cities don't just pop up out of nowhere for the mere sake of being grand and decadent. His points are well taken. However, I note that Tolkien appears to have constructed his geography dramatically. He writes in one of his letters of the impossibility of pleasing at all points, and admits the shortcomings of his geology. In fact, I've always suspected that the parallel lands of Beleriand and Eriador were originally one and the same region, but were separated (and Beleriand submerged) when too many inconsistencies arose. Indeed, Tolkien sometimes takes his correspondents to task for asking questions as though he'd written a history rather than a work of literary art intended for its own sake.
Lin's recommendation to diversify the forms of government is also well taken. Not many people heed this even now, alas. You'd think a lot of writers got all their knowledge of political systems from their sixth grade world history class. Since most fantasy realms are city-states, more or less, I here have to recommend one of my favorite books, J. B. Bury's A History of Greece. It was written back when they could still write all-encompassing histories, as opposed to specialized monographs and revisionist screeds, and as such is eminently readable. It goes into great detail as to the organization of Sparta, the various Athenian democracies, and so on. It's available (used) in a Modern Library edition. Lin uses a de Camp story as an example of governmental variation; it sounds a bit silly to me, but Gene Wolfe's Book of the New Sun strikes me a good example of atypical political world-building.
After this Lin dwells at length on the phony citations that often adorn fantasies. Lin himself delighted in such frippery, of course. Irrepressibly enthusiastic as ever, he can't resist citing examples from his own work:
I learned this trick from Howard, and my Lemurian Books use for their chapter-headings quotations from various historical documents such as "The Lemurian Chronicles" and "The Tsargol Records," magical or occult works such as "The Scarlet Edda" and the grimoire of Sarajsha, epics like "Thongor's Saga," pieces of Lemurian literature such as "Diombar's Song of the Last Battle" and any number of folksongs, sea chanteys, sayings and proverbs, and war or marching songs such as "Drum-Song of the Kodanga Tribesmen," "Battle-Song of the Black Dragons," and "Caravan-Song of the Jegga Nomads."To which I say, just say No. Unless you've actually written the works yourself, and they're actually kind of good, don't, please don't use silly made-up quotations as your chapter-headings. Don't let your characters quote texts and proverbs in dialogue, either, unless you're trying to be funny and are Jack Vance. Trust me. It sounds like something from Star Trek: Voyager.
Tolkien could throw in snatches of rhyme and passages of epic poetry because he wrote that kind of stuff for its own sake. You get the sense of incalculable ages behind the events of LOTR because of the strata of stories he'd written over the preceding decades. When Túrin Turambar is mentioned in passing it's not just a bit of color. Tolkien knew very well who Túrin was, and had known for years.* That's not the kind of thing you can just manufacture. Lin Carter's "Tsargol Records" comes off as pasteboard scenery.
All that said, I hereby exempt the quotations that head Robert E. Howard's Conan stories from this complaint, by a raw exercise of blogging authority.
Lin goes on to recommend hinting at distant regions and peoples to build up a rich background. To which I say, again, don't do it, unless it's really necessary to the story somehow, and you've really done your homework. Otherwise it just sounds phony. The example he gives from his own work exemplifies this. There's an art to it. When I read LOTR, I'm always wanting to get into Rhûn and Harad, as though they're real places that lie beyond the edge of the map. All a matter of style, I suppose.
Lin concludes by pointing out how carefully he distributed his made-up foreign country-names over the alphabet. This brings us to Chapter 10, which I'll cover in a separate post.
* Queen Beruthiel's cats were apparently just a bit of color. Sometimes even Tolkien takes wing shots!