There are, as everyone knows, two ways of doing a thing: one way and the other way. For any given thing worth doing, there may be an infinite number of ways to divide it into two categories; just as there are an infinite number of angles at which you can cut an apple in two. All these lines of division are technically valid, of course, but some are clearly more helpful than others. (Here is an example of an unhelpful division. There are two ways of tying your shoelaces: with a barbecue lighter and without. I think it is safe to say that all the usual methods of tying shoelaces fall into the second category.)
There are, accordingly, two ways of reading books; but infinitely many ways to divide up the act of reading into two classes. One way, which I and others have found useful, is to divide reading into the immersive and the analytic. If you prefer, you can call them ‘reading for the story’ and ‘reading for the text’. The immersive reader dives joyously into the vicarious experience of the story, identifies with the characters, laughs at the funny bits, cries at the moving bits, and generally wallows in the sensuous details of the story-world. The text is translated on the fly into a sort of 3-D movie playing inside the immersive reader’s head. Vladimir Nabokov despised the immersive reader. The analytic reader, who is most often found in academia, stays carefully on the surface of the text, studying the language word by word and sentence by sentence, looking for nuggets of technique and jewels of craftsmanship, and treating motifs and symbols as if they were algebraic variables. Nabokov courted and lionized the analytic reader; which is why Nabokov’s books are read (now that the naughty-naughty of Lolita has been eclipsed by a planet full of Internet porn) chiefly by bored university students labouring their way through the ‘close reading’ of a set text.
It will appear that my sympathies are altogether with the immersive reader. This is an oversimplification. Immersive reading, like genre writing or comedy acting, requires a lot more skill than meets the eye. Every immersive reader is capable of reading analytically. If you want the proof of this, take an average romance reader (who could fairly stand as the paragon of immersiveness), and put a cookbook or an airport timetable in front of her. She will not immerse herself in the lives and loves of the parsley; she will not have raptures and rhapsodies about the three o’clock flight to LaGuardia. She will read ruthlessly and efficiently for information, staying very definitely on the surface of the text, but keenly on the watch for any sign that the narrative (so to speak) is unreliable. Then, once she has extracted the data she wants, she will turn back to her Nora Roberts or Barbara Cartland and lose herself once more in the sensuous delights of the bare-chested hero and the throbbing heart of the heroine. What she will not do is read analytically for fun. That is a specialized taste, and for various reasons, it is a taste cultivated largely by people who have not got the knack of reading immersively.
As a writer of fiction, one can try to appeal either to the immersive or to the analytic reader; seldom or never to both. I have nothing to say about how to appeal to analytic readers: I don’t do that kind of writing myself, know very little about how it is done, and frankly, have not much interest in knowing more. But I can say a little about some of the ways of writing so as to appeal to immersive readers.
There are, of course (sticking to our method here), two ways of writing for immersive readers, and we may once again call them the immersive and the analytic. It may sound odd that you can reach immersive readers by analytic writing, but like many odd things, it is true just the same. The technique is not one I tend to use, but I can at least roughly describe it.
The analytic writer has the same relationship to stories that the food chemist has to cookery. He knows which monkey tricks will produce what emotional responses; he knows the formula for his genre — five parts Cops And Robbers, one part Boy Meets Girl, and three parts Travelogue Through Interesting Places, seasoned to taste with sex and Worcestershire sauce; simmer over medium heat until hard-boiled and cynical. (This, if I remember my Mrs. Beeton’s, is the recipe for a Thriller Soufflé.) This method of constructing stories was widely popular among writers for the old pulp magazines. To make a living from the pulps, you had to be immensely prolific, and writing to a formula is the easiest way to increase your output. Besides, the pulps’ readers were mostly young, inexperienced, and technically undemanding, and would devour reams of third-rate writing without audible complaint; better writing would not exactly have been wasted on them, but it was not worth doing for a penny a word. This analytic method is still the usual technique in run-of-the-mill writing for films and television. It is sometimes called ‘the Old Baloney Factory’ by its practitioners; critics, a less genial breed, tend to call it hackwork.
The advantage of the analytic method is that it is reliable: it leaves little or nothing to inspiration. Writing by this method consists chiefly in the ingenious manipulation of set story elements, mixed together in the prescribed proportions. It is relatively easy to do; if originality fails, the formula will see you through. The disadvantage, of course, is that it is predictable — it leaves little or nothing to inspiration. Once you have seen all the elements of a particular formula half a dozen times, you’ve seen them all, and the stories lose the power to surprise you; eventually they lose the power to move your emotions in any way.
This, possibly, is one of the reasons why some people give up on immersive reading and turn to analytics instead. Having been swindled once too often, they decide that all immersive reading is a swindle, and take all their reading pleasure from the detective game of catching out the swindler. They read in the spirit of the tiresome people who go to a magic show not to be entertained, but only to figure out how the conjuror works his illusions. And like the know-it-all at the magic show, half their tiresomeness comes from their attitude of sniggering superiority to the unwashed masses who actually watch the show to be amused.
The immersive writer, on the other hand, is unreliable and unpredictable. She writes in an approximate reversal of the immersive reader’s process: the story plays itself like a movie in her head, and she tries to capture as much as she can in words. She is never satisfied with the result; the waking dream was so much more vivid than anything that survives translation into language and back again. But she persists, and wrestles thanklessly with her material and her technique, and once in a while, with luck and skill, she catches lightning in a bottle and produces a remarkable and original story.
I do not want to give a misleading impression. Not all immersive writers are eye-minded. Some perceive their stories chiefly in terms of sound; these tend to be heavy on dialogue and short on visual description. Some perceive them in terms of the naked emotions and interactions of the characters; they are, so to speak, possessed by the viewpoint character until his spirit is exorcised through the pen. These writers can be very interesting, because (if skilful) they can generate tremendously powerful emotional effects. What Robert Frost said — ‘No tears in the writer, no tears in the reader’ — is especially and vividly true of such people.
It will appear, once again, that my sympathies all lie with the immersive writer; and once again, it is not that simple. Nothing makes worse reading than an inept immersive writer. In the first place, the movie in the writer’s head may be a hotchpotch of recycled tropes, gleaned not from life (that unpredictable and therefore interesting auteur) but from too trustful a reading of the canned motifs and stock responses vended by the Old Baloney Factory. This is one reason why fiction by young writers is so seldom good reading. Christopher Paolini, in his tender years when he wrote Eragon, was not a hack, but the work of his heart, the interior drama of his soul, was largely strung together from bits and pieces of other people’s hackwork. The immersive writer needs more time and experience to mature than is commonly true of the analytic.
More to the point, just as every immersive reader necessarily has a certain amount of analytic skill, every immersive writer needs an analytic side to serve as her first editor and internal critic. This analytic side is the annoying voice in every well-adjusted immersive that continually whispers: ‘This was fun to write, but will it be fun to read? Does it advance the story? Is it relevant? Does it serve what we’re trying to accomplish here, or are we just being self-indulgent?’ These questions need to be asked. A really fluent writer will learn to operate with a sort of mental bifurcation: the analytic mind learns how to contribute its meed while the immersive mind is creating, and in such a way that the immersive mind does not start crying and hide in a corner from the cold cruel world. Most writers have to exercise the two faculties alternately — first create in passion, then edit in cold blood.
Then, of course, one must bear in mind that I am oversimplifying — making a false dichotomy out of what is, after all, only one element in a tremendously complex creative process. Sometimes you need false dichotomies to make sense of a continuum. Not everything in life is black and white, but if we want to measure a shade of grey, we need black and white as the standards of comparison. When we say that Greenland is further north and Tahiti is further south, we depend implicitly on the North and South Poles to stop where they are, so other places can be located in relation to them. The pure analytic and the pure immersive are rather like the poles: frigid and infertile wastes where nobody actually lives, but terribly handy to the rest of us for giving directions by.
In his talk on creativity, John Cleese said a good deal about the ‘open mode’ and the ‘closed mode’, and said that all creativity has to take place in the open mode. It would be fair to say that the immersive writer works most of the time in the open mode, and the analytic writer tends more towards the closed mode. Of course every creative artist needs both modes; this is a tendency rather than an absolute division. Cleese himself, by his own account, is an immersive writer. He gives away his position by this remark in the talk:
I was always intrigued that one of my Monty Python colleagues who seemed to be, to me, more talented than I was, did never produce scripts as original as mine; and I watched for some time, and then I began to see why. If he was faced with a problem, and fairly soon saw a solution, he was inclined to take it; even though, I think, he knew the solution was not very original. Whereas if I was in the same situation, although I was sorely tempted to take the easy way out and finish by five o’clock, I just couldn’t. I’d sit there with the problem for another hour and a quarter, and by sticking with it, would in the end almost always come up with something more original.
This is the hallmark of the immersive writer whose analytic mind remains always on duty. The job of the immersive faculty, working in the ‘open mode’, was to come up with funny ideas. The job of the analytic side was to evaluate the ideas as they came out, and say — gently, not cruelly — ‘That’s too easy; I know you can come up with something better,’ and send the immersive mind repeatedly back to its work until it came up with something that passed muster. In this process, the analytic mind is not actually operating in the closed mode; instead, it is acting as a kind of valve or gateway between the open and the closed mode, making sure that only the choicest products of the open mode are let through for the mind to work on later in the closed mode. The mind in the closed mode is a factory hand, not an artist, and is strictly limited by the quality of the raw materials you give it to work with. Cleese made sure to give only the best to his.
My own working method is somewhat similar; and since it is seldom or never mentioned in the how-to books for writers, it may be worth describing here. I should caution you that I am not recommending this method to anyone; I would not use it myself if I had a good alternative, because it requires an extremely high level of concentration, which is easily disrupted by distraction or ill-health, and is too tiring to keep up for more than a few hours a day. In fact, I am not operating on that level right now; I don’t generally need it for non-fiction, and particularly not for these informal essais, which are more about exploring my own thoughts in writing than trying to edify others. This is why, when I start up the mill after an interval of writer’s block, I nearly always warm up by writing a couple of essais before I graduate to the more difficult task of composing fiction.
What happens is that I create, as Cleese calls it, a space-time oasis — a set interval of time that I can spend in a place relatively free of distractions. For this purpose I find that working in an all-night diner works well. The light buzz of the customers and staff and background music actually help my mind focus inward on itself: the very light effort of ignoring these noises tends to improve my concentration. Then, too, if I am interrupted in such a place, it is seldom important and never urgent, and I can deal with it summarily (‘Yes, thank you, I’ll have another Diet Coke’) without breaking my immersive state. Cleese recommends making the oasis last for about an hour and a half; I like to let mine run longer — up to three hours or so — chiefly because, as a solitary man, I have the luxury of doing so. The signal to emerge from the oasis is that my laptop battery is about to run out.
Once in this oasis, and once the babble and kerfuffle of what C. S. Lewis calls ‘the voluble self’ have subsided, I give my attention to the immersive experience of the current scene in the story I am writing. I start by re-reading the last bit that I have written, to place myself correctly in the story and remind myself of the setting and mood. Once I can picture it clearly, I let the performance begin to play itself out slowly in my mind. Slowly, because I cannot type as fast as my characters can talk; and still more because I use speech-tags and description and ‘business’ to measure out the dialogue, introducing pauses and caesuras at needful points, so that the characters’ talk comes out at a natural-seeming pace instead of being blurted out in solid chunks. The silences and ums, ers, and you-knows in actual conversation serve the same purpose — to regulate the timing; you might say that I stick in my description of the scene in place of these conversational noises. Often, like John Cleese composing a Python skit, I have to send my characters back to come up with something better to say; something pithier, more emotionally revealing, or more relevant to the scene. It may take several runs through a short bit of dialogue before I get the feel of it right.
As for the ‘business’, since I am a fantasy writer, I am in the position that most writers nowadays find themselves in, and that fantasy writers have been in all along. That is, I cannot measure out my conversations by making the characters play with tobacco. Thousands of fictional conversations have been meted out to the rhythm of sucking on cigarettes, striking matches, and tapping the dottles out of villainous pipes. None of my characters are addicted to recreational fire-eating, so I have to work harder to give them something to do. Facial expressions and body language come in handy here; physical action is still better. If two characters can have a talk whilst travelling across country, or doing something else physically strenuous and requiring a bit of dexterity, this gives me a ready and original stock of business, and gives them a handy way of showing their emotional states without cheap theatrics.
If, as it often happens, the scene is not one involving much (or any) dialogue, I have a similar problem of pacing: how to keep the physical action moving whilst interlarding it with enough static description to keep the reader informed about where things are and what the character’s environment is like. One falls into a natural rhythm here, something like making sandwiches: a slice of description, two pieces of action, and another slice of description on top, with a pickle and mustard shoved in between. This creates a template from which I can suddenly depart at any time to convey surprises and other emotional effects. The techniques of dramatic description and action writing have been discussed and dissected by far abler persons than I; I shall not bore you with my versions of them here. I just find that it is helpful to establish this kind of normal mix or entrelacement of action and description, or dialogue and business, to give each scene a fairly consistent background colour against which the foreground actions will more clearly contrast.
Now, where my technique begins to differ sharply from most writers is that I tend to juggle all these considerations in the first draft. The movie in my head has its scenery and action, its dialogue and business, and I don’t need to put them all across; in fact, it is better not to — to mention only the most telling details, and leave the reader’s mind to supply the rest, in the interest of keeping the story moving. But if I leave a detail out of the first draft, I generally lose it for ever. Sometimes, indeed, I will so compress an action scene that I leave out the action itself, describing only the lead-up and the after-effects; but this is done deliberately and for effect. There is, for instance, this bit from the opening chapter of The End of Earth and Sky:
The stranger made no answer, but brandished his stick like a weapon. There was a flash of light and a loud bang. Bron’s grin vanished as he fell backwards in the thicket. Dropping his pack and drawing a broad-bladed dagger, the stranger sprinted across the meadow to pounce on Bron’s twitching form. He jerked Bron’s head back and slit his throat with one swift stroke. Then he stood up, licked the blood off his blade, and threw his head back in an ear-splitting yell of triumph—
A crossbow-bolt sprouted from his neck like the first crocus of spring. He fell heavily to the ground. Håkar ran at him, screaming with rage, stabbing wildly with his spear. Bright blood fountained from the stranger’s chest. My bow dropped from my nerveless fingers, and I was noisily sick in the grass.
It is only afterwards that the narrator notices that his crossbow has been discharged, and puts two and two together and realizes that he shot the stranger. In the heat of the moment he acted on intuition or instinct, faster than his conscious awareness could follow; and I defer the realization to the time when his conscious mind caught up, instead of giving a blow-by-blow description in real time.
Now, some writers would have written the blow-by-blow description first, and rearranged the order of the narrative in revision (or not). But my method is almost entirely immersive. I put myself imaginatively in the character’s head, drawing on memories of certain fights I had in my younger days (in which, however, nobody was killed or seriously hurt), and played out the whole scene almost exactly as it stands — including the delayed realization. That was there from the beginning; I knew by experience that there was no time for him to notice trivial details like the fact that his crossbow had happened to shoot its bolt. Still less would he have paid much attention to his own actions. At such moments you are painfully aware of what is happening all around you, but you don’t have the close awareness of your own actions, almost pornographic in detail, that writers are liable to impute to their characters when writing from the Old Baloney Factory.
I give this example in some detail, because it is so very unlike the method usually prescribed by how-to-write books, and especially by gurus of the ‘all writing is rewriting’ school. It is, in fact, true that this whole scene was done as a revision: once I had done a draft of the story, I saw that the main conflict was not introduced early enough, and devised poor Bron’s death as a way to bring it forward into the opening chapter. But the scene itself, once written, was hardly revised at all. I cut it a little, and added one or two details that I wanted for other purposes later in the book, but which would be easier to take in if I introduced them at that earlier stage. Sometimes all the revision I do to a scene is to stick in an explanatory sentence, so as to save myself a paragraph of tedious exposition somewhere else in the story.
In a similar vein, I have been known to hold up in mid-sentence for fifteen minutes whilst sending the immersive mind back, over and over, to come up with a better bit of business than yet another facial expression or tone of voice. I know what the characters felt at that point in the scene; but how did they reveal it? What, as a poker player would say, were their ‘tells’? Many writers would be content to put down something pat and obvious at that stage, or leave it out entirely, and go back and fix it in revision. But I find that if I do not get all the emotional ‘beats’ right the first time, the action will not go; the immersive mind becomes lazy and sloppy, and relies on the analytic mind to come up with second-hand ideas to fill out the sketchy scene — the baloney factory again. So, like John Cleese staying long after quitting time until he came up with an original idea, I brood over a single detail until I get it exactly right. That detail, for all I know, could be the key to the character or to the scene. It often is.
Or I will hang fire over a verb — verbs are the trouble in writing; verbs are the killers; there are never enough vivid verbs, even in English, which is why people have such need of adverbs — because I see a bit of action happening in my mind’s eye, I know exactly what it looks like, but the scene will not make sense unless I can describe it unambiguously. And I can’t afford to stop the show while I draw a verbal diagram thirty words long; I have to cramp it into half a dozen words and keep things going forward, or the pacing will flag and the illusion of motion will be lost. It seems like a trivial thing, but every time you take ten seconds’ worth of reading to describe an eyeblink’s worth of activity, you strain your readers’ patience and risk throwing them out of the story. Immersive readers are not a forgiving lot; you may make a hundred mistakes that they will never notice, but as soon as you make the one mistake they do notice, they are out of their trance and may never get it back again. Some readers will be bounced out by things that are not, strictly speaking, mistakes at all — matters of taste, things that call up associations in their minds that you could never have guessed or made allowance for. This is bad enough. The immersive writer’s great and unending task is to avoid giving the reader more excuses for bouncing out of the story and putting it aside.
As you may imagine, it takes a good deal of practice to wear this many hats at once — to be attending all at once to the movie in your head, the flow of your prose, the choice of individual words, and the timing of everything in the scene. To do it all in one draft is an extraordinarily difficult juggling act; I find I can only do it at the peak of my powers. I still do a good deal of revision afterwards: cutting, pruning, inserting bits of foreshadowing, rearranging scenes, and so forth. But if I don’t get all the essentials of a scene to make it work on the first go, I never can manage to fix it afterwards. The structure must be sound or the building will not stand, no matter how many coats of paint I apply to hide the defect.
Since I can only do this for a few hours a day at best, and not at all on days when I am ill or distracted by other cares, I am not (by the standards of analytic writers) prolific. I keep reading advice from writers who think nothing of knocking off a million words a year, and prescribe career plans to other writers based on the assumption that if you don’t write a million words a year, you are a mere piker or poseur, and will never find or deserve an audience. Such writers are either analytics or freaks of nature. Not even Ray Bradbury in his story-a-week days, or Isaac Asimov at his most fecund, averaged a million words a year. The complete works of Shakespeare are less than a million words all told. I am afraid that when I hear these million-a-year folk talk, I am seized with an urge to take up a different line of creative art, and express my cognitive dissonance by performing an interpretive knife dance on their entrails.
That, from my own point of view, is what it is like being an immersive writer — a certain kind of immersive writer, the kind that I am. I don’t recommend choosing to write that way. But if you happen to be afflicted with a process that is naturally similar to my own, I hope I have given you some ideas on how to manage it a little more smoothly; or at least some company for your misery to love. And if you are not so afflicted, I hope I have given you a bit of insight into life on the other side of the fence, where the grass is only greener because we have painted it that way, blade by painful blade.