Infodumps – those long passages of narrative summary that provide a huge wodge of background or plot development or characterization – have an undeservedly bad reputation among would-be writers. The allergy to infodumps is a bit of stylistic advice which is largely peddled to beginning writers, but which is not upheld by looking at real live published fiction. Infodumps that are ineffective, boring, annoying, or unnecessary should be cut, obviously, but that is by no means the same thing as “all infodumps are horrible and a sign of bad writing.” James White, for instance, used infodumps to great effect for decades in the Sector General books; ditto David Weber in his Honor Harrington series, Patrick O’Brian, and many others.
There are three basic approaches to fixing a story when someone has complained that it is infodumpy: 1) You can rewrite to remove the infodump, 2) you can rewrite to make the infodump work in the context of the story, or 3) you can ignore the advice and leave the story alone because you know your critiquer well enough to know that he/she has absorbed the “no infodumps rule” and is therefore not actually assessing whether your infodump really works in context.
Rewriting to remove the infodump is often appropriate if the infodump is ineffective, boring, etc., but in quite a lot of cases, the problem is not that the infodump is the wrong technique to choose, it’s that the particular author doesn’t know how to write good infodumps, or doesn’t know how to make an infodump work in the context of the story and viewpoint he’s chosen. Rewriting to remove the infodump will do nothing to solve this underlying problem, if it is present, since simply removing all infodumps provides no practice whatever in how to write effective infodumps … and odds are that sooner or later, the writer will need to write an effective infodump.
So the first question the writer needs to answer is: What is the most effective way to give the reader the information necessary to understand the story? Should I use an infodump, or something else? This is generally a question of pacing, rather than structure, because traditional infodump mechanisims like narrative summary or two-page blocks of lecture in dialog lay out information a lot faster and in a more condensed fashion than the slow revelation of needed details in a scene or during a conversation.
Assuming you decide that you need the infodump, the next question is how to make it work. One of the most effective ways to do this is to arrange things so that the infodump is of information that the reader already wants to know. This is one of the reasons why long prologues full of the background history of the world seldom work and are commonly cut by editors – when they pick up a book for the first time, most readers are more interested in the story or the characters than in all the cool history the author has worked out. It isn’t until later, when the information becomes important to the characters and the story, that the reader wants to know more about the background.
Another technique is to lay out the story-problem as a central part of the infodump, or (if the story problem has already become obvious), give the infodump information in such a way that it clearly makes the story-problem worse. If, for instance, you have two characters at the top of a cliff preparing to go down, and you need to infodump a bunch of geological and geographic information, you could probably get everything you needed into an infodump that described, in gory detail, just how high and pointy and hard to climb this cliff is (because of relevant geological facts) and just how many other people have died trying to climb down it (for various geo-political reasons). It’s a matter of focus: the information you want to infodump is all still there, but the focus is on some plot-point that increases the tension.
Some voices and stories lend themselves to effective infodumps more than others. James White makes extraordinarily effective use of the sessions in which his doctors are briefed on their new patients. He slides effortlessly from the setup conversation into two- or three-page narrative summaries of the important background information (which the reader already wants to know because of the hints in the setup conversation) and then back to his fully dramatized scene. A first-person narrator can get away with infodumping information a lot more easily than a third-person narrator. The trick here is to make the narrator sound as if she is musing on or analyzing or explaining to herself something that is important to her at the moment. A story that covers a lot of time – months or years – also often requires a good many chunks of narrative summary to fill in what happened. Memoir tends to have a lot more summary than dramatized scenes.
I’m currently getting lots of practice with this because the Frontier Magic books are first-person and an imaginary memoir and covers years per book. When you have thirteen years of a character’s life to cover, which important bits of her life story you show and which ones you summarize turn out to be a lot less obvious than you might think. Pacing and structure and flow become a lot more important, too. Thank goodness that problem will decrease in the next two books, because they won’t be covering fifteen years in one volume.
At least, I think they won’t.