There are two basic methods for describing anything: 1) the lump-of-description, and 2) in bits, over time. They aren’t mutually exclusive, so you could also claim the combination as #3.
Lump of Description
Generally, this is somewhere from a paragraph to several pages of detailed description of whatever the author is describing. The logical place to put a descriptive lump is when the narrator or viewpoint character first encounters whatever-it-is. Usually, this is when the person/place/thing first appears on stage, but sometimes this sort of description gets used when the viewpoint character is remembering something in detail, or when they are describing something in dialog with another person. Theoretically, this will prevent the reader from creating a mental picture of the person/place/thing that is totally wrong in some important way, but this never works a hundred percent. There will always be readers who read “he walked into a cluttered living room” and who mentally picture a Japanese tatami room with minimal furnishings and lots of empty space.
The longer a lump of description is, the more interesting and/or relevant the information needs to be in order to keep the reader from skimming or skipping it entirely. One way to accomplish this in first-person or tight-third person is to periodically provide a character’s reaction to what they are describing.
“He walked into a cluttered living room and wrinkled his nose, half expecting to smell something spoiled that was hiding under the magazines, clothes, and miscellaneous items littering the floor, draped over the couch, and crammed onto every flat surface he could see.”
One could then go on to mention specific items the character considers tacky, or have him wonder why anyone would keep a mug with a broken handle or a bright red parasol with a beaded fringe.
Getting reactions in is easiest if one’s viewpoint character is particularly observant and/or especially chatty. “The minute I walked into that room, I just knew there was something spoiled and rotten underneath all those magazines and old clothes and things. I didn’t smell anything funny, but things don’t smell once they’ve dried out enough, and from the layer of dust on everything, there’d been plenty of time for drying. I couldn’t tell whether the couch was supposed to be cream or tan or gray, there was so much dust on it. Well, on the parts of it that weren’t draped in T-shirts and moth-eaten sweaters and bell-bottom pants straight out of the 1970s.”
Making a lump of description interesting is often a matter of what the writer leaves out. Is it actually important to the story that the living room is cluttered with magazines, old clothes, and tchotchkes? Or is the story going to flow along just fine, even if the reader is picturing a tatami room? This can be extremely difficult for some writers to tease out – they know what the room is supposed to look like, and it bugs them if they think a reader will get it wrong. But “a reader will get this wrong” is not the same as “if readers don’t get this information, they won’t understand the story.” (And even when the writer believes the latter, they’re often wrong.)
One of the oldest and most useful tricks in the book is to describe smells, sounds, and textures – things that appeal to senses other than sight. (Taste belongs in there, too, if you can manage it, but unless you are describing someone eating/drinking, it’s generally much harder to work in.)
Over Time
The second method of describing a person, place, or thing, is to provide details as the viewpoint character moves through a scene and different things become relevant to what they are doing. This can be particularly effective when you’re describing a place or object that your viewpoint is encountering for the first time. “He reached for the doorknob, but stopped when he noticed a green-and-black crust of … something … that covered it. He dug in his pocket for a tissue, and used it gingerly to turn the knob. The door opened a few inches with a rustling sound, then stuck against something. He shoved, then shoved harder. With a crunch and a ripping noise, it gave way, and he took a careful step into the room. Peering around the door, he saw the litter of papers that had kept it from opening, now crushed and jammed against the wall.”
This takes longer than a lump-of-description, but it doesn’t have the stop-the-action effect that a long descriptive lump does, because the action is mixed in with the description. It’s also a lot easier to work in reactions, as well as smells, sounds, and textures, because this type of description doesn’t drop the reader out of the immediate action for more than a sentence or so at a time. The viewpoint character is always front and center, experiencing the place, handling or closely observing the item, evaluating or reacting to the new person or old acquaintance.
Combination
This kind of description starts with a small lump-of-description that sets the scene or establishes a few basic aspects of the person who’s just shown up, then adds details as they affect the viewpoint character/narrator. For instance, the writer gives a brief description of the customer who just entered the office – possibly just height and hair color, or possibly adding one really striking/unusual detail, like the purple suit he’s wearing – and then mentions his nasal voice when he greets the receptionist, his too-firm grip when he and the viewpoint character shake hands, and the way his eyes shift and the tension in his shoulders changes when the POV shuts the conference room door.
Some writers find it easiest to write expansive descriptions and then cut them back in revisions; other writers spend their revising time adding reactions, specifics, and sensory details to phrases like “a cluttered living room” or “a blond guy.” Do whichever seems to work better for you. Sometimes, you will find that you over-describe place and have to cut those, but under-describe people or things, and need to add to those. Or vice versa. The main thing is to pay attention to what you are doing.
One of the things I’m big on in description goes back to my first feeble efforts at description; I started out with sentences like “He had a nasal voice” and “His grip was too firm.” Verbs keep things active, if you use dynamic ones. “Had” and “was” are pretty limp by contrast.
Yours are much better, with “greet” and “shake hands” and the rest: “…his nasal voice when he greets the receptionist, his too-firm grip when he and the viewpoint character shake hands, and the way his eyes shift and the tension in his shoulders changes when the POV shuts the conference room door.”
I prefer to be iceberg-like in descriptions; to have deep, detailed description in-hand, and then only hand out small fractions as needed. But I worry (perhaps excessively) about contradictory descriptions and making accidental, erroneous changes.
Also, something I enjoy reading (and writing, when I can pull it off) is description as an interesting little fictional non-fiction essay, especially when it is about something Not Found In Kansas. The breeding and raising of domesticated griffonettes, for example, or that beautiful chess set of Onyx-wood and Sky Oak and the special techniques needed to carve those materials, or the bath-salts, perfumes, and silken clothing used by Baron Sung of Draco-Agorialith to prepare elven slavegirls for sale to the harems of the Satraps of Ifor.
interesting little fictional non-fiction essay
I enjoy those, too. Done well, they can provide insight into the world-building and/or a tremendous sense of the pov character’s competence in that particular area, as well as being interesting in their own right. (See for example the voice-overs in Burn Notice.)
I have trouble with this. It’s not that I know “what the room is supposed to look like, and it bugs [me] if…a reader will get it wrong.”
It’s that I genuinely have difficulty determining when specific details are important to the story and when they are not. Thus I tend to both over-describe and under-describe.
My beta readers save me, but I wish I had a method that didn’t rely so heavily on others! Any tips?
Here’s how it works for me:
My native mode of writing is fast and flat: characters can think and talk to other people for pages without doing anything other than shrugging or nodding and without interacting with anything at all, so I had to learn to slow down my writing and look around every time I enter a new location and work out what’s actually there, to be in the moment, in the character’s body instead of their head. What I do is draw maps and diagrams of places, and I find that very often story flows from that. (This is why I cannot retrofit description; description changes plot) I walk with my character and work out what they can see instead of just assuming that everyone in a room is free to speak to everyone else.
How far into a room does my character walk? Who or what may they see? Observe out of a window? If they’re in a place they shouldn’t be: how much noise do they make?
If I have more than four people sitting at a table: who can make eye contact with whom? Whisper to someone else without being overheard? If they’re waiting for a message: who keeps their eye on the door and never looks at their friends? Who sits with their back to the door and gets all twitchy? And what do they DO when they’re twitchy? Shred a paper napkin? Unpick a loose thread? Drum their fingers on their leg? And does someone else notice, and how do THEY react? Just by going through this process my story acquires a depth that isn’t there if I concentrate on ‘writing the story with the occasional detail thrown in’. It’s an organic process, a dialogue between ‘what could be there and what does it mean for the character’ and ‘what would I like to be there because it would enhance the story’.
Agreed! Of course, my natural mode seems to be “overdescribe body language/motion, air temperature, and pretty landscapes – because all of that seems extremely crucial to the story while I’m writing the first draft – while underdescribing characters’ appearances unless they’re conveniently unconscious or dazed (at which point their appearance is one of their only halfway-interesting aspects, so I go ahead and overdescribe that, too)”…so my advice is to learn your patterns of over- and underdescription, if any exist, and delete or add stuff accordingly while editing. (Also, to show great appreciation for your beta readers, because good beta readers are invaluable, but I suspect I don’t need to tell anyone here that…)
What I do in practice is describe through the lens of the POV character’s experiences and biases. But I’m not able to analyze what to describe from the stance of what the story needs.
This character notices this and ignores that and revels in the other. But are all those elements necessary to the story? I don’t know!
Ouch. That sounds like a good practice for characterization, but, you’re right, not necessarily a guarantee that each viewpoint character’s interests will be plot-relevant. I’m sorry not to have more useful advice than “Yep, beta readers are wonderful for catching what we can’t easily observe in our own work…”
I’ve both expanded and cut down descriptions BUT if you have two details and know you need one, but not which, put them both in. Much easier to revise one out than remember what the other one was.
is it actually important to the story that the living room is cluttered with magazines, old clothes, and tchotchkes? Or is the story going to flow along just fine, even if the reader is picturing a tatami room?
For me, the answer is it *should* matter. If it doesn’t matter, I have failed to write the story I wanted to write, because most people don’t live in white rooms and hazy landscapes with steady ambient temperatures. Their world is real to them and should be real to me.