Describe this

Can you describe the scene? Look at it for 5 seconds,  look away, try and ‘redraw’ it in words…  (it’s actually not a fantasy cover, but a modern hamlet in the mountains. You can see powerlines, but not at a glance) Yeah. well the same applies to characters or people…

“Can you describe the villain who perpetrated this hit-and-run for us, Ms. Smith?”

“He was tall, blond, and was wearing a MAGA hat.”

The one thing you can be sure of is… if they catch the villain, she will be none of the above. She won’t be tall, as in above average height, she will probably have indeterminate color hair, and most likely won’t even own a hat of any description. Might have been wearing a pinkish-red bandanna.

Was Ms. Smith doing a Jussie?  Was Ms. Smith trying to protect the guilty party?

Quite possibly not.

The honest truth is most of us are not actually much good at describing people.  We may be good at recognizing them and picking up slight differences, but even then we’re not much good at minutely describing them. There are exceptions to this, of course, but they’re exceptions.  In Ms. Smith’s case she saw the incident in a brief period of time, was shocked by it, and in practice retained some brief impressions, which her mind then tied to what she thought a villain would look like, courtesy of her diet of mainstream media, and a fairly limited intellect. The perp may well have been taller than her (therefore ‘tall’ to her) and had light hair rather than dark, and something of a shade that was vaguely red on their head. The sex she assumed, along with their politics.

This of course is not only confined to Ms. Smith.  It’s most of us. I’m not much good at describing people either, and when it comes to ‘what were they wearing’ I almost always get ‘naked or not naked’ right, but I wouldn’t go a lot further than that. Clothes are not a big interest in my life.  I’d be far more likely to notice and remember what someone was reading than what they were wearing.

Of course, for those of us interested in clothing, that could well be different. I have a friend who might not have a clue what the face or body shape was, but would describe the fabrics, the cut, the style, the trim, and the type of stitch and color of cotton used. Our perceptions are shaped by our interests and our background. I have another friend who agrees wholeheartedly that readers (who vote with their pockets) not editors (who at least in theory might know a great deal about books) nor English Lit professors (who once again study books at length) should choose what they want read. But she believes firmly that foreign policy decisions should not be taken by those who voted for them (or their representatives) but be directed by the unelected bureaucrats who study these things.

Because that is her world. In her head the bureaucrats know what is best for us.  Editors and English Lit Profs don’t.

It’s not implausible that editors and English Lit Profs see themselves as the knowing guides who know what is best for us to read, but think elected politicians should make foreign policy decisions rather than bureaucrats – especially if their favored politician is in power, and they disagree with the bureaucrats.

Now I’d come down on the side of letting people choose their own books and foreign policy. If they’re terrible choices… well, there is no real evidence that editors get it right, and I am less than convinced that expert bureaucrats do either. Hopefully next time they’ll choose something else. YMMV.  My point is that the individual’s perceptions (and therefore descriptions) are shaped by the individual and their background. Fact or logic or even the success of books or foreign policy may not even enter the race.

Now, come back to books and descriptions.  We’re not trying to catch perpetrators, or decide foreign policy.  All we’re doing is trying to appeal to readers so we can sell lots of books. And we go into this knowing 1) Readers are individuals. No two are quite alike.

2) They will notice and glom onto details that they are interested in.  And those can very narrow fields.

We’re up shit-creek, aren’t we?

Ah. But there are further points which allow you to use things which seem against you to your advantage — like a judo-master might.

3) They will – despite obvious evidence to the contrary — see what they think they ought to see. 4) They will ignore things, or skim over them, that don’t interest them or comply with their interests

5) While the overall set of their interests is unique, subsets of those interests will be held by many.

6) And really, if I say ‘red-haired woman’ as my description, the image created in my mind is not the same as the one in your mind. The hair could be any one of 23 shades of ‘red’, and be short, long, straight, wavy, curly or even frizzy… And ‘woman’ cover a range of ages, shapes and other attributes. But we both think that description quite adequate, possibly even charming, depending on how much we like red-heads.

Look, your mileage may vary. I expect it to. But what I do – and what I have noticed far more successful writers than me doing, is to make sure I get the details in the main, popular subsets. I have ‘go to’ experts in those areas that I get to check my work (or their interest parts of it) even when my own expertise in that area is fairly extensive. There ARE readers with no interest in guns, boats or horses. They will skim those detailed descriptions (if you use them). The knowledgeable folk will however pan your book even if there is a brief, undetailed mention… which is wrong.

Secondly I try to ‘frame’ any description with a few small precise details… and actually leve the rest rather general (see point 6) unless it is plot relevant.

Thirdly you ARE to some extent working on stereotypes (point three). If you want to challenge/change those, you have to build that in as carefully (and with suitable foreshadowing) as a major plot element. Or you will have (and I have, and I know several dozen other authors report the same thing) the reader (or at least some of them) has reached their own (entirely unrelated to the author’s intent) conclusions. Their image is certainly a million miles from what you were trying to portray, but they can well believe you terrible or brilliant from it. I have been accused of not knowing any South African women, and making them seem just like American women.  (Liz De Beer in PYRAMID SCHEME).  Um. I’m not American and knowledge of American women, in the biblical and other senses) was miniscule when I wrote that. I was in fact modelling the character off two real South African biologists that I knew very well. But that didn’t match the reader’s stereotype of South African women. You can’t please everyone.

Finally… there is a place (and proportion) for precise description. But it really works best not in huge easily skipped chunks. Big lumps are a suitable requirement for vomit, not story telling.

Image by Jörg Peter from Pixabay


  1. I have to say that I pay a lot more attention to scenery pictures than those of people. So in your example I look at everything I can to try and figure out where it is, what country it is in etc. (I’m guessing Switzerland FWIW)

    I seem to recall some well-known author (Lois M Bujold?) say that every book was a collaboration between the author and the reader and that therefore no two readers ever read the same book. I think there’s a huge amount of truth in this and that therefore there’s a lot of benefit to relatively sparse descriptions because the reader can fill them in the way he wants.

    Of course that’s fine until the author needs to add some critical part to the description in a later part of the book and this totally clashes with the image the reader had in his mind…

  2. Most detailed description in fiction is irrelevant or worse. The typical reader isn’t there for your word-pictures, but for an emotional experience — and that comes almost exclusively from what your characters say and do. Thus, fictional description should be limited to those aspects of a scene that the current viewpoint character notices (or should notice) in the course of his actions. That might include details about what another character looks like, but here there be tygers: Readers’ notions about what a character should look like are often very strong, and clash with specific descriptions of characters’ appearances.

    Elmore Leonard was once asked by an admirer why his books contain so few descriptive passages. He replied, “I try not to write the parts that people skip.” There’s a lot of wisdom in there.

    “Everything not essential to the story must be ruthlessly cut away. If in Act One you say that a gun hung on the wall, then in Act Two or Three at the very latest, it must be discharged.” — Anton Chekhov

    1. And one of those tygers might not be obvious: Once you’v given that detailed description, it becomes what the readers expect, and you’ll have to keep track of it…. because if it changes unexpectedly, the readers will be even more put off.

      Prime example: David Weber set the early description of Nimitz the treecat (and ‘cats in general) as being reasonably cute and cuddly. Alien…. but not something you’d be scared of on sight. And the covers he was on reinforced that.

      Then Issue 1 of the Honor Harrington comic hit, and there was Nimitz: jet black, looking like the stereotypical witch’s black cat riding on the end of a broom for Halloween. Nothing you’d let within a mile of kids! I was looking forward to the comic and would have bought the whole series… until I saw that.

  3. I notice a lot of the ‘translation error’ seems to come in translating impressions into descriptions– my husband is good at doing write-ups for stuff he wasn’t there for because he learned how to get people to tell him what their impressions are instead of demanding precise details.

    For example, he would’ve instantly clicked on to the ‘tall’ and flagged it as ‘taller than the speaker or gave the impression, check for elevation differences.’ Likewise “blond” would be everything from white and peroxide to “several shades darker than rich honey, but not dark brown.”

    Oh, and it’s a hit and run, so they were seated- which brings in issues of torso length, pillows in seats, adjustments, the red being something BEHIND the driver….

    A lot of screw-ups happen because either police want more detail than someone is trained in or people THINK they have to be pseudo-precise.

    For clothes, I SUCK at seeing what they were– but I do classify as “nice” (casual office wear, for example), “practical” (Jeans and t-shirt or non-fancy flannel), “trashy” (visibly torn and/or dirty with underwear visible, ranges from ‘flirty trashy’ say thong visible to ‘treat as possibly mentally ill trashy’ with dirty u-top undershirt worn as a shirt, aka, “the wifebeater.”) and “fancy.” (Church clothes, for another shorthand– you’d be a little surprised to see it in an office, but it wouldn’t be unsuitable.)
    Sadly, “telling” might beat “showing” here; part of why I like stuff written in Dresden Files POV is that Harry can interpret a lot of stuff for us. He’s got some blind spots, but he also knows more than I do about magical history. And probably clothing in Chicago.

    FWIW, I recognized the photo as a flattering, modern rural scene– and it made me homesick. 🙂 Getting the shot so you CAN’T instantly see powerlines and such is a pain; I think it’s called soft focus and something to fiddle with the depth perception that helps, but….oh, what a nice lookin’ place.

  4. As a kid, I liked illustrated books, even if such illustrations were infrequent. I am specifically referring to Lattimore’s translation of the Odyssey. Drawings of scenes are sparse, and not so finely done as to oppose any one reader’s visualization of the action.

    1. Huh. I *hated* them. The illustrations were usually insultingly bad, and often had no connection to any nearby text. And worst, they didn’t match my internal view of what was going on.

      1. That would be what makes horse races I guess. On a related note just occurring to me. We all agree that covers, a visual element, are essential marketing tool. Yet internal art is ignored. I’ve read that a space battle, or anything, really, that appears on the cover, need not be faithful to what actually happens in the story. Is internal art held to a higher standard?
        I ask all this for a reason other than to annoy established authors with my noob questions. I’ve been a painter (canvas, not houses) for a fair time. I don’t have any insecurity as to its marketability, as my pieces have sold well in the past, some for thousands of dollars. So if I decide to illustrate my series, it will not be the case that I’d need to hunt down an artist and then pay him or her. It will, of course, add time to complete each book. The question becomes, will it increase the sales of each book?I know any art will not please ALL readers.
        After all, look at the dreck polluting the walls at many (most?) modern art museums. De gustibus non disputandum est. Some folks prefer ketchup over mustard on their burgers. I think they’re nuts, but don’t think this preference is a moral issue.

        1. On internal illustrations – I haven’t researched this, but my general impression is, because they break the flow of story once you get into the story, they need to be much more precise and add something to the story.

          Readers understand, on some level, the difference between cover and story is akin to the difference between movie poster and movie. So they get annoyed when the redhead is actually blonde, but they’re very forgiving in general. However, the purpose of the cover is to get them to buy the book – just as no one stops in the middle of a movie to go out and contemplate the poster that got them in there, nobody breaks the story flow to contemplate the cover.

          Interior illustrations in adult fantasy, though, are expected to add things to the story – the map must be right, and it tells us that this is going to take place across multiple countries.. or, if it’s a map of a city, we expect the layout of the city to be important to the plot. On the few YA I’ve read with interior illustrations, the illustration has often been of a key plot point – the necklace that’s being fought over, the puzzle box…

          On a completely different, but also important, not: one reason indie authors shy away from interior illustrations is that text is compressible to a very small file, with a very low transmission fee, while images rapidly increase the file size and eat away at net profit per book. Several romance authors, in the early days of kindle, discovered to their surprise that they could finally put in all the fancy drop letters and section separators and everything they wanted to their heart’s desire… but since those transmit as images, their profit went from a buck a book to a nickel a book.

          Something to keep in mind, and file away.

          1. Dorothy, thx for the feedback. You made several useful points I would have not thought about.
            My idea about illustrations as filler for stretches of blank whitespace that appear as a manuscript is formatted. A chapter ending on the wrong page, jazzed up scene breaks,etc.
            The increased cost I never thought of. Perhaps only for special editions, contests,and the like? Lattimore’s illustrations were all B&W drawings. How much color, if any, is yet another variable to stuff into the black box.

          2. On my bullet list, for when I am considering maps, is to look into SVG, which the Kindle can handle since the 3rd generation, and most other dedicated ereaders. Being textual, like the HTML that is actually under the hood in every book, it compresses quite well*. Since it is vectors, the reader can also zoom in on the image (the main reason that I still tend to buy Baen books in hardcover is that on my Kindle – and even on my large computer screen – the detail is horrible).

            * Compresses quite well, assuming you have a generator for it that doesn’t go half a dozen times around the block, with a side trip to Alpha Centauri, to define a simple rectangle. I dug into some of the code coming out of MathML once… shudder.

        2. An writer my no-longer-a-kid likes, Alydia Rackham, does interior illustration and decoration in her books. They do add to the reading experience – on paper. When the author did an ebook edition, the conversion was awful. i had a list of about a dozen things that were wrong, (nothing scaled, the drop caps messed up the lines, so the text was (clearly) out of order,; … and eventually found the ‘report problems with this book’ button and did so. The next day it was pulled.

          So, it works on paper. Be wery wery careful when trying the same thing with ebooks.

  5. Well, I mentally swapped the trees beside the house (barn?) with silos to make it more of an American dairy farm, and added a cut in the ridge where none exists, but go the rest pretty much correct. However, my research field is land use and water, so that’s what I key onto. Slightly rolling, well watered terrain, with ridges behind, so we are at the edge of an upland, probably have relatively fast flowing streams. The ridges seem abrupt, not obviously age softened, so either you have very hard rock, or a “young” landscape, possibly recently glaciated . . .

      1. Could also have been southern Germany or Austria… I almost thought that you had the exact same image as one I used for a cover. (Being rather OCD, I painfully airbrushed out the power lines in that one…)

        I am also horrible at describing people, but found a way to improve. I went to the local mall on a weekend (probably not available to everyone, but another public gathering place should work). I simply sat at a table in the food court for three or four hours each day, looking at the people coming through, and writing down a description for whoever caught my eye. Since people tend to linger in a food court, I could get a description down, look up again, and see where I screwed up.

        (Recommendations – make sure you buy something every so often from one of the vendors. I didn’t have any problem, but you also probably want to be ready to get up and walk casually away if the security people start giving you the stink eye.)

  6. > Can you describe the scene?

    What I see: water stains. Maybe some mold too.

    Staring at it a while, it *might* be stripes of dirtcolor as a highly abstract landscape. There might be mountains or clouds in the background.

    There’s not enough visual information there to make any further description useful.

    No, I’m not exaggerating for effect here. I’m used to seeing panels or blotches of random shapes and colors that have no meaning, or any apparent meaning is unlikely to be intended.

  7. And sometimes it doesn’t matter how you describe something.. the reader’s first impression (which may come from nothing at all) sticks regardless. I experienced this with the Deryni books: author makes a big deal of Morgan being very very blond; it’s even an occasional plot point. BUT when he first came onstage, in my mind’s eye he had black hair, and so it stayed. She could describe it on every page, and I could make a concerted effort to change his appearance, yet the moment I stopped straining to make him look as described (which felt like sand in my brain), he reverted to that originally-perceived black hair. Eventually I gave up and just let him be a mismatch.

  8. Remember that your characters are like your readers. I was once very annoyed by the way two military characters trudged along the road in dangerous territory without once noticing whether the terrain was good for ambushes, or whether there were any defensible positions where two men could hold off a larger band.

  9. One thing writers can do that a photo or a painting can’t, is load the language. Sure, you can’t control what color the reader sees when you state that a gown is red, or even adjective red. But it does matter a lot whether the adjective is ruby or blood or rose or fire.

    (Was recently rambling on on that, here.)

  10. So I was doing a work errand downtown yesterday and there were two young men with dogs. One dog was a beautiful black Great Dane, slender and sleek. The best girl. The other dog was a Siberian Husky. Pretty and fit. Happy dog, fluffy but slim, perhaps older than the Dane.

    The men? I *think* that the fellow with the Dane was black. The guy with the Husky was white. He *might* have had facial hair. Brown hair. Both likely around 30.

    I passed them twice before I noticed that the guy with the Husky was wearing a service pistol, some full-sized black semi-auto… could have been a 1911 but I don’t think so, in a tied down, two black nylon web straps, holster on his thigh. Which, once I noticed it is the only reason I can tell you that he was wearing faded blue jeans.

    Passed them twice… total situational awareness. LOL.

Comments are closed.