Show and Tell

by Brian on July 12, 2003

Andy Egan at Philosophy from the 617 responds to some of the debate Henry’s Harry Potter post produced, and in doing so brings up an interesting point about how we judge fiction. Lots of people say that in fiction, especially visual fiction but also in written works, the author should show the audience what happens, not tell them what happens. But what exactly does this rule mean?

bq. One first-pass thought is that it’s something like this: when you’ve got a choice between making something true in the fiction directly (by, e.g., writing down a sentence that expresses the proposition you want to make true), or making it true by saying a bunch of lower level stuff that entails it (or demands that we imagine it, or whatever), you should always do the second. In other words, it’s always better to force the higher-level facts by explicitly fixing the lower-level facts. But that’s pretty clearly crazy- it calls for books written entirely in the vocabulary of microphysics, which would be incredibly long and boring and incomprehensible and awful.

Andy goes on to argue that there’s some privileged intermediate level of detail that’s appropriate for writing at in particular the “level of description that our imaginative faculties operate at”. I’m not entirely sure what that is. Are concepts like CHAIR – things I can easily imagine but which I always imagine as having more detail than just being chairs – at the “level of description that our imaginative faculties operate at”? I don’t know, but perhaps we can work out an appropriate privileged level.

(In these cases I always think the class of monomorphemically lexicalised maximally precise concepts can play a crucial role. But that’s probably just because I (a) read too much Fodor and (b) use too much jargon. So ignore that suggestion.)

But I think the better response is just to accept the allegedly absurd conclusion that Andy offers. The “show, don’t tell” rule does imply that everything should be written in microphysics. But it isn’t the only rule authors should follow. Among other salient rules, there’s the “keep it simple, stupid” rule, not to mention the rule against repetitiveness. If we wanted to try and find one rule for writing it would look something like “Show rather than tell as much as possible consistent with keeping the work accessible to the average reader and not being boring and …” Or, in slightly fewer words, “Show rather than tell as long as ceteris are close enough to paribus”. But once you’ve said that you may as well go back to the simple “Show don’t tell” rule, remembering that if you forget the other rules it will have fairly silly consequences.



Avram 07.12.03 at 7:34 pm

I think a better phrasing, rather than “always use the lower-level stuff”, is “rather than telling your audience what you wish them to observe, tell them what you think will enable them to deduce what you wish them to observe”. This implies a different level of detail for different assumed audiences. If you’re writing for people who can deduce human-level activities from observation of subatomic particles, fine, but most books aren’t written for that audience.


Richard Johnston 07.12.03 at 11:43 pm

It’s a lot simpler, isn’t it? ‘Show, don’t tell’ always strikes me as a handy way of helping writers to remember to not treat the reader as a fucking moron. (sic., but I split my infinitives without any guilt.)


Matt Weiner 07.13.03 at 12:01 am

Brian, I think you’ve quantifier-shifted Andy’s conclusion. It’s not that there’s a single appropriate level of description at which all fiction should operate–it’s that, for each bit of fiction, there’s an appropriate level at which it should operate.

So you could say something like, “Mauritania and Ruritania had been at war for three years, and the 17th regiment had been bivouacked in town for a week. A soldier was sitting outside Laura’s door. He jabbed his bayonet into the ground, pulled it out, and jabbed it into the ground again. The front lawn was covered in uprooted flowers and dirt.” (Of course, you would want to make it much better.)

This has some very high-level description of countries and regiments, and some low-level description of the soldier’s actions and the state of the lawn. It skips the intermediate-level description of the soldier’s boredom and the chaos inflicted on Laura’s tidy existence by the occupation. This is appropriate–well, it would be appropriate if this were actually good fiction–because the reader’s imagination operates at these different levels. We can imagine Laura sitting in her house, thinking about the war, and watching the soldier.

Incidentally, it’s worth noting that rules for good fiction are incredibly loose. “Show don’t tell” is really writing-workshop talk for “Move to a lower level of description.”


Shieva 07.13.03 at 8:01 am

Here’s my thought:
Perhaps the “Show, Don’t Tell” rule is grounded in the subjective paradigm (which, as far as I understand it, is the idea that the truth cannot be communicated, or that a person cannot learn the truth from someone else, but rather must experience it for themselves, and therefore the best way to teach someone is to create a situation conducive to their having such an experience). If the rule is based on this, then perhaps what an author needs to do in order to follow it isn’t to express propositions by stating other propositions which entail what they wish to express. Instead, it would be by doing whatever they need to within the writing to allow the reader to experience the truth of the proposition they want to communicate. So if they want to convey that ‘the woods are dangerous,’ some bad ways to do it would be to simply state the proposition, or to describe the ways in which the particles composing the wood and those composing the characters stand in relation to one another. But perhaps a good way to do it would be to create suspense, by drawing out descriptions of events, or giving the reader plot-relevant knowledge that the characters lack, or by making the descriptions of the characters stand in stark contrast with those of their surroundings . . . the use of symbolism, foreshadowing, and also manipulation of the medium of communication (altering sentence length, or using assonance, alliteration, etc.) are some other ways an author might create an experience for readers, rather than simply telling them about it. This is what I think the “Show, Don’t Tell” rule is getting at.


Angus 07.13.03 at 2:53 pm

Maybe I’m wrong, but I’ve always assumed that “show don’t tell” was a reference to the distinction in classical poetics between mimesis (“showing,” ie what drama does) and diegesis (“telling,” ie what narrative does). All fictional narrative is, strictly speaking, diegesis (except for quoted speech), but the “show don’t tell” maxim seems to me to express the realist view that fiction should. as far as possible, approach the condition of drama by making the mediating voice of the narrator as “invisible” as possible. (As such, of course, it’s a highly tendentious and disputable maxim.)


Matt Weiner 07.13.03 at 3:32 pm

Richard Johnston, posting after I stopped typing, demonstrated his superior grasp of one of those other maxims: “Be concise, dammit.”


Andy 07.13.03 at 4:13 pm

Bringing in other rules would definitely do the trick, but to the extent that I’ve got intuitions about this, it seems like there’s just no pressure at all to run all your descriptions in absolutely bottom-level terms. Not clear how much weight that ought to have, but it’s something.

Matt’s right about what the claim was supposed to be- the right level of description’s pretty clearly not constant. There’s actually a pretty nice explanation of that if what’s going on is multiple-rule-interaction type stuff, but it’s harder to explain on my (ripped off from Ishani) not-yet-a-theory. And we have to cash out “level of description that our imaginative faculties operate at”, which I happily admit is a pretty appalling hand-wave as it stands. Still…


derrida derider 07.14.03 at 12:07 pm

Why all the fancy analysis? People are wired to be curious – and “show, don’t tell” caters for this. People enjoy using their mental models of the world to fill gaps – it’s just going with the flow of their neurology.


Ruth Feingold 07.14.03 at 3:23 pm

“Show, don’t tell” was a maxim of the New Critics, and a principle valued at mid-century (that would be the 20th). Flannery O’Connor, for example, used it as her mantra.

The point is not only that different kinds of writing work well in different contexts, but that both literary AND critical styles change. What’s considered good literature now may not be “good” in 50 years. So saying what authors should or shouldn’t do is a tricky proposition.

Oh dear — I’m sound pretty Postmodernist, aren’t I?

Comments on this entry are closed.