From author Charles Stross:
I get mail. And sometimes I want to share it with you. Especially when it’s email like this one, from Jacques Mattheij:
Question for you: One HN thread caused me to wonder about this: What would a technological society look like that somehow managed to side-step the written word? Would such a thing even be possible? If not why not?
Just to keep you awake at night 🙂
This question caught my attention like a snagged fingernail, and it’s still pulling at me: here’s my first cut at an answer. I’m taking the no-writing parameter seriously as a limiting condition: what level of technological society can emerge in conditions which preclude writing—for example, if it’s forbidden for religious reasons? I’m going to treat this as holy writ for purposes of this thought-experiment: rules-lawyering around the no-writing rule in the comments will be treated as Derailing and deleted, with one special sort-of-exception which I’ll explain near the end because it opens up a bunch of interesting consequences.
My rule of thumb answer is: it wouldn’t be possible for human beings to develop a technological civilization—at least anything beyond roughly 17th century levels of energy utilization and mid-19th century levels of agriculture—without some form of record-keeping technology. And without writing they might never get that.
The reason is memory capacity. Yes, we can memorize lengthy texts when assisted by verse metrics as a form of mnemonic—the Iliad and the Odyssey, the Koran—but the format is error-prone, transcription is at least as time consuming as copying a mediaeval illuminated manuscript, and the “books” are high maintenance (they need food, clothing, and shelter). I don’t know how many books one human being can memorize, but even if the number runs as high as two digits (which I think would require a very rare level of memory) you’re then faced with the problem of what to do if one of your books gets cancer or dies of old age. So not only is copying more expensive than in a mediaeval monastery’s scriptorum, but the substrate onto which “books” can be copied is extremely expensive (because we’re coming at this from a pre-industrial situation where agriculture is labour-intensive because there’s no copious supply of cheap energy). To put it in perspective, if one “book” can memorize five texts, then those five texts represent an entire productive human lifespan’s worth of opportunity costs.
. . . .
And then we run into mathematics. Assuming they figure out binary, integer arithmetic on fingers and toes gets you a long way for basic counting, multiplication, and optionally subtraction and division. But I’m not sure how they’d explore reals, let alone algebra or calculus, in a notation-free environment. I imagine tally sticks might work if our sophonts have opposable thumbs, but then we’re cheating and getting into writing systems by the back door.
. . . .
Law and arbitration is going to be problematic. The Mediaeval Icelandic parliament is said to have started each session with a recitation of the legal code; any law that no sitting legislator could remember was deemed to have passed beyond the sunset. This is thus shown to work, after a fashion, for non-literate societies up to a mediaeval level. However, reliance on memory means that a case-law system simply can’t develop, except in the sketchiest of ways.
Link to the rest at Charlie’s Diary
Here’s a link to Charles Stross’ books. If you like an author’s post, you can show your appreciation by checking out their books.