Read & Forget

Adventures in the world of electronic books, and how it relates to productivity

This month’s Law & Liberty feature from me is on Louise Perry’s The Case Against the Sexual Revolution: A New Guide to Sex in the 21st Century. Law & Liberty is a US magazine; the book was released there just last week. This kind of deliberate coincidence is all very standard in the world of publishing.

I’d have quite liked my piece to coincide with the week of release, which, given I read the book in the last week of August, should have happened. That it didn’t requires some explanation.

For a range of reasons, Perry’s publisher wasn’t able to get me a printed copy until today. To make up for this, they sent me an ebook version in advance of the print edition. Which I read with great interest (even making copious electronic marginal notes).

And then promptly forgot.

I don’t mean “forget” in the sense that I’m unable to tell you what the book’s about. I can spit out something approximate to what’s on the flyleaf easily enough. What I can’t do is write the sort of detailed, analytical book review for which I have become modestly famous.

This has happened once before with an ebook. I accepted an electronic review copy of one of Douglas Murray’s books a couple of years ago, just to see what it was like. I then had to read the e-book version twice — pausing to take handwritten and not electronic notes — before I could even begin to put something down on paper. The whole process took about three times as long to produce finished copy than it otherwise would.

I did not, for this reason, make the same mistake again.

At the time, I was tempted to blame this on my mild dyslexia (something I’ve written about elsewhere). Other people, surely, didn’t forget ebooks as soon as they’ve read them. I felt vaguely guilty at demanding hard copies for review purposes. They clutter up my house and and can’t be donated to charity or given as presents because they’re full of my marginal scribbles.

However, because reviewers, writers, lawyers, and academics often talk to each other, I discovered I was far from alone, and that there were related disabilities. I met the managing partner who had to print out commercial contracts and go through them line-by-line with a blue pencil. Then an academic who printed out a set of proofs for her forthcoming book on contractual remedies only to discover a large number of egregious errors she’d missed while reviewing it online. Finally, a physics teacher who banned laptops in his classes because students didn’t seem to remember notes they took electronically.

Newton’s laws of motion, it seems, need to be written out by hand in order for them to stick.

The print edition of Perry, which arrived today.

Now Perry’s book has arrived in print form, I’m going to read it again. I suspect this means the review will run next week. Apologies for the delay, Louise.

More widely, I would like experiences like mine to be researched by the sort of people who take an interest in how humans process information, because I suspect the puzzle as to why the arrival of computer technology in many workplaces over the last 30-40 years (from PCs in offices through to computerised production lines) has not had a measurable impact on the rate of economic growth may thereby be solved.

There is an ongoing debate among economic historians over why productivity of this type has stalled. One useful comparison made by an economist of my acquaintance is to the introduction of electricity last century. Electrification started in the 1880s and was more or less complete in Britain and the United States by the late 1940s.

It did not have a measurable impact on economic growth for most of this period, however. My friend thinks this was because it took a long time for electricity to prompt wider changes in industrial organisation.

Pre-electric factories had a single steam-powered engine which powered a series of belts running through the length of the building. These belts were connected to large machines on each floor, tended by workers on a rota.

At the beginning of electrification, many owners just took out their steam powered engine and replaced it with an electric one. The costs and benefits of doing this mostly balanced out (they ran more cheaply but broke down more often) so there was little impact on growth.

It wasn’t until much later that people realised each machine could have its own power source, doing away with the belts altogether. But then the whole factory floor had to be redesigned, and workers needed special training for each machine — and on and on it went.

It took such a long time for the full implications of electrification to be realised that the dividends it paid in economic growth only started to appear after WWII, by which point it was allegedly already “finished”.

So it is quite possible that, far from computerisation being more or less a finished process, we are actually in the fairly early stages of realising its full implications for many kinds of workplaces. A simple example of this is that PCs in offices don’t clearly improve worker productivity compared to earlier tools like typewriters.

Basic computerisation is a process that started in (arguably) the early 1970s, nearly 50 years ago, mostly in universities and large companies. The arrival of AI, peer-to-peer networks and similar innovations — together rather grandly called the “fourth industrial revolution” by some commentators — is comparatively very recent.

It’s not a revolution at all. It’s just the latest attempt to reformat our workplaces in order to realise the full growth potential of computers. And it’s entirely possible we’re going about it all wrong. Maybe there is something different about the way human minds interact with digital text as opposed to printed pages or handwritten notes.

And the sobering fact is that if any such digital productivity revolution succeeds, it will be the first success.