Epistemic Status: Personal anecdote with reasonable sounding theory.

Background

I love to read. There is no activity I like more than it, or one I want to do as often. I’ve been blessed with an intensely vivid imagination, and twenty minutes spent reading a well-written short story can elevate my mood through the whole rest of a day. For that reason, you would think I would try to make sure my reading was as high-quality as possible.

At the beginning of 2020 I set out to find out just how much of my reading was high-quality. I started a spreadsheet to make a record of everything I read during the year. Excluding only news articles (which I began the year with a plan to avoid as much as possible – HA!), I recorded every fiction and non-fiction book, graphic novel, pen and paper RPG rulebook/supplement, and essay I finished, noting the author, title, and date of completion. From the data I hoped to be able to determine roughly much time I was spending reading worthwhile things, and conversely just roughly how much time I was wasting on things that were not worthwhile – on the fair assumption that if I was reading X at time T, I could have substituted reading Y at time T.

I should say a little about what I mean by ‘worthwhile reading’. When I say a piece of writing is worthwhile, I mean that it possesses at least one of the following traits:

(a) It is intrinsically enjoyable to read (every short story, full-length book whether fiction or non-fiction, and graphic novel met this criterion, as I would not have finished it otherwise), or

(b) I learned a new concept or mental model from reading it (for example, the fourth essay I read about an emergency room struggling through the spring surge of Covid-19 cases was not worthwhile on this metric, nor the fifth article about how the U.S. government’s pandemic response was dysfunctional, but Tomas Pueyo’s “Coronavirus, The Hammer and the Dance” or The Zvi’s “Immoral Mazes” sequence was)

Trait (b) is a bit vague, but like (a) I think it is one of those “you know it when you feel it” things. Anyway, my goal was to ensure as much of my reading during the year matched at least one of the two traits.

I collected some preliminary results on November 1st and was a disappointed. During that span I read 15 non-fiction books, 15 novels, 33 short stories, 42 graphic novels, 1 gamebook, and 617 essay and articles. I wasn’t so much disappointed with the makeup of my reading as I was with its quality (1). Reading over the list, I made of note of which articles I read met the worthwhileness criteria, and only 33.9% of them did (240 worthwhile items / 708 total). When I consider how many news articles I read – almost none of which was worthwhile in the slightest, even in the sense of being usefully informative and action informing – on top of the articles and essays, I wasted a huge volume of my year’s reading time.

Analysis

With the data, I was able to look at what I wasted my time on, and when. Mostly they were essays about contemporary politics or the pandemic, or reviews of B-movies and 8-bit video games, and very few of them were worthwhile by any metric. I also read increasingly more of them as the pandemic wore on. Why then did I spend so much time reading that?

Practicing mindfulness during the subsequent days revealed what the main driver of the poor reading selections were. I reached for inferior reading choices – articles in monthly magazines, inconsequential reviews of 8-bit video games – when I was frustrated, bored, or fatigued. And I was those three thing, either individually or all at once – many times through the past ten months, as I am sure many of you were as well. The ironic thing was is that I was prepared for these lulls – they had happened before in previous jobs and I had large lists of short stories, books, and essays I had prejudged would be worth reading during those down times. Cross checking those lists against what I did read, I can see that I read almost none of them. What made the difference?

My theory is that I was a victim of the Paradox of Choice. Even with the prejudgment that a story or essay on my list would be worth reading (H.P. Lovecraft’s “The Whisperer in the Darkness”, or example) it was one among dozens of potential things to read, not to mention the functionally infinite amount of stories I could read easily at the click of a mouse through the Internet. Faced with too many options, the attractiveness of any single option was so flattened as to be indistinguishable from any other.

The primary factor that drove my reading choices was novelty. Logging onto a monthly magazine’s website would introduce me to things I had not previously considered reading, and the mere fact of novelty was enough to make it attractive enough to go through with reading. Needless to say Sturgeon’s Law held: “ninety percent of everything is garbage”, and when you let novelty be your primary guide, you will read a lot of garbage.

Random Selection as a Means of Overcoming the Paradox of Choice Working from the observation that (1) I was facing a Paradox of Choice and (2) my reading was being guided by mere novelty, I decided to see whether I could simulate novelty. I created a spreadsheet with my lists of stories and essays, divided up into sub lists, then wrote two linked random functions. The first random selection picks the sub list, and the second selection picks the work within the sub list.

The reason for a two-stage selection are to further enhance novelty. One of my sub lists is a list of over a hundred classic science fiction stories, from the 1900s to the 2010s. Another is a list of Chekhov’s complete short stories. To include the two lists together with the other volumes would ensure that I read a lot of classic SF and Chekhov – not a terrible outcome, but not preserving enough of the novelty I want to use to drive my selection.

Results

At the end of a month using my new system, I’m very happy with the change. Personally, the commitment of will to open the spreadsheet and read the selected item is smaller than it is to pick a story deliberately from a list and follow through on reading it.

It also turns out that my instincts about what would ultimately be worthwhile to read are very often accurate, and of the 80 items I have read since November 1st, 62 of them have met at least one criteria of worthwhileness (77.5% success rate). The few items I judged not to be worthwhile during this time were primarily pieces I was sent to read by others through e-mail or texts, and read to be able to converse with them. I am going to continue with this system through the New Year, and see whether it works for full-length books as well as essays and short stories. I’ll also be expanding my tracking to include originating links (for online works) and word counts (to be better able to estimate reading time), as well as making sure to include news stories.

Concluding Thoughts / Possible Applications

• This technique may be useful for studying, particularly for long lists of material that do not have to be read in a particular sequence (for example, survey courses where a syllabus lists a large number of works to be read before a seminar).

• It may be useful for getting through a full inbox of mostly unrelated emails, rather than addressing them sequentially.

• I am thinking about turning this into a simple mobile or browser app for personal use, complete with links and an in-app browser to rapidly access the randomly selected material.