In my quest to be a better user of Classics Twitter, today I responded to a poll on whether faculty should encourage or discourage their students to use electronic dictionaries. I immediately tried and failed (as it turns out if you reply to a text notification it only shows up on Twitter as a reply to yourself! live and learn) to create a thread on this topic, as I have given it quite a lot of thought.
Since this is an area that I can’t do in 140 characters, even in a string of them, I figured I’d come back to the blog (which has been on hiatus this year while I’m on sabbatical and working on other projects).
First: I would say that in my general embrace of all things metacognitive, I have for some years now asked (upper-level) students to do a little of their own research: a week of Perseus use and a week of physical lexicon use, with some checking to see how well they retain vocabulary and how quickly they read in each case. What has become clear, as you could probably have predicted, is that a) they read more slowly with a real (what’s the term for this? physical? IRL? analog?) dictionary, but b) they retain more successfully the vocabulary they look up the hard way. (This was confirmed for me recently in a really interesting TCL piece language learning and memory which you should read if you haven’t; check out p. 118) So I have come down to just having a meta-discussion with my students about this: sometimes it is beneficial to go more slowly; sometimes you want to read quickly. You need to choose the appropriate mode for your goals.
However: I have a dream. What if there were some kind of browser gateway though which you entered Perseus, which would track all the words you clicked on in a given session… then, at the end of that session, it would generate a little quiz for you. Maybe it would choose, say, five words at random from those you’d looked up. Maybe, if it were a little more sophisticated, it would choose the words most likely to recur in the text you’re reading, and quiz you on those. Or maybe it would use high-frequency lists like the Dickinson College Commentaries core vocabulary lists. Even better: the next day, when you started in to read some more of that text, it would start by giving you a little quiz on what you had looked up the last time. It’s well-researched that testing is a good way to consolidate knowledge: why can’t we leverage that, along with the data that must be recoverable on what words students click on, to make Perseus into the learning machine it could be?
I have spoken to some computer scientists at my institution about building such a thing: they all think it would be perfectly doable. None of them has yet had time to do it, though. If I succeed in goading them into it, I will report back here. And if any of you have the skills or connections to make my dream a reality — please let me know!
Update: There’s now a great blog post from Hannah Čulík-Baird (@opietasanimi) who first posted the Twitter poll on dictionaries — lots more interesting responses to the question!