An earlier post discussed Malcolm Parkes’s idea of sense transfer units, that is the amount of text a scribe retained in memory when transferring his/her attention from an exemplar (the text to be copied) to the scribe’s own copy (the text the scribe is writing).
During a recent meeting, I found myself copying sentences from the handout to my notebook and in noticing some patterns began to imagine a sort of trial to see how Parkes’s principle might work in practice.
Using a Livescribe Pulse pen, which is essentially a digital pen that records, I sat down and copied a few lines (as naturally as possible) from a nearby book. Because it records, the Livescribe pen allows you to playback your writing in individual sessions. In everyday life, this feature is meant to be used by students who, using the pen to record lectures while taking notes, can then tap/click a certain part of their notes (either on the special paper or on a monitor after they’ve been uploaded to a computer) and then hear what the lecturer was saying at the time the student was writing.
For the purpose of this exercise, the recording pen allows us to see my copying in ‘real time’ (more or less, I don’t know enough information on the recording speed and processing time to talk about the pen’s reading/recording precision and accuracy). You can see the pauses in my copying and their duration.
The exemplar
This session lasts about four and a half minutes (to copy about 8.5 lines from a printed book) and I must admit that it was begun with a more attentive mind than it was completed. After a few lines, I was copying while my mind was wandering to other preoccupations. Under these conditions, I did pause at the end of words (at the end of a transfer unit), often for an extended period of time as I had to go back and re-read to find my place in the exemplar again. Especially pronounced after “Jausse”.
However, when I was copying with full(er) attention and was working more quickly and regularly (arguably conditions that lead to more accurate copying as well), then I would pause a few letters before the end of the last word of a phrase or string of words. At that point I would look quickly at the exemplar and with relative fluency return to copying. In this case, the transfer unit as written down, that is the words and partial words set down on the page of the copy, does not match the conceptual unit, that is the phrase or string of words that are actively in the forefront of my mind while copying.
For example, I pause near the end of “Adorno” (bewteen “n” and the final “o”) in the string “like Adorno” because my hand and mind know how to complete this word in order to minimize hesitency I look back at the exemplar before “Adorno” is finished. Whether or not these pauses lead to anomalies in alignment within a word is not clear to me. Indeed, even in more extended pauses at the ends of words later in the passage I don’t see misalignment, but then I am working on pretty clearly ruled notebook paper (the ruling does not show up on the recording but with the exception of very small and light dots (used to orient the recording pen to the paper) the Livescribe notebook looks and feels like average college-ruled notebooks); for the medieval scribe ruling may well have been less prominent on the page. Perhaps, more precise measurements are needed for this. Also, more advanced study might use eye-tracking technology to see how the eyes move in relation to the copyist’s writing activity.
For now, it seems that conceptual and transfer units are useful principles as long as we recognize that in practice the way these two relate to one another and to the process of copying are variable, both from scribe to scribe, but also possibly within one scribe’s own work.