The Life of Words

Two subprojects concerning OED quotation metadata are now near enough to complete to present some preliminary results. They concern the sex of the authors quoted in the OED, in both the first edition (1928) and the later Supplements (1933, 1972-86).

The most focused work on this question so far has been Baigent, Brewer, and Larminie, “Gender in the archive: women in the Oxford Dictionary of National Biography and the Oxford English Dictionary” in Journal of the British Records Association (2005), which did a sample survey of 300 quotations in Letter A, and found 14 (4.7%) to have been written by women.

I’ve now marked about 70,000 female-authored quotations in OED2, which I believe is pretty near to a complete count.

[Edit 8/24/2020 – turns out I was overoptimistic about this. Having spent this summer consolidating various updates to my OED metadata, I now count over 82,000 female-authored quotations in OED2. That extra 10,000 can make quite a difference to some of the percentage counts, given some of the relatively low totals. The data presented here can therefore no be considered longer reliable (especially where the Supplements are concerned). Stay tuned.]

This implies that just less than 3% of all OED2 quotes are by women. But of course this is still a crude number, and not of much use. At least two additional factors need to be considered.

The first of these is the date of the quotation. We would naturally assume for instance (and correctly so), that quotations drawn from the early modern period will contain a lower proportion of women authors than those from the nineteenth century, a time when women wrote more, and women’s writing enjoyed broader publication, larger audiences, and higher social and cultural prestige.

Next, we might naturally assume that the edition would make a difference: not only did the Supplements (the “2” part of OED2) have several decades more of modern (and increasingly equitable, perhaps) publication history to draw from, but we might also expect that those editors were more receptive to those female authors’ works that might have been ignored the first time around.

My initial analyses bear out the first but not the second assumption. Below are two graphs showing the number of male- and female-authored quotations by date (trailing 25-year count), with the female percentage of the total running to the right-hand axis.

The first is for the 1928 OED1 [click it to big it]:

This shows just what one would expect for OED quotations drawn from before the 18th Century: nearly no representation of women authors.

What’s happening in the spike up to 1800, you ask? Not much, it turns out–the period (roughly 1760 to 1800) does include (parts of) some of the OED’s most quoted women: Frances Burney, Ann Radcliffe, and Jane Austen among them (see fuller lists at bottom of post). But really the female proportion is helped by the relatively slow growth of male-authored quotations during this period.

What is more revealing, I think, is that later, as the total number of OED1 quotations rises sharply, the percentage of female quotations rises as well, up to around 6% at the peak period of OED1 quotation collection, from about 1850 to about 1875. It then goes down to around 4% for the period leading up to the year I’ve cut this chart off, 1928, the year OED1 was completed.

4%, it turns out, is exactly where the Supplements end up too, after some topsy-turvy ups and downs. I thought it would be higher. The peaks before this are based on so few quotations as to make them statistically suspect–we can see that in the bulk of the quotation-collection period the proportion goes between 3% and 5%.

Given that they used a very small sample, and didn’t control for date, I’d say that Baigent et al. were fairly lucky to have hit on 4.6% as a female percentage, which seems about right for the bulk of the OED’s modern quotations, where the question of female representation is most acute.

However, these numbers do not bear out Baigent et al.’s tentative suggestion that the Second Supplement may have increased the representation of female authors.

Top 10 Female Authors in OED1

Top 10 Female Authors in Supplements

*one final note – as is evident from these lists, OED1 tended to employ Miss, Mrs, etc. for female authors, whereas the Supplement eschewed this. Except, it seems, in the case of Elizabeth Cleghorn Gaskell. OED2 increased the number of quotations by Gaskell by about 60%, but retained the old title.