Behind the Dictionary
Lexicographers Talk About Language
Inside the OED, Part 3: Across the Alphabet
In parts one and two of our interview with Oxford English Dictionary editor at large Jesse Sheidlower, we talked about how the the OED is being transformed by new electronic research methods and the creation of a continually updated online edition. In our final installment, Jesse explains how OED editors are taking a fresh approach to revisions for the dictionary's Third Edition, focusing on particularly interesting entries from across the alphabet.
VT: With the revisions that are being made for the Third Edition of the OED, there have been some major changes in the way things work. Could you first describe the way things had been done previously?
JS: When we first started, we were just revising the dictionary alphabetically. We started editing at the beginning of the letter "M." When people ask why "M," we say, "Well, that way we get to 'Z' faster!" But the real answer is that you typically begin a dictionary with the letter "A," and that was the case when the OED first started, and when the Supplements were published in the 1970s and '80s. The result is that the earlier parts of the alphabet have somewhat different editorial practices than the later ones. It took a while to figure out how things were going to work. By the time the original editors got to "M" or so, the dictionary was felt to have reached a level of editorial stability that we could comfortably work in. So we started our revision at "M." And when we initially started publishing entries for the Third Edition in 2000, we were working just on those entries — editing every single entry in the dictionary in alphabetical sequence, completely rewriting every part of every entry.
Shortly thereafter, we started publishing new words out of the alphabetical sequence, so that if there were words like Internet or bling, let's say, that we thought were sufficiently important that they should appear now rather than having to wait 20 years for us to get around to that portion of the alphabet, they could be published then as well. For a number of years we did those two things together, so every quarter there would be about 2,000 revised entries published and about 250 new words, from anywhere in the alphabetical sequence that they appeared.
In March, we started with a third method of publishing: revising words out of sequence, so that words that were already in the dictionary could be fully revised right away. This is necessary because most of the established entries in the dictionary have been fundamentally untouched for 100 years or so. There might be new senses added to the existing entry, but if an entry was originally written in the 1890s or 1910s, the odds are very good that the core of it has been untouched since that time.
So what we started doing in March is publishing important entries from anywhere in the alphabet, so the things that have a great deal of intrinsic interest or have changed the most, or that people are interested in looking up can be fully revised right now, without having to wait many years before we get there.
VT: How did you decide what was important or intrinsically interesting?
JS: There were a number of different things we looked at, including how important we felt the word is as part of the language, whether it's very common, has a lot of senses, and whether there have been important changes in how the word is used over time. And by important, I mean that anyone looking at the entry would think that this is a word of great interest. One of the examples is the word cancer, where the entry was originally written in the 1880s and has been largely untouched since. And clearly this word has changed a great deal since then and needed a significant amount of revision. This is true of many other technical terms as well.
We've also looked at the sort of words that people tend to look up. It's not a popularity contest, but if there were words that people tend to look up frequently, we'd consider revising them out of sequence because we're trying to help readers get the answers they need. For example, affect and effect were revised because there is a usage issue, and the entries themselves are rather complicated. The usage history of this was not explained well, and we had those to try to clarify that as much as possible because it's something people always look in the dictionary for.
VT: Affect and effect are also two of the most looked up terms on Merriam-Webster Online. Why do you think people are focusing on affect and effect so much?
JS: There are number of usage issues like that, but this is one where a lot of people know that there is a difference, but they don't know what the difference is. And I think that contrasts with words where people don't know that there is a difference, such as insure and ensure, or alright and all right. People are just unaware that there's any difference at all. So if people think there's some kind of problem, they might know to look that up, but in other cases they might not know to look it up.
VT: What are some other words that have been recently revised?
JS: The entry for gay had to be revised, including the senses relating to homosexuality, where there's a lot of research that's been done on the early history of the word. And it was necessary to clarify the history of the senses in detail.
First of all, there was an example that was given as the earliest attestation in the Second Edition, which is now thought not to exemplify the word. A book on underworld slang from 1935 defines geycat as "a homosexual boy." But the belief now is that this is basically a coincidence and that the gay element in that word did not in fact refer to homosexuality.
There are more examples from the 1930s, which are ambiguous, and there has been a large amount of scholarly discussion about how to interpret these examples. And there are still earlier examples that some people have interpreted as referring to homosexuality because of associations with the authors' sexuality or with the context of where it's being used. But again, it's more likely that these are later projections onto these examples that didn't originally have those associations.
VT: So there's a lot of gray area?
JS: Yes, and also significant antedating to some of the senses. For instance, gay as a noun was brought back from the early 1970s to the early 1950s.
Another newly revised word where it's difficult to draw a dividing line between an old sense and a new sense is computer. The nature of what a computer is in the sense of an electronic device has changed so much even in the 30 years since the sense was first added to the Supplement that it was necessary to discuss that in great detail. A great deal of research was done. There were earlier senses referring to machines that perform or facilitate calculation, which had to be teased out from the modern sense. Because the ways that computers are used now are so different from how they used to be even 30 years ago, what was there was not very accurate, in terms of what people now understand a computer to be. There's a long note describing how computers have developed and a bit about the history of it with many different links to different types of computers.
VT: So what's the dividing line? When did the new sense officially come on the scene as far as we know?
JS: First of all, we never use the word "official" for anything. The earliest example we have of the modern sense is from 1946, although we do have a quotation from 1945 in a paper by John Von Neumann, talking about a device used for calculations. It's sort of ambiguous, but we put it in a note there because he is referring to an automatic calculating machine. That was referred to as a computer, but they were not really regarded as precursors of the computers in the modern sense. They're not versatile or programmable in the way that we regard computers as being now. So there are examples of things like these in the 1940s, but they were different. [Compare the revised entry and the unrevised entry for computer.]
VT: Any other words that were particularly difficult to revise?
JS: Another one is love. People often note that love is one of the words that is very difficult to write a definition for. There are different senses and different nuances that are hard to cover, encompassing what people talk about when they mean love. One part of the entry that I worked on extensively is the phrase make love. In the early sense, make love meant "to court" or "to woo." But there isn't a clear dividing line between that and the newer sexual sense. Once again there's a lot of gray area.
[Read more about the changes to OED's revision plan in this update from the dictionary's chief editor, John Simpson.]