The DMLBS is mostly based on quotations taken from texts that have been printed and can be readily consulted. Some of the documents and records that we use, however, are only available in manuscript and can only be consulted in the relevant archives. Two or three times a year, therefore, we send a team of lexicographers to The National Archives in London to check that our quotations from sources held there are accurate. Here is Shelagh in action:
The task of locating and verifying quotations is not straightforward, partly because many of the words are abbreviated. Here, for instance, is one that we had down as espereveriis:
This is from an account of falconry expenses from 1286. There are two abbreviations – per is written as a p with a line cutting through the tail, and ver is written as a u followed by a backwards loop – and the vowel between the two is almost certainly an o rather than an e. This is an unusual spelling of (e)spervarius, a sparrow-hawk; interestingly, it predates the Oxford English Dictionary‘s first attested spelling of spar(r)ow with an o by 50-odd years.
Another difficulty is the diversity of handwriting. As an illustration, here is the first known appearance of tobacco in British Latin, squeaking into the DMLBS with a date of 1598:
This is written in a very different hand, and the letters a and r in particular may come as a surprise. With a bit of effort, though, we can decipher it as unum barellum de lez tobacco.