Faculty Profile: Peter Grund
An Interview with Peter Grund
on the Salem witchcraft trials and Checkley, an open source handwriting description database.
Interviewed by Brendan Allen, November 2012
Could you introduce yourself, as well as your field of study?
My name is Peter Grund, and I’m in the English department. My areas of research are the history of the English language and the intersection between language and society, culture and historical setting.
Could you describe your current project?
The one we’re working on now – I’m working with two colleagues (Margo Burns and Matti Peikola) – is a large scribal database of all the recorders who took down documents during the Salem witchcraft trials. There’s about a thousand documents from the actual trials. We were part of an earlier project that put together a large edition of the Salem documents, but at first we didn’t really think a great deal about the people recording the documents – we were primarily interested in the content of the texts.
Then we started noticing that there were so many different recorders, so many people taking down the documents in various contexts. Previously, people have thought that there were just a few court clerks writing the legal documents, but that was clearly not true. The three of us did some initial charting of different scribal hands, but we couldn’t do very much within the context of our previous project. So, now we’re revisiting that idea and going back to look at who these people were and what the implications are of knowing who they were.
One of the things we’ve found is that many of them were just ordinary villagers, inhabitants of Salem and neighboring communities. Some of them were closely related to people who were accusers or the accused during the trials, so there are all kinds of things coming out that we didn’t know before.
The digital humanities aspect of the project is that it will all be available online as a searchable database, so you can search the recorders by name or by scribal features. If you’re interested in who was using a particular abbreviation, you can go in and look at who was using an ampersand, for instance, and what their particular ampersands looked like. Or you could look at where these people came from. Did they come from Salem or one of the neighboring communities? The database will be available for free once were done – but it’s obviously going to take a long time, perhaps five or six years.
You mentioned being able to search by aspects of writing. Does that mean you have a catalogue of different types of writing units, such as ampersands?
We will! What we’re doing is conducting handwriting analysis. So, we’re going in and looking at these documents and analyzing letterforms, abbreviations, punctuations and spelling, and we’re trying to separate the different handwritings on the basis of those features. Once we’re done, there will be a full database of all these different features. Say that you want to know how many recorders used the spelling “H-U-R” for the word “her,” which we now spell as “H-E-R.” You can just search for that and get all the recorders who used that particular spelling.
Part of our larger interest is to look at literacy – who could write at this particular point in time? How were different handwriting models transmitted? How were they taught? By looking at who have similar handwritings, we want to see if there are networks. Does a particular family have the same type of handwriting? Or do you see other connections between people in the same village? We don’t know exactly what will come out of this, but we will once we get all the patterns.
How does this analysis really work? Are you working with physical documents?
This is one type of document. In this one you have four different people writing in one document. You can look at different letterforms. For instance, there are two different types of ‘e’ here – this is an initial ‘e’, and this is the ‘e’ we’re used to today. And then there’s a third kind of ‘e’, which looks like it’s backwards. These different letter forms were present in the period, and by looking at who uses what, we can start separating the writers.
Of course, it’s a simplification to just look at one letter; you have to look at a number of different letters at the same time to see any patterns in the recorders. You just start by collecting different features. So this is what the ampersand looks like for this particular document. [points at document]
Oh wow, that doesn’t look like anything I’ve ever seen.
Exactly! And sometimes they look much more like the modern versions that we’re used to. But there’s a whole range of different kinds of ampersands, so they’re really a good marker that can be used to separate one recorder from another. Or you can look at if they used abbreviations – this is an abbreviation for the definite article “the.”
It looks like a Y with an E above it.
Yeah, and there are historical reasons for why it looks like that. Over here someone just writes out “the” – so there are different ways of doing it. It’s a lot of collecting data at first, and then we can see the patterns once we’ve collected it.
As you compile these samples of handwriting, has it been possible for you to actually identify the specific authors?
Yes. Some of them are easy to identify, because they sign their names now and then. One person we know is Thomas Putnam, who was the father of one of the accusers of the alleged witches. He wrote about 100 documents, he was really involved. There’s also Simon Willard, who writes part of this document. We know that because we have his signature, not here, but in other documents.
Of course, we have handwriting that we’re not exactly sure who it belongs to, because they’ve never provided a signature. What we’ve baked into our project is that we will go to some of the archives in New England and look at documents that are outside the thousand legal documents from Salem. We can then compare the handwriting of the Salem documents with contemporaneous handwriting to see if we find a particular handwriting elsewhere, and if it is signed.
We’ve actually managed to do that already – we had a seed grant from the Hall Center, and that enabled us to go to Boston. There we made several discoveries. For example, we found text in the same handwriting as the first hand in this document, and that text was signed by Thomas Flint, who later was a parish clerk in Salem. We would not have been able to find that out on the basis of the Salem material only, but we found it when we went outside and looked at other documents. No one has thought of doing that before.
These documents, of course, have been studied over and over again because there’s so much interest in the Salem trials. But this is a completely new aspect – it’s very interesting to know who these people were, because then you get a new sense of who was involved. Did they have a particular interest in the trials going a certain way? Surely some of these people did. Especially Thomas Putnam, whose daughter was highly involved – what kind of implications does that have for our understanding of the trials?
This doesn't sound like a very easy process.
No, it’s not! It will take a number of years, and you have to be very systematic. But this is one of the advantages of the digital humanities aspect – Margo Burns, who is our programming wizard, has come up with a database where we can collect all of this material – it’s a lot of data – and then sort it and store it in different ways. Without that, this project would not have been possible. We would have had to use index cards, and you can’t search index cards. An electronic tool is crucialboth for the presentation of the final results and the analysis. We couldn’t do this without it unless we wanted to spend 30, 40 years on the project.
You’ve said several times that, once this database is complete, you’d like to give open access to anyone who would like to use it. Does that mean other people working on their own projects would be able to fill this type of database with their own information?
Yes. There will be two products coming out of the project – first, the scribal database with all of the Salem material in it. But there will also be a reusable “shell” of the database, where people can basically use our methods and analysis and plug in their own material. So there will be both the content and no-content versions, and people can do whatever they want with them. They can even tweak the shell further, if they wanted to.
It will be available on GitHub, where people can download it and develop it further if they want to – or they can just take the shell as we’ve developed it and plug in their own handwriting data. It should work for pretty much any period and probably for different languages too, although different languages will have different requirements – people might have to tweak the database further. There are letterforms that won’t appear in English, but that people working with Spanish might want, for instance.
It’s important for us that the result is publicly available, and freely available. That we’ve said from the very beginning.
And why would you say that’s so important?
Because we really want it to be used. If you restrict it, if you make people purchase a license or something, it’s not going to be as widely used. It’s going to be used in libraries, in particular. They’re going to buy a subscription, whereas other people might not be able to afford to. But there are so many people interested in Salem. There are people interested in witchcraft in general, but also family history – many people trace their ancestors back to Salem. They’re family historians, and they’re not going to be able to afford this kind of product. We really want it to be freely available for everybody. That’s the main point, we feel very strongly about that.
Where do your collaborators live?
Well, I’m here in Kansas, obviously. Margo Burns is in New Hampshire – she works at St. Paul’s High School. And then Matti Peikola is in Finland, at the University of Turku. We all started collaborating on a different project and continued because we have very similar interests.
So how does this process work, when you’re analyzing these historical documents from the Salem Witch Trials, in tandem with these people from New Hampshire and Finland? That doesn’t sound like an easy process either.
No, it’s not. It’s quite a challenge. When we’re trying to set up a Skype meeting, we have to find a day that works for all of us. It’s a seven-hour time difference between here and FinlandOnce we get to the actual classification of handwriting, it doesn’t matter as much because we can work through the online database whenever we have time to do it. We will divide the manuscripts into chunks, go through and analyze, and then we’ll swap. We have a very rigorous schedule set up for how that works.
Does this database have a specific name yet?
It’s called CHECKLEY. It’s taken from the name Anthony Checkley, who was the prosecutor for most of the witch trial cases in 1692. We have his handwriting in quite a lot of our documents because he wrote indictments, in particular. It’s a fun pun, an academic joke – “checking” with CHECKLEY.
I remember hearing you mention that this project began when you heard about a certain grant that you wanted to apply for.
The Hall Center advertised a seed grant for collaborative research, and we said, “Okay, we’ll try and see what happens.” It was a way for us to think through what we were doing – and we just happened to be successful, which was obviously gratifying. That seed grant allowed us to do a lot of different things in preparation for writing an application for a larger grant. We were able to read up on some of the background literature that we hadn’t looked at before, and also go to New England to look at some of the archives.
You can imagine what would happen if we had a longer period of time in which we could look at these documents and see how many correlations and correspondences exist. Obviously the Salem scribes were writing things outside of the Salem trials as well. New England people were very litigious – they sued people left and right, so they left a lot of legal documents. Back in those days, you had to write down your own witness depositions – or you went to someone who knew how to write, and they would take it to the court authorities. So we’re confident that we’re going to make many new discoveries.
It sounds like this research wouldn’t even be possible without digital tools.
Just the publication of the data – it would be something like 10 books, and it wouldn’t be very usable, because if you wanted to find correspondences between different people, how would you go about doing that with ten volumes? You’d need one gigantic index for that, and it still wouldn’t capture the complexities. With a database, of course, you can search for multiple things at the same time and look at the combinations of handwriting features. That would just be impossible in a book format. A lot of the value will come out from the fact that you can search and apply different filters for these searches. But again, we don’t know completely what to expect, so it will be interesting to see it when it’s all done.
What kind of potential uses would these filters serve? What kind of criteria will you be able to search through this database with?
That’s something that’s up in the air. We’ll have it tested to see what different people are looking for – what would an archivist be looking for, what would a librarian be looking for, what would a Salem expert be looking for, etcetera – and try to accommodate those things. But we assume that people would want to search for names, residences, who was an accuser in a particular document, the type of document, and also any type of letterforms, abbreviations, spelling.
Spelling is of particular interest for people who do the things that I do – the history of the English language – so who was spelling a word a particular way, at this particular point? Were they all doing it? Was it part of a standard? Is there a particular social class using this particular spelling? There are a number of different correlations you can look at. Did people of higher social status use more or fewer abbreviations?
So this could potentially branch out into other fields, such as etymology and sociology?
Absolutely, that’s part of our point. It sounds like it’s only about the Salem recorders, and that we get to know the names of these people, but there’s so much more coming out of this project and there are so many more potential uses. We want to look at literacy and transmission of handwriting models, but who knows what other people would want to look at?
Is there anything else you’d like to say about your project and the realm of digital humanities?
I think that digital humanities help open up new perspectives on old issues. It’s well worth revisiting manuscripts. People often take manuscripts for granted as scholar have produced book editions of the documents for a long time – people keep using these books and forget about the original manuscripts. But the manuscripts are still there, and it’s important to go back and make them available in different ways digitally. Not only with facsimiles, but also with this kind of handwriting analysis. The manuscripts contain a great deal of information about the situation and context of writing, about the scribes who wrote the documents, and other issues that cannot be seen in editions.
It’s important to revisit the manuscripts not only for scholars, but also for students – students tend to not know about manuscripts because they use modern editions. They don’t know what the manuscripts and the visual aspects of the original materials are like. We can get blinded by the book editions. Digitial humanities approaches on the other hand provide the opportunities to make the original manuscripts more accessible and a window into the world of text making in earlier periods.