Tag: Transcription

Finished, but only just beginning

Yesterday I completed transcribing the burial records for Green-Wood Cemetery. By the end I’d managed to improve my time from four hours per page to two. Interesting entries included two children killed by gunshot wounds, one in Brooklyn and one in Manhattan. I’m curious as to whether the child in the city—a five year-old girl—was caught in crossfire during the Draft Riot, which was taking place when she was killed. Unless I can find an notice in the newspaper about it, it’s unlikely I’ll ever find out.

There were also two soldiers killed at Gettysburg. But it seems that those casualties were either buried there or were shipped up later in July if not sometime after, beyond the range of the records I have.

I have flagged names and other details that were hard to read for further investigation and will use other sources like directories, municipal death records and death notices to make corrections. I’ve also discovered that the cemetery’s burial records do include some of those in the registers, despite the early date. It’s very inconsistent. Perhaps it was a project someone began and then abandoned. I’ve also found that commercial sites have been helpful for some of these details, as well as locating the lots in the cemetery sections. Of course, I have the information but the information about where the lots are within the sections are another story, and the cemetery’s signage is very inconsistent as well. I also need to confirm how many of the burials with lot but not grave numbers, which are numerous, are in private/family plots. I’ve also flagged records that list an unusual cause of death, perhaps from an accident. On one of the previous sheets I transcribed, there were several listed as “casualty,” including a child. I have not been able to find news of any accident in the city that might have caused those deaths.

A few interesting things I’ve noticed:

  • Many of these burials are in the same lots, which were probably public lots.
  • A sizable number of people buried in these lots were later moved to what must have been private lots or family plots.
  • One lot—448—seems to have been a designated temporary lot. In every case the number was written in ink next to the deceased’s name in the ledger and the location data left blank until a later date, at which time it was pencilled in, along with the date of reburial.
  • Without an accurate count yet, I think it’s safe to say that about 75% of the burials at this time were children under 10. Most died from childhood diseases and other conditions that today could be treated. These include marasmus, which would now be termed “failure to thrive,” and was most likely a problem metabolizing nutrients; cholera infantum (which, unlike cholera, is not contagious) and similar enteric diseases that were perhaps all the same disease but diagnosed differently; hydrocephalus; and typhus. Given the sample of 347 deaths, the number of stillbirths and premature births was quite surprising. I’d like to know if the enteric diseases were more common in the summer than other months, but unfortunately the data set I was able to get can’t tell me that.

With this first step complete, the bulk of the project is just beginning. I have learned a lot already, and most importantly I’ve discovered that there are some formidable obstacles in my way, most caused by the pandemic and my inability to access the records at the cemetery. The archivist has not responded to any of my emails and I’m reluctant to call him as a result.

I know that Green-Wood just got a major grant to digitize all of these records and make them searchable. That is a huge undertaking, and I’m sure will take years. I believe this is what he’s been working on since last summer.

NYCDH Week just concluded and I took a number of workshops that were very useful, including one on a platform called Fair Copy. It is, essentially, a TEI word processor. A TEI markup of ledger or portion of a ledger is something I’d wanted to include in my project as part of the proof-of-concept. Now that I have a Fair Copy account I may be able to do that, although it would require formatting one of the image files to comply with what I learned in another workshop, the International Image Interoperability Framework (IIIF). I don’t know if this is something I can do on my own. (Fortunately, I’ve just discovered a workshop covering this very thing on GitHub!) Otherwise, if I include this sample treatment in my project, I’ll need to do the markup “by hand.”

Immediate next steps (not necessarily in this order):

  • Read the cemetery’s regulation and policy books from the period. These provide information such as costs, procedures, etc. There is also a history of the cemetery produced mainly for tourists that I’ll read. Numerous copies are digitized and available on archive.org and other library sites.
  • Standardize and correct any questionable data in the spread sheet
  • Locate section numbers for as many of the lot numbers as possible
  • Play with it to see what I can see
  • Choose a portion of the records to use for biographical research (home addresses, occupations, cause of death, etc.).
  • Choose another portion for site surveys. Some of these records may overlap with those for biographical research, such as people buried in family lots or mausoleums, and others with an unusual cause of death, such as those killed in the riots.

Not-so-immediate next steps:

  • Conduct site surveys. Hope against hope that by then I will be able to access the cemetery’s more detailed burial maps. I will use the DEBS standard to record data and rent a GPS receiver for the surveys.
  • Determine which platform(s) and methods I will be using to present the project.

Even further next steps will get into the technical work, some of which I can do easily and some of which will pose a challenge:

  • Locating addresses on georeferenced period maps of Manhattan and Brooklyn
  • Locating burials on a georeferenced map of the cemetery
  • Creating map layers for both to enable sorting to visualize data like age and cause of death
  • Network analysis. Given that the time span for these records is only three weeks, I don’t think extensive network analysis for something like extended family relationships is relevant.  Married women, for instance, were often buried with their husbands’ families, not in their own family plots, so those relationships would be traceable across a cemetery. However, there are so few adult married women in the sample that it wouldn’t be a useful visualization. Nevertheless, I think the project should include an example, since it intends to show what can be learned from standardized cemetery data analyzed with open-source software. An analysis might work for people buried in private lots and mausoleums who were from families that had some degree of wealth. It would be interesting to see who, among these people, had mutual connections to particular church congregations or other social groups. This would require more extensive biographical research.

Three down

Engraving on a hotel in a landscape
The Relay House near Ellicott Mills, MD.
Engraving by artist T. Moran and engraver J. Karst from Picturesque B&O, 1882 by J. G. Pangborn. This Hotel/Station, built in 1830, was located north of the present hotel building near the Viaduct Monument, visible on the left in this engraving. Image: Internet Archive

My facility with burial ledger transcription is improving. Today I did an entire page—57 entries. It doesn’t sound like much but a lot of time is eaten up with deciphering handwriting, and manipulating an oversized image on my desktop while navigating a very wide spreadsheet on my laptop. Plus there is that nagging case of desk elbow. So one afternoon with a 20 minute break is a new land speed record, and I didn’t even go down any rabbit holes today, like city directories and death notices to verify name spellings.

Page three listed the first Gettysburg casualty, and another soldier who died of his wound in New York. As usual, many babies and small children succumbing to malnutrition, cholera, and childhood diseases. Two women committed suicide. Two German men in their 50s died of delirium tremens, one at the Relay House in Maryland.

Last weekend, in anticipation of a much-needed long walk, which we took at the cemetery, I sorted some of the data into burials grouped by lot number. I realized I could use Green-Wood’s Civil War and WWI veterans bios pages to search for lot numbers; any matching numbers in the listings there would also include section names, so I could at least correlate some of the lot numbers to sections. I had even better luck by searching death dates and names at Billiongraves dot com, which I have otherwise tried to avoid. But those workarounds indeed worked, at least in theory.

On the ground, the lot number signage is so minimal within sections that I gave up after trying to find two with no luck. Lots can be large and what I really need is a more detailed burial map, which can only be accessed in the [still closed to visitors] cemetery office. Moreover, I get the impression that the numbering system may have been revised at some point. But the grounds people have to know where these people are, so there must be a way.

What I’m noticing, however, is that many people in the records, so far, were interred in the same lots, especially 4196. And of those, a number were removed and reinterred elsewhere in the cemetery. I would like to learn more about the history of the cemetery, and if this was a public lot. Most of the burials took place within a day or two of death, and this was high summer, which leads me to wonder if the removed burials were first put here until family or church-owned plots were could be acquired. Not only that, but the cemetery was very busy, with over 20 burials on July 1 and over the next week and a half, between 10-15 burials a day.

I need to get back in touch with Jeff Richman, the historian, with the hope he can answer these questions about usage. Meanwhile I still have had no response from the archivist about the confusing ditto marks in the ledger. Perhaps it’s time to crowdsource help on Archivist Twitter.

 

 

19th century advertisement for a new kind of egg beater

Transcribing…

I spent about 4 hours transcribing burial records today, completing the first of six pages in the process. 40 new records are in the spread sheet now. Some names and addresses required cross-checking with Brooklyn and New York City directories, which took up time, and which was not always successful.

In the process I discovered the names of some long-lost streets, such
as Mansfield Place (perhaps a mews or private courtyard?) on W 51st St. between Fifth and Sixth Avenues.

The sponsored notices in the back of these directories run from Yellow-pages style business listings to product advertisements.

19th century ad for prosthetic limbs

The Civil War was sadly creating a growing market for Palmer’s prosthetic limbs. From Trow’s New York City Directory.

 

Transcription, marasmus, and a hydra.

I’ve been slowly transcribing the 347 records from Green-Wood’s 1863 burial ledger. So far, a 20-record session takes approximately 90 minutes—if I don’t get sucked down an archival rabbit hole. Common rabbit holes are the Manhattan and Brooklyn city directories to verify names, and other online sources that list antiquated terms for diseases that either are known by their modern names, or are seldom seen anymore. Marasmus, for example. I have transcribed about 40 records so far. There is a lot of work ahead.

I never thought this process would be easy, but I should have assumed it would be more precarious. Beyond the limited data set I am working with—as a result of the cemetery shutting down non-essential operations like archival research—a hydra of other data issues has emerged. It’s not just a matter of accessing information, it’s a matter of trusting it.

There seem to be an alarming number of inconsistencies in how the burials were recorded. Ditto marks ( ” “) appear where a blank or “~” would better indicate “no information available”— mostly in age months and days, marital status, and lot number columns. Blank spaces and “~” also appear in these columns, which do seem to indicate where no information was available or applicable (the record of a stillbirth was one instance where age information was sadly unnecessary). Ditto marks also appear where they make sense—location of birth or death, for example (mostly New York). In addition, many plot numbers are missing. In my transcription spreadsheet, all such information is currently recorded as Null. However, this may change. Is the meaning of the ditto marks contingent on the data column? Do I just need to make my best guess? Or is it all entered correctly? I am waiting for Tony Cucchiara, the cemetery’s archivist, to respond to my questions about this.

If there’s no way to clearly resolve the inconsistencies and verify the information, then this will winnow out a lot of my burial data. But how much would remain? Probably less than a third of the burial records.

Perhaps the most problematic of my discoveries is that, while lot and plot numbers are (theoretically) recorded, cemetery section numbers are not. Yet, no cemetery maps, including those dating back to the 1860s or prior, include lot numbers. There are 190 sections of the cemetery. There are many thousands of lot numbers. They are not all contiguous. This could pose a major obstacle, because if I can’t locate the graves I can’t produce site surveys.

On Monday I got in touch with Jeff Richman, Green-Wood’s historian, to ask about maps with lot numbers. He told me these are available (in very large format hard copy) in the office, and also in digital format on the office’s computer system. Unless the cemetery administration will grant me access, though, I’m not sure how I can create a single database with both site survey records and the burial ledger records.

An alternative is to make use of the Civil War Veterans research that Greenwood has put online, which includes lot numbers. In this case, I would be using the cemetery’s biographical data and completing the site surveys on my own. But except for a couple of Civil War veterans in my data set, none of these records will correspond to anything in the ledger.

I’m resolving myself to the likelihood of this outcome. My hope, however, is that once all or even most of the burial records are transcribed, I can sort them by various criteria, select a much smaller subset of graves than I’d planned, and ask if Jeff or another person at the cemetery could send me images of the lot maps from those locations. Or, if the transcription and additional biographical data collection (occupation, for instance) takes longer than I have planned for, perhaps by spring the COVID vaccine will make visiting the cemetery offices in person possible. That is something I can’t bank on, though, and I don’t want to drag this project out “just in case.”