In the 1960s, the UK Public Records Office (now the National Archives) decided to sample and then destroy a large collection of shipping papers--crew agreements which were labour records documenting British imperial maritime work from 1850 to the 1970s.

Several historians at Memorial petitioned to save the collection. British archives were allowed to take what they wanted and MUN would get the rest--six shipping containers (20,937 boxes) of documents covering the rise and fall of British shipping. A goldmine of data.

Part of the preservation effort was the understanding that the crew agreements, particularly the lists of vital personal and labour data of the working men and women who went to sea, would be ideal for new computer-based research programs.

A team of researchers was gathered at MUN--archivists and historians--who combed through the papers, created important databases and began the Atlantic Canadian Shipping Project to understand the decline of shipping in Atlantic Canada.

The computer which facilitated this project occupied the entire basement of a recently build St. John's apartment tower and the information was entered using punch cards. When I worked there, we found one of the cards--it's just astonishing to think about today.

Even so, six shipping containers of papers were still too much data--the project had to narrow its focus to input a dataset that would be useful. Only ships registered from certain Atlantic ports were counted--Saint John, Windsor, Halifax, and Yarmouth.

The full promise of the project was barely realised--two of the historians died tragically young--David Alexander in 1980 and Keith Matthews in 1984. A handful of books were published before many of the scholars moved on to other projects.

But also, the project itself was mired by optimistic ideas about what a computer could provide to historical research. There was an understanding, and this persists today, that computers are somehow objective.

The last ten years of history should have devested us in this fallacy, but the message had already been delivered by the 1980s by historians who realized after two decades of working with cliometrics that the programs they built were influenced by researchers and their data.

Research questions, funding, time and technological limits, archival organization, the original record-makers, even the very presentation of the documents themselves influenced the Atlantic Canada Shipping Project and therefore the information which it produced.

My research there, on the clothing of merchant seafarers, relied heavily on a component of the crew agreements called the official log which was almost entirely ignored by the ACSP because, unlike the lists, these booklets were not written in a format that was easily computed.

But still, I bought into the idea of computing my research and I tried to build a database of sailor information and clothing to see what would happen. It was a mess.

My MA turned into a jaded rant about how the inventories of clothing were written--OS and AB clothes lists were extremely vague, so the men I wanted to know the most about had the least information.

Then all the data was skewed by one steward, immigrating (we suspected) to Australia with four pages of possessions including 150lbs of cheese and 100 paper collars. And by a dead captain whose inventory was so lovingly detailed that the writer even pointed out his dirty laundry.

Apprentices didn't receive standard wages and neither did captains. Some capacities were simply more likely to die than others, meaning that I could hardly find inventories for engineers and officers.

The wealth of information of men's clothing was there, but it had to be understood through qualitative research using diaries and other scholarly works on 19thC clothing with the quantitative data acting in a supporting role, not as the central research device.

I had to understand what was there and also what was missing. The men were buried at sea in clothes that had to be additionally accounted for. Different capacities were allotted different amounts of space. Racist ideas about Indians/Asians meant they were allotted even less.

Women's possessions were almost never inventoried because their clothes could not be resold to their male colleagues. Non-Europeans were segregated to different agreements so that the rights and protections won by British sailors did not carry over to their racialized colleagues.

I understand why quantitative research is so alluring. Wouldn't it be great if information could just be punched into a computer and it would reveal the trajectory of our lives?

But what information do you use? The data can itself be biased, political, and limited, as I learned. But also, what questions you ask also changes the project and influences the answers received.

So, in conclusion:

Check out More Than a List of Crew to find out more about the Maritime History Archive and its contents: https://www.mun.ca/mha/mlc/toolkit/archiving/

Also, if you are interested in one of the premier archives of British imperial working men’s history, check out their website: http://www.mun.ca/mha 

And finally, myself and three other historians wrote about the MHA and our research using its amazing documents in the May issue of the International Journal of Maritime History https://journals.sagepub.com/toc/ijha/31/2