“Crossed Wireless” after the Titanic’s Sinking: our Neatline exhibit.
Project Description and Methodology
We (Paul and Caitlin) chose to map conflicting wireless messages sent in the wake of the sinking of the Titanic, on the evening of April 14, 1912. Our method relied on newspaper headlines, such as this from the morning of April 15:
How did this headline happen? What were the sources of wireless and telegraphed communication that allowed this false hope. Vancouver World, April 15, 1912, from newspapers.com
We began by gathering data – creating our “text” to map. We searched “Titanic” and the phrase “lives lost,” in order to capture the differing reports of numbers of those saved and lost over the first few days after the sinking, in four historic newspaper databases (Chronicling America, Newspapers.com, GenealogyBank, Newspaper Archive). We only ended up using information from twelve newspapers from just Chronicling America and Newspapers.com – the amount of material published on the Titanic is massive, even when limited just North American news sources, published on April 15 – 16, 1912. The mini-corpus we created could definitely be added to and mapped further; what we created during the week of DHSI is just the tip of the iceberg. We created a Google spreadsheet to log the data we were extracting from these contemporary news sources.
Screenshot of our Titanic data spreadsheet, showing the different categories of information we aimed to extract from the newspaper corpus we created.
While we began conscious of conflicting accounts from one paper to another, each newspaper also printed openly conflicting accounts that contradicted the editor’s choice of headline. We thus specifically gathered unreliable information with deliberately multiple ways transforming and mapping into a narrative. For one, the narrative itself is in flux, as the information being transmitted via wireless and then published in newspapers was often conflicting, and the facts that would form the backbone of the historical narrative were yet to be firmly established (see Stephen Hines’ Titanic: One Newspaper, Seven Days, and the Truth that Shocked the World for a detailed account of how one newspaper [London’s Daily Telegraph] reported on the numerous narratives). What was the “text” we were mapping? What was possible using Neatline, and what was the most interesting story to tell?
After a lot of research and discussion, we chose to map the spatial and time relationships between the Titanic, other ships in the area (the Carpathia, the Virginian, the Parisian, the Olympic, and later the Californian), Marconi stations on shore (Cape Race, Saint John, Halifax, Sable Island, and Siasconset), and the shipping line offices (White Star Line in New York, and the Allen Line in Montreal), not showing the subsequent mass transmission to thousands of newspapers via newswire. We layered a small sample of wireless transmissions to rough locations of the ships – i.e., when the Carpathia initially confirmed that they had picked up Titanic survivors, that message was relayed from the Carpathia to the nearby Olympic, then the message was picked up by the Marconi station at Siaconset (Nantucket, Massachusetts), and then onto the White Star Line offices in lower Manhattan.
Screenshot of the “array of false hope” event on our map/timeline – we coded a series of hopeful messages that ended up being false, in an attempt to keep the map readable for users.
A key point of our Neatline mapping exercise is beginning with a collection of historical documents, each partial, ephemeral, conflicting, and unreliable, and then having to discern the “text” and the “narrative” as our own interpretation across the historical material. In a sense, the mapping exercise is a visualization of the implicit ur-text synthesizing the various partial newspaper accounts, each of them in turn listing conflicting and unreliable information. Our sense of others’ efforts to map literary texts seemed on one level to aim for the opposite process: taking a presumed reliable narrators’ account in a singular text into constituent implicitly contradictory and latently conflicting patterns. In a sense, literary uses of mapping seemed to deconstruct narration, while our historical project mapped a deceptively coherent narrative onto conflicting accounts. Our approach thus had the inadvertent effect of our map connoting the empirical facts of the historical events, rather than adequately conveying our intentions of narrating the conflicting flows of radio wireless and telegraph wire information.
Reflections: Limitations and Lessons
We successfully emphasized the simultaneous reception and interception of conflicting wireless signals and their retransmission over telegraph wires. This was our main narrative, and this was achieved relatively well.
Yet, in two other key respects, the character of our map fails to convey the Titanic rescue efforts. First, the ships were all moving in the ocean, but we depicted them as static points. Second, wireless signals transmit in an indiscriminate radius from the ships at sea, but we depicted messages as point-to-point signals. Our various efforts to display a circular radius from ships sending signals proved difficult to put into action, just as our displaying the original routes of each ship made their voyages seem more spatially bounded than a sailing route could possibly be. The waves of ocean transport and the airwaves of wireless radio signals both appear earthbound on our map.
Some of the difficulties we encountered in mapping our text corpus have to do with the course’s choice of tool, Neatline. While low-barrier tech-wise and (mostly) user-friendly, Neatline limited some of our mapping possibilities, due to either display or functionality. Any tool selected would have shaped how our map turned out, and while we weren’t able to achieve everything we wished, the data visualization of the wireless messages does shape a narrative and provide a different, more spatially-based, perspective than the written text corpus.
Text Mapping as Modelling 