Category: Digital Humanities Coursework

Zotero is probably the most exciting bit of software this program has thus far introduced me to. I began by creating a History and a Political Science collection in the library to split the works I would be collecting between my two primary fields.

Once the core categories were set up, I started populating them with literature on violence by non-state actors, my chosen subject of study in both Political Science and History. At this stage, I was still separating them out into the same two categories.

Throughout this process, I was expecting to need to intervene in the citations Zotero was providing, but I never experienced any issues. The software seems to be reading WMU’s library website perfectly and the one article I took from another cite (a dissertation from Murray State) seems to have loaded correctly as well, missing only the abstract. The one issue which appeared consistently was the capitalization in article and journal titles; notably, this was the WMU Library’s fault rather than the software’s. Helpfully, Zotero had a tool for this, allowing me to set any field to “Title Case,” automatically capitalizing items in that field.

The auto-generated tags were unfortunately not the most useful items. They were too numerous and too specific to be helpful. For example, having added four articles on the topic of Chechnya, the tag “Chechnya” brought up only two of them and the tag “Chechen” brought up only one. Still, the tag system itself seems useful once I delete the existing items and add my own.

Once my initial twenty-five items were set up, I started to look towards creating subcategories. I mainly wanted to separate out History titles dealing with Late Antiquity (my specialty) from a handful of other titles I had collected.

Once this was done, my library was effectively complete for my purposes, although I’m certain more subcategories will need to be set up as I progress through new projects. Thus far, I’m deeply impressed with this software. Zotero has delivered citations practically ready for use with every book and article which I tracked down.

Thoughts on AI Analysis

AI-driven analysis comes with all of the positives and negatives of other computer-based styles of analysis. It has the capacity to generate and sort enormous quantities of data at speeds alien to human researchers. It can also, however, incorporate errors, inconsistencies, and irrelevant information seamlessly into these datasets, forcing human researchers to check the computer’s work or even toss out the data entirely.

What appears to be a new development in AI’s case is its unique ability to source data from third-party sources. This introduces a new set of complexities; this data cannot be fact-checked by human researchers if they are unsure where the information was gathered. The larger the project, the easier it becomes for such problems to slip into the finished product. As such, I am cautious to speak positively of their research potential at this time though, as I have mentioned at previous points, the potential for an exclusively academic-run AI is enticing.

Project Proposal

Searching through the Cultural Analytics Lab’s pages, I came across an article titled “Seeing a Century Through the Lens of Sovetskoe Foto,” by Alise Tifentale of the City University of New York. This project, carried out alongside Lev Manovich and Agustín Indaco, analyzed Soviet photography based on the covers of the USSR’s state-sponsored photography magazine over the years it was published.

I suspect that a similar project could be applied (with considerable additional effort) to the material culture of far older groups. I (cautiously) suggest that a database of surviving artifacts from Antiquity could be formed in three dimensions, allowing AI to search for comparisons across cultural groups. With appropriate safeguards in place for false positives, this could become a particularly useful tool for trade-heavy periods and regions in which similar items in different cultural styles frequently crossed borders. The AI could potentially identify cultural exchange and the drift of artistic style.

The Co-Citation Network for Philosophy is a sprawling graph which demonstrates trends (rather than absolutes) in citation between publications in top journals. The image below shows the complete graph. Any point on this graph can be dragged out of the way to reveal overlapping text and possibly freeze your browser.

A user unfamiliar with the field of philosophy could reasonably expect to get some useful information out of this graph. Critically, for someone like me who has no idea who the key authors are in the field, this is a decent way to locate some of the trend-setting texts for introduction (although, in fairness, gives no suggestion on whether other, prerequisite texts might need to be read before them. For example, as the graph’s designer notes, David Lewis is frequently found at the center of citation clusters, marking him as a key figure in English-language philosophy.

On the opposite end of the spectrum, some clusters form isolated, operating as small, closed discussions (some overlap may still exist; the graph only manages relationships over a specified threshold of citations). These discussions, were I to speculate on them, likely represent authors who are part of particularly niche debates or overlap strongly with other, non-philosophy fields, thus preventing them from forming citation chains with authors in more central positions on the graph.

While useful, the graph has strong limitations. The creator notes that the project is not in a finished state and that it remains, at this time, primarily “descriptive” in nature. As such, more complex data must be sourced from elsewhere. This does not inherently render the graph useless for such efforts, though. For example, while the graph offers no data on items such as gender, race, or seniority, the most-cited authors are made very clear. Comparing them with data on these subjects gathered from other sources could create a fairly clear (if time-consuming) picture of the biases at play in English-language philosophy.

Introduction

The Stanford Spatial History Projects seem to be in a rough state without Adobe Flash. Three projects I found interesting turned out to be malfunctioning, but I eventually tracked down two which I could draw a comparison between. I landed on “Geography of Chinese Workers Building the Transcontinental Railroad,” an effort to reconstruct the areas Chinese workers were present in during construction, and “Richard Pryor’s Peoria,” a documentation project surrounding Peoria, Illinois during the time Richard Pryor lived there.

Geography of Chinse Workers Building the Transcontinental Railroad

The Railroad project is designed to allow users to follow the exact route of the Transcontinental Railroad. The webpage repeatedly provides 3D overviews of the route through Google Earth, which can in turn be accessed through the link on screen, allowing users to jump to the same view and manipulate the map to further study the modern terrain.

Interspaced between these assets are contemporary photos of the period described in the text which are as closely associated with the mapped locations as possible. These images have no interactive component but provide important context to the original terrain prior to the construction of the railroad or other infrastructure.

Throughout all of this, users are given the option to toggle a map which will track their physical progress along the railroad route as they scroll through the project page. This seems to be the most unique and interesting feature implemented on this project and is remarkably helpful once users get used to it.

Richard Pryor’s Peoria

The Peoria website is an archive of source documents stretching from 1919 to roughly 1980 (while a few documents postdate this period, they are collected in a brief closing section). From the home page, the site allows users to filter based on several overlapping criteria: People, Places, Eras, or Themes (the “More” heading, confusingly, acts as the home page).

The site is focused definitively on Pryor himself, with most of the documents provided linking to him or framed by his life. There are a few documents which have little connection to Pryor; the project seems to have been willing to include any additional information they received on the topic of Peoria even if it didn’t personally impact Pryor, but it does not appear to have such sources out. The below diagram, for example, displays that all individuals in the “People” heading are of some relationship to Pryor.

Comparison

The two websites discussed above serve vastly different purposes. The former is a mapping project overlaying historical activities onto the current physical world. The latter is an archive of sources on a particular city. Both are “spatial” history projects, attempting to make an area more legible to modern users, but they have vastly different approaches.

The user interface for the Railroad project has a slight learning curve and can appear almost broken at first; the various links and hovering mapping feature do not move the way the webpage seems it should. After a couple minutes practicing, however, the benefits of their system far outweigh the time spent to learn it, allowing users to see multiple types of information on screen at the same time.

By contrast, the Peoria website is remarkably user friendly, allowing documents to be found by effectively any criteria. The multiple approaches create redundancies, but the site’s relatively few documents prevent this from becoming a problem. While it would likely scale poorly, the current layout of the Peoria website is excellent for its purpose.

To locate files with good metadata attached to them, I thought first of visiting government websites. I visited the Irish government’s website and selected a PDF outlining part of their housing plans, specifically discussing the effective use of existing properties. The first page of this item is pictured below.

Entering the metadata was mostly straightforward for this document, although I will confess that the creator and publisher become a little blurred when dealing with a government resource. I decided to enter the listed department of the Irish Government as the creator and the Irish Government in general as publisher, which seemed most logical given the way the website was set up.

I wanted to use a different file type for my second entry, so I looked through several areas of the Irish Government’s website and eventually, by way of the Cork City Council’s site, found the website of The Butter Museum in Cork. From their gallery, I selected a jpg image of a butter wrapper, which I have included below.

This turned out to be a slightly more difficult item to include metadata on. The original image had been shared to Tumblr by the museum and included little information about its background. Even the location is not terribly clear; County Tipperary helpfully contains two towns by the name of “Ballingarry.” The southern one seems more likely, given the mention of the nearby town of Thurles, but that remains speculation.

Beyond that, I encountered few problems. Once I was finished, both items were visible in the listings on the Omeka site.

I still feel shaky on some of the metadata entries, but this project has demystified the subject a good deal. I should be able to work with it again in the future.

Exploring the Library of Congress’s categorization system has been a useful experience for me. What has and has not changed in the past decade tells a lot about the priorities of the archivists and the volume of information moving through the Library of Congress. I initially thought to look into Private Military Security Companies (PMSCs) since the practice has seen significant continuing evolution in recent years, but it turned out (to my surprise) that the Library of Congress had made no changes in the period of 2012 – 2022.

Curiously, however, I noticed that not far away, “privateering” of all things had seen revisions. Compared to the 2012 version, the 2022 entry had added a section on privately operated prisons, created a new subheading on the legal status of privateering, and altered the heading on “pirates” to instead reference “piracy.”

After this, I turned my attention towards my Late Antique work. For the most part, little had changed. Merovingian Period history is not prone to updates and had not undergone any revisions when I checked into the dynasty or its principal sources. While looking into Gregory, Bishop of Tours, however, I found that the entry for Tours itself had undergone some revision and expansion. This was not focused on the city, but rather on types of tours available for the public, with the concept of “ghost tours” being added in the 2022 version.

While my specific areas of study appear to have been little altered in the past decade, the entries around them have undergone changes. I found the change from “pirates” to “piracy” particularly insightful, indicating a shift from group focus (piracy as an occupation) to activity focus (piracy as an act).

Learning the basics of Markdown was an interesting and surprisingly quick experience. The system becomes intuitive after a relatively short time. For some of the more basic elements, I used this as an opportunity to brainstorm while working on my M.A. thesis. The first item I formed was a possible chapter/section breakdown organized under some of the list features, as seen below.

I then transplanted a paragraph from the introduction and applied bold text to all of the names in the paragraph, noting that it retained its italics from the original text. I also tracked down an indentation command so it would appear as it does in Word.

Next, I used the block quote feature to create a comparison between a Latin copy of one of my sources and an English translation of the same passage.

For the fourth and final setup, I tried out the link feature, linking a pair of websites through Markup.

Overall, I would call the system as a whole fast and remarkably user-friendly. I picked up the basics immediately despite having no real familiarity with Markup. I am not yet sure of the utility it will have in the future, but I said the same about Latin when I first learned it. I look forward to seeing where Markup will crop up as I move forward in this course.

I’m Samuel Talbott, a graduate student currently working on an M.A. in History at Western Michigan University. I completed my undergraduate degree in History and Government & Politics at Millsaps College. I have created this website as part of a digital humanities course and look forward to maintaining it as a resource in the future.

History cannot be approached as an isolated field. I apply the tools of political science to my historical studies, using these methods to improve my understanding of government and politics across the human past. This is only one set of methods, however. A comprehensive approach to history must include multiple disciplines from both the humanities and the sciences, necessitating collaborative projects between academics.

Description

This is a multi-day annotation project which displays both the use of the Hypothes.is browser extension and the value it has brought through the repeated annotation of a single PDF. Over three rounds of annotations, my thinking developed along new lines guided by the comments I had left myself earlier. Each set is shown below with screenshots for context.

Round 1

My initial commentary largely described the utility I saw in the use of Private Military Security Companies (PMSCs) and the examples of their use which seemed valuable to my research. Much of this ties into other works I have read or previous papers I have written.

Round 2

My second round of annotations expanded upon the first. I became interested in the comparison between PMSCs (primarily employed by NATO factions and Russia) and the unofficial maritime militias used by China. My original plan had been to study the PMSCs used in the context of the Russo-Ukrainian War and the legal status of foreign legion troops, but the Chinese side of the equation became much more interesting at this point and drew my attention away.

Round 3

The final annotations are focused on deepening the lines of inquiry I set up during the second round. I continued on the topic of Chinese maritime militias, wondering how far the contracts they operate under are from those given to western PMSCs. I also revise a previous comment I made about the nature of PMSCs. While these organizations are on long-term contracts to the states they are based in, this does not mean that their members are citizens of that state. In that regard, the charge of mercenarism is still very much on the table for PMSCs taken prisoner in modern conflicts.

Conclusion

Using Hypothes.is has been both new and familiar to me. In the past, I have normally taken notes in a separate document open to one side of my screen, so the act of electronic note-taken is nothing out of the ordinary for me. Annotating directly (and, more importantly, repeatedly) onto a PDF feels different in part because there is less impetus for me to keep things tidy – the program is doing that for me. At the same time, this method forces me to scroll back through the PDF to view my annotations, making accessibility more difficult in post. I can see myself using a combination of the methods in the future, perhaps alternating depending on the type and density of the text I’m working with.

Algorithms of Oppression offers insight into the ways in which efforts to optimize search engines have undermined the accurate dissemination of information to the detriment of its users. I have never found search engines particularly useful for scholarly purposes and this text goes a long way towards explaining why. These engines are not simply affected by confirmation bias, but actively (if perhaps unconsciously) implement it. Simple answers friendly to existing user beliefs are promoted; complex answers that would be valuable to scholars are pushed downward.

When considering the effect this has on my own field (I specialize in Late Antique History), the “Fall of Rome” came to mind. Few subjects in Late Antiquity elicit such consistent debate, of course, but I specifically recalled a conversation from a year or two ago with my grandmother who, attempting to learn more about what I was working on academically, had looked into the Roman Empire’s collapse on the internet. The searches she conducted provided her with a simple and painfully outdated explanation fond of the term “barbarian invasion.” With her experience in mind, I typed “Fall of Rome” into three search engines on my own. I have included screenshots of the initial search results below.

Google has tossed its “AI Overview” at the top of the page. The description of what the Empire’s collapse was is decent enough, but the “causes” section becomes concerning rapidly, listing out unchallenged a number of highly controversial reasons that remain under intense debate in the scholarly community. Each item the AI provides has a link attached to it, but a user must follow it to its source website to see the citation; none is provided written out on the search page.

Bing, to its credit, offers a centralized set of information all pointing back to the subject’s Wikipedia page. Short of diving into real scholarly works, this is probably the best set of information it could provide a user on the subject. Things fall off quickly afterwards, however. Beyond the initial Wikipedia information, the search results provide a set of sponsored websites, all offering tours of the city of Rome.

I expected to see some kind of modern political content crop up given the prevalence of the “Fall of Rome” narrative in American conservative circles, but Ecosia has proven my point better than I had anticipated. The top slot on this search was occupied by none other than Hillsdale College attempting to market a box set of DVDs on Roman history. I will let the description the institution has provided in the search results speak for itself.

In each case, the search engine has done the user hoping to find information on the collapse of the Roman Empire a disservice. At best, the engines are burying them in almost entirely unrelated advertisements; at worst, directing them to modern propaganda. It is no wonder that my grandmother found little useful information on the subject. For all the convenience they provide, these search engines have done little to make information on this subject more accessible to the public.