X is for Dog

X is for Dog is a creative experiment in which I use Digital Humanities-adjacent tools and techniques to assemble, probe, manipulate, and visualize four different systems of domestic dog classification.

Research Agenda

Dogs are more than just animals; they’re also man-made technologies! For this reason, dog breeds have historically been classified as breeds rather than subspecies. In other words, the mainstream methods for classifying dog breeds don’t conform to taxonomy conventions from the natural sciences. Instead they tend to group dogs according to the “design intent” behind the development of the breed. Curiously, the taxonomies in use today also exclude breeds that are extinct or seriously dwindling in number, another red flag that these systems of knowledge might not be so scientific.


Because people’s relationship to technology changes quickly, frequently, and radically, many different taxonomies have been constructed to categorize dog breeds over the centuries (compared to a non-domestic/non-technological animal, for example).


I want to argue that taxonomy is at least as much art as it is science, which is a point of view that isn’t unique to me but that I do think is worth reiterating. I think domestic dog classification is a good way to illustrate my argument because it plays assumptions about the “hardness” of classification methods in the natural sciences (dogs are animals, therefore breeds of dog are probably categorized as subspecies of the domestic dog) against an easy-to-demonstrate but seldom-spoken condition (they aren’t, because dogs aren’t strictly animals: they’re also man-made technologies).



Research Question


What’s different and similar among my selection of taxonomies, and what further meaning can be extracted from these findings?



The Datasets (or Captasets)


Here’s what you need to know about the datasets:

  • You can check out my spreadsheet here.

  • They were made by me. Not as in I made the data up, but as in I compiled the data by hand rather than starting with a prefabricated spreadsheet of data.

  • They come from four different sources (shown above), one for each school of classification I looked at. The data structures of the sources were all very different.

  • I treated the extinct breeds list as its own marginalized, monolithic classification scheme, since these dogs aren’t documented in any others.

Therefore-

  • They’re in no way comprehensive or authoritative.

  • I’m hesitant to describe them as data. Johanna Drucker’s alternative term, capta, might be more appropriate. But for the sake of accessibility, because everyone knows what data means and most people probably don’t know what capta means, I’ll just say data from here on out.

This project is maybe more of a proof-of-concept for some future project that would need to be done by a team of people.

What I Did

  • Firstly, I needed to build my datasets. This basically meant studying each of my sources and distilling the most essential information into a spreadsheet.

  • At this point I began to appreciate that spreadsheets really don’t want to deal with fuzzy or qualitative information, so I imported my spreadsheets into a data asset management program called Omeka. This enabled me to do two things:

a.) create a satellite data container that can happily contend with issues like tags, multiple or uncertain information values, and bodies of text, but is still linked to my spreadsheet

b.) run the data through some of Omeka’s many plugins to perform different analyses on it. You can check out the Omeka site here.

  • I subdivided the data into several sets to do more specific, comparative analyses on them, and ran these datasets through a bunch of filters to generate visualizations.

  • I examined the visualizations and built personal interpretations out of the results.

Visualizations + Interpretive Findings


INTERPRETIVE FINDING #1

Studying the natural history of domestic dogs could be a wellspring of information about human history.


INTERPRETIVE FINDING #2

Studying the history of record-keeping about the natural history of dogs could be a wellspring of research material for sociologists and psychologists.

The existing record-keeping about the natural history of dogs strongly suggests the prioritization of white, European history and disregard for others.


INTERPRETIVE FINDING #3

This selection of descriptions and classification styles supports the argument that the Western mainstream prioritizes the role of dogs as technologies above their role as animals.



Media

  • Omeka Classic via Reclaim Hosting

  • Omeka plugins:

- Neatline

- Simile Timeline

- Waypoints

- Ngrams

- CSV Import

  • Hypothes.is (Chrome extension)

  • Microsoft Excel

  • Google Sheets

  • Voyant

  • RAWGraphs

  • WordWanderer

  • Adobe Photoshop

  • Adobe Illustrator

Additional Inspiration: