The Political Compass 2.0: Mapping Memetic Tribes

Figure 1: Collections of political compass memes on 4chan (left) and Instagram (right), and a failed yet serendipitously mesmerizing Gephi graph (middle).

Team Members

Hailey Beaman, Nicola Brignoli, Anthony Burton, Eleonora Cappuccio, Simon Copland, Carmen Ferri, Kim de Groot, Sal Hagen, Emilija Jokubauskaite, Stijn Peeters, Ilir Rama, Jack Wilson, Daniël de Zeeuw.

Contents

Summary of Key Findings

Instead of doing a traditional digital methods data sprint, this project formed a more exploratory, creative, and artistic endeavour. Specifically, the goal was to develop new frameworks for mapping and interacting with new political subcultures online, or “memetic tribes”, using the political compass meme as a starting point for digital and artistic research methods. It resulted in a qualitative mapping of online political subcultures, original empirical research, an interactive 3D universe with “memetic tribe planets”, and a role-playing card game.

1. Introduction

In a recent article published in New Models, New York-based artist Joshua Citarella writes: “The Overton window for Gen Z online political spaces is infinite. The cultural nichification of the internet is producing communities so polarized that they have almost no concept of a shared reality or grand narrative” (2019). At the boundaries of the liberal-democratic center where politics has grown stale and is reduced to a form of authoritarian technocracy, what is left of the political splinters into a thousand blistering fragments, creating new pockets of radical thought masquerading as political cosplay. In these liminal spaces outside of the purview of the “lamestream” media, where the basic institutionalized reality-consensus starts to unravel, the lines that separate left from right, conservative and progressive, traditionalist and futurist, start to grow porous and open to new invigorations: Randtian eco-capitalists, anarcho-primitivists, parallel-worldists, anti-trans feminists, solar-fascists, anti-Civilizationists, queer animalists, alt-rightists, libertarian postcolonialists, neo-Judaist flat-earthers, blockchain-dystopianists, constitutional globalists, pro-extraterrestrial life transhumanists, anti-illuminati, and bacterial-rights activists (only some of these were made up).

Figure 2: Memeified and abstracted political axes.

One of the ways in which these strange new political subcultures can be divided up is by a political compass that creates a multi-dimensional space consisting of three oppositional axes: left-right, authoritarian-anarchist, and traditionalist-futurist (more dimensions could obviously be added). Moreover, this space allows for the mapping of individual or collective journeys from one position to the other (or from occupying one to multiple positions, moving from relative consistency to paradox, and back again).

2. Initial Data Sets

As a starting point for our own classification of niche political subcultures, we used a spreadsheet of memetic tribes compiled by Peter Limberg and Conor Barnes (2018). This spreadsheet lists specific groups, including their “chieftains”, “existential threats”, “campfires”, and more.

For the empirical research, we used datasets of posts made on Reddit, Instagram, and 4chan. For Reddit, we used the Pushshift Reddit data archive (Baumgartner 2019) uploaded on Google BigQuery. We collected Instagram posts through the DMI Instagram Scraper. 4chan posts were collected through 4CAT (Peeters and Hagen 2018).

3. Research Questions

• What are the various memetic tribes of the weird internet?

• How can we “map” these tribes?

• How can we empirically research them with the various digital methods tools at our disposal?

4. Methodology

Mapping the tribes

With the memetic tribes spreadsheet as a starting point, we compiled our own tribe list. We added tribes that were absent, like anarcha-feminists, and added their logos, ur-texts, and mottos. This was a highly qualitative and non-rigorous practice, but we also added data metrics to these groups, such as the amount of posts that appeared in the subreddits we deemed as their “campfires”.

To “map” the tribes in a political space, we reused the concept of the political compass. Instead of the usual libertarian-authoritarian, left-right axes, we “upgraded” the compass. Considering the tribes occupied radically different positions on a spectrum of Internet nicheness, we added another axis: dank-normie. We also changed the libertarian-authoritarian axis with anarchist-authoritarian: many tribes were not primarily concerned with economic laissez-faire policies already captured in the left-right axis, but rather a complete absence of centralized state power. In specific, we understood the axes as follows:

• Anarchist-authoritarian: Concerns the question of where power was located in a given ideology. That is, whether the individual is operates with absolute autonomy with regard to their being in the world a-la the myriad ideologies with the ‘anarcho-’ prefix, or whether they are entirely subject of a near-or-actually-trascendental authority, like in the case of Esoteric Orthodoxy.

• Right-left: Concerns the configuration of the economy. In particular, dealing with the question of what group or class owned the means through which value is produced. For example, where the rightmost case ‘anarcho-capitalism’ would inevitably see a concentration of ownership among a small hypercapitalist class, the leftmost case - anarcho-communism - sees a diffusion of ownership and means throughout the body politic to such a degree that the property form loses its meaning.

• Dank-normie: Concerns with how 'online' a given tribe is - that is, to what degree do the adherents of an ideology articulate/enact their politics online. Somewhat ironically, this found the Ted Kazinsky-influenced ‘anarcho-primitivism’ as the dankest case, where the primarily offline ‘Christian Right’ was the most normie.

The coordinates were assigned qualitatively per axis, per tribe. A group of domain experts engaged in a moderated strictly structured discussion, working per one axis at a time, going through the list of all tribes and placing them on a hand-drawn scale from -100 to 100. These numerical values were only assigned after the subculture was physically placed on the scale by group consensus to avoid basing the discussion on individual assessment. A strong factor in the discussion was also referential and relative placement amongst the remainder of the subcultures, oftentimes manifesting in focusing on finding tribes that were believed to be located close to either pole of the spectrum as well as the null value, before proceeding with placing remaining tribes.

Use of Instagram hashtags by memetic tribes

In order to take a closer look at how memetic tribes position themselves in relation to one another within a platform, we performed a co-hashtag analysis on Instagram using the DMI’s Instagram Scraper tool, using anarchist subcultures on the platform as a case study. We also sought to analyze how political subcultures that share a larger ideological framework or idea (i.e., anarchism) might distinguish themselves from other “tribes” through the means of positioning afforded by the platform, like hashtags.

In his article “Politigram and the Post-Left,” Joshua Citarella cites anarchist-affiliated subcultures as well-represented within the Politigram issue space, ranging from anarcho-capitalists to egoists (2018). We determined that the prominence of this community would make it a feasible case-study to explore how the subcultures within it position themselves through hashtags.

In this case study, we interpreted hashtags as constitutive of an “issue space.” Discussing the use of hashtags and other such “digital objects” on Twitter and their role in discursive practices on the platform, Richard Rogers states that the issue space is “where publics organize and compete to establish what is at stake” (2019). Hashtags are used similarly on Instagram, both as a means of organizing content for the sake of searchability and of entering a conversation around an event or theme. Furthermore, Rogers points to co-hashtag analysis as a means to study the “twinning of concerns,” or those ideas that appear frequently together through hashtags (2019).

In applying a similar methodology to anarchist subcultural hashtags, we wanted to investigate what the usage of hashtags within anarchist subcultures or memetic “tribes” might reveal about how they position themselves in relation to each other and how we might map them relationally in turn. Using the DMI Instagram scraper, we scraped posts from the following anarchist hashtags:

• #anarchocommunism

• #anarchoprimitivism

• #anarchotranshumanism

• #anarchocapitalism

• #anarchosyndicalism

• #agorism

• #anarchopacifism

• #mutualism

• #anarchofeminism

• #egoism

• #anarchoindividualism

• #queeranarchism

These communities were determined by previous research to be the most prominent affiliations within the anarchist political space on Instagram. We scraped up to 10.000 posts per hashtag to get as much of the total corpus of posts as possible. Finally, we visualized the output of the co-hashtag analysis in Gephi (Bastian et al. 2009).

Subreddit Referrals: How do campfires refer to each other?

One of the core questions of this group was how these memetic tribes interacted with each other. Which tribes are linked, how closely are they linked, and can we map these links across the political compass?

To do this, we created subreddit referral networks. Beginning with the list of memetic tribes as explained earlier, we engaged in a qualitative process of linking tribes to Reddit subreddits. We searched Reddit using key terms for each tribe. This provided us with a list of 74 subreddits. Some memetic tribes had no subreddits associated with them (as they were often too extreme to be able to have a subreddit), while some tribes had multiple subreddits associated with them. Overall however we managed to choose subreddits across both the political spectrum, covering all points of the political compass.

From this list we then scraped how many other subreddits were referenced by users within these subreddits in the subreddit comments. We conducted a second scrape examining all of these subreddits to see how often they referenced the original list of subreddit tribes. We then created a network using Gephi (Bastian et al. 2009) to analyse the links between tribes. The subreddits in this network were coloured based on modularity ranking.

Exploring the Political Compass 2.0

With the finalized coordinates, a Web application was developed letting one navigate along the axes to explore what memetic tribe is placed where. 3D rendering library (three.js) was used in order to map in 3D. A polysemic theme of “space” was incorporated by mapping every tribe as a sphere. This was apt in a broad sense because of a long tradition in relating politics, opinions, and ideology to space or “maps”, but also materialized in simpler ways like backgrounds for the axes, such as the Death Star for Authoritarian. Further information was plotted per each “planet” sphere, including mantra, image, ur-text, Wikipedia page, and coordinates. The planet sizes were derived from the amount of comments on related subreddits, all from their inception until April 2019 using Google BigQuery.

5. Findings

Instagram Co-Hashtag Networks

The co-tag analysis of eleven hashtags affiliated with anarchist subcultures revealed a dense and yet fairly polarized network. From the visualization found in figure 3, we see that the network is largely grouped into three distinct clusters colored in green, pink, and blue. These clusters were determined using the modularity class algorithm in Gephi (Bastian et al. 2009), by which nodes are clustered based on their density in relation to one another. In other words, nodes within a certain modularity class (like the pink, blue, or green clusters in figure 3) have a markedly denser connection to those other nodes within that cluster than those outside of it, thus forming a kind of “community.”

Figure 3: Instagram co-hashtag network based on seeds of eleven anarchist tribe hashtags.

While all of the hashtags used as seeds for the tribes all share an anarchist ideological affiliation, there appear to be noticeable differences in the other themes that also appear alongside these hashtags. Furthermore, the hashtags included in these clusters appearing alongside the anarchist hashtags reveal what we might call a “distribution of concerns” across these spaces that seems to gesture to the ideas of main concern to these subcultures. As mentioned above, while all the anarchist tribes share an anarchist frame of reference, they appear to cluster primarily based on tags relating to what their ideal society would look like and what issues might be addressed in their specific anarchist program.