On Digital Irony, Sincerity and the Rhythm which Binds the Two*

Amidst the rise of the far right and subsequent post-truth narratives in 2016, there have been multiple accounts which amplified what we can call ‘ a crisis of irony’. The premise of the crisis is that factions of the far right were ‘weaponizing’ irony to further their agendas. Amongst many things, this referred to incidents such as the radicalization of the online meme depicting a hate-filled right-leaning version of Pepe the Frog. However, the discomfort with the alt-right’s use of irony goes beyond its re-appropriation of ironic discourse; Irony is complex, relative and emotion fuelling. Irony, in this post-truth climate, has become a major cogwheel in the mechanisms of concealment. Irony works silently in the background, hidden by the camouflages of double-meaning and cultural specificity. The crisis of irony is that when it comes to mass media, we tend to understand the uses of irony without understanding its form. As the Chicago School of Media Theory puts it:

“A definition of irony gets into trouble once it goes beyond this nature of having double meanings, because the way of arriving at the ‘real’ meaning can differ from person to person, and some people might take the ‘face value’ meaning for the ‘real’ meaning — in other words not find the message ironic at all”

Another tool for irony detection?

This disconnect has led to sparse and mechanized investigations in the field of data science for the sole purpose of detecting irony and sarcasm. Accounts like [1] [2] [3] look at Machine Learning (ML) methods of detecting sarcasm. The underlying assumption about irony which all these accounts make is that it’s absolute and self-contained. They isolate single statements for analysis, disregarding the context in which the statements were said. For example [1] uses ML classifier to detect the difference between an apparent sentiment and a database of objective truths as a marker for irony. For example, “ I just love being stuck in traffic” provides an apparent contrast between “I love” which is a positive sentiment and “being stuck in traffic” which is an objectively negative experience.

However, as online discourses and cyberspaces have increased in complexity, more ML efforts attempted to center context-aware understanding of irony . For example, [3] looks at specific user histories in order to define and understand each user’s attitude towards irony. Meanwhile. [4] tries to look at the corpus of nouns from an entire community to understand the underlying sentiment of the community. As some of these approaches might be costly, and still gamble on cultivating a coherent understanding of the lingo and subculture of each of the cyberspaces in question, a different framework was needed to better grasp the network nature of online irony.

Here debuts the notion of Lefebrvian rhythm; French philosopher Henri Lefebvre in his posthumously published book Rhythemanalysis provides a meditation on everyday rhythms and how they reflect and influence social processes. His study of rhythm has inspired multiple investigations in the field of urban studies, reclamation of space and collective memory. Seeing as cyberspace is also a space where rhythms contest, compete and reflect techno-cultural battles, and that irony is a multi-use cultural marker, how can we better expose the relationship between rhythm and irony?

The scope technocultural rhythms would be anything that falls within the human-machine collaboration that makes the web 2.0 possible; from the flow of packets as regulated by a protocol, to the sorting of a feed of information by a social media algorithm, to the flow of clicks, reactions and comments.

two different rhythms imposed on the same content [source]

A digital study of rhythm as pointed out by Shintaro Miyazaki can bean important analytical tool in studying the techno-cultures of the web. What can rhythm teach us about the way irony is used online?

Figure 1: a rhythmic experience of irony

Figure 1 is an illustration of a single user’s experience of irony; the user in question is experiencing a feed of posts. Based on the user’s cultural understanding of the content, they’re bound to project their own un(ironic)meaning on the post. Since Irony is also a shared and networked experience, the variety of ways in which different users can experience the same feed of content must also be considered [figure 2].

Figure 2: Networked rhythm of Irony

This sparsity of cultural understanding, and the different projection of rhythms when experiencing a feed of content teach us two things:

• When looking at weaponized irony — mere irony detection is insufficient. Irony, or its lack thereof, is often expected apriori.
• Irony and rhythm online are a shared and co-created experience.

A closer look at networked irony:

A case study was needed to further bind the understanding of networked irony together. Pro-life memetic spaces on Facebook was an interesting digital site to consider, taking into account it’s a politically charged, health-policy related space. Consider this particular habitat of interlinked facebook pages in particular:

Figure 3: Habitat of interlinked pro-life meme pages

To further zoom in on the memetic spaces, two pages were chosen in which studying rhythms would be illuminating:

• Dank Pro-Life Memes: To be dank is to deliberately detest and reject both structure and rhythm. Dank memes champion Hito Steyerl’s Poor Image. They also champion non-substantial engagement with politically charged content. A Dank meme is also a meme of low circulation value (see this work on the tracking of value and on Meme Economies ). How does a dank meme page balance between its projected ‘dankness’ but also effectively lobby for policy change and push forward its social cause?
• The Fluffington Host: As the name suggests, the page is a parody of the Huffington Post. Its content comes mainly in the form of newspaper headlines. This page proliferates non-visual memes or repeating tropes in liberal media coverage to deliver content.

Finding Rhythm and Uncovering Sincerity

By looking at both pages and how they construct their online identity, it becomes clear how their rhythms of posting content differ in form; the earlier uses images of dank memes as a medium, while the latter uses fake text-based headlines. It also becomes clear that both pages subscribe to the larger postmodern moments of post-irony and new sincerity. That is, the ionic content is riddled with sincere moments that might be helpful to uncover.

Multiple questions arise:

1. How can we define the shared experience of irony amongst the audiences of each group?
2. How can rhythm and its distribution help us understand the nature of irony and the underlying sincerity in each case?
3. How can we better reveal the rhythms to a wider audience?

It’s possible to consider audience interaction metrics as indicators of the shared experience of rhythm. Facebook’s reactions provide a richness of prescribed cultural meaning and affect. Practically, Facebook exposes reactions to its public posts through its API, making reactions an accessible analytical tool in understanding identities of facebook pages.

1. To define a rhythm based around audience reactions, a scatterplot was made for 6 months worth of content and their relative emotional response from each page. The plot then allows for the visual inspection of the posts and their relevant outliers.

Figure 4: scatterplot and outliers of the Fluffington Host page

Figure 5: scatterplot and outliers of the dank life meme page

From the scatter plots, multiple candidates were identified as outliers that were, in essence, a disruption of rhythm. These were posts which either illicit too many reactions or too little reaction compared to the flow of reactions on the surrounding posts. Take for example the following post:

Figure 6: outlier meme

“Most of the time, I have a good time finding a terrible pro-choice argument, finding a quick way to refute it, and making a meme to mock its silliness. And other times… the reality of abortion really hits me”

This post by far is a clear disruption of rhythm as it solicited numerous sad reactions. This interestingly coincided with the sincere nature of the post, where the OP (original poster) of the meme posts a heartfelt reflection on his memetic practices and exposes the underlying cause of being pro-life.

Similar patterns were found in the Fluffington Host page, where the rhythm was often disturbed by non-sarcastic comments annotated with “- staff” or “(not fake news)”

Figure 8: Outliers in the Fluffington Host Page

The Fluffington Host page has actually weaved these distributions into its rhythm as if to cause what Lefebvre would’ve dubbed as a Eurhythmia or the constructive interaction between rhythms. The page would routinely publish staff annotated comments for the purpose of redirecting the irony whenever it was felt that the audience would lose touch. This demonstration of ‘careful irony’ is particularly helpful when understanding the weaponization of irony. Just as files are routed and rerouted by the different nodes in a network, online irony is also often disseminated by a conscious perpetrator, distilling the irony with sincerity as it navigates different modes of cultural understanding.

To make the rhythms accessible as a tool for a wider audience, an experiment was conducted to sonify and visualize the flow of interactions in each page. Every interaction was turned into a track, and the tracks were overlayed on top of each other to expose the rhythm of emotional interaction on the respective page. The results can be seen in the videos below:

A side-by-side visual of each track can be seen below:

Although the investigation was explorative, it provided valuable insight into the nature of irony and the uses of irony. It also set a precedent for a methodology for rhythm studies on social platforms. It also opens the floor for further explorations.

• How can we identify and study different Facebook-specific rhythms, beyond likes and reactions?
• How can we better define and study user-imposed rhythms (such as likes and shares) with algorithmic, platform induced rhythms (such as information sorting)?
• Can we repeat the same investigation across platforms?

This investigation is far from conclusive; it only brushes the surface of conceptualizing irony and networked humor. It provides insight on how irony can take shape on digital mediums, where its projection is the result of the interplay between human and machine. This work, as it stands, is the story of irony and sincerity as told by rhythm. It exposes one use of irony through a methodology which it hopes is transferable to other uses. Amidst claims of weaponization, this work still needs to directly address the question of power, how are power structures maintained by this ironic discourse. Even more importantly, how does irony shape the relationship between the oppressor and the oppressed?

• The work in this article was done in collaboration with Iskra Ramirez, Beatrice Gobbo and Daniel Leix-Palumbo, as part of the DMI Winterschool 2019

The Digital Society School is a growing community of learners, creators and designers who create meaningful impact on society and its global digital transformation. Check us out at digitalsocietyschool.org.