Why is online communication so toxic? Why does it often feel like different groups are talking past each other? One reason might be the type of writing digital spaces encourage. Simply put, much digital discourse—on blogs, news platforms, social media, etc.—is marked by the use of restricted (rather than elaborated) codes centered around emotionally laden, community-specific keywords. As we’ll see, digital space incentivizes this sort of writing. Indeed, restricted codes are often an efficient means of communication. But (and this is important) they are not well-suited for communication between groups. Therefore, the use of restricted codes may contribute to the sense of disconnection we often experience online.
The distinction between restricted and elaborated codes comes from sociolinguistics. In a restricted code, certain words act as a form of shorthand—pointing to an entire complex of ideas. This allows meaning to be conveyed with fewer words. Such writing can be used when the communicating parties share a set of assumptions or experiences. Elaborated codes, on the other hand, are more explicit. Terms are defined and the connections between ideas articulated. The text, in other words, provides more direction as to how it should be interpreted, thus allowing similar meanings to be made by people with different sets of assumptions.
Digital writing is often highly restricted. This makes sense for a number of reasons. First, reading on a screen is hard. It’s physically taxing (compared to print) and within the digital environment, there’s a lot of other stuff vying for our attention. The writer, therefore, to get read, needs to condense meaning. It also takes time and energy to make a text more elaborated. Digital texts are often produced quickly; both tech platforms and the current media environment push writers to emphasize quantity over quality. Once again, this equals more restricted texts.
Finally, and perhaps most importantly, it seems to me that in the digital world, venue increasingly helps determine meaning. As noted, restricted codes work when the writer and reader share a set of assumptions. In a world where each reader controls her information flow, the mere selection of a venue should equate, roughly, to a defined set of assumptions. If I’m an urban liberal, for example, I’m probably not going to read Brietbart. A Brietbart writer, in turn, because he doesn’t need to “speak to the other,” can use a more restricted code. A sort of feedback loop develops. As the Brietbart code becomes more restricted, it becomes less likely that liberals will choose to read the Brietbart text. The Brietbart writer is thus free to make his code even more restricted, and so on and so on.
From the above analysis, we can see why, when trying to communicate online, different groups might seem to be talking past each other. As noted, an elaborated code can “stand on its own”—it contains all the information a reader needs to make meaning. To understand a restricted code, on the other hand, a reader needs to hold a certain set of assumptions. If she doesn’t, the text can only be read with great cognitive effort and/or is more likely to be misinterpreted.
Let’s take a look at how the above ideas play out in practice. Twitter missives are perhaps the ultimate example of a restricted code. Consider this tweet:
This tweet received over 800 likes, indicating it successfully conveyed meaning to its audience. The nature of this meaning, though, is not apparent on the face of the text. This is because apart from the seemingly illogical “on the contrary,” no direct guidance is given as to how to connect the tweet’s two sentences. A shared definition of the term “toxic masculinity” is needed to establish this connection and make the message coherent. The tweet’s intended audience, or many of them anyway, can access such a definition. For them, the tweet reads something like:
It’s so sad to see [a societal demand that men not show emotion] destroy relationships between dads and their children. on the contrary: [displays of emotion such as] kindness, love, and tenderness are foundational to being a good man and father.
Understood in this way, the tweet makes a coherent claim. Integrally, though, if a reader doesn’t have access to the shared definition—if they are operating with another set of assumptions towards the emotionally charged keyword “toxic masculinity”—they will not have access to this claim. The message will likely appear illogical or as a cliched “attack on men.” In other words, because of the highly restricted nature of the code, only readers who already basically agree with the writer can grasp her meaning. As such, this sort of writing acts as a barrier to communication between groups.
Interestingly, the same keywords often circulate among different communities, where they have different, but equally implicit definitions. Consider a recent Fox News article entitled “This Father’s Day let’s call toxic masculinity what it is.” Though longer than the above tweet, this piece is also written in restricted code. And again, it is almost incomprehensible to an outsider. The author’s primary claim is/seems to be that the conventional ideal of manliness is good, needed and under attack. The discussion turns around the term “toxic masculinity.” The academic left accuses manly men of having “toxic masculinity.” They insinuate that manliness is to blame for the actions of sexist workplace monsters. Apart from this “they insinuate” there is no further attempt to define what “toxic masculinity” might mean to the writer or his opponents. Again, I’d argue, an explicit definition is unnecessary. The article’s intended readers, by virtue of having chosen to read a Fox News op-ed about toxic masculinity, can be assumed to share an implicit definition of the term. A simple reference to it invokes an entire complex of ideas, emotions and identities (undoubtedly negative), which the writer can then work against.
The tribal nature of behavior online has long been noted. The above analysis indicates that the very language we use might be fueling this tribalism. As noted, the use of highly restricted codes is driven by structural incentives. That means that for people (like me) interested in curbing social conflict, there is no easy solution. It does seem, though, that a general recognition that certain types of language are more likely to be ignored or misinterpreted by those different than us is an important first step. Perhaps our best option is to learn to read our own texts better—to recognize the implicit definitions and missing connections—and start to take pride in being (more widely) understood.