Featured image: “The Programmers Cockpit” by Jürg Stuker, CC BY-NC 2.0
New week, new blog post. This time I read “A Theory of Persuasive Computer Algorithms for Rhetorical Code Studies” by Estee Beck.
In this essay, Beck analyzes computer algorithms and how persuasion uses them. Beck uses two examples at the beginning. First, a job applicant named Joe Smith tried to apply for a job in the academic job market. The University rejects him not because he failed the criminal and financial background checks (he passed both of them). The University’s algorithms detected that Smith has anti-social and hostile behaviors on social media.
Another example involved a woman named Jane Johnson who bought an item from a store. A week later, Johnson receives a message saying her credit line is going down by more than a half. This was because the people who shopped at that store have had a poor payment history. This is despite the fact that Johnson had never went to that store before. The first story is not exactly true. It’s an amalgamation of reports that companies use algorithms when hiring which include social media activity. The second story however, was actually a true story back in 2009. Beck also uses examples of how algorithms best cater to each user. Like how Facebook curates certain friends’ posts. Amazon curates the best shopping recommendations. Google personalizing the best search results.
Beck defines persuasive computer algorithms by explaining three different features: algorithmic processing, algorithmic inclusion/exclusion; and algorithmic ideology. Algorithmic processing is how information is process and organized in a way to prioritize how to show information. An example given is when Facebook developed a newsfeed feature. It’s want to prioritize which posts would appear in a user’s feed more than others. Algorithmic inclusion and exclusion is how the algorithm will prioritize data that best fits their structure. If not, it will discard data that doesn’t fit that structure. An example given is how an algorithm will only allow data collection from websites that have at least 20 hyperlinks. So websites with less than that amount are not in that dataset.
Algorithmic ideology is what happens the creator of these algorithms insert their biases into said algorithm, whether intentionally or subconsciously. They can include social, cultural, and/or political ideologies. They can range from gender to race to class to values to their organization and many more. One example is the microwave. It was originally marketed as a hi-tech product for men to cook quick meals. After it failed to appeal to men, it was re-marketed as a household appliance for homemakers. Another example is how modularity in UNIX connects to segregation in the United States. UNIX programmers were situated in a powerfully raced and classed political landscape of their time, which unconsciously influence their encoding of UNIX.
Recently, a field called critical code studies has been emerging. It has provided critical theorists a sub-discipline to investigate the relationship between computer code and cultural, social, and political content. Beck differentiates critical code studies from rhetorical code studies by explaining that while critical code studies focus more on hermeneutics, rhetorical code studies focuses more on social science methodologies. Beck believes that rhetorical studies should be used to examine the relationships between machines and human behavior.
My Thoughts:
Another wordy essay. Seems to be a common thing on Enculturation. Not as difficult to understand as Ethos, Hexis, and the Case for Persuasive Technologies, but a bit more difficult than Sneaky Rhetorics, Habitual Media, and the Affective Politics of Lurking on Facebook. I do think that me reading both of those essays over the past two weeks help me understand this one better than if I read this one first.
This oddly seems to be an extensions of both the past two articles. I say oddly because while Ethos was published the same day as Beck’s essay, Sneaky Rhetorics was actually in 2022, six years later.
I notice that both Ethos and Algorithms talk about how digital rhetoric can influence and cater to our behaviors. Both Steven Holmes (the writer of Ethos) a nd Beck use Amazon as an example of how digital rhetoric influences what we purchase. It’s not surprising as Amazon is one of the biggest companies in the world. Everyone has an Amazon account, including me.
As I said before, Algorithms was published before Sneaky Rhetorics, but it feels as though Beck was responding to the authors of Sneaky Rhetorics even though obviously that isn’t the case. The authors of Sneaky Rhetorics note that they frequently went on their Facebook pages and try to find out why. Algorithms expands on that question with Beck giving a theory as to why social media and other computer algorithms can influence decisions.
Now for me, I’m no expert on computer programming and algorithms so I can’t say if I agree or disagree with Beck’s theory. Some of stuff Beck writes does make sense or at least I can see happening. Like how algorithmic inclusion/exclusion is about how datasets will prioritize or discard various pieces data depending on the programmer. Because things get so big, there are gonna have to be points where not all information matters or certain does matter so much it needs to be there. There’s Beck’s example in the essay but I’ll add another example, me. While i’m not writing an algorithm, then I write these summaries, I need to know what parts of these essays are important to understand the overall meaning and what isn’t necessarily important so I’m not just writing a cliff notes version.
I will also agree that programmers put their biases into algorithms whether they wanted to or not. Like I said, I’m not a programmer but from my ventures in the arts and just everywhere that involves making something, biases will show up. Another example I’ll add to Beck’s is, me again. I chose to read Sneaky Rhetorics last week because of my own bias that I already had knowledge in social media use and “lurking”. In fact, the reason I’m bringing these parts of this essay specifically is because of my own biases.
Overall, I think this is a good theory with evidence that can cited from just personal experience. However, like with Steven Holmes, I feel like I need to do more research in this topic before I can give my formal opinion. Beck does give a good stuff to start off of and you never know. Maybe after my own research, I can create my own theory.
Leave a Reply