I’m teaching an upper-level rhetorical theory class this semester, and this week we’ve been discussing Bitzer/Vatz and the idea of the “rhetorical situation.” In case you’re not familiar, Bitzer thinks rhetorical situations are a somewhat more external, agreed-upon set of events that are defined primarily by the presence of three interconnected elements: exigence, audience, and constraints. Vatz, on the other hand, tends to place more emphasis on the subjective nature of rhetorical situations, arguing that the construction of “the situation” as such is a highly rhetorical and discursive process. In short, Bitzer sees the situation itself as the primary influence of rhetorical activity while Vatz views the rhetor themselves as the primary factor. Although both approaches have their shortcomings, and subsequent theories of the rhetorical situation are perhaps more compelling (Biesecker; Rice), I find Bitzer useful in pedagogical contexts as a kind of quick heuristic for analyzing the complexity of rhetorical discourse.
Bitzer’s article was first published in 1968 and is easily one of the most influential pieces of rhetorical scholarship. So this got me thinking about my previous post about the use of AI text generation. Because so many people have written about Bitzer’s theories, I was curious to see how the AI would respond to prompts that ask it to define, contextualize, and perhaps even apply his theories to various contexts.
First, I asked it to define Bitzer’s theory in simple terms:
Bitzer’s theory of rhetorical situation states that the effectiveness of a rhetorical situation is determined by three factors: the exigence (the need or problem that is motivating someone to communicate), the audience (the people who will receive the communication), and the constraints (the limitations of the rhetorical situation, such as time, space, and available resources). Essentially, the success of a rhetorical situation depends on how these three factors interact and how they can be used to create an effective message.
Not bad, but probably not too difficult considering that definition is maybe the easiest function of AI. Similar results would most likely appear if I typed the same command into the Google search engine.
But here’s where things got…interesting. OpenAI can not only produce information but actively adapts the text style of its output based on the input commands. Playing around with this function, I asked openAI to do the same task (define Bitzer’s rhetorical situation) but this time in the style of a punk anthem. Here’s what I got:
Verse 1:
Discourse is the key, to solving any issue
Bitzer's theory of rhetorical situation is the way to use it
It's a dynamic process, that starts with a need
To create a context for effective communication to proceed
Chorus:
Rhetoric is the power, to shape any debate
Context is the tool, to make a strong case
Bitzer's theory of rhetorical situation, will help you through
When you're trying to make your words come true
Verse 2:
The audience is the focus, of any good discourse
It's the factor that will determine, what words will work of course
The exigence is the problem, that you're trying to address
And the constraints that you face, could determine your success
Chorus:
Rhetoric is the power, to shape any debate
Context is the tool, to make a strong case
Bitzer's theory of rhetorical situation, will help you through
When you're trying to make your words come true
Bridge:
The speaker, the audience, the exigence, and the constraints
Are the four components, that Bitzer explains
It's a powerful tool, to help you communicate
Not exactly groundbreaking from an intellectual standpoint but kind of mindblowing from a stylistic one (at least for an AI). It not only has rhyming elements but actively links these rhymes in ways that make sense conceptually with Bitzer’s theories. It’s not perfect of course, and some bars don’t totally make sense (context is the tool?), but I can’t help but be impressed with this result.
Finally, I asked the AI to actually apply Bitzer’s theory to a given rhetorical act: President Biden’s acceptance speech.
President Biden’s acceptance speech is an example of a rhetorical situation, in which the speaker is trying to influence the audience towards a specific action. According to Lloyd Bitzer’s theory, the three key elements of a rhetorical situation are an exigence (a problem or pressing need), an audience, and a purpose or goal. In President Biden’s speech, the exigence was the current state of the nation, which he addressed in his call to unite and heal the country. The audience was the American people, and his goal was to inspire them to work together to create a better future. All three elements of the rhetorical situation were present in President Biden’s speech, making it a successful example of rhetorical situation.
Definitely the least impressive considering that there is little incorporation of Bitzer’s actual terminology here. It seems that the AI is mostly latching onto generic notions of rhetoric to conduct this analysis, which is unsurprising considering the prevalence of bland, generic rhetorical analysis essays that are circulating around online.
To go meta with this post, what does AI technology mean for the kind of rhetorical situations writers find themselves in? Which component of Bitzer’s theory are most transformed by this technology (exigence, audience, constraints)?
There’s an argument to be made for all three, but I would say that it most immediately transforms a writer’s constraints, particularly at the level of process (time), stylistic output, and genre familiarity. As of now, these tools can be most easily exploited by writers who already have a strong grasp of the information and texts that they are asking the AI to generate. This still requires a keen editorial eye, and perhaps even expert knowledge, of the ideas within the AI generated text. But it undoubtedly expands this expert’s ability to craft their message in a variety of styles, formats, and genres, almost instantly and for nearly any occasion or audience. As such, it not only removes stylistic constraints but more material constraints like time, factors that (in a capitalist society) place strong pressures on the genres of writing that come to define various industries and contexts.
Considering Microsoft’s recent $10 billion investment in ChatGPT, it’s likely that these tools will soon find their way into word processing programs in the not-too-distant future. Like I mentioned in my last post, resistance to AI-generated text is futile. We need instead to figure out how this reshapes the boundaries of what we mean by “writing” and “rhetoric” and how our research and teaching practices might adapt in response.