What is Writing? Why We Misunderstand the Coming of ChatGPT
Is high-school English dead?
A Dec. 9 article published in The Atlantic by Daniel Herman, a high-school English teacher, says yes. Herman asserts that the new AI chat program ChatGPT drastically changes the nature of education, especially the teaching of writing. The software can respond to prompts of almost any kind—even very complicated ones—in a convincing human-like manner.
Specifically, Herman questions whether academic writing remains a relevant and teachable skill—or even a fitting metric for intelligence—now that ChatGPT can autogenerate essays that are better than most student writing. And education isn’t the only area that may be impacted by the program’s impressive abilities. The concern is that any industry, task, or job involving writing could be affected.
Certainly, ChatGPT comes to us as a force to be reckoned with. It will be harder to detect cheating, for example, and information online must be treated with even more skepticism. But this sudden outburst of concern over the future of education and other intellectual pursuits because “the computer can do it better” makes little sense. These cries of dismay could only come from a society that has lost track of what intelligence, thought, creativity, and writing really are.
Let’s take the field of education, for example. “It’s no longer obvious to me that my teenagers actually will need to develop this basic skill [of writing],” writes Herman. Toward the end of the essay, he asks a fatal question:
Many teachers have reacted to ChatGPT by imagining how to give writing assignments now—maybe they should be written out by hand, or given only in class—but that seems to me shortsighted. The question isn’t ‘How will we get around this?’ but rather ‘Is this still worth doing?’
These words are tinged with a note of despair: What’s the point, anymore? Computers have finally overtaken us in the realm of abstract thinking, writing, and art, just like they did in mathematics, chess, and scientific modelling.
Like Herman, I am a high-school English teacher, but questions like these never even cross my mind in connection with AI chatbots. In my teaching, I am not looking only at outcome or end product, but at purpose, intention, and process.
Of course high-school English is not dead. Of course writing and creativity remain the province of human agents. The fact that such questions can even be asked reveals that we have, as a culture, long misunderstood the nature of the liberal arts and even intelligence itself.
We are dealing with a denial of the soul and free will, a failure to distinguish between man and machine. Herman doubts the usefulness of his profession because we can get the same or better end product using AI as we can get when we teach kids how to write on their own. So why teach writing? This scientific, output-oriented approach looks at the two end products—one written by a human being, the other by a computer—and compares them as though they are comparable. But in reality, they bear no similarity because the agents and processes are completely different.
Writing encompasses a thought process and a communication. One reason to write is that writing helps us learn how to think even as we share those thoughts with others. Writing is a message with a meaning. But in order for there to be meaning in something, someone must put it there—a conscious, rational mind, using its free will, must put sense and intention into the message.
That meaning is then extracted by another conscious, free mind. What the computer produces is not actually a message by that definition. In a very real sense, the computer’s “essay” is meaningless because there is no conscious intent behind it by a rational, self-aware mind. There was only a complex, lifeless algorithm of some sort created by a team of programmers. It simulates a communication, a message, but it is just another one of those simulations that plague our modern world and try to deprive us of the real thing. There is nothing real behind it.
The computer has no intentions behind its messages because, lacking a soul and a conscious mind, it has no intentions at all. It has been “trained” by scanning countless texts written by real people and “learned” the patterns of our language—patterns used by those who are actually communicating.
The computer just blindly apes this process by using statistics to predict what kind of word should go where, based off the millions of model messages it has “read.” But it does not know what they mean. So when we talk to chatbots like this, we are not really communicating: We are only hearing an echo of our own, human thoughts, a kind of reconstruction of fragmented sentences from the millions of sentences by real people swirling around on the internet. These are our own words.
As far as actual education is concerned, such a tool is useless. Using ChatGPT to “write” for you does not teach you thinking or communication. And it’s absurd to equate an AI generated text, however seemingly eloquent, with something written by an actual student. In the case of the student, actual thinking and communication (however fuzzy) happens. In the other case, only a convincing illusion of communication occurs.
Perhaps the ills connected with ChatGPT are our punishment for forgetting what we used to know about language. In 1967, Roland Barthes wrote an essay called “The Death of the Author” that called into question traditional literary interpretation and, indeed, the nature of communication in general. Barthes argued that we can’t rely upon the intention or biography of an author to find the “ultimate meaning” of a text. We can never know what intention or objective meaning lies behind a message. For Barthes, there is no “ultimate meaning” because the meaning of a text is created by the reader, not the author. And the author—like Victor Frankenstein—vanishes and dies, overcome by the Thing he has created, which takes on an autonomous life. Meaning escapes from the hand of the author and, at best, is recreated in hundreds of mutated ways by the reader. In such a scenario, true communication becomes impossible.
We can thank ideas like death of the author for corrupting our understanding of what writing is. These ideas prepared the way for us to view a computer-generated text and a human-generated text on the same plane. Since we have denied the relevance and even existence of a rational intention behind communication, what difference does it make if the message comes from man or machine?
Drawing upon the nihilism found in the Eastern philosophy he teaches, Herman expresses this despairing relativism succinctly in the conclusion of his essay:
Everything is made up; it’s true. The essay as a literary form? Made up. Grammatical rules as markers of intelligence? Writing itself as a technology? Made up. Starting now, OpenAI is forcing us to ask foundational questions about whether any of those things are worth keeping around.
But, of course, for those of us who believe that meaning can and must exist in language and communication as a product of intelligent and free minds, it remains essential to keep writing. If we hope to think well and deeply, and share our thoughts and our hearts with other intelligent, conscious beings, then we must keep teaching and learning to write. In a society that equates man and machine as well as matter and spirit, the art of writing—true, human writing, messy, disorderly, but carrying all the heft and blaze of actual human thought and feeling—is needed now more than ever.
—
Image credit: Flickr-Brett Jordan, CC BY 2.0