By Micah E. Weiss
Warning: the following essay contains verbosity, convoluted syntax, and long discursive sentences written expressly for the purpose of making you slow down.
There is always the temptation to “strike while the iron is hot” when it comes to controversial topics that are sweeping the zeitgeist down some speculative wave. This function of the media—intellectual punditry in a swarm of hasty commentary—is a sorting mechanism for determining acceptable opinions that will then move into the two Overton windows to save most individuals the trouble of having to think for themselves should they want to avoid that calorically costly process. Thus new ideas, technologies, candidates, etc., can be placed in the context of the orthodoxies of our increasingly polarized elite cultural wings with something that looks like public debate. ChatGPT appears to be an exception to that sorting rule.
The controversy came over the newsstreamsinternets without the usual polarizing clash, and is now being absorbed into the public consciousness, fully digested, with public opinion reduced to a handful of acceptable ideas regarding the inevitable future outcomes of this “revolutionary” technology–few of them associated with a political wing. How odd. However, if one searches around the usual publications for some intellectual insight, one may find the responses to be rather knee-jerk and predictable, from Wired’s expected “don’t panic” variety, to Slate’s “you’re not going to like it, too bad,” to the inevitable assertion of decline and decadence of our society (a general point I don’t disagree with), to the most recent cold glass of water from Ross Douthat. While not expressly about ChatGPT, I consider Mr. Douthat’s thoughts to be just downstream of the fallout of the new panic-inducing tech because it reconsiders our own complicity in the decline of a lettered society. We were, for a brief period, a culture of printed texts, while now we are a culture of tweets and texts, and soon, we could very well be a culture of AI generated texts. And so the universities that have been using printed text for several decades now are appropriately shitting themselves.
Under threat is the college paper. The essay, the five paragraph analysis, the term paper, the final paper, the mid-term paper, the thesis statement, body paragraphs, and elusive conclusion, supporting textual evidence, the “Where’s my goddamn MLA handbook?!” Can you taste it, former English majors? Can you feel the confidence of the midnight run for cigarettes, the jolt of 2:00 a.m. coffee, the bleariness of 4:00 a.m. progress, the despair of the 6:00 a.m. “not done yet,” the triumph of the 8:00 a.m. printout, the joy of sleeping 12 hours the next night? I’m sure majors from History to Philosophy to even those wimps in the hard sciences (pfff, labs? Concrete evidence? Come on! Some of us don’t need “reality.” We have “theory,” dammit!) have similar shared experiences. And ChatGPT is coming to wipe it all out. From the Slate article:
“Consider the flood of essays that would have us believe that not only college English courses but in fact the entire education system are imperiled by this technology. In separate pieces, the Atlantic proclaimed “The End of High-School English” and announced that “The College Essay Is Dead.” A Bloomberg Opinion column asserted that ChatGPT “AI will almost certainly help kill the college essay.” A recent research paper tells us that GPT-3 (a precursor to ChatGPT) passed a Wharton professor’s MBA exam.”
All of the articles cited are articulating a serious problem; however, the solutions in the public sphere have not been inspiring. You might not be shocked to learn most solutions suggested have involved more tech. Tech got us into this mess, I’m sure it will get us out! Historian Ronald Wright coined the term “Progress Trap,” defined as the process through which human ingenuity in solving problems inevitably leads to greater problems. One of his best examples explains how when the first humans came to the North American continent they quickly learned how to run entire herds of megafauna off cliff sides in order to eat like kings. This method likely hastened the extinction of most of those species of megafauna, which necessitated a much less prosperous time for those early travelers. Life on the prairie was rather bleak for the tribal groups that inhabited them until a herd of horses in Santa Fe got loose; horses that were European in origin because one of the species of megafauna that went extinct after humans arrived on the continent was the American Horse.
For a more recent example which maybe you are already considering, the burning of fossil fuels has been the central factor in enabling the greatest growth in wealth in human history, and also just so happens to be the leading cause of greenhouse gas emissions. We are already trapping ourselves further in the attempt to fix that problem by replacing gasoline powered cars with EV technology which, in many geographies, aren’t any better, and indeed may be worse for carbon emissions than highly efficient internal combustion engines. Those are debatable points to be sure, but the long term prospects for EVs are terrible when you consider the sheer amount of mining that is necessary to build the batteries to power them, to say nothing of the frames to house the batteries of an entire fleet of EVs to replace our current gasoline-powered automobiles. Then there is all the lightweight material needed for EV interiors, and the generation of energy that powers EVs, and the lack of nuclear power plants being built, and the fact that windmills are difficult to recycle and maintain, and, and, and. . .I almost get exhausted by the number of progress traps I now see everywhere I look.
And if thou sayest fusion to me, I will slap thee with my glove, sir or madam, and challenge thee to a duel of honor! The most miraculous thing about fusion in the last 50 years of our brightest minds trying to invent it is fusions ability to perpetually remain 20 years out from the present.
ChatGPT is a progress trap of the first order. As a society that requires expertise in a variety of thought-based realms, what are we to do when those hired to think can’t be assessed reliably because there is a mechanical shortcut? A shortcut designed to make thought-based work more efficient. The solution? Most proposed solutions in the public sphere came in the form of more technological and mechanical shortcuts. Apps that surveil students, AI to hunt and find the AI, digital “watermarks,” new and improved firewalls. The solutions laid out in the Slate article were nearly parodic in their lack of detail:
“Instead of ‘imagining’ what ChatGPT might do to the classroom, educators have to adapt discussions, activities, and assessments to the changed environment that it creates. Some of that work is exciting, like when many of us began to bring social media into the classroom to connect our students with outside thinkers or collaborating in real time on a shared document.”
Wonderful! What a bunch of non-suggestions, except that last one—and what do the authors think the students will be doing on their phones while they collaborate on that shared document open on their laptop?
Instead of insisting that we need more progress to get out of the progress trap, I might tread lightly into the discussion and suggest a modest solution from a bygone era not actually that long ago, and the structures to support that solution, as we move into the brave new utopia that advances in technology will surely deliver for us. In-class, in-person, real-time assessments, on paper, and physical books. This is where we need to double down, before it’s too late, and students forget how to use a pen entirely. The classic Blue Book pictured above is still being sold in bulk, so they are still being used by some teachers and professors. Students are still being taught to write by hand, though penmanship as an assessment category is long in decline in elementary schools. It shouldn’t be difficult to roll back the clock for the sake of an honest assessment practice to sort the future teachers, thinkers, leaders of our country. I, for one, would like them to be able to write a sentence, by hand, without the help of Grammarly, and to think on the spot without the help of cliffsnotes.com. If the Blue Book is still in use, this cannot be an absurd suggestion.
So, naturally, I have to make it worse. Handwriting. This is a skill that nearly everyone can master given practice and discipline from an early age, utilizing drilling, and drilling, and more not-fun drilling. Elementary schools must return to heavier emphasis on handwriting. To do it correctly, we’d need to reboot penmanship as a culture, and likely move back towards cursive as the standard, a form of writing that is not just faster than simple print, but also has the added benefit of being more beautiful.
There are serious downsides: not everyone has a natural talent for fine motor skills like writing, and I am one of them. I attended a Waldorf School where they still operated in the dark ages (by choice) and so I was taught to write in cursive. I did not take to it very well; it did not come naturally to me, but, for a brief part of my childhood when I was being drilled in my writing day after day, my handwriting was improving from actually legible to nice to look at. Then I moved, transferred to a public school, and no one cared anymore. The steady decline (more like a fall off the cliff) of my penmanship skills never recovered, and to this day my students squint at my whiteboard notes. It’s too late for me, but the handwriting issue could be centered in anticipation of a push in universities to get a little old school when it comes to the variety of assessments they utilize to gauge learning in their students.
This provoked in me a reminder about where the power in this educational structure really lies. The universities themselves should be dictating standards from the most advanced students down to the least. I think education departments have forgotten this, if they ever knew it. It is high time that the universities reassert their authority over the lower levels of education. From a market perspective, graduate schools are the customers of undergraduate programs, undergraduate schools are the customers of the secondary schools, etc down to Pre-K. If higher educational institutions say they want students with good handwriting, the lower schools will provide.
I have issues with the educational system that run rather deep, but as I mentioned in my first essay, we did build the most powerful and prosperous nation-state in the history of mankind with the help of an educational system that looks somewhat like the one we have now, so it must have worked in the past. We should remember what worked, and hold onto it, or bring it back.
Aside from handwriting and Blue Books, my final argument is related to the Douthat piece cited above: Victorian syntax. ChatGPT may be able to mimic a Dickensian sentence, and certainly the AI of the future will, but we can counter its effect on us by encouraging the verbosity of previous epochs hitherto forgotten in the break-neck process of updating and “decolonizing” curricula. The politics of these changes I find abhorrent, but those that pay the true price for the loss of voices of the past in favor of politically acceptable voices of the present are the students whose ability to think in layered, complex, and multivalent ways is materially diminished by lack of exposure to the greatest practitioners of the art of perspicacious voluble word-craft from eras in which that craft was most prized (I know, too much, but I couldn’t resist). There is extraordinary value in requiring a young mind to slow down, re-read, and decipher a long and complicated thought.
Before we jump into either radical acceptance or panicky anxiety of new tech, and definitely before we listen to the second-string ‘journalists’ the intelligentsia sent out to cover ChatGPT and its consequences, we should remember that we’ve been doing some of the tasks we prize as a civilization for hundreds or even thousands of years, and, as the adults in the room, we set the parameters that students have to accept. In this case, there isn't much to this issue that can’t be addressed with a pen, a Blue Book, and a well read copy of Great Expectations.