Centering Humanity in the Age of the Chatbot

8

Cincinnati, OH. “In the beginning was the Word, and the Word was with God, and the Word was God.” So begins the Gospel of John in the 1611 Authorized King James Bible. This landmark translation commissioned by King James I of England and VI of Scotland took some fifty scholars seven years to produce. According to one count, the results revitalized the English language with at least 257 new idioms, more than any other source including the complete works of Shakespeare. Not bad for crowdsourcing. Subsequently disseminated throughout the British Empire by legions of Protestant missionaries, this linguistic haul helped to transform English from a peripheral European vernacular into a global tongue on which markets and machines run.

While such globalization would not have been evident in the seventeenth century, the King James Bible revealed the Word made flesh in all its potential. At face value, for example, the blunt English word is a pale shadow of the majestic Greek logos; a different Greek word, lexis, more logically conveying word’s narrowly grammatical usage today. But the translators confidently grasped the Word of God as both the Platonic, immutable logos from which all reason derived, and the active, redeeming logos incarnate in the flesh of Christ. That such a double meaning came easily to the translators of King James’s day is less striking than the dwindling of the Word in our own. Our poor, desacralized Word is more lexis than logos: a dying scrap of verbiage more commonly evoked in supplication to Alexa or Siri than to the Almighty.

Whether or not we share the faith professed in the King James Bible, we may share some concern at the erosion of the Word, as disruptive technologies call into question the centrality of humanistic values to language. A December 2022 New York Times article by technology critic Zeynep Tufekci asks: “What Would Plato Say About ChatGPT?” The question, though tongue-in-cheek, is of particular urgency given the emergence of ChatGPT and other software applications destined to follow in its wake. This Orwellian-sounding technology, explains Tufekci, is “a conversational artificial intelligence program released recently by OpenAI.” ChatGPT is “a significant advancement that can produce articles in response to open-ended questions … comparable to good high school essays.” This is not some remote hypothetical technology. Nor is it comparable to familiar writing assistance apps, such as the ubiquitous cloud-based writing assistant Grammarly. Rather, ChatGPT is a free, open-source platform, readily accessible to even the most indolent, and generating original (read unplagiarized) content with only minimal prompting. You may not yet know ChatGPT, but to paraphrase the Irish poet W.B. Yeats, this beast has already slouched towards Bethlehem to be born.

Returning to Plato, who notoriously “mourned the invention of the alphabet,” we can only wonder with Tufekci how he might have deplored Artificial Intelligence. Like today’s AI, the alphabet of Plato’s day was a wildly disruptive technology, provoking both enthusiasm and dismay at its potential consequences. In Plato’s dialogue “Phaedrus,” Socrates assumed the voice of Thamus, the Egyptian king of the gods, rebuking the techno-optimism of Theuth, the Ibis-headed patron deity of the useful arts. Responding to Theuth’s claim that writing will make the Egyptians “wiser and give them better memories,” Thamus replied:

O most ingenious Theuth, the parent or inventor of an art is not always the best judge of the utility or inutility of his own inventions to the users of them. And in this instance, you who are the father of letters, from a paternal love of your own children have been led to attribute to them a quality which they cannot have; for this discovery of yours will create forgetfulness in the learners’ souls, because they will not use their memories; they will trust to the external written characters and not remember of themselves. The specific which you have discovered is an aid not to memory, but to reminiscence, and you give your disciples not truth, but only the semblance of truth; they will be hearers of many things and will have learned nothing; they will appear to be omniscient and will generally know nothing; they will be tiresome company, having the show of wisdom without the reality.

In light of the Golden Age that followed, we might easily dismiss Plato’s conservatism as a forerunner to Luddism. In its classical heyday (480-323 B.C.), the city-state of Athens produced a short-lived but seminal democracy, an admired inspiration for the United States and other future nations. The writings of Aeschylus, Aristophanes, Euripides, and Sophocles shaped the forms of modern drama. Fellow Athenians Herodotus and Thucydides invented history as we know it: the critical interrogation of the past, not just mere chronology. Plato contributed to this milieu not just as a philosopher, but as a teacher and institutional founder of higher education. Foremost among the students of his Academy, Aristotle first categorized logos, or reason, as one of the three forms of rhetoric, along with ethos (authority) and pathos (emotion). Ever since, Aristotle has been to thank or to blame for the most soaring Churchillian oratory on the one hand, to the most plodding high school essay on the other. Athenian progress emerged in a city with a peak population of fewer than 150,000 residents, empowered by the information technology of the Greek alphabet. In this light, Plato’s ingratitude resembles a Silicon Valley billionaire complaining that folks need to reconnect with their hunter-gather roots.

And yet, what if Plato was in some respects right? For all the wonders of literature and literacy, what if the written Word has forged forgetfulness in our souls? Plato is rarely given sufficient credit on this consideration. To be sure, writing gave us Milton, and Dickens, and Harry Potter. But oral culture gave us Gilgamesh, and Homer, and Beowulf. Writing also gave us headaches that Plato never had to contend with: online income tax submission, junk mail, and pop-up ads for pills to cure erectile dysfunction. In all seriousness, it is hard to deny we relinquished something vital by outsourcing the power of human memory to writing, a loss subsequently compounded by revolutions of print and digital technology. The fact that Plato, some two-and-a-half millennia ago, diagnosed this loss should give us pause. As stubbornly as we cling to our faith in technological progress, we have created technologies which roust us from the saddle of human control and concern. This sundering by technology explains Plato’s distrust of the upstart alphabet, which threatened to estrange his generation from “the wine-dark sea” of Homeric memory.

As Plato understood, those of us exiled from the resonant world of our ancestors risk becoming tiresome company. This tediousness of written culture likely began in Plato’s day. It only deepened with the advent of print culture and popular literacy, as novels and newspapers supplanted older texts, those more rooted in oral culture and discourse. But the deadening hand of the written Word has become face-plantingly obvious in today’s wordscape of tweets, emoticons, and inane memes. Thus, having the show of wisdom, the troglodytically-online lack the reality of the world beyond. Claiming the truth, they possess only its semblance; hearing many things, they learn nothing; they seem omniscient, but they are blinkered. And in their frustration at human limitations, they lash out like spoiled children. Witness the superficial manner in which partisans clash over the burning political issues of today—fought online as the planet burns—face mask mandates, critical race theory in schools, preferred gender pronouns, Brexit, and the War on Christmas among them. As Zeus remarked in Homer’s Odyssey: “Look you now, how ready mortals are to blame the gods. It is from us, they say, that evils come, but they even of themselves, through their own blind folly, have sorrows beyond that which is ordained.” How loudly these words must have rung in the mouth of Homeric bards, when gods walked the earth. How weakly they fade in comparison on the printed page or cell phone screen, as we nod in intellectual assent but lack the righteous anger these words should inspire.

Technology has not yet supplanted human memory entirely, but the technological erosion of the logos, the living Word that fulfills our human potential, seems ineluctable. Oral culture still flourishes in corners of the world, but it has been mostly uprooted in the West and is being vigorously suppressed in countries such as China. The ongoing cultural genocide by the Chinese Communist Party against the ethnically-Turkic, Muslim Uyghur majority in Xinjiang Province is a case in point. Over one million Uyghurs are miserably confined in concentration camps. Xinjiang’s remaining Muslim population is virtually incarcerated by a brutal surveillance state, which employs methods ranging from torture and rape to forced sterilization and the destruction of mosques in its assertion of a dominant racial, cultural, and ideological orthodoxy. One cornerstone of this destruction has been China’s suppression of the Quran. Originally preserved through recitation, along with the accompanying hadiths, or traditions and sayings of Mohammad, Islamic scripture is upheld as the inspired Word of God, revealed to the Prophet by the Archangel Gabriel. Unlike the emphatically literary Bible, organized around the written Word, the Quran is formally untranslatable. Even the Quran’s Arabic transcription serves primarily as a double-entry ledger for the oral tradition conveyed in religious madrassas, for which the Communist re-education camp now functions as a surrogate.

Closer to home, the stakes seem less drastic, but China’s cultural genocide in Xinjiang is a salutary warning against the potential demons of technology. Regardless of outward ideology, all technocratic societies converge towards the same ends, faststreaming the rational over the sublime, the efficient over the traditional, the utilitarian above the aesthetic. In the United States—as elsewhere—such impoverished thinking generally goes unchallenged, reducing public policy to a mere function of technological imperatives (in this regard, American politics is truly bipartisan). To give only one illustration, US Secretary of Education Miguel Cardona recently tweeted: “Every student should have access to an education that aligns with industry demands and evolves to meet the demands of tomorrow’s global workforce.” Should they, though? What does Cardona’s statement even mean? What is “tomorrow’s global workforce”? If it lies beyond the horizon, by what oracular magic are we to discern its demands? And if we conflate today’s “industry demands” and the “demands of tomorrow’s global workforce,” what kind of chimera will we be feeding?

Cardona’s tweet, though vague, reflects federal policy, specifically federal subsidy for state-level provision for Career and Technical Education (CTE). A new bottle for the old wine of vocational training, the CTE model originated with the Carl D. Perkins Vocational and Technical Education Act (1984), which tied secondary and post-secondary technical education funding directly to economic outcomes. The act has since been strengthened and reauthorized five times by administrations of both major parties, most recently under the Trump administration (2018) with overwhelming bipartisan support. While no federal equivalent of CTE exists for the humanities at the secondary and post-secondary level, federal funding for agencies such as the National Endowment for Humanities has declined significantly in recent years, reflecting broader cultural trends.

Few would deny that a technically proficient workforce is in the best interests of the American economy, but should skills-based, vocational training preclude the fruits of higher education that have enriched America’s middle-class over the past several generations? A plumber well-versed in Emily Dickinson and the Bhagavad Gita is not a worse plumber, and is certainly a better citizen, while an elected leader who has heard of neither has no cause for self-congratulation. Meanwhile, a 2022 analysis of over 51 million US job postings, conducted by Harvard Business School and the Burning Glass Institute, paints an ominous picture of the American jobs market, showing employers turning away from four-year degree requirements, while struggling to fill the “soft skills” gap traditionally plugged by humanities grads. While a shift towards basic skills and a growth in vocational “micro-credentialing” competes with more formal academic qualifications in job listings, contradictory trends in the jobs market suggest misalignment between “industry demands” and future realities. The nationwide study projects that some “1.4 million jobs could open to workers without college degrees over the next five years.” This projection likely masks the full crisis of higher education as AI supplants human labor, and obsolescent technological credentials fall by the wayside. Paradoxically, the oral and written communication skills taught by the humanities retain their value in the long-run, but as employers are discovering, competent writers and communicators are a vanishing breed thanks to the hollowing-out of the humanities.

Enter ChatGPT. As a history teacher, I dread that even a handful of my students might abuse this technology, feeding assignment prompts into the algorithm like ground pork into a sausage machine. The prospect of individual dishonesty troubles me less than the erosion of trust between teacher and student, given the stealthy footprint of AI. Leaving aside the thorny question of what constitutes a “good” essay, the advent of ChatGPT comes at a time when we need basic writing skills more than ever. ChatGPT’s mere existence is at once an aggravation and an inescapable reality, demanding frank discussion among academic administrators, faculty, and students. Only through such dialogue, ChatGPT might actually inspire a revival of pedagogy rather than its decline. Ironically, the solution may be less technology rather than more. Just as Plato eschewed gratuitous scribbling in his Academy, prizing oral debate over written exposition, a renewed emphasis on dialogue, discussion, and debate may be demanded of students, encouraging valuable communication skills. Written assignments should reward formal citation, source critique, and close reading over the glib generalization at which AI excels. Even the much-maligned blue book exam merits a second look. A generation of students unaccustomed to in-person, face-to-face interaction (or even active online discussion) may balk at such renovations. Some coaxing will be warranted, and more importantly, clear goals and expectations arrived at through classroom consensus.

In the long run, attempts to contain the disruption of AI in liberal higher education may prove a lost cause, yet resistance seems preferable to acquiescence. The perils of applications such as ChatGPT extend far beyond the classroom. Automated text composition threatens to corrupt every form of written expression from scholarly research, literature, and professional journalism (where algorithms are already supplanting humans) to business communication and technical writing. Nor is such generative AI limited to writing. The visual arts are revealing in this regard. The first prize in art at the 2022 Colorado State Fair went to an illustration of a cavernously fantastical palace yawning out to an alien dreamscape. Artist Jason Allen used the Midjourney AI platform to design his piece, entitled Théâtre D’opéra Spatial. He claimed to have broken no rules in submitting his entry, to the chagrin of fellow competitors. Though generic, the digital artwork is visually arresting in the style of Dune or similar science-fantasies. As technology marches on, AI will open new frontiers. How long before machines produces baroque cantatas in the manner of Bach, or soulful blues in the style of Nina Simone? To deny the possibility of either scenario implies the privileging of an essential human “soul” distinguishing humanity from artifice. To do so on purely technological grounds looks increasingly like willful thinking, as the trajectory of automation begun with assembly line robots in the 1960s threatens to consume artistic and intellectual endeavors still jealously imagined as human preserves.

Though the metaphor sounds alarmist, an unimaginable tsunami is barreling down on a complacent world. We may have time to adjust, who knows? On the optimistic side, Zeynep Tufekci compares ChatGPT “to previous technological advances, like mechanization,” which entailed public policy reforms and labor protections that allowed “more people to live with dignity as a basic right, even if their skills had been superseded.” She contends: “With so much more wealth generated now, we could unleash our imagination even more, expanding free time and better working conditions for more people.” Though sympathetic to Platonic contrarianism, Tufekci insists: “The way forward is not just to lament supplanted skills … but also to recognize that as more complex skills become essential, our society must equitably educate people to develop them. And then it always goes back to the basics. Value people as people, not just as bundles of skills.”

Valuing humanity for humanity’s sake, Tufekci concedes, “isn’t something ChatGPT can tell us how to do.” If humanistic society has any future at all, we must hope for teachers like Plato who understand the word as a means to personal wisdom—a logos rather than lexis. At least Tufekci recognizes that any return to basics begins in the classroom. Otherwise, we shall be reduced to asking ChatGPT what Plato would think about AI.

By way of follow-up, I asked ChatGPT precisely the question Tufekci asked in her title: “What would Plato say about ChatGPT?” Its response was telling:

Plato believed in the power of reason and the pursuit of knowledge, and he may have seen the development of ChatGPT and other AI systems as a way to explore and understand the world in new ways. He may have appreciated the potential of these technologies to enhance human understanding and improve our lives.

However, Plato also believed that the material world is not the true reality, but rather a shadow or imitation of a higher, eternal, and unchanging realm of Ideas. In this sense, he might have viewed ChatGPT and other AI systems as being limited by their reliance on material processes and unable to access the true nature of things. He may have cautioned against relying too heavily on them and losing sight of more fundamental truths.

The flesh-and-bones Plato would likely have scorned the opening part of this response. But surely, he might have agreed with its cautionary conclusion.

Image Credit: Winslow Homer, “To the Rescue” (1886).

8 COMMENTS

  1. Lots of good ideas here, Matthew. I think this part touched on a very deep issue:
    “What is “tomorrow’s global workforce”? If it lies beyond the horizon, by what oracular magic are we to discern its demands?’

    That’s the essential conundrum of education: how can elders predict what youngsters will need? We can’t. We can only pass along what we already know.

    In that light, the old model was well-rounded education which is why the standard HS curriculum during most of the 20th century included history, English, math, science, and foreign language, plus physical education (now deemed an insurance risk by administrators, a form of oppression by wokeists, and a waste of time by parents who want their kids to be good little robots who study all the time). The idea was that general education prepared students for many possible future paths.

    I wasn’t aware that Plato opposed the alphabet, but I had read years ago that the habit of writing erodes basic memory, not only ancient poetry but also shopping lists! On the other hand, it does enable transcendence of time, place, and culture by readers distanced via all three factors.

    “With so much more wealth generated now, we could unleash our imagination even more, expanding free time and better working conditions for more people.”

    This quote from Ms. Tufekci sounds like a rehash of an old techno-capitalist bromide used for decades to justify job loss via automation. Sure, new jobs will be created, but those 40 and over are much less likely to become qualified for them. I think even Dept of Labor has produced stats showing that “new jobs” are divided starkly into low skill and high skill, with fewer each year that lie in between.

    Further, the obsession with display screens on personal phones or as huge electronic ads, should give us pause regarding the idyllic promise of “unleash our imagination” that dates back at least a half-century (e.g., a 4-day work week).

    Finally, I feel compelled to point out a non-sequitur in your wonderfully wide-ranging overview: cultural genocide linked with “potential demons of technology”. China’s suppression of Islam, Uighur culture, and other abuses of human rights is not motivated by technology, any more than the manufacture of a kitchen knife is a “potential demon” enabling murder.

    I’m also unsure whether you intended the following sentence to suggest that Islam is anti-technology. (Leave aside the word “surrogate” that implies indoctrination by China is replacing indoctrination by madrasahs.)

    ‘Even the Quran’s Arabic transcription serves primarily as a double-entry ledger for the oral tradition conveyed in religious madrassas, for which the Communist re-education camp now functions as a surrogate.’

    Here in Indonesia, countless Muslims of varying degrees of orthodoxy use their cell phones to receive prayer-time alarms, zhikr, sermons, and inspirational music. They don’t see such technology as an obstacle — most see it as enhancement.

  2. Thanks Martin, for these thoughtful comments. Regarding China, I view the Uyghur genocide as embedded in technology (or technocracy) in a general sense, rather than motivated by specific technologies. But you raise some excellent points!

  3. Thanks for this reflection on yet another depressing disruption from Silicon Valley. I’m a history/literature teacher at a classical Christian school. During the Covid era, I grew dismayed at my students spending so much time in front of screens, so I almost exclusively assign handwritten assignments in the XL Moleskine notebooks that are organized for the Cornell Notes method now. We also do a weekly formal discussion that prizes oral discussion and debate. I teach 9th grade and they don’t do research papers in my class (they take a writing class as well), so I realize I’m not in the same position as many teachers with the capacity to do this and still meet the goals set for the course.

    I was talking with friends about it, and I think the best I can do to summarize the pit that forms in my stomach whenever I think about where this could lead is say that it reminds me of the kind of dehumanizing betrayal that we see in a sexbot. Berry, of course, already summarized the problem well a few decades ago in “Feminism, the Body, and the Machine”:

    “. . . The spiritual, physical, and economic costs of this ‘freedom’ are immense, and are characteristically belittled or ignored. The diseases of sexual irresponsibility are regarded as a technological problem and an affront to liberty. Industrial sex, characteristically, establishes its freeness and goodness by an industrial accounting, dutifully toting up numbers of ‘sexual partners,’ orgasms, and so on, with the inevitable industrial implication that the body is somehow a limit on the idea of sex, which will be a great deal more abundant as soon as it can be done by robots.

    This hatred of the body and of the body’s life in the natural world, always inherent in the technological revolution (and sometimes explicitly and vengefully so), is of concern to an artist because art, like sexual love, is of the body. Like sexual love, art is of the mind and spirit also, but it is made with the body and it appeals to the senses. To reduce or shortcut the intimacy of the body’s involvement in the making of a work of art (that is, of any artifice, anything made by art) inevitably risks reducing the work of art and the art itself. In addition to the reasons I gave previously, which I still believe are good reasons, I am not going to use a computer because I don’t want to diminish or distort my bodily involvement in my work. I don’t want to deny myself the pleasure of bodily involvement in my work, for that pleasure seems to me to be the sign of an indispensable integrity. . . .”

    • Matt,

      Really appreciate your comments here. I teach middle-school literature and composition online and in-person, and I’ve been moving more and more to assigning handwritten work. Of course, it’s been easier to assign handwritten work to my in-person students, but I would like to try requiring everything except final drafts to be handwritten for my online class. I’m also working on integrating more oral examination (esp. à la the Charlotte Mason vein of classical ed).

      More and more I think that the 7th-9th grade is a very defensible stage in which to require handwritten work and to practice thoughtful oral discussion (especially when I hear of the travails of friends who teach undergraduates!).

    • Matt, thank you for bringing up the physical aspect – the body. Quoting from Berry’s essay “Feminism, the Body, and the Machine” stirred my memory, and I will return to that work for more guidance on this issue of artificial intelligence.

  4. Thanks, Tessa. Totally agree. I think we’re going to need to do as much as we can to protect time for true face-to-face discussion or quiet, non-screen writing time in the classroom. If you can require purchase of the XL Moleskine notebooks, I think the format really helps make writing more rewarding for both students and the teachers that read their work. It’s one small challenge to throwaway habits, and they are simply more beautiful than looseleaf paper in a plastic binder or composition books that are too small.

  5. Thank you for this essay, Dr. Smith! Insightful, especially regarding Plato’s disdain for the alphabet and technology. Also, your distinguishing between logos and lexis put light upon a truth I had not considered until this essay. It’s tragically ironic that we tend to rebel against Divine authority while treating our fellow humans not as image bearers of the Divine, and yet we want our tools to have divine authority over us.

Comments are closed.

Exit mobile version