The Ancient Case Against Programming “Languages”
When Apple CEO Tim Cook addressed an audience of tech executives, venture capitalists, and policy makers at a “startup fest” event in Amsterdam, he took the opportunity to weigh in on the future of language instruction: “[coding] is just another language, and just like any other language it should be taught in schools.” Similar rhetoric can be heard in state legislatures across the country. The Florida senate passed Bill SB 468 last year allowing students to satisfy “language” requirements with coding classes, and similar measures have come up in Texas, Kentucky, and New Mexico.
Curricular decisions will always need to prioritize some subjects over others — there is a limited amount of instructional time available and choices need to be made. What is pernicious about this French vs. Python or Japanese vs. Ruby conversation is that it is based on a false equivalency hinging upon the slipperiness of a shared word: language.
Learning a second language and learning to code in fact play different roles in training a student to think critically. The first provides the basic building blocks of expression—that is, the construction and arrangement of words, sentences, paragraphs, and so on—while the second helps us to organize these raw materials sequentially as well as test premises and their consequences for validity. In antiquity, the former would be considered “grammar” and the latter “logic.” Far from equivalent skill sets and substitute goods in the educational economy, grammar and logic each contribute fundamentally to the development of a well-ordered mind.
Others have defended language instruction against incursions from the programming world. What I would like to add to the argument is the observation that we already have in the classical tradition a model for understanding how these two separate and different types of mental training can work together as foundations of a liberal arts education — the model of the trivium.
The trivium, as its name suggests (Lat. tri- = “three,” via = “way”), consists of three paths along which the mind can develop: grammar (the study of language), logic (the study of sound argument), and rhetoric (the study of persuasive argument). A supplementary course of study, the quadrivium, adds four subjects of increasing abstraction, namely arithmetic, geometry, astronomy, and music. The seven disciplines taken together are the traditional formulation of the liberal arts.
While the name trivium is likely a coinage of the 8th-century Carolingian Renaissance, this tripartite division of cognitive faculties has deep roots in the Greek intellectual tradition. The sophists of 5th-century Athens built their itinerant curricula around these three subjects. The Attic orator and schoolmaster Isocrates would assign rhetoric the central role in the formation of young Greek minds. Aristotle systematized both the fields of rhetoric and logic, with his Prior Analytics in particular leading some to call him the creator of modern computing.
The Roman polymath Varro and other giants of antiquarian philology like Cassiodorus and Isidore of Seville all had a hand in shaping the trivium. But the author who was most effective in raising its profile and securing its elementary status in the medieval curriculum was Martianus Capella. His Marriage of Mercury and Philology written in the 5th century CE would become the definitive guide to the liberal arts for centuries to come. The Marriage mashes together Olympian mythology, Neoplatonic allegory, and educational philosophy in an obscure half-prose, half-verse style that would lead its English translator, W. H. Stahl, to remark:
The reader is immediately at a loss to explain how a book so dull and difficult could have been one of the most popular books of Western Europe for nearly a thousand years.
Martianus writes about the wedding day of Mercury and Philology, a brilliant young woman (virgo docta) who rivals her messenger fiancé because she too, as the embodiment of human intellectual excellence, jetsets around the globe. The seven Liberal Arts, as Muse-like personifications, each toast the bride and groom. Grammar and Logic get the night underway by wishing the newlyweds well with a pair of the most boring wedding speeches imaginable. Grammar rattles off lengthy descriptions of how the Latin consonants sound, list after list of verb conjugations, and even a brief digression on irregular adverbs. Logic brandishes her collection of “knotty puzzles” (contorti nodi) and treats the crowd to a greatest-hits performance of Aristotelian syllogisms. While nuptial allegory often obscures the details of Martianus’ pedagogical principles, the larger point of his guest list is clear: each of the Liberal Arts contributes in its own way to Philology. As Grammar says on her own behalf in the Marriage: “My duty in the early stages was to read and write correctly; but now there is the added duty of understanding and criticizing knowledgeably.” Logic replies: “Not even Grammar can unfold her subject without using my reasoning.”
Where Martianus recognizes the space for both grammatical training and logical training in basic education, modern-day advocates of “language” substitution have chosen to conflate the two by taking convenient advantage of shared vocabulary. But in hindsight, the rise of the programming “language” in the first half of the 20th century is neither obvious nor inevitable.
The term has clear roots in the the formal languages of mathematically minded logicians from Leibniz to Frege. Yet, in the earliest stages of what we would now call computer science, these instructions were referred to by a matrix of words such as calculus, system, assembly, scheme, plan, formula, and, sure enough, code. “Machine language” appears early, but widespread adoption of the word would take time. Certainly, by 1959, the development of COBOL, or Common Business-oriented Language by the Committee on Data Systems Languages (note the plural) suggests that this was the default term. The exact process of its popularization is difficult to trace. Noam Chomsky’s algorithmic “descriptions of language” clearly exerted influence, but it may also have been spurred on by Grace Hopper’s introduction of English keywords and syntax to computer programming as she sought to replace math-heavy commands. Hopper’s instincts were correct and coding has moved increasingly towards human-readable “languages.” ALGOL, SNOBOL, SQL, Tcl, HTML, and perhaps Perl— they all hide the victory of “language” in their acronymic ells, and it is this victory that has given policymakers license to exploit semantic slippage for their own curricular ends.
So back to the trivium. Education theorists from antiquity on have understood that our facilities for critical thought are honed by a set of interdependent, but separate skills. Logic equips the mind with something that grammar does not. Yet grammar retains a presence in elementary and secondary education through English and Language Arts programs. (Of course, the direct instruction of English grammar is a debate of its own.)
Logic, on the other hand, is rarely taught as a subject in K-12. Thinking logically and “[building] a logical progression of statements to explore the truth of their conjectures” is outlined in the Common Core guidelines, but the national curricular standard offers little in the way of formal guidance on how to arrive at these outcomes. Instead, it is as if logic is so natural and instinctual that it requires no formal training at all. At the same time, one could argue that addition and subtraction are also instinctual, yet hundreds of instructional hours are devoted to basic arithmetic competence.
So logic becomes an assumption rather than a component of our ability to think critically. Yet, as Michael Shenefelt and Heidi White make clear in their recent social history of logic, If A, Then B, the mechanisms of logic cannot be assumed. They explain how the discipline has evolved — whether Socrates responding to the weaker argument in the Athenian assembly or Jeremy Bentham responding to fallacious arguments of an upper class threatened by the rise of commerce — to provide people with the tools necessary for maintaining orderly reasoning. Accordingly, logic deserves a place on the curriculum, and coding seems like a good place to start.
While this may appear to be simply splitting hairs over the word “language,” I would point out that the stakes over this distinction in educational policy are high. These policies, founded on the false equivalency that “(natural) language = (programming) language,” could result in reduced funding for secondary language programs and further chipping away at their already tenuous curricular footholds. Under this specious rhetoric of substitution, coding courses would be built on time, money, and students siphoned from traditional language programs. This is exactly against the spirit of the trivium. Grammar and logic are not mutually exclusive, but mutually beneficial.
Language courses can impart the best of what grammar, in its trivium sense, offers: an understanding of the building blocks of expression, an appreciation of this expression in others, and the pleasure of its greatest achievements in literature. Moreover, our political environment suggests that our students need secondary language instruction more than ever, not only for the utilitarian purposes of international commerce and diplomacy, but also for deepening empathy and, as Martha Nussbaum has written, fostering “cultural humility.” If coding classes sneak into the curriculum under the banner of “language,” grammar’s greatest rewards will suffer.
Moreover, as someone who has spent the better part of the last decade teaching Classics, I worry that Latin and Greek may be especially vulnerable to reforms that put language and coding courses in direct competition. The following statement, for example, is not uncharacteristic in defenses of learning Latin:
The study of Latin provides training in logical thinking, boosting cognitive processes essential for math, science, and engineering. Latin has been said to cultivate such mental processes as alertness, attention to detail, memory, logic, and critical reasoning.
The AP Latin guidelines list “use of coherent structure and logical development” as a primary learning objective for analyzing texts, but this goal could apply equally to any coding class. This is why I prefer Llewellyn Morgan’s take on Latin’s “logic” problem. Morgan once described the language as the “maths of the Humanities,” but qualified this by adding that it is also “maths with goddesses, gladiators and flying horses.” In defending classical languages against computer languages, we may well find it more advantageous to promote their rich literary traditions — again the domain of Grammar in the trivium — over their reputation for building critical thinking skills. Make Apollo conquering Python more of an asset than our students “conquering” Python3.
The trivium from Late Antiquity until the Renaissance formed the basis of a liberal arts education, that is the education that Seneca the Younger defended as a prelude to philosophy as opposed to an education based on “moneymaking skills” (meritoria artificia). In this system, logic stood as an equal partner to grammar and rhetoric. By following Seneca’s definition, proponents of language education would be wise to brand not only their own discipline as a liberal art, but also coding, as media theorist Douglas Rushkoff has suggested, and resist arguments that reduce programming and computer science to technical training and therefore utility. In so doing, we promote a culture of education where both secondary language instruction and computer programming work not in competition but rather in concert as the prelude to all future study. (That said, we should be careful not to assume that the teaching of the liberal arts was ever completely free from instrumentalism — for example, the reinvestment in the trivium during the Carolingian renaissance was tied to the training of preachers capable of elucidating Scriptural nuance.)
Priorities shift over time. Scholasticism witnessed the rise of dialectic, a debate-driven style of logic, while the studia humanitatis of the early Italian Renaissance — the seed of our modern-day “humanities” education — returned the emphasis to grammar. But the basic framework of the trivium has proven to be a useful abstraction for over a millennium and a half, and in this framework, grammar does one thing and logic another. This is why the debate over the substitution of second languages and computer languages is built upon a faulty premise. As Martianus’ textbook demonstrates, there is a place for both foundational training in the arrangement of words and in the arrangement of ideas. We need to preserve the grammar-fostering role of secondary language instruction in our schools. We also need to carve out a space for logic. Both are necessary for intellectual growth but at the end of the day code is not grammar. Rather code is applied logic for our generation — let’s call it what it is and return logic to its rightful place in the curriculum, one of three pillars of a 21st-century trivium.
Patrick J. Burns works as Digital Projects Associate at the Institute for the Study of the Ancient World in New York City and is contributor to the Classical Language Toolkit, an open-source platform for ancient world language research. He writes about Latin and digital philology on Twitter at @diyclassics.
