Alright, so there are more than a few things going on here, so I’ll try to go kind of one thing at a time and then - eventually, I promise - we’ll get at why, since nobody liked it, nobody actually got around to making those changes.

Let’s start with the Nobody Liked It part. To be clear before we go too far, this is actually, if anything, something of an understatement. The Roman schoolroom was almost universally hated in antiquity, and criticizing or complaining about it was something of a literary topos of its own, especially because highly educated Romans (and Greeks) also often expressed a fair amount of contempt for elementary school teachers on the basis of the more basic nature of their curriculum, and these two things sometimes get lumped together. But what was actually disliked about it?

You’ve hit on one side of it (useless memorization, and a lot of it), but there was another aspect about the Roman schoolroom that is actually the much more frequent target of criticism in our written sources: the Roman classroom was also proverbially violent. Anecdotes scattered into the literary record from the second century BCE (some of the earliest literature we have in Latin) to the fourth century CE depict students, angry at their teacher’s mistreatment of them, getting angry enough to turn classroom objects, like tablets and styluses, into weapons to be used against their instructors or destroying the tools themselves. But the force of these anecdotes is in their reversal of status quo: normally, it was teachers who used violence against their students. Hence Juvenal, in a critique not of the classroom’s violence but of the poet population of Rome’s boring, predictable, and overly pedantic choice of material, notes that “I, too, have snatched my hand out from under the cane” (Juv. 1.15), a shorthand for ‘I have gone through the same education the rest of you have’. Juvenal doesn’t mention whose cane he might have been trying to avoid, and he doesn’t have to. A Roman reader will already know that he’s talking about the classroom.

To add to the ever-present threat of physical violence, yes, the practice of Roman education also involved a lot of memorization and was incredibly (and boringly) repetitive, something which I’m sure anyone who has ever worked with children can confirm is not going to be particularly popular with anyone. This was true at all levels of Roman education, but I’m going to focus on elementary education here, because that level shows all the same issues surrounding the intersection between theory, practice, and historical precedent that played out at all levels of Roman education and for various reasons there’s just more to go on.

In Rome, it was thought that the best way to learn to write was to learn first individual letters, then syllable combinations - and I mean all possible syllable combinations - from the most basic to the complex in order to build up to words and sentences. This is the recommended approach from the theoretical side, mostly Quintilian, but we also see that people did actually follow it in real life from extant school exercises. Note, there isn’t total accord between theory and practice here: we have exercises written on ostraka of shaky beginners’ hands writing out their names before they go through all the steps Quintilian outlines first. This shouldn’t surprise us too much, as being able to write one’s name is an important skill to have, especially for students who might not have stayed in schools long enough to go through all of the theorists’ recommended steps. And while this is a very specific example, I mention it to highlight that when our written sources talk about education, they often do so from a more theory-informed perspective. What we see when we have actual extant sources to go from sometimes confirms what they’re telling us, but sometimes it doesn’t, and while beginner sources highlight this distinction more visibly, we cannot assume that what happened in the classroom always looked like Quintilian and others tell us that it ought to go.

But in general, when it came to learning to write, students learned letters then syllables then words, going through the repetitive and not particularly interesting process of learning letter combinations they might not even ever need to use. So: if people hated this sort of highly memorization-dependent model of pedagogy so much, why do it?

The obvious answer would be that it’s because this is what pedagogical writers said was the best method for learning, but if we blame it all on the theorists we’re going to end up in a bit of a recursive argument. In actuality, the learn-by-syllables approach has a long history in Italy that serves as the likely model for the Roman adoption of it in their own schools.

In Etruria, a number of abecedaria and other writing exercises have been found in burial contexts, including one that dates from the middle of the seventh century BCE - either before or more or less contemporaneous with the Praeneste Fibula, which represents the earliest extant writing we have in Latin. Venetic exercises clearly modeled on Etruscan rote-memorization alphabet exercises have also been found that date to the early republican period in Rome, showing that the Etruscan pedagogical model spread elsewhere in Italy. Since the Etruscan alphabet also formed the model for a number of other alphabets that developed in archaic Italy, including Venetic and Latin, we shouldn’t find this particularly surprising. Roman abecedaria along this model, then, likely represent an adoption of an existing method of elementary education in Etruria - the dominant cultural power in Rome’s immediate sphere and on its border in the archaic period. To return briefly to the Praeneste Fibula, the Latin found in the inscription there shows a number of influences from Etruscan writing practice (among other things; for the linguists there are a lot of fun controversies in the Praeneste inscription) on early Latin, a number of which disappear relatively quickly in the early epigraphic record but which preserve an important early stage in written Latin all the same.

And if we think of this rote learning model in the context of Etruscan writing practices, it starts to make a bit more sense. Etruscan writing used punctuation to mark out irregular (i.e. not consonant-vowel) syllables, a system that was also adopted by other Italic cultures following the Etruscan model (note that one of the Praeneste Fibula’s Etruscan debts is its punctuation, which includes, in addition to word division also a three point divider, albeit one used somewhat creatively according to a recent proposal, which has been found mainly in Etruria and the surrounding areas, notably at Caere and Veii, both in southern Etruria and close to Rome). Learning in syllables, then, makes a lot of sense.

Note also that the Romans eventually adopted a system of writing we call scriptio continua, “continuous script”, from the Greeks. They didn’t mark punctuation, and they didn’t use breaks between words. In the same way that marking punctuation, as the Etruscans did, requires attention to syllable structure, so too does the total not-marking of syllables. One way we can sometimes even tell apart early educational texts from others is the presence of breaks between some of the words. Learning how to read continuous text with no punctuation or breaks was a skill that had to be taught and learned, which means you need to know your letter groupings without any kind of marking.

And while all these things do help rote learning by syllable grouping make a little more sense, at both the elementary and higher levels of education we can’t underestimate the inherited tradition. Once this approach was engrained within Roman educational practice, people used it. Etruscan abecedaria show that this system was present in Italy for at least as long as the earliest Romans were writing Latin at all. It’s the system Roman teachers learned with themselves, and it’s the one they used with their own students.

I know this has focused more on rote elementary language learning than the memorization that happened in rhetorical education, but all of the same principles apply up as well. Rote memorization was set as the way you learned things, was written about as the best method, so you continued doing it throughout your education, and, if you became a teacher at any level, this was the method you used with your own students.