The Trouble with "The Lost Tools of Learning."

“That I, whose experience of teaching is extremely limited, and whose life of
recent years has been almost wholly out of touch with educational circles,
should presume to discuss education is a matter, surely, that calls for no
apology. It is a kind of behaviour to which the present climate of opinion is
wholly favourable.” Dorothy Sayers in “The Lost Tools of Learning.”

I would like, first, to preface this by saying that I have nothing but respect for Dorothy Sayers as an intellect. She was a truly exceptional woman and scholar and I have benefitted enormously from her work. However, I am going to explore in this post some of the issues presented by “The Lost Tools of Learning,” though it is not easy to describe these issues because they are problems stemming from her idea, and then our misinterpretation of it.

Since the 1990s the principles espoused by “The Lost Tools of Learning” have been put into effect by hundreds of schools and co-ops and programs, and written about in dozens of books. But it has not come without its own issues.

As most educational theorists do, Sayers lays out all the woes in the sphere of education as they were perceived at the time. She especially highlights the problem of high ‘susceptibility’ to propaganda. While everyone can read, in her view, very few people seem equipped to think through their consumption of propaganda. In other words, people are literate but not learned. They can read words on a page, but they cannot think. And this occurs in spite of a profusion of subjects that students must study in school. Miss Sayers’ solution to this is a return to a what she presents as a medieval formulation: the Trivium. She says:

“I want to inquire whether, amid all the multitudinous subjects which figure in the syllabuses, we are really teaching the right things in the right way; and whether, by teaching fewer things, differently, we might not succeed in “shedding the load” (as the fashionable phrase goes) and, at the same time, producing a better result. This prospect need arouse neither hope nor alarm. It is in the highest degree improbable that the reforms I propose will ever be carried into effect (2-3).”

She then goes on to explain how the trivium is a ‘method of dealing with subjects,’ and correlates its individual parts to the ‘ages and stages’ in child development. These stages are, rather, identified via stereotype and observation and certainly do have a basis in observable reality.

‘Now it seems to me that the lay-out of the Trivium adapts itself with a singular appropriateness to these three ages: Grammar to the Poll-parrot, Dialectic to the Pert, and Rhetoric to the Poetic Age.’

She then goes on to argue that Latin should be the ‘grammar’ that one learns in the poll-parrot stage, because it

‘…cuts down’ on the difficulty of logicking through English grammar by ‘fifty percent.’ Notably, she also goes on later in the essay to speak of the ‘grammar of Mathematics, beginning with the multiplication table….’ (13).

Thinking of “grammar” as merely a stage or even a way to think about the foundation stones of a topic isn’t inherently bad, but it has been misunderstood. And I think she misses that there is an underlying principle of the “grammar” part of the Trivium that goes beyond just the learning of Latin (more below).

The bigger problem is that her presentation of the Trivium has been built into a mechanistic explanation of human learning which has become the basis for much of what calls itself “classical” education in the United States. The Trivium, historically understood, is about language as the first basis of education. Historically, it has very little to do with our ideas about the mechanisms of human learning.

In the following quote from “The Lost Tools,” we see some of these problems concentrated:

Now the first thing we notice is that two at any rate of these “subjects” are not what we should call “subjects” at all: they are only methods of dealing with subjects. Grammar indeed is a “subject” in the sense that it does mean definitely learning a language—at that period it meant learning Latin. But language itself is simply the medium in which thought is expressed. The whole of the Trivium was in fact intended to teach the pupil the proper use of the tools of learning, before he began to apply them to “subjects” at all. First, he learned a language: not just how to order a meal in a foreign language, but the structure of language—a language–and hence of language itself—what it was, how it was put together and how it worked. (pg. 7)

Some of this rings true, some of it does not. “Grammar” “dialectic” and “rhetoric” absolutely were subjects to be studied as a part of language; the three together make the foundational subjects of all learning. Historical authors wrote about grammar, logic, and rhetoric as subjects to be learned. We cannot learn about arithmetic or music without first having a grasp of good language, whatever that language is.

She is right that one must learn the structure (i.e. the logical structure) of the language, but approaching the Trivium as mere “tools” is incorrect compared to the historical presentation of it (see below). It is not merely a methodology, and we are misled in our application of it.

So the problem is twofold: she presents a misleading picture of the Trivium, and then we in turn have confused the tool for actual learning (and the motivation to learn).

And this is the essential trouble with this essay: not everything she says is wrong, but much of it can be easily misconstrued and misapplied. Before I explain more on the Trivium specifically, I did find a remark in one of her letters which can help us understand what was really happening here.

Sayers’ letters offer some interesting clues as to her real thoughts about education, and what she may have actually intended in The Lost Tools. Perhaps one of the most surprising statements is here in a letter in 1940-41 (several years before “The Lost Tools of Learning” was delivered):

‘You are bothered about education. So, in a sense am I – my trouble is that I don’t really agree with any of my wise contemporaries. I find it difficult to say that I gained very much from it myself, and I cherish a heretical notion that it can do little for one except (a) implant a desire for knowledge and (b) to give one a rough machinery for finding out for one’s self the things one wants to know…I learned a little at home – except, indeed, how to speak French; nothing at school – except, indeed, how to pass examinations and at Oxford the only things I think I can say I was taught were the rudiments of Old French grammar (which you wouldn’t think very useful) and a general attitude to knowledge which I can only call a respect for intellectual integrity. And between you and me, I doubt greatly whether a liberal education can impart much more than those two things: a reverence for the working mind, and a knowledge of method applied to some subject, no matter what. But if I said this in public, all the advanced people would be shocked to the marrow.’ (Sayers, Letters: 223).

This passage potentially explains much. Formal education is very limited in what it can do, according to her. She goes on in this letter to express some concern for the twisting of the facts that were occurring in the Nazi youth programs in Germany at this time, and how this twisting of facts meant an unmooring from any factual basis for history. However, if this is her real attitude about education, and these are the very firm limits she places on formal education, then I think “The Lost Tools of Learning” essay is better contextualized for us.

In her letter, she refers to a ‘machinery’ of learning and then later a ‘method’ of learning, which introduces a very modern and mechanical vocabulary to the discussion of education which I think runs completely counter to the original idea of what the “liberal arts” are about. But at the same time, in her view, formal education is meant to give a student some idea of how to learn something that they are already interested in learning. She sees formal education not as imparting actual knowledge but merely imparting the ‘tools’ by which someone might learn something. That is where we misunderstand her. The Trivium itself doesn’t impart knowledge, in her view, it is merely a way of approaching knowledge that one is already interested in (and she concludes the whole speech by saying this).

In turning “The Lost Tools of Learning” into a 3-step method that one must undertake in order to learn school subjects, 21st century educators have missed the forest for the trees. In her view expressed here in the letter quoted above, formal education does not impart knowledge per se but merely tries (imperfectly) to give students the means by which to acquire knowledge that they already desire. At best it imparts a desire to learn, and a general sense of how to learn something.

She herself says it: “…and the Trivium is by its nature not learning, but a preparation for learning.” Now, I agree that it is not learning by its nature. But I disagree that it is merely a “tool” or a mechanism. It is part of an imaginative picture of the totality of human learning and the relationship of the branches of learning to one another. Reducing it merely to a methodology robs it of its richness.

Can understanding the Trivium as a “3 step method” to approach new knowledge be helpful? Sure. But let’s not pretend that that’s the totality of what the Medievals were doing with it.

Another aspect of Sayers’ thought here is the concern with rising subjectivism and the unmooring of rational thought from feeling, leading to a situation not unlike our present one where some aspects of reality are defined by one’s feelings rather than the plain facts. C.S. Lewis foreshadows this in The Abolition of Man as well.

For this reason, Sayers appears concerned with ‘re-forming’ the Trivium into something that gives students a sense of what is ‘factual’ and will enable young students to avoid the pitfalls of subjectivism (especially those which are introduced into society via propaganda and mass media). But the problem remains that if a person does not wish to learn, there is only so much we can do to arm them against subjectivism.

If she is indeed correct that formal education does not impart knowledge but at best a love of learning and the ‘tools’ by which a person can self-educate, then our entire understanding and application of “The Lost Tools of Learning” is wrong.

The brunt of my criticism, rather, lies more with her followers who have reified what she intended as a thought experiment into something it was never intended to be. We think that if a student follows the three-step method, they will have gained knowledge and become educated. That is not what she is saying.

At best, a student may become self-motivated to learn on their own, and then have a sense of ‘methodology’ to use when they want to put their hand to a new subject.

But when we leave it merely with that, we have instead gone down an entirely different path that I do not think Sayers herself would have endorsed. As Dante said once, ‘ché la diritta via era smarrita’ (La Commedia Divina 1.3). ‘The straight path has been lost,’ as often happens in important debates that have tremendous societal importance.

The modern understanding of the Trivium as a rather mechanical technique has persisted now for the better part of a century after Sayer’s original publication of her essay, receiving another revitalization with the 1989 publication of Doug Wilson’s Recovering the Lost Tools of Learning, which has been taken as a guide for most of the classical education movement that has grown in the last 20 years. There has been a steady stream of books published on the topic which assume this vision of the Trivium as a mere tool or technique, and further divorce us from the imaginative vision we ought to have.

In her book The Well-Trained Mind, Susan Wise Bauer summarizes the Trivium thus:

This is the classical pattern of the trivium, the three-part process of training the mind. The first years of schooling are called the ‘grammar stage’ – not because you spend four years doing English but because these are the years in which the building blocks for all other learning are laid, just as grammar is the foundation for language…By fifth grade, a child’s mind begins to think more analytically…the second phase of the classical education, the ‘logic stage,’ is a time when the child begins to pay attention to cause and effect, to the relationships among different fields of knowledge, to the way facts fit together into a logical framework…The final phase of a classical education, ‘the rhetoric stage,’ builds on the first two. (Bauer, Mind: 43).

Glass writes in her book Consider This: Charlotte Mason and the Classical Tradition:

Modern educators, looking for something better than our woefully inadequate progressive practices, have built her [Sayers’] non-traditional ideas into an entire educational movement commonly called ‘neoclassical,’ which distinguishes the newer idea of stages from traditional, historical classical education. Unfortunately this view places the burden of rote memorization, often without understanding, upon young learners who would be better served by a relational, synthetic approach to knowledge. (92).

In the preceding pages of Consider This, Glass outlines the issue at stake which is what I wish to emphasize in this article: the central impulse of a truly “liberal arts” education is that it seeks unity of knowledge and uses the imagination to emphasize that unity. Glass lays out a paradigm between ‘analytical’ knowledge vs. ‘synthetic’ knowledge which tends to be missing in typical presentations of the Trivium.

“Analytical knowledge” breaks everything down into its component parts and views them as singularities, and not as part of a larger whole. “Synthetic knowledge” seeks to view things as a whole and part of a greater unity. This ideal of unity and wholeness is the thread that ought to connect us to our educational past.

So, then what?

The first issue, in my mind, is unpacking what is historically meant by the “Trivium.” Again, I would have fewer problems with the way we understand the Trivium now if we weren’t claiming to return to some Medieval, Renaissance (or even Ancient) paradigm. We aren’t, by and large. The way most classical education providers interpret the Trivium is to understand it as a mechanistic process of input vs. output with Christian morality sprinkled in. At best, it doesn’t produce miraculous results in the students but mediocre ones; at worst, it becomes rank legalism that puts terrible pressure on the students to perform and conform.

This has been damaging to the movement because it is simply not true to what a “liberal arts” education was actually meant to be.

As for the precise formulation of the “liberal arts” as Trivium and Quadrivium and its appearance in the historical record, Henri-Irenee Marrou in his landmark book A History of Education in Antiquity asserts that the seven liberal arts as such were finally formulated somewhere in the 1st century BC. Somewhere between the time of the grammarian Dionysius Thrax and the eminent Roman polymath Varro, the subjects of the seven liberal arts, broken up into the Trivium and Quadrivium, were solidified.

This formulation was taken over more or less entirely by the Carolingians. Marrou writes at the end of the book:

'...the material components of the classical tradition were enabled to survive the destruction of the classical Form and be used again in a new synthesis that appeared as mediaeval culture, with its Christian inspiration. This first began to manifest as its own distinctive form in the Carolingian Renaissance…however original its first inspiration may have been, however much opposed to the spirit of the old humanism, it was not fundamentally different from it. It was not an absolutely new beginning, a fresh start. From the beginning and throughout its subsequent history it drew so much from the old classical sources that it seems to appear as a continual Renaissance. Despite the barbarian interlude there was a certain continuity of matter, if not of form, that made Western man the heir to the old Classicism.'

What we recognize now as the “liberal arts” with its terminology of “Trivium” and “Quadrivium” is a Medieval formulation drawn from a classical well and is continually renewed depending on the priorities of the age.

The idea of the ‘Trivium’ is that it is a ‘meeting of three roads.’ One might envision a Y shape. The ‘Quadrivium,’ on which I will be focusing less, is a ‘meeting of four roads.’ For this, one may imagine an X to represent the idea. The subjects that make up the Trivium and Quadrivium are conceived as a unity, and once that unity is apprehended, we find ourselves at the very heart of what they represent.

The Trivium is the language basis required for all subsequent learning. The traditional subjects of the Trivium are: grammar, dialectic (logic), and rhetoric, while the traditional subjects of the Quadrivium are: arithmetic, geometry, astronomy, and music.

The idea behind all of these seven liberal arts is not that a student will know everything that can be known, but that they will have a more-or-less all-encompassing understanding of the underlying principles of how the world works in abstract terms. It is an imaginative picture of well-rounded knowledge.

Now, we know that some aspects of the Quadrivium especially are outmoded. No one thinks that the planets are producing harmonized tones anymore (or at least, not in the same way) and our understanding of the physical world has advanced beyond what the Medievals could have imagined. However, I am not saying we need to return to any particular aspect of medieval knowledge but to the unchanging principle that underlies the ideas, which can be adapted to the knowledge we have acquired since the advent of modern science. These seven subjects were the subjects that the Ancients and Medievals believed represented all the important abstract knowledge about the world and human life, but the underlying principle is that we ought to continue trying to seek those areas of knowledge that best promulgate a humane intellectual life.

Sister Mary Joseph in her book The Trivium: The liberal arts of logic, grammar, and rhetoric writes this:

The utilitarian or servile arts enable one to be a servant – of another person, of the state, of a corporation, or of a business – and to earn a living. The liberal arts, in contrast, teach one how to live; they train the faculties and bring them to perfection; they enable a person to rise above his material environment to live an intellectual, a rational, and therefore a free life in gaining truth (5).

This sort of humane intellect is very conscious of what the late antique scholar Cassiodorus calls the ‘divine structure,’ which are best apprehended by means of the liberal arts in their totality. Even despite our advances in science, we still know that the universe is anything but random and unstructured. Of the seven liberal arts, the Trivium subjects remain untouched in their essence, as they are not ‘sciences’ in quite the same respect as the subjects of the Quadrivium which have undergone more of an evolution over the years. The language basis for learning remains unchanged. We have changed in our attitude towards it, but perhaps not for the better.

The medieval ‘liberal arts’ education is a series of subjects that will enable a person to live the ‘contemplative’ life that is praised by Plato, Aristotle, Cicero, Quintilian, Plutarch and many others throughout the centuries. Language and literature are at the very heart of the contemplative life, even if one does not believe in outmoded astronomical models anymore.

But to narrow the vision again, I will discuss “grammar” as part of the Trivium, particularly as it pertains to the understanding of language itself as the focus and the root of all education. Now, I am highlighting grammar here but I have a post elsewhere that discusses how the three members of the Trivium are always working in tandem whenever language is at work. It is impossible to engage in language without the right combination of grammar, logic, and rhetoric.

The Canadian literary critic Northrop Frye says in his essay “The Double Mirror,” ‘What words do with the greatest power and accuracy is hang together’ (Frye, Mirror 36). Words create verbal (logical) structures, and these structures generate meaning. This is why ‘grammar’ is about more than merely morphology or the forms of individual words; it has even more to do with how words ‘hang together.’ And another question is: do these words hang together in particularly impactful ways (rhetoric)?

The most powerful verbal structures that exist are found in literature, and students were put into early contact with those structures as their first education. In the secular world, the literary models students would study were the great poets and prose-writers of antiquity. In the Christian world, perhaps best represented by Cassiodorus, Christian learning involved studying the Biblical canon and secular works together. ‘Grammar’ as such revolved around learning language through exemplary models (Institutes Book 1).

Quintilian in Book 1 of his Institutio Oratoria advises that once a boy learns to read and write easily, his first task is to learn literature. In fact, the precise terminology used to refer to his study of literature is “poetarum enarrationem,” or the “narration of the poets.” In the course of reading widely, a student will learn the conventions of his language and the elements of style, beginning with the reading of Homer and Vergil, who were considered the cornerstones of all literature. It is worth bearing in mind that to the ancient mind, poetry was the pinnacle of the literary experience. Quintilian also makes it clear that other language-related subjects are studied simultaneously such as what we consider “grammar” today and the correct pronunciation “...recte loquendi scientiam.”

While Quintilian spends the majority of Book 1 of the Institutio on elements of proper grammar and diction, he also says in 1.10.2 that all the aspects of a literary education will take place concurrently: while a student is learning his grammar and literature, he is also learning about the elements of oratory.

Most interestingly, Quintilian states that the subjects that we would recognize as the ‘liberal arts’ are insignificant in and of themselves and yet together help to create the fully educated person (Quintilian 1.10.6).1

By contrast we see in Cicero’s de Oratore 1.187:

“...in grammaticis poetarum pretractio, historiarum cognitio, verborum interpretatio, pronuntiandi quidam sonus.”

“...in the study of grammar, the thorough knowledge (pertractatio) of the poets, the knowledge of history, interpretation of words, and a certain sound/manner of pronunciation.”

A knowledge of “poets” and “historians” is considered the pinnacle of one’s “grammatical” education, wherein you learn your language through studying the best examples of it.2

For an even fuller treatment, Robert A. Kaster’s book Guardians of Language: The Grammarian and Society in Late Antiquity also affirms this view of antique education as a primarily literary pursuit that was called “grammar.” The “grammarian” was the one responsible for teaching young students their command over the canonical classical texts, and once that mastery of those books had been achieved, they could then be expected to convert that knowledge into skillful speech and composition (dialectic and rhetoric which is learned by habituation in “grammar” subject).

Sayers does get this right:

In English, verse and prose can be learned by heart, and the pupil’s memory should be stored with stories of every kind—classical myth, European legend, and so forth. I do not think that the Classical stories and masterpieces of ancient literature should be made the vile bodies on which to practise the technics of Grammar—that was a fault of mediæval education which we need not perpetuate. The stories can be enjoyed and remembered in English, and related to their origin at a subsequent stage. Recitation aloud should be practiced—individually or in chorus; for we must not forget that we are laying the groundwork for Disputation and Rhetoric. (12)

This is much closer to the historical understanding of what was historically meant by “grammar,” but Sayers belies it a little bit by saying that students must learn Latin. I think the principle can be applied more broadly to one’s own language (e.g. English). In addition to that, students are simultaneously learning the ways of logic and rhetoric simply by inhabiting a world of language and stories.

What is clear is that this particularly literary formulation of “grammar” emerges from classical antiquity, but it will be fossilized in a little-known work that I will discuss in a second post: the late antique author Martianus Capella and his work The Marriage of Mercury and Philology or The Seven Liberal Arts. Martianus Capella’s work was the textbook on education in the Middle Ages.

But returning to the topic at hand after much digression, Sayers presents us with a very thin and modern idea of what the Trivium could be, and it is not particularly historically-grounded but it is based on child psychology which is not inherently invalid but it is not really getting at the core of what the Trivium is.

The bottom line is this: Sayers herself was not saying that the Medieval Trivium was itself the producer of knowledge but merely a means (hence tools) for knowledge that one already wants to acquire. I do not believe this explanation of it is sufficient to the task, but we need to understand what she did and did not mean by it.

And I will end this post with an apt quotation from Charlotte Mason. “There is no education but self-education.”

We can use the modern Trivium as a modern set of tools (with all the subtlety of a pneumatic drill), but we cannot use it to make students desire to learn.

And on this point, Sayers agrees with Mason at the conclusion of “Lost Tools:”

“For the sole true end of education is simply this: to teach men how to learn for themselves; and whatever instruction fails to do this is effort spent in vain.”

In Part 2, I will look more closely at Martianus Capella himself, whom Sayers does not directly name, but I suspect he was at least one of the names she had in mind when she alludes to “Medievals.”

Bennett, Beth S. ‘The Rhetoric of Martianus Capella and Anselm de Besate in the Tradition of Menippean Satire.’ Philosophy & Rhetoric, vol. 24, no. 2, 1991, pp. 128–42. JSTOR.

Bauer, Susan Wise & Jessie Wise. The Well-Trained Mind. W.W. Norton, 1999.

Cassiodorus, Institutions of Divine and Secular Learning & On the Soul. trans. James W. Halporn. Liverpool University Press: 2004.

Chang, Han-Liang. ‘The Rise of Semiotics and the Liberal Arts: Reading Martianus Capella’s ‘The Marriage of Philology and Mercury.’’ Mnemosyne, vol. 51, no. 5, 1998, pp. 538–53. JSTOR,

Colson, F. H. “The Grammatical Chapters in Quintilian I. 4-8.” The Classical Quarterly, vol. 8, no. 1, 1914, pp. 33–47. JSTOR, http://www.jstor.org/stable/635705. Accessed 19 July 2024.

Curtius, Ernst Robert. European Literature and the Latin Middle Ages trans. Willam Trask. Harper and Row: 1963.

Desmond, Marilynn. ‘Gender and Desire in Medieval French Translations of Ovid’s Amatory Works’ in Ovid and the Middle Ages eds. Frank T. Coulson, James G. Clark, Kathryn L. McKinley, Cambridge UP: 2011.

Frye, Northrop. Anatomy of Criticism. Princeton UP: 1973.

–. ‘The Double Mirror.’ Bulletin of the American Academy of Arts and Sciences, vol. 35, no. 3, 1981, pp. 32–41. JSTOR.

Jaeger, Werner. Paideia: The Ideals of Greek Culture Vol 1. Trans. Gilbert Highet. Oxford UP: 1939.

Jeauneau, Edouard. Lectio Philosophorum: Recherches sur L’Ecole de Chartres. PDF on Scribd.

Joseph, Sister Miriam. The Trivium: The liberal arts of logic, grammar, and rhetoric. Paul Dry Books, 2002: reprint.

Kaster, Robert A. Guardians of Language: The Grammarian and Society in Late Antiquity. U of California P, 1988.

Kristeller, Paul Oskar. Renaissance Thought and Its Sources. New York: Columbia University Press: 1979

Lewis, C.S. The Allegory of Love. Oxford UP: 1936.

– Abolition of Man. Oxford UP: 1944.

– The Discarded Image. Cambridge UP: 1964.

Marrou, Henri-Irenee. A History of Education in Antiquity trans. George Lamb. Sheed and Ward: 1956.

Mason, Charlotte Maria. Parents and Children. United Kingdom, Kegan Paul, Trench, Trubner, & Company, 1904.

– School Education. United Kingdom, Kegan Paul, Trench, Trubner, & Company, 1905.

Relihan, Joel C. ‘Martianus Capella, the Good Teacher.’ Pacific Coast Philology, vol. 22, no. 1/2, 1987, pp. 59–70. JSTOR, https://doi.org/10.2307/1316659.

Sayers, Dorothy L. ‘The Lost Tools of Learning’. Essay presented at Oxford University: 1947.

Stahl, William Harris, Richard Johnson, with E.L. Burge, Martianus Capella and the Seven Liberal Arts vol. 1 & 2. Records of Western Civilization Series, Columbia UP: 1991.