The Lost Tools of Learning: Sayers's Trivium and Her Stages

Dorothy L. Sayers read this paper at a vacation course in education at Oxford in 1947. Methuen printed it as a pamphlet in 1948. She calls her own experience of teaching "extremely limited". Her right to speak, she says, is common to everyone: "if we are not all professional teachers, we have all, at some time or other, been taught." The essay is short, sharp and openly speculative. Its diagnosis has worn better than its timetable.

The diagnosis

Sayers starts from symptoms she saw in adults. Writers "fail to define the terms they use". Debaters talk past each other, each taking a key word in a different sense. Mass literacy without training in argument leaves people exposed: "By teaching them all to read, we have left them at the mercy of the printed word." Her summary of schooling is the line in the hook. Pupils learn subjects, but "They learn everything, except the art of learning."

The tools she wants taught

Her remedy is the medieval syllabus. It had two parts. First came the Trivium: Grammar, Dialectic and Rhetoric. Then came the Quadrivium, which "consisted of 'subjects,' and need not for the moment concern us." The Trivium was "intended to teach the pupil the proper use of the tools of learning" before the pupil turned to subjects. Two of its parts are "only methods of dealing with subjects". Grammar is partly a subject, since it "does mean definitely learning a language".

She wants each tool taught in a concrete way:

• Grammar. "I will say at once, quite firmly, that the best grounding for education is the Latin grammar." With it go memorised facts, dates and verse.
• Dialectic. Formal logic is the "key-exercise". Pupils learn to "detect fallacy, slipshod reasoning, ambiguity, irrelevance and redundancy".
• Rhetoric. Pupils learn expression, and "each pupil should learn to do one, or two, subjects really well".

The aim comes in her last sentence: "the sole true end of education is simply this: to teach men how to learn for themselves."

The stages: a proposal, not a law

Sayers then matches the arts to ages: "Grammar to the Poll-parrot, Dialectic to the Pert, and Rhetoric to the Poetic age." In the Poll-parrot stage "learning by heart is easy and, on the whole, pleasurable; whereas reasoning is difficult". The Pert stage brings "contradicting, answering-back, liking to 'catch people out' (especially one's elders)". Of the Poetic stage she writes: "It is self-centred; it yearns to express itself". Her timetable gives "Grammar from about 9 to 11, and Dialectic from 12 to 14", with Rhetoric "at about 14".

It is tempting to read this as a law of child development. Her own text does not support that reading:

1. Her evidence is one child. "My views about child-psychology are, I admit, neither orthodox nor enlightened." She builds them by looking back on herself, "the child I know best and the only child I can pretend to know from inside".
2. She frames the plan as imagined. "Let us amuse ourselves by imagining that such progressive retrogression is possible." She expects no school to adopt it: "It is in the highest degree improbable that the reforms I propose will ever be carried into effect."
3. She lets the stages bend. The Pert stage "naturally, overlaps" the Poll-parrot one. The move to Dialectic comes "so soon as the pupil shows himself disposed to Pertness and interminable argument". A fixed law of age would not wait on the pupil.
4. Her numbers carry no source. Even a rudimentary knowledge of Latin, she says, "cuts down the labour and pains of learning almost any other subject by at least fifty per cent". The essay offers no study or measurement for the figure.

So the stages are a teacher's hypothesis: Provisional. They may hold roughly. Nothing in the essay tests them, and this page cites no study that does. The medieval schools she admired did not obviously separate the arts by age either. John of Salisbury describes Bernard of Chartres pointing out "grammatical figures, rhetorical embellishment, and sophistical quibbling" as he read the authors with his pupils (Metalogicon I.24). In that account, one reading carried all three arts. The Trivium as One Art: John of Salisbury and Sister Miriam Joseph follows that thread.

How it bears on the seven arts and the five acts

Sayers keeps the Trivium and sets the Quadrivium aside for the moment. This site keeps all seven arts at the centre. Its five acts begin with her three tools, renamed: understand exactly (grammar), examine the reasons (dialectic), express it truly (rhetoric). Two more follow: check it against evidence, and act and review. That five-part form is the site's own synthesis, not Sayers's and not the medieval schools'.

The bigger difference is timing. The site mixes the acts from the first level, two or three to each check, at a rising standard, rather than one art per age. The Levels path assigns her essay early in Foundations and returns to it at the end of that tier.

Each tool has a page here. Her dialectic leads to Proof and Precise Reasoning, on validity and counterexamples. Her rhetoric leads to English Expression Mastery. Her case for Latin can be weighed against Choosing a Language for Reading. Hugh of Saint Victor made an older arts-first argument; see Hugh's Didascalicon. And her fifty per cent is a claim waiting for a test: Hypotheses, Predictions and Tests shows how to write one.

Where this could be wrong

The strongest objection. Calling the stages "a proposal" undersells them. Young children do seem to memorise readily, and young teenagers do seem to love an argument. A rough match of art to age may be good practice even without proof.

The best reply. A proposal can be good practice. The risk lies in treating it as a law: withholding argument from a curious nine-year-old, or memory work from a fifteen-year-old, because a chart says so. Sayers's own text allows overlap and follows the pupil.

What would settle it. Teach matched groups of children with and without the age sequence. Test memory, reasoning and writing a few years later. No such study is cited here, so the stages stay Provisional. What Sayers says is Established from the 1948 text.

Try this

1. Outline the essay's argument as numbered steps. Mark each step as a claim of fact or a proposal.
2. Pick one text you are reading. Do one exercise per tool: parse its hardest sentence (grammar), name one ambiguity or fallacy in it (dialectic), and rewrite one paragraph for a named reader (rhetoric).
3. Take her "fifty per cent" claim about Latin. Write what result would support it, what would refute it, and how you would measure "the labour and pains of learning".

Further reading

• Dorothy L. Sayers, The Lost Tools of Learning (Methuen, 1948). It is short: one sitting.
• Sister Miriam Joseph, The Trivium: The Liberal Arts of Logic, Grammar, and Rhetoric (Paul Dry Books, 2002), a full textbook of the three arts.
• John of Salisbury, The Metalogicon, trans. Daniel D. McGarry (University of California Press, 1955), Book One.