Learning to read before spelling became standardized: 1. Quintilian & Coote

In the 16th and 17th Centuries, English spelling had yet to become completely standardized, but it was not a free-for-all. To take one example, the word ‘unless’ might have been spelled unless, onless, unles, onles, unlesse, onlesse, onnles….etc. There are options, but they are bounded ones.

The same book might well have three or four different spellings of even the most common words. This meant that readers must have taken a sound-based approach to reading, because the whole word could differ so much and so often. An outline of a word, as posited by whole-word methods in the past few decades, would be useless; and even learning a whole word would be unhelpful. The ability to break the word down, to understand the different spelling options for individual sounds, is crucial in recognizing unles and onelesse as the same word.

So, how did they learn?

It seems that the approach was based on a very ancient set of practices indeed, tracing its origins in Ancient Greek and Roman elementary education, with the traditions continued into and through the Middle Ages by means of the Church.

Although there will have been all sorts of differences in different places at different times, there is one process which is common to both the Roman elementary practice and the teaching of English reading prior to spelling standardization: the progression from learning letter-sound correspondences to recognizing and reading syllables, to putting those together into words.

Quintilian, writing in the 1st Century AD, is aware of many of the pitfalls of early literacy. In The Orator’s Education, he says:

‘At any rate, I do not like the procedure (which I see is very common) by which children learn the names and sequence of the letters before their shapes. This is an obstacle to the recognition of the letters, since they do not when the time comes pay attention to the actual outlines, because they follow the promptings of their memory, which runs ahead of their observation. This is why teachers, even when they think they have sufficiently fixed the letters in a child’s mind in the order in which they are commonly first written, next reverse this, or muddle it up in various ways, until the pupils come to recognize the letters by their shape and not by the order in which they come. It will be best therefore for them to be taught the appearance and the name side by side: it is like recognizing people.

But what is an obstacle in learning letters will do no harm when we come to syllables. Nor do I rule out the well-known practice of giving ivory letter-shapes to play with, so as to stimulate little children to learn—or indeed anything else one can think of to give them more pleasure, and which they enjoy handling, looking at, or naming.

Once the child has begun to trace the outlines, it will be useful to have these inscribed as neatly as possible on a tablet, so that the stilus is guided by the grooves.’


[‘Neque enim mihi illud saltem placet, quod fieri in plurimis video, ut litterarum nomina et contextum prius quam formas parvoli discant. Obstat hoc agnitioni earum, non intendentibus mox animum ad ipsos ductus dum antecedentem memoriam secuntur. Quae causa est praecipientibus ut, etiam cum satis adfixisse eas pueris recto illo quo primum scribi solent contextu videntur, retro agant rursus et varia permutatione turbent, donec litteras qui instituuntur facie norint, non ordine: quapropter optime sicut hominum pariter et habitus et nomina edocebuntur.

Sed quod in litteris obest in syllabis non nocebit. Non excludo autem (id quod est notum) irritandae ad discendum infantiae gratia eburneas etiam litterarum formas in lusum offerre, vel si quid aliud quo magis illa aetas gaudeat inveniri potest quod tractare intueri nominare iucundum sit.

Cum vero iam ductus sequi coeperit, non inutile erit eos tabellae quam optime insculpi, ut per illos velut sulcus ducatur stilus.’]

Quintilian The Orator’s Education tr. Donald A. Russell (Loeb Classical Library 124) (Cambridge, Mass., 2001) Bk. I ch. i pp. 76-8.

On syllables, he says:

‘With syllables, there is no short cut. They must all be learned; there is no point in the common practice of postponing the most difficult questions relating to them, to be discovered only when we come to write words. We must beware also of trusting the first memory too readily: it is better to have repeated syllable-drill over a long period, and not be in a hurry to achieve continuity or speed in reading either, unless the sequences of letters are produced without hesitation or doubt, and anyway without the child having to stop and think. Only then let him begin to construct words with the syllables themselves and form connected sentences with the words. It is unbelievable how much further delay in reading is produced by haste. The result is hesitation, interruption, and repetition, because they are venturing beyond their powers, and then, when they make mistakes, losing confidence also in what they already know.’

NB The original text does not specifically say ‘syllable-drill’, but rather than you need to repeat and reaffirm/emphasize the syllables in order to learn them.

[‘Syllabis nullum compendium est: perdiscendae omnes nec, ut fit plerumque, difficillima quaeque earum differenda ut in nominibus scribendis deprehendantur. Quin immo ne primae quidem memoriae temere credendum: repetere et diu inculcare fuerit utilius, et in lectione quoque non properare ad continuandam eam vel adcelerandam, nisi cum inoffensa atque indubitata litterarum inter se coniunctio suppeditare sine ulla cogitandi saltem mora poterit. Tunc ipsis syllabis verba complecti et his sermonem conectere incipiat: incredibile est quantum morae lectioni festinatione adiciatur. Hinc enim accidit dubitatio intermissio repetitio plus quam possunt audentibus, deinde cum errarunt etiam iis quae iam sciunt diffidentibus.]

Bk. I ch. i. pp. 78-81.


Latin was pretty much a transparent language (i.e. there was mostly a one-to-one correspondence between spelling and sound, though there were long and short vowels to contend with, and other quirks which evolved over time). This meant that the syllable units tended to be orthographically clear, so it made some sense to teach this way, though the actual method sounds potentially mindbendingly dull.

(For more on the details of Latin pronunciation, which, being a living language, varied over time and in different places, see W. Sidney Allen, Vox Latina: The Pronunciation of Classical Latin (2nd ed. Cambridge University Press, 1978).)

Yet Quintillian does not seem to expect abstract repetitive drilling. In relation to the beginnings of literacy, he says:

‘I am not so careless of age differences as to think that the very young should be forced on prematurely, and that set tasks should be demanded of them. For one of the first things to take care of is that the child, who is not yet able to love study, should not come to hate it and retain his fear of the bitter taste he has experienced even beyond his first years. Let it be a game; let him be questioned and praised and always feel glad that he has done something; sometimes, when he refuses a lesson, it should be given to another child, of whom he can be jealous; sometimes he should compete, and more often than not think he is the winner; and finally, he should be encouraged by rewards suitable to his age.’


[‘Nec sum adeo aetatium inprudens ut instandum protinus teneris acerbe putem exigendamque plane operam. Nam id in primis cavere oportebit, ne studia qui amare nondum potest oderit et amaritudinem semel perceptam etiam ultra rudes annos reformidet. Lusus hic sit, et rogetur et laudetur et numquam non fecisse4 se gaudeat, aliquando ipso nolente doceatur alius cui invideat, contendat interim et saepius vincere se putet: praemiis etiam, quae capit illa aetas, evocetur.’]

Bk. I ch. i pp. 74-5


When applied to English, with its less predictable orthography, the syllables to be learned were less easily demarcated or learned (because they could be spelled more than one way), and this led to some interesting adaptations.

It seems clear though that in English schools the usual approach, where learning went beyond the alphabet, was a progression from letters to syllables, and from syllables to words: but that there was a sense of progression within the syllables, too, from least to most complex. As a result of not having a completely standardized orthography, that is, the learning of the syllables also required a strong emphasis on sound-symbol correspondence.

An extraordinarily clear example of this is in Edmund Coote’s Schoole-maister of 1596. I have transcribed the first couple of pages, and will gradually add to them as time allows:

Coote p. 2

Note ‘Fa fe fi fo fu’, which I suspect might be the origin of ‘Fe Fi Fo Fum’! The earliest example seems to be in Thomas Nash, Have with you to Saffron-Waldon (1596):

I wonder you will prate and tattle of sixe and thirtie full points so compendiously trust vp (as may bee) in sixe and thirtie sheetes of paper, when as those are but the shortest prouerbs of his wit; for he neuer bids a man good morrow, but he makes a speach as long as a proclamation; nor drinkes to anie, but he reads a Lecture of three howers long, De Arte bibendi. O tis a precious apothegmaticall Pedant, who will finde matter inough to dilate a whole daye of the first inuention of Fy, fa, fum, I smell the bloud of an English-man.

Nashe implies that the phrase is very old, or that its origin is lost: this has been interpreted as an ‘obscure’ source, but Nashe’s words work just as well as an in-joke about all-too-familiar and long-established classroom practice. His use of it, and an appearance in King Lear as well, does suggest that this was a familiar pattern of syllables. The Jack and the Beanstalk story appears much later.

In Love’s Labours Lost v. i. 45, there is a joke on the use of syllables in teaching, where the page says of says of the schoolteacher Holofernes (the ‘Pedant’):

Pag. Yes, yes, he teaches boyes the Horne-booke: What is Ab speld backward with the horn on his head?

Peda. Ba, puericia with a horne added.

Pag. Ba most seely Sheepe, with a horne: you heare his learning.

Peda. Quis quis, thou Consonant?

Pag The last of the fiue Vowels if You repeat them, or the fift if I.

Peda. I will repeat them: a e I.

Pag. The Sheepe, the other two concludes it o u.

‘Ab speld backwards’, of course, being ‘ba’ (and ‘baa’ being the noise of a sheep). Holofernes does not get the joke; the boy calls him ‘a most silly sheep’.

The other joke about vowels (is it ‘I’ or ‘u’ (= ‘you’) also relates to the first pages of the school books collected here.

[The version of Shakespeare’s text given above is the one from the Folio edition of 1625, STC 774.]

Coote p. 3

As Coote’s book goes on, he introduces CVCC, CCVC, CVCCC words progressively, with passages to read as examples.

The first passage is:

If we do ill: fy on vs all: Vp go on: lo I see a py,
Ah is it so: is he my fo? So it is, if I do ly
Wo be to me, if I do so. Wo is me, oh I dy
Ye see in me, no ly to be. p. 2

His next little passage reads:

By, go thy way vp to the top of the hill, and get me home the bay nag, fill him well, and see he be fat, and I will rid me of him, for he will be but dull, as his dam, yet if a man bid well for him, I wil tel him of it: if not I do but rob him: And so God will vex me, and may let me go to hel, if I get but a iaw-bone of him ill.

p. 4

Then the next one introduces CCVC:

I met a man by the way this day, who when he saw me, hit me a blow, that it did swell: for that I did not stir my cap when I met him. But I fled from him, and ran my way: Then did he fret and out-ran me, and drew out his staf, that had a knot on the end, and hit me a clap on the scul, and a cros-blow on the leg, so that I did skip at it: yet was I glad to know and to see as in a glas by bad spot: And I will pray him, that if he shall see me so gros, and so far out of the way, that he wil whip me wel, that so I may know, what I am to do.

p. 7

He then introduces CVCC words, and finally a section which, he says,

Teacheth words ending first in three, then in fower consonants, containing the hardest syllables of all sorts, with practise of reading the same.

p. 11

Somewhere along the line, we lost all this awareness, which perhaps had become too closely associated with rigid teaching methods to survive.

[To be continued…!]


7 thoughts on “Learning to read before spelling became standardized: 1. Quintilian & Coote

  1. Interestingly, Greg Brooks writes about syllables in his book ‘Dictionary of the British English Spelling System’.

    He says that syllables are difficult to define rigorously, even though they are intuitively obvious and children can easily be taught to count them, whereas phonemes are much less intuitive and more difficult to count.

    He writes on p 458
    ” in strict linguistic terms, therefore, as just implied, only spoken words have syllables, and written words do not.”

    That seems rather a sweeping statement to me.

    However, he goes on to justify it with the word ‘extra’. It is ex-tra in written syllables but /ek – stra/ in sounds.

    He states two reasons for his conclusion.
    “imagining that written words have syllables further confounds already confused attempts to predict word stress from written forms of words”

    2. “it is often difficult, and sometimes impossible, to say where the boundaries between written syllables are.”

    The key question he asks is “how can you tell which syllables are stressed and which are not” when trying to arrive at the pronunciation of a written word you have never seen before.

    Some of the words he cites are ‘cotoneaster, machete, oesophagus, Yosemite and misled’.

    His answer is that it will take him volumes of books to explain.

    So when I saw the quote:

    “With syllables, there is no shortcut. They must all be learned.” … It made me think that the author must be referring to transcribing spoken syllables into print first , and only after that going onto words.

    I am still thinking ‘wow’, and trying to get my head around it. I had to tell you about Greg Brooks because I’m still trying to get my head around what he says too.

  2. This is why it’s so important to remember the differences between Latin and English orthography. Not only was Latin transparent, its syllables must have been more ‘transparent’ in a way, too, because while each unit was basically phonetically transparent, it was also morphemically transparent in many cases too. If morphemically is a word. Think of trans-par-ent, and how closely the syllables match the morphemes. English, as ever, is less consistent (especially outside of Latin & Greek derivatives).

    I suspect that syllabification in Latin, for both reading and writing, was eased by the morphemic element far more than it is in English.

    Overlaying English on a system which was built for a transparent language was bound to have caused difficulties, and required work-rounds. My search for this sort of thing is still in process!

  3. I am also noticing that the way the syllables are built up. ( Coote links). It’s almost like building pseudo words in modern phonics programmes. However, I suspect the underlying reason is different.

    Because the basic unit of articulation in normal speech is the syllable, I suspect the scholars were being taught how to put letters together so that they could easily pronounce what they had written.

    If we also consider Vygotsky’s theories that the smallest unit of meaning is a word ( I know it’s technically a morpheme) and that we think in words when we develop verbal thought, (to the extent that without meaning we can’t think and, thus, we can’t learn), then there must be some impact of having an alphabet more closely related to morphemes than the English alphabet.

  4. Sorry, just to clarify, I don’t think the alphabet is more closely related to morphemes in Latin, but that syllables often are; and the discussion in both texts relates closely to reading, from learning letters onwards.

    Both Quintilian and Coote specify reading as the aim in their descriptions of the children’s work, though other parts of what Quintilian says make it clear that he is talking about spelling and composition as well. The work on the syllables in each chapter of Coote is a preparation for being able to read the examples at the end. Once I’ve done a bit more transcription, I think this will be clearer – but it’s a very slow process!

    So I’m not sure what process you envisage when you say ‘I suspect the scholars were being taught how to put letters together so that they could easily pronounce what they had written’. They are explicitly being taught to read by means of these syllables, though of course this would then have had an impact of their spelling.

    There is a difference, presumably, between the processes by which Latin native speakers, English native speakers learning Latin, and English native speakers, would have experienced the syllabification method, and differences between using is for English and for Latin.

    There *was*, in pre-Reformation English schools, at least those which taught choristers, a stage during which children had been taught to ‘sound out’ Latin without understanding it: they were taught Latin language/grammar in the next stage of their education There would have been a practical aspect to this: if you want children with very high voices, and to train their singing early, they will need to be able to sing in Latin before they have learned to understand it. In some cases it seems that they were taught by ear, and in others by sounding out syllables. But they weren’t writing their own text.

    NB Understanding was considered necessary for clergy: in order to prove ‘Benefit of Clergy’ (ie that you should be tried by an ecclesiastical court, not a lay one) people were sometimes asked to read a text – and to test whether they were actually reading or remembering, the book was sometimes given to them upside down. There were regular complaints that monks did not in fact understand what they were saying or singing (this was a particularly common charge by Reformist authors in the mid-sixteenth century).

    There’s some really interesting stuff about elementary education, singing, etc in _Singing the New Song Literacy and Liturgy in Late Medieval England_ by Katherine Zieman http://www.upenn.edu/pennpress/book/14400.html.

    NB The children under discussion are the same sort of age as children learning these things now: around about 5-8 years old.

    Another important thing to remember is that these Elizabethan & Stuart school textbooks were part of the process by which English orthography became standardized. The writers were having to make decisions about spelling which would then be passed on to children in the schools which used their books.

  5. When I was thinking about the scholars learning syllables first, I was thinking about chanting in church services. It occurred to me that maybe the scholars were taught syllables so that they could chant. Even if there was no meaning to the syllables they were chanting, they would be practising transposing from sight to sound whatever was written down. I was thinking of this process as being equivalent to ‘decoding’. (As you write ”sound out ‘ Latin without understanding it.’) Then from this base they could go on to become literate clergy.

    Sorry, if I am interpreting this all wrong. I did ‘do’ Latin at school, but it was Latin written with a modern English alphabet. I’ve forgotten almost all of it anyway. But we had to be able to read it out loud when our turn came in class. When we had finished reading our section out loud, we then had to be able to translate it.

    …. And that reminds me of the Rose Review and Milton’s daughters.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s