I have been reading Suzanne Reynolds’ ‘Medieval Reading: Grammar Rhetoric and the Classical Text’ (Cambridge 1996). She has studied twelfth century glosses on a number of manuscripts of Horace’s Satires, which is known to have been used as a teaching text, in order to cast light on teaching strategies of the period and contemporary theories about grammar. It is a fascinating, if challenging, read.
Language teachers know that the middle stages of language learning are often difficult. The learner may find the learning of words and phrases relatively easy but combining these elements to express her own views is much more taxing. We know that young monks started with knowing the psalters by heart, then learnt letters and then the rules of morphology, probably from Donatus. We know that scholars regularly corresponded with each other to develop their ability in Latin even further. But how did they get from ‘amo, amas, amat’ to the kind of skill we see in letters by Alcuin or Lupus? Reynolds helps to fill this gap. She quotes Alexander Nequam’s account of progression in Latin: from the very basics to Donatus to selected classical texts to Priscian, who, alone at the time it seems, included syntax as well as morphology. We have of course no way of knowing how general a pattern this was, but Reynolds’s analysis of glosses by different teachers in different places on the same text does strongly suggest that in this case the theory does reflect practice.
Reynolds speaks of three kinds of glosses: Those that render the text accessible, those that aim to develop the learner’s command of the language and those dealing with the theoretical side of grammar that the learner will meet at the next stage of his course.
At this point, she deals with another language teacher’s bugbear: teaching a mixed ability or mixed experience group. She has found glosses that distinguish between the ‘pueri’ and the ‘provecti’: most of the ‘access’ glosses are aimed at the former, the others at the more advanced. Since all three can appear in the same manuscript, it seems, perhaps we are talking about some kind of mixed grouping here. Already a clue as to how progression was managed.
Access glosses helped the learner make sense of the text. Occasionally there are translations in the vernacular, more often paraphrases. They give the context – explaining what ancient Roman custom lay at the heart of the satires – and, where the syntax is especially difficult, insert words which make the text more readily understandable to someone whose vernacular was quite different from that of Horace.
It is easy to see how the access glosses could shade into the development glosses and leave room for the emphasis to shift according to the needs of the learner. Having given help with a word’s meaning, the glossator may then relate the actual word form used to the root form, offering opportunity to practise declensions and conjugations. Reynolds points out that the words ‘hic, hac, hoc’ are used with a new noun, not so much to replace the definite article (missing in Latin but present in Greek and Old English and Old French), as to remind the learners of gender. The glossator may show how other words can be formed from the word which happens to occur in the text. Easy to learn by heart and thus increase vocabulary if you’re a puer, a hint as to how a similar activity might be undertaken in a new text with an unfamiliar word if you’re a more advanced student. Finally, many glosses address the problem of word order. Sometimes there is an alphabetical superscript indicating what was called the ‘ordo naturalis’, sometimes this is done by lines. Either way it performs three functions: it makes the texts easier to translate, it shows a methodology for deciphering similar texts and it points the way for the creation of Latin text. By all means sketch it out in ‘natural order’ to get the grammar right, but then consider how it might be recast…
Reynolds again compares practice with theory by relating these observations to a twelfth century text by Alexander of Villa Dei, which explains how a learner should ‘construe’ a text: First identify and place the vocative case, then the nominative, then the main verb and then the rest. This still works today but requires the learner to be able to identify reliably the different parts of speech and their cases, tense etc. But nowadays we have parsing dictionaries which do the identifying for us. In their absence, it is easy to see why access to classical texts relied totally on excellent recall of morphology even if you hadn’t done much syntax.
The theoretical glosses go a step further. The classical texts did not always conform to the rules of grammar. This is particularly true of word order but can also apply to morphology. Since the auctores were the model from which the boys were to learn, these discrepancies had to be explained, at least to the ‘provecti’. If it was poetry, all could be forgiven. But elsewhere, interpretation had to stray into the territory of Rhetoric, and Reynolds goes to some lengths to uncover the contemporary mental distinctions between the two.
There is much more in this book, but it has answered many of my questions with an authority which I had not dared to hope for. What I think is now needed is to try out these ideas against another annotated text and some thinking about the way the manuscript itself was used. If it was an aide memoire for the teacher, which seems likely unless two or three gathered around the manuscript, then we must acknowledge that it is only part of the story and the evidence for classroom practice is more likely to be found in letters and vitae.