To the first question above there is no simple answer, even in an institution such as Cambridge where almost all the students have gained an `A' grade in `A' level Latin. As Langslow points out, "even an `A' grade in `A' level Latin is, in itself, of very limited value as evidence of successful learning of the Latin language". Even if there were only one Latin `A' level syllabus, with no options to make the `A' level more literary or more linguistic, one could not easily define the level of competence that candidates taking the examination were being expected to achieve. As it is, teachers can choose to enter their pupils for any one of three (soon to become two) different Latin `A' level syllabuses, each with its own slightly different aims and objectives - though these are in the process of being standardised by the School Curriculum and Assessment Authority. And not all the differences are slight: one board, for instance, permits the use of dictionaries in the unseen paper.
The picture is further complicated by the wide diversity of Latin provision now offered in schools. Before the 1960s, it could be safely assumed that students had studied Latin for at least seven years before coming up to university. Now there is a vast gap between the traditional pattern of provision still available to some pupils in the independent sector and the minimal provision offered in those comprehensive schools which have managed to retain Latin in some form or other in spite of the pressures of the National Curriculum and tight financial constraints.
At one end of the spectrum there are the boys (and I mean boys - this pattern is confined almost exclusively to boys' preparatory schools) who start Latin at seven, know their Kennedy from start to finish by the time they take Common Entrance, take their Latin GCSE a year early (and still find it much too easy) and in the sixth form read a broad range of authors over and above the prescribed `A' level texts. At the other end, there are the boys and girls who take their GCSE after three years of studying Latin outside the timetable, are justifiably delighted when they gain an `A' grade, and spend most of the sixth form desperately trying to plug the gaps in their knowledge caused by the speed with which they have had to rush through the GCSE syllabus. Through no fault of their teachers, these students are very unlikely to have read more than the minimum number of `A' level texts, and may well focus on verse authors such as Virgil and Ovid rather than prose authors such as Cicero or Tacitus, because of the linguistic difficulties associated with the latter.
Given this wide variation among first year students in terms of their previous experience, and taking into account variations in aptitude and motivation, one can see how difficult it is to develop an effective and accessible programme to raise the general level of language competence. And what level of language competence are the first year students expected to achieve in Cambridge? During the first year they are required to read the equivalent of seven `A' level set texts in Latin, and a similar number in Greek (unless they are on the Intensive Greek programme). For each prescribed text there is a lecture course - which may be supplemented at a college level by specific supervisions - but students are generally expected to read, i.e. translate and come to an initial understanding of, the texts on their own. Therefore the level of language competence expected of these students with regard to their preparation of set texts might be defined as that of fully independent readers, by which I mean readers who are able to read, understand and appreciate a range of classical authors without the excessive use of dictionary and/or reference grammar (to adapt one of the definitions quoted by Langslow). This represents a significant step-up for students who have received the minimal school provision I outlined above, not necessarily in terms of the difficulty of the authors (though Horace and Tacitus, for instance, are likely to cause problems) but in terms of the sheer quantity of texts to be read and the degree of independence with which they must be tackled.
It was with these students in mind that I set up a Latin consolidation course to help bridge the gap between school and university and give them the means to become more independent as readers. In an attempt to provide the students with a fresh way into material that was already familiar to many of them (and in line with Langslow's argument that developing students' active competence in Latin can be the quickest way to develop secure recognition skills), I have built the course round a series of cloze exercises requiring the students to use active recall of Latin syntax and accidence (but not vocabulary) to revise key points of grammar, starting with sessions on basic language principles such as adjectival agreement and the sequence of tenses before moving on to all the main constructions they are likely to meet in the course of their first year reading.
Each week the students are given 15-20 pairs of sentences on a specific point of grammar. They are also given a sheet with a brief explanation of the grammar point in question, which they can keep for later reference. Each pair of sentences comprises an English sentence followed by a Latin translation with key endings omitted, presented in the following format:
The king was advised not to attack the city.
rex mone___ ___ urb___ oppugn___.
While these exercises are obviously too easy for students who have already done prose composition at school, they provide a reasonably demanding challenge for most other students Not only do such exercises provide the student with constant reinforcement of basic grammar within a systematic framework, they also provide the lecturer with a valuable assessment tool. Consider, for example, the following three answers:
Example: We saw the farmer working.
agricol___ labor___ vid___.
Answers: agricolam laborare vidimus.
agricolam laborans videmus.
agricolam laborantos vidimus.
The first answer is not absolutely right, nor is it absolutely wrong: at least the words make a possible Latin sentence. The second answer might be the work of a careless student who knows that the present participle is required and knows what it looks like but does not remember to make it agree with the noun (videmus for vidimus or videbamus also suggests carelessness). The third answer is perhaps the most worrying, partly because its author clearly has confused ideas about how the present participle declines, and partly because -os seems such a surprising accusative singular ending. In all three case, of course, further evidence will be needed before one can make a more precise assessment of the quality of the individual's work.
In order to make the teaching materials more flexible - and therefore of greater benefit to a wider range of students - I have experimented with an electronic version of the cloze exercises, providing the students with exactly the same information on each topic, and with exactly the same examples to complete. Once they have filled in the endings on the screen, they can check for errors: a click of the mouse reveals the correct endings and highlights any mistakes they have made. And at any point they can receive a record of their score.
From the students' point of view there are a number of advantages in having the materials on computer. They can choose which grammar point to revise, where to revise it, and when to revise it; they can enter their answers without any fear of being exposed or humiliated, however bizarre their answers may be; they can receive regular feedback; they can return to exercises at a later date to see for themselves whether they have improved (or not!). But there is a major drawback: the computer can only tell the students what they got wrong, not why they got it wrong. It will be some time before computers can carry out the sort of sensitive and precise assessment we considered above.
Can cloze exercises contribute to students becoming fully independent readers of Latin? Students for their part have indicated in formal evaluations that the consolidation course has made them feel more confident of their knowledge of Latin grammar. Certainly, the fact that cloze tests require students to make active use of Latin accidence does improve their ability to make effective use of dictionaries and grammar books, and this in turn should improve the speed with which they can read their texts. Finally, at the most basic level, all the students taking the consolidation course develop a greater awareness of the importance attached to the ends of words in Latin, which is no bad thing when it comes to studying an inflected language.