I developed this flowchart as a way to visualise how a student could use explicit knowledge of Latin case endings to arrive at a set of possibilities for what those forms could signify.

I said ‘explicit knowledge’. I’ve been thinking more about the role of explicit knowledge in second language teaching. We shouldn’t reject a potentially useful tool based on ideological opposition to the concept, but we should weigh up the relative merits and drawbacks of the tool, and act in the interests of our students within the imperfect educational systems we have (while also working to make those educational systems better). Proving that input is sufficient for language acquisition is not the same as proving that explicit knowledge is always useless. And on the other hand, if explicit knowledge may be useful in some circumstances, that does not necessarily mean it is useful in all.

There are limitations to the usefulness of explicit knowledge of the cases, but knowledge of the cases does provide some information that is useful for comprehension.

Latin case endings are very often ambiguous if seen in isolation. They give incomplete information. Is ‘-ae’ genitive or dative singular, or nominative plural? The endings alone will not provide the answer. That’s why it is important for students to interpret words in meaningful contexts. The context gives a range of possible meanings, which then intersects with the range of possible meanings indicated by the morphology. The two sources of information (top down and bottom up) need to be combined and confirmed with each other for processing and comprehension of the morphology to occur.

Even when a student narrows the possibilities down to one particular case, they have not necessarily removed all ambiguity of meaning. Individual cases like ‘genitive’ and ‘ablative’ could support a range of possible meanings, which again have to be decided based on context and the likely content of the message being expressed. Reading experience helps greatly.

While explicit knowledge is not sufficient in isolation, being able to recognise possible cases based on noun endings is a useful tool within the larger set of tools readers use to comprehend texts.

I’m not arguing that the noun cases must only be learned explicitly: is possible to gradually acquire Latin cases without explicitly learning the endings of all nouns. The -er/-or/-ir nominative marker, for example, is almost never explicitly taught in tables, or it is explicitly taught as ‘not an ending: the ending is put after the -er’. Despite being sidelined in explicit instruction, this nominative indicator is gradually picked up intuitively after sufficient reading experience.

But noun cases are not all naturally acquired in all students before students need the knowledge. In school settings with high stakes competitive examinations, one of the students’ priorities is to maximise their performance in comprehending Latin within the limited hours of instruction they can receive before sitting a final exam that affects their university entrance scores. Within those constraints, it is helpful for students to be able to use all tools at their disposal to comprehend the most they can out of unseen passages. This includes both their internalised knowledge of the case system and their explicit understanding.

A reader can ‘rule out’ certain interpretations by noticing that the case endings do not support their hypothesis of what the sentence means. They can do this if they know what to look for and have sufficient time to try out multiple hypotheses in reading a difficult sentence until the most likely meaning becomes clear.

These circumstances appear in high stakes examinations, which are not a natural phenomenon, but it is an unavoidable reality for a large number of students who will be assessed competitively on their performance at comprehension well before they have completely acquired the language.

But outside of school, readers also benefit from being able to get themselves out of incomprehension whenever their reading material presents significant difficulty. The counter argument, that readers should never let themselves read anything which presents significant difficulty, is unrealistic. When readers select at-level texts, there will be parts within those texts which are either accidentally harder or easier to read, because texts are not completely homogenous. The only way to guarantee that you will never encounter significant reading difficulty is to only select texts well below your reading level, meaning you will never allow yourself to read the classical canon until your at-level difficulty is… above the most difficult surviving parts of the classical canon? (Has any Latinist ever gotten there without reading any mildly challenging texts along the way?) In reality, independent readers tend to select texts which interest them, often in the process reaching for texts which would be considered too challenging, because of their personal preferences and enthusiasm. I don’t think we should pry enthusiastic readers away from reading the texts that interest them.

The other counter argument is that when presented with difficulty, readers should let go and be fine with not comprehending difficult parts in what they read. This is useful language learning advice, because going down rabbit holes to learn tiny insignificant details will take away from the time spent actually reading. But this advice needs to be held in tension with need for the reader to actively engage in trying to comprehend what they are reading. If it only takes a few more seconds, a couple of re-reads with thoughtful application of some small nugget of explicit knowledge, for a reader to gain more comprehension out of a difficult spot in the text, such an unobtrusive tool could help the reader stay engaged with the meaning of the text and not need to either skip parts of their text or spend significant time away from reading.

If some level of challenge is a part of the normal reading experience, being able to comprehend more of the meaning relatively quickly in occasionally difficult passages is advantageous. Being able to test hypotheses on the basis of explicit knowledge of the noun cases is one way (among many) for a reader to get more value out of reading some difficult passages, which in turn provides comprehensible input for items not yet acquired.

Explicit knowledge, when it is taught in such a way that it is applied in reading and is not simply sitting inert as knowledge for knowledge’s sake, supports comprehension. Comprehension in turn supports acquisition. While declarative knowledge has no direct pathway into implicit knowledge, the knowledge used in the reading process helps students gain experience, and they gain implict knowledge from the sum of their experiences.

The question is, when and how do students ‘use’ what is taught explicitly? If there is no clear way for the average student to make a connection from chanting a noun table to reading a text, we as teachers haven’t made the noun table useful to the student. It sits in its own silo as inert knowledge, and thus becomes a waste of time. If we are going to take up students’ time in explicitly learning noun endings, we should either make it clearly oriented to the process of reading and comprehending, or else excise it completely. If noun endings are useful to comprehension, we had better make it obvious to students how they are useful.

In CI classrooms, explicit grammar is not completely removed; it is taught in small pieces in service to comprehension. In ‘pop up grammar’, it is explained in response to student questions. In some lessons, small grammatical rules of thumb are explained immediately before they are used in communicative tasks. How different is this in practice to PPP – presentation, practice, production? In my experience, when explicit grammar is removed from early years, students in later years tend to ask for lessons on it. I think sometimes as CI teachers, we invest too much of our identity in being opposed to ‘explicit grammar’, when in reality it is not the enemy of comprehension.