I read with interest the reaction to Mounce's vlog by Daniel and Tonya at Hebrew and Greek Reader (parts 1 & 2). I read their blog with some interest and respect. Mike Aubrey specifically asked me to post some thoughts. In this post I will explore my own thoughts on these issues, with some interaction with D&T. I would like to highlight my respect for them (so far as their blog has presented them to me), whilst you will note my disagreement with their position.
First, I want to propose a broad purpose for which classical (a term I prefer to 'dead') languages are learnt. Our purpose should guide our method in this case.
The purpose to which most students seek to learn a classical language is for the study of ancient texts. That is to say, few are interested in the linguistics of ancient languages in and for itself (though it often seems otherwise); few are interested in communicating in ancient languages for the sake of conversing with contemporary people; most are looking to read ancient texts with insight and understanding. This generally leads to carefully, methodological, studious reading of texts with precision and rigour.
Second, I want to suggest what method(s) will best achieve this purpose.
It seems axiomatic to myself that you will improve in what you practice. If you practise soccer, you will improve primarily as a soccer player. If you do a lot of work with flashcards, your memory for flashcards will improve. The mind, like the body, tends to be lazy and specific – it will accustom itself to what you accustom it too.
Therefore if reading and analysing ancient texts is the primary purpose, the main method to be used should be the reading and analysing of ancient texts. And yet it's at this point I think our analogies break down and mislead us.
There is a vast gap between language acquisition and meta-language learning. To put it another way, it's one thing to learn a language, it's another thing to learn about a language. And most time spent in a traditional Grammar-Translation language class is spent learning about a language – analysing and memorising grammatical and linguistic knowledge about the language. Now, this in itself is not a bad thing, but time spent learning about a language is not time spent learning a language, not at all. Those who only learn to analyse the grammar of a language, when they come to texts, will do what they have learnt to do: analyse the grammar. If they acquire much language ability, it will be a by-product, a kind of cross-training, much like playing a lot of soccer might help you as a runner.
This is why I advocate reversing the pedagogical assumptions above. It's far better to acquire the language you are studying, and then to learn about it. It's even more superior to acquire the language and learn about the language while using it (ie, to discuss Greek grammar in Greek, Latin rhetoric in Latin, and so on). It's superior because the time and effort reap a better outcome.
That is to say, if you can pick up and read a Greek text, for instance, without needing to parse every word, and without needing to translate it, you're actually reading Greek as Greek. Which is what you want to be doing if you want to have much hope to provide a decent account of how it would read in its original socio-historic setting. Having a grammatical meta-language is going to help you work out why it does that, but having that grammatical knowledge is not essential to knowing that it reads a certain way. The meta-language of linguistic description is a tool for analysis and for explaining to others, but it's not a necessary tool for understanding. Plenty of English-users understand English perfectly well without being able to explain the simplest grammatical structures. There's no reason a person's acquisition of a classical language need be any different (except that we may well want a student to acquire exactly that meta-language skill).
All parts of this series: