I’d like to survey two matters at issue: one would be to ask Latin teachers what other languages do they know and how did they learn them. An example of why it is important to exclude teachers not raised in the U.S.: one of my colleagues was raised in Latin-America and complains that American students don’t work hard enough to learn Spanish, citing her own efforts in going to English-language movies, reading magazines, and so on, all the things we know people highly motivated to learn English to do. Once in a while, I’ve run across someone who was taught a fl in the U.S. via what we would call a communicative method and even CI (I recall a teacher at my last school who had a German nun for German class and she sang with them, told them stories (!!!!) in German, and so on, with little grammar, and when she went with her school mates to Europe and each student had to interpret for the group depending on the language they had taken, she was the only one who could function in her language), so we have to exclude them b/c they do not “prove” we need explicit grammar instruction to learn a language. With such a survey, we could determine how effective the typical American fl classroom is most U.S. teachers use as their base a method called Cognitive Code, though few recognize the term). What I think the survey would reveal is that few Americans actually learn a fl via classroom instruction unless they teach it!
Another survey could be of high level high school students in fl classes, incl. Latin, to get at exactly what they do and just how competent they are as they perform at this high level. I would love to know just how they use their paradigm knowledge to read a novel or short story or an article on the bee population. Bill VanPatten has pointed out in several books the typical process by which students use lexical, not grammatical, knowledge to “figure out” what a sentence says. How exactly is this grammatical knowledge supposed to work? Do students read along and come across an ending that doesn’t immediately communicate something to them and then flash through their memorized paradigms to locate it and then attach a meaning to it? I can speak for myself (I’m at various stages of acquisition/learning mainly learning in several languages and do a lot of reflection on what and how I learn and I’m proposing the same for a large number of advanced students). One more bias we have to interject here specific to Latin…. actually two biases: the first is that we translate into English a text in a fl and that constitutes “reading”. No. The second is that languages not highly inflected are somehow easier to learn than highly inflected languages. A top American scholar in the Urdu language told me Urdu was so easy b/c it wasn’t highly inflected; she is the product of American schools. Once I pointed out just a couple of complications e.g. to say “to tell” you put “to hear” in the causative so it means “to make to hear”…….. that’s the idiomatic way of saying it, along with a lexical verb, a complex predicate with “do”, and an honorific from either Arabic or Persian. “Oh”, she wrote, “I see”.
So, controlling for all the biases and complications, we might be able to tease out exactly what is going on among our highly proficient students. I’ve done this sort of survey for years now on a very casual basis and my “findings” back up the contention that, as one teacher put it at a conference, “Teaching grammar is a waste of time”. She was talking about novices, which would perhaps include all students with less than a university major in the TL.