Horse-racing people like to say that it’s “horses for courses”. Certain horses perform very well on certain courses, and perform poorly on other courses. Same horse. Same quality of breeding, care and training. It’s just that the purpose to which the horse is put isn’t the one that suits it best. A quarter horse will beat a Thoroughbred over a short distance, but not a longer one.
Methods of language teaching are like that, too. The variable, IMO, is comprehensibility. TPRS demands 100% comprehensibility (or as close as humanly possible). The definition of “comprehensible” these days for other methods is “kinda-sorta comprehensible” — certainly far less than 100%. People who train teachers in these methods often stress that students “need to be prepared for the real world where people won’t tell them what things mean” or that they “need to deal with ambiguity”. From the perspective of a 100% comprehensible method, this is throwing students in to do things for which they have not been given the tools yet. TPRS puts off the “real world” while equipping students to deal with it — not with a huge vocabulary, but with the structural knowledge that makes them micro-fluent and able to deal with unknowns as a native speaker does.
TPRS skills can be categorized like this, I think, in order of simplicity:
1. Staying in bounds (use the right language)
2. Circling the language that is in bounds (manipulate the right language)
3. Making the circling unpredictable (manipulate the right language in an unpredictable way)
4. Creating a story out of the circling questions as you go along (manipulate the right language unpredictably while also creating a bare narrative line)
5. Providing a flood of 100% comprehensible language in the process of doing 1-4. (manipulating the right language unpredictably in a narrative line that is extended with details and class talk which is also in bounds)
Part 5 is where many teachers get mired down and “decide” that TPRS doesn’t work so well. Kids will acquire language from 1-4, but they will astound you with what their brains figure out and what their mouths say if you can get to part 5.
Most “communicative” or project-based teachers fail at step 1, because the language demands are just too complicated. Project-based learning uses language as a tool to learn more language. That’s really good for situations where students have some language in the first place. Can it work with zero-language students? Yes, but not as well as TPRS. Does TPRS work with advancing students with most of the structure of the language acquired already? IMO not as well as task-based learning. Each is a tool that has specific strengths and specific weaknesses. The challenge is for folks to put aside their emotional responses to statements like “can’t teach beginners with…” and “can’t teach advanced students with…” and restate these ideas as “the most effective way to teach beginners is…” and “the most effective way to teach advanced students is…”
Why does task-based “suddenly” become a good tool for students who are moving up in their language proficiency? Because at that point, they are able to handle less-comprehensible input. They are able to use the language they have acquired through 100% comprehensible input as a foundation to attach and relate less obvious meanings to. Ambiguities that would stop a beginner cold can be handled by more advanced learners. Imagine that you know 9 words, and then get an unknown word. That’s 10% of your total language. But if you have 99 words, that 1 unknown word is 1%. It has a much less terrible impact on your ability to comprehend what’s going on — and remember that no matter how you slice it (communicative, task-based, TPRS, whatever) the acquisition mechanism is the same: matching language to meaning and generalizing to form the mental grammar.
The reason TPRS has such power for beginners is its high level of comprehensibility (makes them able to understand), its personalization or customization (keeps them listening) and its repetition (gives them enough exemplars to make a permanent link between meaning and form). Thus far, I have not seen any other method that comes close when used on post-childhood (teen and up) novices — including immersion.
I completely do not look down on teachers who don’t use TPRS or CI. I just really wish they would come and teach my advanced students (the ones who have had 2 years of CI-based Chinese) with the methods they’re really good at, and let me have the beginners. Not because of my preferences or experience, but because of the methodological strengths involved.