One friend told me that, as a child, he mistook the line from the old Christmas song which goes, "While shepherds WATCHED THEIR FLOCKS by night..." and always ang "While shepherds WASHED THEIR SOCKS by night..."
But I think that we can say that incidental learning actually produces massive results. A tremendous amount of learning is taking place without it being taught, not only new language patterns but reinforcement and greater development of language already acquired. And I think we can say that nearly all of this learning is accurate.
This is where Mert Bland's Concept Pod Theory comes in. It starts as a nucleus of one single idea of a word. Over a period of time more and more understanding and definition and application for the word is added through a vast number of contacts and encounters with the word and experiences attached to the word. It's something like a snowball effect as more and more ideas get stuck on the word and our understanding of the word grows.
Christine Tierney is right. Some of it needs to be corrected if it is mistaken when acquired. Thus, direct teaching has a role to play in correcting bits and pieces of this massive amount of material we have not learned correctly.
Now Christine Tierney's students' word, "firstable", was a concept nonetheless. They had a clear idea of what they were trying to express with that word and the concept was valid. Thus, I think according to Mert's idea of a Concept Pod, the Concept Pod for that word was begun. They had ideas of how to use this word and how not to use it. It's complex growth had begun. However, a bit of correction was needed to the Concept Pod to realize that actually the correct thing to say is "first of all", not "firstable". On the other hand, perhaps we are watching the birth of a new word in the English language.
Here in China, almost all students have trouble with the word "colleague". They want to say "col-lea-gue" as opposed to "col-league".
As Krashen says, "The study of grammar has value, however: Even those who are well-read may have small gaps in their writing competence, and conscious knowledge of some grammar rules can be helpful in filling some of these gaps (e.g. the it’s/its distinction)."
Proponents of the value of indirect learning (Comprehensible Input, Extensive Reading, what I call "Extensive Contact", etc) are not saying that all direct teaching should stop and all learning should be done indirectly. It's just that we realize the massive amount of language our students learn without it being directly taught and also realize the difficulty students have to learn even simple language rules that are directly taught, so we raise the question if "direct instruction" should support efforts towards "indirect learning" and not the other way around.