A C.I. Rebuttal by Eric Herman


Eric Herman for the last couple of years has offered quality information about teaching with Comprehensible Input.  He is able to articulate the finer points of TCI because of his obvious passion for the research behind language acquisition.  One of the things I admire about Eric is that he also is an excellent practitioner of TPRS and often shares his work with others.

The following information was written by Eric in a discussion about “Acquiring and Errors” from the more TPRS Help list serve.  I asked Eric if I could share his work here because his response was kind, extensive, and very helpful for understanding the world of using Comprehensible Input.

Eric writes….

If you are challenging the distinction between learning and acquisition, then there’s tons of research to debate (e.g. Krashen’s entire life’s work). And this distinction is at the core of this group’s philosophy. I see some confusion possibly being caused in this discussion, because terms (e.g. practice) are not being defined and mean different things to different people.

There are other ways to refer to this distinction that are used by SLA researchers: explicit vs. implicit knowledge. For some of the support for it’s existence I recommend finding access to read Ellis, R. et. al. (2009).Implicit and Explicit Knowledge in Second Language Learning, Testing and Teaching. We must distinguish between teaching, learning, and knowledge. The result of explicit instruction can be implicit and explicit knowledge and vice versa.

VanPatten clarifies Krashen’s terminology in (2007). Consciousness in SLA. page 28-29.


From that article, this was quoted:

Krashen (1982): “The first way [to develop competence] is language acquisition, a process similar, if not identical, to the way children develop ability in their first language. Language acquisition is a subconscious process. . .The result of language acquisition, acquired competence, is also subconscious.”

In essence, acquisition and learning refer to process. The product is acquired competence and learned competence, respectively. The terms can be distinguished based on context: acquisition deals with the traditionally non-classroom, natural conditions under which we develop acquired competence in L1. And both terms can be distinguished based on focus of attention: meaning or form, respectively.

If you are arguing against this paradigm then some will listen to your anecdotes and discuss them, but unless you have some compelling, research-based evidence that this is a false paradigm, then attempts to change opinions is likely in vain.

Ultimately, we share the goal of developing a students’ ability to effortlessly and unconsciously communicate, internalizing the linguistic system and automatizing its use. A skill-building paradigm proposes it is the rule -> practice that develops this (declarative -> procedural knowledge), while those of us of the CI paradigm see it as the result of getting comprehensible input and the effect on one’s Universal Grammar. We are averse to terms such as “practice” and “teaching” because they have connotations of skill-building. To distinguish what we do we say we are providing CI and using implicit teaching.

The CI paradigm does not mean that we see no role for grammar instruction, whatever it’s type. See Long’s distinction between focus on form and focus on formS.http://eltj.oxfordjournals.org/content/56/3/303.full.pdf+html

TPRS is more like the former, drawing attention to form amidst the CI rather than fronting our instruction with explicit instruction and then trying to design output-based practice. We simply see that rule instruction leads to explicit knowledge (learning) and it does not convert into implicit knowledge (the non-interface hypothesis). Of course, this is highly controversial. This line of thinking is highly influenced by all of the Natural Order of Morpheme and Developmental Stages of Syntax studies showing that there is an INalterable order of acquired competence. That explicit grammatical knowledge has a limited role in real-life oral interaction – a monitoring role. There are people who are under/over/optimal monitor users. Knowing the rules and practicing may help turn someone into an optimal monitor user. I am probably stating what you already know, but it is our perspective that unless the 3 conditions for monitor use are satisfied (knowing the rule, thinking about the rule, and time to apply the rule), then that explicit knowledge will not aid communication.

The other role of grammar pop-ups in language development is to make the input more comprehensible. This is to accept a weak version of the comprehension hypothesis. There are numerous ways to pop up grammar. We could say the “o” at the end of the Spanish present tense verb means “I” or we could say: ” ‘como’ means ‘I eat.’ ” I’m not sure how much evidence we have to show this works, but accurately processing the form-meaning link is prerequisite to acquiring that aspect. That still may not change the fact that the person is not “ready,” i.e. not at his/her i+1. I bet many of us believe it can help acquisition, at least be fuel for the monitor, and for some it will satisfy a need to understand how the language works. By the way, I liked your term, comprehensible grammar (CG).

What makes this a hard discussion to have is the inability to control all the variables so as to be able to say: “It was the focus on form that developed the unconscious language communication.” The other problem is that loosely defined term “proficiency.” Proficiency tests vary in the amount of implicit and explicit knowledge they measure, because there are occasions on proficiency tests in which the monitor is available and even encouraged. In general, if the test demands a focus on meaning and happens “online” (real-time processing speeds), then it’s more likely we are measuring an acquired competence. The ACTFL’s OPI is more of this type. But teacher-designed assessments frequently are not. Thus, when a teacher claims to have developed “proficiency” I want to know how that was measured in order to know more about relative implicit vs. explicit knowledge.

For a really fascinating study, see: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3558940/

This study found that the type of instruction, implicit or explicit, lead to equal performance of “proficiency” but that it was implicit instruction that developed native-like brain responses. That’s some brain-based evidence that the distinction between learning and acquisition is real.

As far as your point about rules not being accurately taught, you are right. In fact, textbooks are guilty as well. See Hastings: http://focalskills.info/articles/implicit.pdf

As he concludes: There seems to be a paradox in teaching grammar. In order to state the rules correctly, a number of complex grammatical concepts are required, and most students won’t understand them. In order to make the rules simple enough to teach, they have to be reduced to formulations that fail to state the facts correctly.”

So then, if rules are taught inaccurately, how is it that they we have students who have acquired them? Also, how do students acquire what has not been taught?

VanPatten argues that textbook rules are not represented that way in the brain. Google search “against rules vanpatten.” I don’t know his motives for including explicit grammar instruction in his textbooks. Again, we are not against teaching grammar explicitly (including using a focus on formS approach). We just do not see that as becoming an acquired competence and view it’s role as limiting. 

In VP’s brainchild (Processing Instruction – PI), the first step of the method is to explicitly teach the grammar aspect and to explain to the learners the faulty processing strategies they bring to the task. And VP is clear that PI is a form of input-based grammar instruction. Interestingly, PI studies have been run that exclude the first explicit instruction step and it doesn’t change the results.

And lastly, as for a sub-fluent and sub-accurate language teacher . . . we view that teacher as developing more acquired competence in his/her students than a more fluent/accurate teacher who is not trained in teaching with comprehensible input methods. The teacher with lower acquired competence does still communicate, right? If the kids acquire his/her language, then they too will be able to communicate, albeit less perfectly. As Terry said, there are TPRS curriculums, such as LICT, that include a lot of pre-scripted language for the teacher to follow if needed. And this teacher can make use of other resources (e.g. reading) to provide the kids with higher quality CI. In the meantime, that teacher can seek out more CI to further his/her personal acquired competence. If you’ve followed all that has been said above then you would NOT recommend this teacher compensate for his/her lack of acquired competence by including more grammar instruction. Perhaps for teachers who are good monitor users, that could help. That is NOT a shortcut to acquired competence and I fear would only perpetuate the cycle producing future FL teachers with lower acquired competence.

And if you are interested in reading more rebuttals to grammar instruction see (both can be read online):

Krashen, S. (2002). The Comprehension Hypothesis and it’s Rivals.

Truscott, J. (2007). Grammar Teaching and the Evidence: A Response to Nassaji and Fotos (2004). IJFLT.

There was also mention of the role of error correction, which is another highly debated topic in SLA. John Truscott is the guy to read if you want to read the side of the argument against error correction.


Eric Herman

About This Author


You can post comments in this post.

  • Mike, I love what Eric has to say here. There are many great points, and of course I’m on this particular train all the way. I want to comment on the “teacher-learner” piece, because I have been amazed by how teaching with CI has changed my language abilities.

    When I started teaching with CI, I could hang in on a Russian conversation, but most of the time a native Russian speaker who knew English would prefer to speak with me in English if the conversation was important or if we were meeting socially. At that time, I had taught for 23 years.

    To teach Russian with CI requires nearly constant reading and searching for new sources, because there is almost nothing out there for beginning readers. Then the teacher has to keep creating: writing embedded readings, stories, movietalk support, and so on. Every Storytelling teacher has to write up class stories, of course. In the beginning, I sent every single paragraph I wrote out to a rotation of Russian friends. Every paragraph came back almost completely re-written at first. It could have been disappointing, but I knew I was a marginal writer. On top of that, my students seemed to be truly acquiring whatever I gave them, and I wanted them to acquire correct Russian.

    Fast-forward about a year and a half, and my Russian friends who hadn’t talked with me in a while started to ask me whether I’d been in Russia for the last year. An OPI training session yielded a result some four levels above my last test. And when I ask now for writing corrections, the main issue is my punctuation.

    I work just as hard to supply materials for my classes as I did earlier to create activities, but the reading and the writing at the high school teaching level, put together with the huge increase in using Russian in class, improved my Russian to a huge extent. I believe that Eric is right about a less-fluent CI teacher’s being able to help students acquire language better than a fluent one who uses other techniques, but I also believe that a CI teacher who puts in the time to prepare for lessons by reading and writing extensively, and supports that by speaking only in the target language, will find that her/his own acquisition will speed up as well.

    Michele Whaley 3 years ago Reply

  • Michele,

    Thank you so much for sharing that story. The Spanish and French resources seem to dominate the TPRS/TCI classrooms. Teachers and students alike need access to materials that develop further language development. Especially language that is appropriate for emerging learners.

    I agree, things like Embedded Readings, short stories, TPRS novels and MovieTalks are all things that help students develop language competence. They also are helping me develop my Spanish and French and German (and Mandarin when I see Linda Li).

    I hope many new and veteran teachers read your message because we all have our weaknesses in our second languages and there is nothing wrong with admitting that we are less than perfect because who is?



    Michael Coxon 3 years ago Reply

Post A Reply