Comments on "A Historic webinar with Chomsky and Krashen: Modern Linguistics" video

From a recent Krashen-Chomsky video (hosted by the Higher School of Technology, Moulay Ismail University, Morocco), Stephen Krashen says:

"We acquire language only in one way, and that's when we understand it. There is no other way in which this can happen" (19:33, emphasis added)

He demonstrates a mini-lesson by pointing to parts of his face and labelling them in German. He then goes on to claim:

"The amazing thing whether you want to or not. It happens without effort and it happens without worrying about it and you can't stop it." (20:05, emphasis added)

Later Krashen describes practicing his Spanish with a local store worker but claims that his improvement in proficiency was solely due to what he does at home.

One piece of evidence he submits (on Twitter) is a case history:

"Sawako, a native speaker of Japanese, clearly did better attending a special class that provided rich, comprehensible, and interesting input (Krashen, 1981, 1982, 1985, 2003, 2004, 2011) than she did working or even studying in an environment where English is used. We describe Sawako’s experiences, covering six phases of her linguistic path since 1993, when she entered a college in Japan as an English major. For each phase, when possible, we included her scores on the TOEIC examination, taken at the beginning and end of each phase." (Mason & Krashen, 2019: 1)

It is not clear from the paper which TOEIC test was used. [Update: Krashen says on Twitter they used the listening/reading test]. An input/comprehension based approach would be expected to improve the receptive skills of a learner, so on the listening and reading tests of the TOEIC we would see this. However it is not obvious that similar improvements would be found in scores on the productive TOEIC tests of speaking and writing.

Meta-analyses comparing comprehension to production approaches supports the common sense expectation that comprehension based instruction (CBI) would improve scores on comprehension measures and production based instruction (PBI) would improve scores on production measures. For example see this summary of one meta-analysis: Input, output, shake it all about-put?

By comparision two studies have found that production based instruction improved performance on both comprehension measures and production measures. They draw on the testing effect (recalling items from memory leads to more gains than recognising items) and prediction generation (comparing output with feedback received). The first study (Hopman & MacDonald 2018) showed the effectiveness of PBI in an artificial grammar design and the second study (Keppenne et al. 2021) showed this effect generalised for learning grammatical gender in German.

Krashen has said that:

"Output has a contribution to make to language acquisition, but it is not a direct one: Simply, the more you talk, the more people will talk to you! Actual speaking on the part of the language acquirer will thus affect the quantity of input people direct at you." (Krashen 1982: 60)

but this is hard to reconcile with the more direct link via prediction generation. Further Krashen's compartmentalisation of comprehension and production is at odds with language processing models that

"suggest production and comprehension are tightly interwoven and that similar mechanisms are active during both production and comprehension processes" (Keppenne et al. 2021: 5)

One thing I did appreciate and chuckle at from the video is Krashen's gentle rebuke to Chomsky's alarmism about student reading:

"The level of reading has very sharply declined. You see it in the school educational programs"(45:49)

"Students are very bright, they haven't read a book, they hardly know what it is to read a book unless it's assigned in class. On their own they wouldn't know how to pick up a book." (46:23)

Krashen responds:

"If this is true it's a recent phenomenon...my last paper was 10 years ago and the title of the paper was "Stop scolding teenagers" (48:21)

References

Hopman, E. W., & MacDonald, M. C. (2018). Production practice during language learning improves comprehension. Psychological science, 29(6), 961-971.

Keppenne, V., Hopman, E. W., & Jackson, C. N. (2021). Production-based training benefits the comprehension and production of grammatical gender in L2 German. Applied Psycholinguistics, 1-30.

Krashen, S. (1982). Principles and practice in second language acquisition.

Mason, B., & Krashen, S. (2019). Hypothesis: A class supplying rich comprehensible input is more effective and efficient than “immersion.”. IBU Journal of Educational Research and Practice, 7, 83-89.