Interview by Ken Schmidt

The two views differ in important ways.

1.	According to the skill-building view, the cause of our competence is direct instruction. Reading only helps make our performance smoother. According to the comprehension hypothesis, reading is the cause of both our competence and our fluency.
2.	According to the skill-building view, reading for meaning is only possible after a certain amount of language has been consciously learned. According to the comprehension hypothesis, real reading can occur as soon as texts are comprehensible.
3.	Skill-building assumes that our previously learned grammar and vocabulary determine the comprehensibility of a text. The Comprehension hypothesis says that we understand texts based on our previously acquired knowledge of language, what we have read so far, and our knowledge of the world.
4.	Skill-building says: teach vocabulary and grammar, in order to make reading possible.
5.	Comprehension says: facilitate and encourage reading in order to promote acquisition of vocabulary and grammar. Acquisition takes place when we understand text that contains aspects of language we have not acquired but are ready to acquire.
6.	Skill-building thus thinks that the use of context to aid comprehension is cheating. The comprehension hypothesis says it is an important part of the acquisition process.
7.	Skill-building limits texts to items that have already been taught. The comprehension hypothesis only insists that texts be comprehensible. If they are, they will contain aspects of language that readers are ready to acquire.
8.	Skill-building claims that improvement is only possible with study. The comprehension hypothesis predicts that a reading habit will ensure continued progress.

Many ER proponents actually hold more to the skill building hypothesis, claiming that learners need to learn the rules and forms first and that ER mainly just develops fluency. They don't ascribe nearly as much power to reading. They say a balance of intensive reading (IR) and ER is necessary. They complain that much of the research on ER is flawed, but continue to tell people to implement ER in language programs. If they don't trust the research, how can they recommend it? They say that ER is good, but they say that Krashen is still wrong.

Many people propose ER PLUS other things - summary writing in English or Japanese, cloze exercises, vocabulary list learning, retelling of the story, summarizing of the stories read in the previous week in writing, discussing the content of the story, etc. I tried these activities. A colleague suggested that I do some of these things. I too thought that these other things might help ER be more efficient (I knew already then that it was significantly more effective than traditional skill-based approach). So, after trying them for several years, I found that these things were a waste of time.

There are two reasons why they are a waste of time. I found these reasons from my own research. One is that students don't like most of the skill-based activities, and two, forced (learned) knowledge is forgotten very quickly.

At this point, I need to be clear and qualify what I've just said. I am not saying that everything else is a waste of time and all students will ever have to do is reading or listening to stories in language classrooms. A teacher who wants to teach poetry can teach poetry in the third year, two years after students build a foundation of language competence. The problem we face right now in Japan is to help low level students to move up to intermediate or advanced levels hopefully within a couple of years. Most college students in Japan are at the low level. I have come to the conclusion that what these students need most right now is a lot of comprehensible input. Reading (and Listening) alone is necessary for efficient language acquisition and sufficient for them right now.