Do antibiotics work? Well, that rather depends on what you’ve got. If you’ve got a viral infection like influenza antibiotics will be useless. To fight viral infections you need to use antiviral drugs. Does that mean antibiotics don’t work? Of course not. If you’re suffering from a bacterial infection like brucellosis then an antibiotic might well be effective. This, I hope, is straightforward.

So if I conducted a piece of research which found that antibiotics are ineffective because they don’t cure viral infections that would be a bit stupid, right?

Well, for some reason, professor of education Stephen D Krashen seems to have done something very similar. He got in touch in the comments section on my last post to tell me the following:

I concluded that the published research shows that intensive decoding instruction only helps on tests in which children pronounce words presented on the list. It does not contribute to performance on tests in which children have to understand what they read. [His article is available here.]

Professor Krashen has found that decoding instruction doesn’t contribute to reading comprehension. Frankly, I would have been surprised if he found anything else. Just in case anyone is unclear, word recognition and language comprehension are two very different aspects of reading. Reading depends on the ability to render graphemes (squiggles) into phonemes (sounds) and to turn these into morphemes (units of meaning).

cfde6d1d0f0dfb91a39dabd9fb56217a

These two strands of reading rely on different mental abilities. Word recognition depends on our  visual-auditory systems to pay attention, block out distractions, apply rules associating letters to sounds and to perform saccades (tiny eye movements between words). Language comprehension requires us to think about the meaning of words, utilise our semantic & grammatical systems, make inference and hypothesis as we read and also to anticipate what’s likely to come next.

Systematic synthetic phonics is an excellent way to store alphabetic principles and spelling-sound correspondences in long-term memory, but of course, this isn’t everything that’s needed. Krashen says that “intensive decoding practice is only the first step, necessary but not sufficient, and it needs to be followed by a great deal of practice in applying the principles learned.” Well, of course. Children  also need to store vocabulary, stories and as much knowledge of the word as possible, leaving working memory free to do the hard work of trying to work out what a text means. The trouble is, while we seem to be hardwired to learn vocabulary and stories, we have no such capacity to pick up decoding. If children don’t have fast, automatic access to spelling-sound correspondences they will expend precious working memory resource trying to decode instead of on making inferences, clarifications, hypotheses and predictions needed to be able to take pleasure in the act of reading.

Once a child has learned the sounds made by letters and combinations of letters they can turn writing into speech, but if they don’t already know what a word means the ability to decode is unlikely to magically assist with their understanding. As a fluent reader, after only the briefest pause, you will, no doubt, be able to pronounce a word like adscititious, but obviously that doesn’t mean you will know what it means. For that you will require a definition.

Similarly, you might struggle to discern the meaning of this passage from Kant’s Critique of Pure Reason

A manifold, contained in an intuition, which I call mine, is represented by means of the synthesis of the understanding, as belonging to the necessary unity of self-consciousness; and this is effected by means of the category.

Even though you probably know what all the words mean, without a fair bit of background knowledge you won’t have a blessed clue what the paragraph is about.

Krashen concludes that “the path to reading proficiency is not through worksheets but through books and stories.” This seems reasonable. I hope nobody has ever argued that children will becomes proficient readers without ever reading books and stories, but just in case anyone has, let me be clear that I think the widest possible range of books and stories is absolutely crucial. If phonics is the antibiotic, then a broad general knowledge is the antiviral needed to make sense of what we read.

The bit I struggle with is that Krashen appears to be of the opinion that although a little of phonics right at the start is helpful, a “high level of proficiency” in decoding actually makes children worse at language comprehension. Not only is there an absence of evidence for this point, it appears profoundly illogical sense to suggest that the ability to decode print interferes with our ability to understand what we decode. If he’s right then clearly phonics really is the Great Evil it’s made out to be by Rosen et al. Needless to say, I’ll need some convincing.