Agreeing to agree

At least mostly.

Sam Bommarito replied to my earlier post “Progress maybe …” by providing a link to a post on his blog in the comments section of my post. I’m replying here to the post that’s on his blog, to which he provided a link in the comments (to my post). Now if you can follow the thread . . .

Wait, if this is too confusing, a visual might help. If you’re not confused, and a visual is just clutter, pls accept my apologies, and skip over the screen shot (the visual below) of my “Progress maybe…” post, showing where Sam put a link to his reply on his blog at the very bottom (where it says, “Discussion about this post”). Got it?

Good.

As you can see from the screenshot, my post was mercifully and uncharacteristically brief, so you can read my post here and don’t need to shuttle back and forth between Sam’s site and mine, posts and comments, which few people would have the patience to do. I wouldn’t.

Before going any further, pls re-read the “Progress maybe ….” post above. Remember it’s very brief, and then Sam’s replies (that I’ve cut and pasted below, interleaved with my replies) will make more sense.

Ok, now… here’s Sam’s reply, which he posted on his blog, with a link to it in the comments on my post. Sorry, I’m repeating myself again.

What he wrote in his reply is in

blocked quote.

Mine is not.

Common Ground

I find [Claude’s] statement #1 around the use of context particularly hopeful. He says that he and Shanahan feel that the current laws removing the use of context from word recognition are a bad idea. I’d love to hear more from him on that point (and I’m posting this blog entry on his site to see what he meant by that).

Simply that laws removing the use of context from word recognition are a bad idea. They are not helpful, because context is helpful, sometimes essential, for word recognition.

I believe the problem stems from the people writing the legislation (and many others as well) confounding “decoding” and “word recognition.” They are not the same.

Decoding involves using phonics and orthographic patterns to say aloud (or to yourself) what the written graphs (“cues,” if you like) represent, which are the sounds within the words of the spoken language. That is it. That’s what “decoding” is. The letters—graphs—represent sounds. Decoding allows you to pronounce the sounds (phonemes) associated with the letters (graphemes). Word recognition, however, requires identifying a word as a word and, ideally, having some sense of its meaning.

Do a morphological analysis: Re- means “again”; cognize comes from cognoscere, meaning “to know.” You have to know a group of letters constitute a word in order to recognize it, which means, literally, to know it again.

You can decode a nonsense word, as long as it’s decodable, eg, presling. (I’ve looked it up, and it’s not a word.) In fact, having learners decode decodable nonsense words is a good way to assess decoding skills.

But you can’t recognize a nonsense word, except maybe to recognize it as a nonsense word. And even then you can’t be sure it’s a nonsense word unless you know every word in the language and can determine whether what you’ve decoded is a word or if it has no sense; it’s “nonsense.” Presling makes no sense because there is no meaning associated with it.

The exception here is if you’re a mature reader and come across a grouping of letters that is orthographically impossible, eg, bnptwzmn. You don’t need to know every single word in the English language to be sure that bnptwzmn is a nonsense word. (I didn’t need to look it up.) It’s also basically unpronounceable, having no vowels, something that early and beginning readers will not necessarily know immediately disqualifies it from being a word.

So to get back to the point Sam wanted to hear more about, prohibiting 3-cueing or use of context for word recognition is ham-handed and self-defeating. The real question is not whether context plays a role in word recognition—it does. The question is what role does it play. One way to think about it is that decoding is the on-ramp to word recognition. Context then plays a role in confirming or disconfirming whether the word has been accurately read and recognized.

One last thing, not to make this more complicated than strictly necessary. As readers become more proficient, and they “mature” as readers, context and decoding change their relationship with each other and relative importance. I can say more about that, but this is already too long. The point is that for beginning and early readers, we need to help them get on the on-ramp but not deprive them of what they need (“cues” or whatever) to get where they need to go—accurate word recognition.

Rather than focusing on claims and counterclaims around “The Three Cueing System” (BTW, it is NOT a system of instruction), I think his point #2 establishes the need to stop using the term cue and to start discussing what information readers use with decoding. Most importantly, what information do readers use when the graph phonemic information fails to produce a meaningful word? That can sometimes be because the word is not spelled the way it sounds (irregular word). It can sometimes be because it is not in the reader’s current listening vocabulary. I’d especially like to clarify his views about this in light of the concept of set for variability.

I don’t think there’s any point in discussing whether three cueing is a system of instruction or what it actually is. Whatever you call it, as generally interpreted it has helped create a misleading and dysfunctional approach to teaching word recognition (and reading more generally) and to assessing whether students’ WR skills and understandings are on track. Pls see my longish posts on this here (be sure to read the footnote on Lucy Calkins and 3-cueing—or not), here, and here. Feel free to follow up if further clarifications would be helpful.

Point 3 is one on which I hope teachers from many points of view can agree. Students should use grapho phonemic information first. Other information can be used to confirm the grapho phonemic information or, in the case of irregular words, can be used to start a problem-solving process.

Ok, except that even “irregular” words have at least one or more “regular” elements, so there is no reason not to alert learners to those as part of the word recognition process, which is what I assume you mean by “problem-solving” process.

Ground for Future Discussion

Point 4 indicates that establishing a “take-off point” or a “self-extending system” is less important than establishing practices that get us there. This misses a very important point. Folks like Mesmer indicate that using decodables is a temporary practice and should end fairly early in the reading process. Shanahan has indicated that those advocating using decodables for years and years and years based on “research” are simply making that stuff up. Ultimately, I think the time comes to take off the training wheels and start riding the bike on your own. We don’t want to under-scaffold or over-scaffold (see figure 1 at the start of this blog). I’ll end with the thought that I would like to continue the discussion with Dr. Goldenberg about the best ways to ensure we promote the creation of engaged readers. I think that is best done by allowing teachers to use both the art and science of reading to achieve that end.

I’m looking forward to the future discussions.

I have no disagreement with any of this. Just the caution that it’s not always clear where the science ends and the art begins, or the other way around, or how exactly to describe the relationship between the two. The invocation of “art”—undoubtedly a part of effective teaching, as it is of effective medical practice, building design, and devising empirical studies—as being in conflict with or a counterpoint or complement to science is not new. Check out C.P. Snow’s famous but not universally accepted statement about the “two cultures” if interested.

A topic with lots of devils, gods, and details, but pretty interesting anyway.

As always, pls…

if you’ve not yet (still free) and or…