Science of Reading: The Podcast

S10 E7: Syntax and comprehension, with Julie Van Dyke, Ph.D.

• Amplify Education

Share episode

In this episode of Science of Reading: The Podcast, Susan Lambert is joined by research scientist and professor Julie Van Dyke, Ph.D., who explains why syntax instruction may be the missing piece in our mission to improve comprehension outcomes for all students. Together, Julie and Susan discuss why syntax is the part of the language system that matters for comprehension, how the same systematicity and rule governance that you find in teaching phonics also exists in syntax, and how explicit syntax instruction could be the next breakthrough in evidence-based literacy education.

Show notes:

Quotes:

"In English, syntax is word order. Syntax is the relationship between the entities in a sentence." —Julie Van Dyke, Ph.D.

"If you want to increase comprehension, you need to be explicit in syntax because that's the part of the language system that matters for comprehension." —Julie Van Dyke, Ph.D.

"Comprehension is the glue between the words. It's the process of gluing the words together, each word as you go." —Julie Van Dyke, Ph.D.

Episode Timestamps:
00:00 Introduction: Syntax and comprehension with Julie Van Dyke, Ph.D.
06:00 Nervousness around syntax instruction
11:00 Comprehension is the glue between words
15:00 The difference between grammar and syntax
19:00 How the brain learns language and how syntax is related to that learning
24:00 Oral language is much less complicated than written language
30:00 Explaining regressions
33:00 The need to be explicit in syntax instruction
36:00 How we develop fluency as syntax
44:00 Closing thoughts: Syntax can move the needle on the nation's report card

*Timestamps are approximate, rounded to nearest minute



[00:00:00] Julie Van Dyke: We still have two-thirds of students who are unable to read what we call proficient, and I think the only way that we can really get that number to be closer to the 95% of what we really want to see is to do something new. And syntax is the new thing.

[00:00:19] Susan Lambert: This is Susan Lambert and welcome to Science of Reading: The Podcast from Amplify. Today we begin a two-part exploration of syntax and its critical role in the comprehension process. To kick off our deep dive into syntax and comprehension, I'm thrilled to welcome Dr. Julie Van Dyke to the show.

As you'll hear, Dr. Van Dyke makes a passionate case that increased attention to syntax is absolutely critical for increasing students' comprehension and improving reading. Dr. Van Dyke holds a joint appointment as an associate research professor at the Institute for Brain and Cognitive Sciences, University of Connecticut, clinical assistant professor at the Yale University Child Study Center, and research scientist at the Yale Yukon Haskins Global Literacy Hub. She's also the inventor and chief scientist at Cascade Reading. On this episode, we cover why syntax is so crucial for comprehension, the systematicity and rule governance that also exists in syntax, and how explicit syntax instruction could be a breakthrough for literacy instruction.

Also, we would love to know your questions on syntax. If you have a question for Dr. Van Dyke or a question about syntax, please share it at amplify.com/SORmailbag. And stick around to the end for a preview of part two of our syntax deep dive. And now here's Dr. Julie Van Dyke.

Dr. Julie Van Dyke. It's so great to have you on today's episode. Thank you for joining.

[00:01:59] Julie Van Dyke: Thank you for having me. It's a real pleasure.

[00:02:01] Susan Lambert: We're going to talk a little bit about syntax in a minute with you, Julie, but before we get there, I would love it if you could tell us a bit about who you are and how did you come into this world of literacy?

[00:02:12] Julie Van Dyke: Uh, yeah, right. So my training is in cognitive psychology and computational linguistics. I was a senior research scientist at Haskins Laboratories for about 22 years. Uh, Haskins has recently reorganized, so I feel like I'm still a Haskins Laboratories person, but the institution has changed. But I did not initially plan on being involved with literacy. I started my career, as I mentioned, in computer science back in the '90's when artificial intelligence was really a thing, not like it is today.

[00:02:44] Susan Lambert: Ya.

[00:02:45] Julie Van Dyke: But back then it meant modeling what the brain does in the way that the brain does it.

[00:02:51] Susan Lambert: Okay.

[00:02:52] Julie Van Dyke: And so we were very interested in the cognitive processes that humans use in order to solve tasks and to use language. And so I began my career studying artificial intelligence and natural language processing. And then cognitive science kind of evolved and I evolved with it. So that's sort of how we, I got into that.

[00:03:12] Susan Lambert: That's so interesting to think about AI way back then.

[00:03:17] Julie Van Dyke: Yeah. Yeah. And then the bridge to literacy, you know, everybody has a journey. So that was my journey. So I moved on to study computational linguistics and cognitive science to Carnegie Mellon in Pittsburgh. I intended to study under Allen Newell, who was one of the founders of cognitive psychology, cognitive science, and had a cognitive architecture, which is a, a theory, an implemented computational theory of how we use language and how we solve problems.

[00:03:45] Susan Lambert: Okay.

[00:03:45] Julie Van Dyke: So I went to Carnegie Mellon in 1993. Just before that, Allen actually passed away. So the program that I was in changed very much because his influence wasn't there. And many of the faculty people moved to other places, and I wanted to continue.

Also, just the field of artificial intelligence was evolving away from really cognitive modeling of things the way that humans do it, to the more engineering-type approaches that you see today, which is where it looks intelligent, but it's not doing it the same way that humans would do it.

[00:04:22] Susan Lambert: Interesting.

[00:04:22] Julie Van Dyke: So I was always interested in brains. So I, I was in Pittsburgh and I wanted to stay in Pittsburgh, so I shifted to cognitive psychology and I studied under Charles Perfetti, Chuck Perfetti, who is a unique person, very influential in the field of literacy, but he's unique because he did research in a lot of allied fields, one of which was the type of sentence processing, formal linguistic, motivated language processing that I came from. And he was able to bridge that world and also the world of literacy and reading. And so it was really coming to his lab that was my first step.

Then 12 years later I had a daughter who ended up having dyslexia and developmental language disability and dyscalculia and ADHD and autism. And it just became, um, you know, all of those personal things and living, having a lived experience with a child who struggles to use language and struggles to read, and here I am. So the universe sort of brought me to this place and I have a unique perspective, and that's what I'm sharing.

[00:05:27] Susan Lambert: It's so interesting when we ask people that question because, you know, nobody wakes up one day and says, "I'm going to study syntax," quite the way that you end up doing it. Right? But all of these things that have come together for you are pretty unique.

[00:05:44] Julie Van Dyke: Well, Susan, I, that's such an insightful question because I think part of what we're looking at now when people are nervous about syntax, I mean, I've been, been doing a lot of speaking lately and trying to encourage people to bring syntax more into their classroom practice.

[00:05:58] Susan Lambert: Yeah.

[00:05:58] Julie Van Dyke: And there is a lot of nervousness and I think that it comes from not really knowing what it is. It's not part of teacher training a lot, and the truth is, if somebody knew how beautiful and systematic the language was, I think you would have people rushing when they wake up to go study syntax.

[00:06:19] Susan Lambert: Maybe by the end of this podcast episode you can convince people the love of syntax and it's so important. And you know, you said something about a bridge, but you actually mentioned to us in the pre-call that you do see yourself as a bridge between sentence study and the reading community. Can you tell me a little bit more about that?

[00:06:41] Julie Van Dyke: Well, it really is, uh, sort of going back to this unique background that I have. So at Haskins for the last 23 years, I was there as the so-called Science of Reading, that we really think about, has evolved. And that's been driven mostly on, we'll just make reference to Hollis Scarborough, my Haskins mentor, um, her Reading Rope, which now everybody knows about.

So you know, you have the reading side and you have the comprehension side of the rope. And the Science of Reading as we typically think about it, really evolved to focus on decoding and word recognition. Haskins is where a lot of the first neuroimaging studies were done to really show that phonics-based interventions can change the way the brain processes language. There were leaders at Haskins, like Hollis Scarborough and Margie Gillis and, and Sue Brady, who were really, you know, developing new ways to teach decoding and phonics in the classroom. But I was always on the language comprehension side in the same building. And so I, you know, I saw all this unfold and became aware that there's just so much on the language side that isn't part of the conversation.

[00:07:51] Susan Lambert: Yep.

[00:07:52] Julie Van Dyke: And I honestly think that as a field, we really weren't ready to hear it yet because there were still so much argumentation and just working out, how do you teach kids to decode? How do you teach kids to learn how to Uh, I don't want to say learn how to read because for me, reading is comprehending.

[00:08:08] Susan Lambert: Mm-hmm.

[00:08:08] Julie Van Dyke: But you know, when we say learn how to read, you know, a lot of times it's really talking about decoding. But I think that now we know how to do that.

[00:08:16] Susan Lambert: Yeah.

[00:08:16] Julie Van Dyke: Right? Like now the issue is just implementation and getting people to adopt these methods. So I think now in time is the perfect time to really start to focus on all that other language side of the equation that we really haven't brought out.

And for that, you need to be able to access these research communities, which just aren't involved with literacy because they don't see that as part of their mission.

[00:08:40] Susan Lambert: Hmm. Yeah, yeah.

[00:08:41] Julie Van Dyke: And I stand in both worlds.

[00:08:42] Susan Lambert: Yeah.

[00:08:42] Julie Van Dyke: So here I am.

[00:08:43] Susan Lambert: Yeah. So here you are. And that's why you're here. I'm so excited about this. So excited to talk about syntax, but you know, this is set in our greater conversation this season about comprehension, and I would love it in your own words, if you could define comprehension for us.

[00:09:01] Julie Van Dyke: Yeah. Well, it's such an important question. So I think it's really important to be clear about the process of comprehension and the products of comprehension. So very often I think that educators are focused on what I would call the products of comprehension, meaning the comprehension questions, the gist statement, the summary, all of those things. In order to do those things, you have to query the results of the comprehension process.

[00:09:31] Susan Lambert: Yep.

[00:09:31] Julie Van Dyke: Right? The process itself is the moment by moment, really millisecond by millisecond action of gluing words together in order to create a meaning. And that is very much a process. It's guided by language knowledge, chiefly syntactic knowledge, and those decisions are happening really within a hundred milliseconds.

And it's just part of the process also that sometimes we can be wrong and we correct it. But ultimately you come up with an actual representation of whatever it is that you were reading, and then you can say, "Oh, so now tell me who did this." Right? So that's a whole separate thing. So your product, your question, or your gist statement or whatever, is only going to be as good as your process was.

[00:10:19] Susan Lambert: Hmm. I love how you said that.

[00:10:20] Julie Van Dyke: Yeah.

[00:10:21] Susan Lambert: I love it.

[00:10:21] Julie Van Dyke: So I've, I've said, just to, to give it a quotable quote, I've said this in some of my talks, that "comprehension is the glue between the words. So it's the process of gluing the words together, each word as you go."

[00:10:34] Susan Lambert: Hmm.

[00:10:35] Julie Van Dyke: And there's just tons of research, this all coming from the language processing community, that we do make these decisions on a moment-by-moment, word-by-word basis.

[00:10:44] Susan Lambert: Hmm. So interesting.

[00:10:45] Julie Van Dyke: So we're not waiting till the end of the sentence or the end of the paragraph.

[00:10:48] Susan Lambert: Right. And this, just making a little connection for our listeners, is why word level, so the bottom part of the rope, right, getting really automatic with words, is so important to the comprehension process because we need to get to that so that we can actually get to the words to glue them together as part of this comprehension process.

[00:11:08] Julie Van Dyke: Yeah. I mean, I think that's right, but I think another big takeaway from language processing is that there's an interaction from the bottom, what I'll call the words themselves, and the top, which is the language.

[00:11:20] Susan Lambert: Okay.

[00:11:21] Julie Van Dyke: So as you are decoding a word.— and, and I'm actually afraid to finish this thought because I don't want anybody listening to believe that I have any, that this sounds like three-cueing. It sounds a little bit like it, and I actually made a specific effort to make sure that we, we distanced this when we discussed this in the Perspectives articles, but what we actually do is we use our language knowledge to make predictions about what's coming.

[00:11:48] Susan Lambert: Okay.

[00:11:48] Julie Van Dyke: So based on your syntax, if you know that, you know, you've got a definite article, "the" you've got an adjective, "happy," you know what's coming next is a noun. So that information is going to help you with the decoding process. If you're even the least bit, not automatic, you can still use that to help you to bootstrap into what that word recognition is.

So that's what happens for somebody who doesn't have, like, automatic decoding. And it happens for all of us when we're coming across a word that we don't know.

[00:12:18] Susan Lambert: Yeah.

[00:12:18] Julie Van Dyke: Right?

[00:12:18] Susan Lambert: That makes sense.

[00:12:19] Julie Van Dyke: So those two things work together, but this is so far away from three-cueing, because three-cueing didn't take into consideration the, the reliance on that language skill.

Like you need to build the language skill. You need to have syntax. You need to have, for the, for decoding, you need to have phonological knowledge. And it's not just guessing, it's not just queuing, it's really using the information you have available to make the best product, the best guess of what's going on.

Again, I don't even want to use that word, but...

[00:12:54] Susan Lambert: I understand.

[00:12:54] Julie Van Dyke: It's a slippery slope because this is not three-cueing. Please don't let any know that. Nobody's quoting me saying I'm supporting this. But it is a process grounded in language knowledge, whether it's phonological knowledge or whether it's tracing knowledge.

[00:13:08] Susan Lambert: Yeah, that makes sense. And it's really funny because this reminds me, I think I have this right, but the very first book chapter where Dr. Scarborough introduced the Reading Rope, there is a quote in there where she explicitly says something about, "Don't think of these as separate processes, they develop together and they support each other in the development."

[00:13:34] Julie Van Dyke: Yeah.

[00:13:34] Susan Lambert: The top down, bottom up kind of idea.

[00:13:36] Julie Van Dyke: Yes.

[00:13:36] Susan Lambert: That they're both important.

[00:13:38] Julie Van Dyke: Yes, that's right. Yes. I mean, Hollis, I, I think, you know, she can speak for herself and I know that she has, um, I think that she was ultimately quite surprised that the Reading Rope has ended up being as influential and used the way it is, because she never intended it that way.

[00:13:53] Susan Lambert: Yeah.

[00:13:53] Julie Van Dyke: But, uh, yeah.

[00:13:54] Susan Lambert: Yeah, it is a helpful model. Okay, before we get into more of the details of syntax, I know that people often conflate grammar and syntax. Can you help us understand the difference between the two?

[00:14:09] Julie Van Dyke: Yeah, I mean, I just think of grammar as a larger term, so my answer is informed by linguistic training, like formal linguistic training.

[00:14:16] Susan Lambert: Yep.

[00:14:17] Julie Van Dyke: So I'm not thinking specifically about English or what you would put into your ELA block in school or whatever. I'm thinking about language in general, meaning the entire world's language.

[00:14:28] Susan Lambert: Okay.

[00:14:29] Julie Van Dyke: So when I talk about the grammar of a language, I'm talking about all of the knowledge that includes not just syntax but also morphology, even phonology, even discourse level things.

[00:14:40] Susan Lambert: Oh, okay.

[00:14:41] Julie Van Dyke: It's all those rules that define that language.

[00:14:45] Susan Lambert: Yep.

[00:14:45] Julie Van Dyke: And then syntax is sort of a sub-piece of that that is specific to the argument structure of a sentence. So, like, who did what to whom of a sentence. And in English, we use word order for that but other languages absolutely do not. So it's not even syntax, like, you know, subject, object, verb. The way we think about it in English.

[00:15:06] Susan Lambert: Yep.

[00:15:06] Julie Van Dyke: It's more abstract. And it has to do with who did what to whom or what the relationships within the particular sentence are. And it's very much at the sentence level.

[00:15:18] Susan Lambert: Yeah. So it's, is it right to say that a syntactical structure is the word order?

[00:15:25] Julie Van Dyke: Only in English? In English, we use word order to communicate who did what to whom.

[00:15:31] Susan Lambert: Okay.

[00:15:31] Julie Van Dyke: But in other languages, you could have other word orders. So like Japanese puts the verb at the end. Right?

[00:15:37] Susan Lambert: Yep.

[00:15:37] Julie Van Dyke: And that has really important implications for how you process those. How a Japanese brain is processing language compared to an English brain.

[00:15:46] Susan Lambert: Ooh.

[00:15:46] Julie Van Dyke: The Japanese brain... like in English, you get your subject, you get your verb, and you know immediately, "Okay, I'm getting an object next." Right?

[00:15:53] Susan Lambert: Okay.

[00:15:54] Julie Van Dyke: But in a, for a Japanese brain, you would get your subject, you don't know what the verb is, but now you have your object. So at a moment in time, you don't even know what the relationship is between those things.

[00:16:03] Susan Lambert: Okay. And I'm, I'm going to put a, like a point in that one. So essentially what we're saying is syntax is word order, but languages have different syntactical structures based on the language.

[00:16:15] Julie Van Dyke: Can I just, I, I, I don't want to fully agree with that, can I...

[00:16:19] Susan Lambert: Perfect. I love that.

[00:16:20] Julie Van Dyke: I want to say that in English, syntax is word order, but syntax is the relationships between the entities in a sentence. So, in English, you, will use word order to say what is the subject? What is the object? Right?

[00:16:39] Susan Lambert: Okay. Got it.

[00:16:40] Julie Van Dyke: So, you know, the sentence is: Susan interviewed Julie. Right?

[00:16:44] Susan Lambert: Mm-hmm.

[00:16:44] Julie Van Dyke: So, Susan is the subject, so that's positioned first. Julie is the object, so Julie is positioned after the verb, right?

But in Japanese, we would be saying, Susan Julie interviewed, right? Because the language uses other things. So it's not just Susan Julie interviewed. How is your brain going to know that? The language now adds morphological markings. So it's like there'll be certain markings that say, "Oh, I have these two nouns here. I've got to know who did what to whom." So the language helps you by giving you markings on that.

So different languages work differently. It's very abstract. But, but the reason, the reason, I mean, why, why bother with this? Like, I'm just being academic and trying to be fancy. Right? No. The reason it matters is that we have so many multilingual learners in our classrooms now to be aware that those languages that are in those brains in your classroom are processed in a different way from English. And so this is even more impetus for bringing syntax out in instruction in an explicit way. Because the child sitting back there in the left hand corner might not even be thinking about language the way you're thinking about it.

[00:17:57] Susan Lambert: Right. Right. Yeah. Oh, that's so important.

[00:17:59] Julie Van Dyke: It's just underscoring why it's really important for syntax to be explicit.

[00:18:05] Susan Lambert: Yeah.

[00:18:05] Julie Van Dyke: I think that, you know, we now as a field, as a literacy field, understand the importance of being explicit about phonics and phonology.

[00:18:13] Susan Lambert: Yep.

[00:18:13] Julie Van Dyke: And even morphology. And you know, we should be boosting it up even to the syntax level, but except for syntax, for whatever reason, makes people nervous.

So we need to find ways to get over that.

[00:18:27] Susan Lambert: We're going to change that. We're going to change that in one podcast episode.

[00:18:31] Julie Van Dyke: I think that's what we need to do. Absolutely.

[00:18:33] Susan Lambert: So I would love then to take a little step back now. You mentioned a little bit about, you know, the brain learning language, but can you help talk us through how does the brain learn language and then how is syntax related to that?

This feels complicated.

[00:18:49] Julie Van Dyke: Yeah. So, I mean, you're asking a huge question, and in fact, when I teach this in my, my psycholinguistics classes, you know, it takes about, you know, six lectures in order to fully cover this. So, uh, so.

[00:19:01] Susan Lambert: On your marks, gets set, go, right?

[00:19:04] Julie Van Dyke: So let's just highlight some core principles and focus on the syntax component of this, because...

[00:19:10] Susan Lambert: That sounds great.

[00:19:11] Julie Van Dyke: ...how we learn language— like, we learn language, we're also learning semantics and phonology and like all of the different levels of languages are all at the same time.

[00:19:18] Susan Lambert: Yep.

[00:19:18] Julie Van Dyke: But the main thing to know is that our brains are incredibly advanced statistical processing machine. And we can track statist, even babies, a baby who can't even talk to you yet, is able to track statistical regularities in the sounds around them.

[00:19:37] Susan Lambert: Wow.

[00:19:37] Julie Van Dyke: And this is work that has experimental work that has been around for 30 years or more. You know, the question is, how does a baby understand where the spaces are? Like it's hearing oral language. We don't speak with pauses. In fact, when you look at the spectograph that shows the language, very often our words are run together, you can't really tell like which word is which. But yet a baby is able to pick out those spaces in between the sounds. And it does it by tracking the statistical regularities. So the transition within a word like, pretty, is regular. Like every time I get a "pri," I'm often going to get a "ta," "t" after it. "Pri t," right?

[00:20:22] Susan Lambert: Mm-hmm.

[00:20:22] Julie Van Dyke: Because that's a word. So that's like systematic. But the transitions between words are not regular because you can have any word. So you could have, like, the end of the word followed by any other word. So there's no regularity there. So when you track those systematic pieces, you begin to get chunks. Like in the word, pretty, "pre" and then "t," right?

[00:20:45] Susan Lambert: Mm-hmm.

[00:20:45] Julie Van Dyke: Or however you want to pronounce it. I'm probably not doing a very good job. So pretty, "pre" and "t." That is a unit in the world. And the baby recognizes it because on some percentage of the time that you hear the sound "pre," you're going to hear "t."

[00:21:00] Susan Lambert: Okay.

[00:21:00] Julie Van Dyke: Right?

[00:21:01] Susan Lambert: Yep.

[00:21:01] Julie Van Dyke: Pretty. So you can figure that out in between words. It's random, right? Because you can have pretty followed by any word. You can have any sound because pretty could be followed by dogs and cats and girls and ribbons and bows. Like it could be anything. Right?

[00:21:14] Susan Lambert: Right, right.

[00:21:14] Julie Van Dyke: There's no regularity there. So that's how it figures out the word, pretty.

The exact same method is done to pick out syntax. We start to see that things in the world like nouns, are so, are depends on your language, right? So in English, the noun would be followed by a verb. And the baby is able to sort of start to learn. We call them transition probabilities.

How often a noun is going to be followed by a verb. How often a determiner is being followed by an adjective and then being followed by a noun, that kind of thing. Because these things, these syntactic categories relate to things in the world, you know, so nouns are things and adjectives are properties of things and all of that.

So it can notice those categories and then notice the relationships between them. You know how they occur together, and you start to build up statistical chunks.

[00:22:08] Susan Lambert: Okay.

[00:22:08] Julie Van Dyke: Then, sort of, it just keeps going. And eventually you learn the rules that, you know, a noun is always followed by a verb. So that's my rule, right? A determiner is always followed by a noun. Like that's my rule. It just sort of builds up that way. So it's really complicated. I'm not sure if that's something that is helpful or not.

[00:22:27] Susan Lambert: It's helpful. And you know what it reminds me of is, we talk a lot about the importance of oral language development.

[00:22:35] Julie Van Dyke: Yeah.

[00:22:35] Susan Lambert: And how it impacts later reading comprehension and written composition. It seems to me that this syntactical sort of process that we're, we're learning when we're young, relates directly to that?

[00:22:53] Julie Van Dyke: Well, yeah. I mean, this is, again, this is really so important. We learn what we're exposed to basically, is what I just described in a really complicated way, right?

Like you're exposed to sounds that go together. You're exposed to grammatical categories that go together. Okay? And you remember them and you create these kind of like patterns, like, you know, eventually your pattern is going to become a rule if you see the same pattern that often.

[00:23:17] Susan Lambert: Yep. Okay.

[00:23:18] Julie Van Dyke: The problem is, is that the language we're exposed to is not the same as the language that's in the classroom in a written text.

[00:23:27] Susan Lambert: Mm-hmm.

[00:23:27] Julie Van Dyke: Right? Like the language that you use around your dinner table to talk about what you did today, it's not full sentences, it doesn't have relative clauses in it, it doesn't have subordinate clauses in it, right? It's vastly different from the language that you have to comprehend when you're sitting in third grade or even fourth grade, you know, trying to learn something.

So that's the mismatch. So the importance of oral language, you have to like, really intensify exposure and exposure to language that is different from what you're going to get just casually talking. That's the way to really kind of build language ability that's going to help towards advanced reading.

[00:24:05] Susan Lambert: Okay. So this brings me to a next question, but I want to summarize some of the things that I think I heard, and hopefully this will come out in a, in a way that makes sense.

So, first of all, the exposure matters; oral language is much less complicated than written language; and when we think about this— earlier, you talked to us a little bit about comprehension processes, which happen while we're reading this sort of word by word level— um, sentences become really important as we're cruising through that text.

And I think one of the things that you really push back on is the idea that comprehension is only at the level of the text or only at the level of the paragraph, but because these comprehension processes are super important to that product, syntax is really important that the level of the sentence.

[00:25:11] Julie Van Dyke: Yeah, I mean, let me just give you an example.

So, you know, we can go to the lab and you know, I can hook up the eye tracker and I can watch how people read.

[00:25:19] Susan Lambert: Okay.

[00:25:20] Julie Van Dyke: And you don't even have to do it in reading. You can do it also in what we call visual worlds, where you've got a few things on the, on the screen and you look at where the eyes are looking. Right?

[00:25:29] Susan Lambert: Okay.

[00:25:30] Julie Van Dyke: And you have statistics about how the language works, how words work together. So if I said to you, um. Just a minute. Let me think about this. I want to get my example, right. This is always my challenge, is coming up with examples on the fly.

[00:25:48] Susan Lambert: On the fly.

[00:25:48] Julie Van Dyke: I'm not very good at it. Yeah. Let's take just the case where I'm watching your eyes, right? And you've got stuff, this is how we do experiments, right? You set up sort of a, a table with things on it. Right? We want to see how people are going to use, are going to process things at a word-by-word level. Okay? So I've got a table and it's got, you know, a stick and, and, and a bear and just random things there, right?

And I say to you, poke the bear with the stick. Okay?

[00:26:14] Susan Lambert: Okay.

[00:26:14] Julie Van Dyke: So the classic example is when you have a bear and then a stick and then a bear holding a stick that's on your table, okay?

[00:26:24] Susan Lambert: Okay.

[00:26:25] Julie Van Dyke: So if I say to you, poke the bear with the stick. Are you going to look at the bear with the stick, or are you going to look at the stick first?

[00:26:32] Susan Lambert: I don't know. What am I going to do?

[00:26:34] Julie Van Dyke: The minute that you hear poke, you're going to look at the stick.

[00:26:38] Susan Lambert: Okay.

[00:26:38] Julie Van Dyke: Because the word poke includes in its meaning that you're poking somebody with something.

[00:26:45] Susan Lambert: Okay.

[00:26:46] Julie Van Dyke: You see?

[00:26:47] Susan Lambert: Mm-hmm.

[00:26:47] Julie Van Dyke: So your knowledge of that guides you even before you've heard the word stick.

[00:26:53] Susan Lambert: Hmm. Okay.

[00:26:54] Julie Van Dyke: Okay?

[00:26:55] Susan Lambert: Yeah.

[00:26:55] Julie Van Dyke: You're already looking for, "what's the thing I'm going to poke with?"

[00:26:59] Susan Lambert: Yep, I got you. Yep.

[00:26:59] Julie Van Dyke: Okay. You see?

[00:27:01] Susan Lambert: Yep.

[00:27:01] Julie Van Dyke: So I say to you the word poke, your brain is already thinking about "where's my instrument? Where's my thing I'm going to poke with?"

[00:27:08] Susan Lambert: Mm-hmm.

[00:27:08] Julie Van Dyke: Right? Even though I have a picture of a bear with a stick. Right?

[00:27:12] Susan Lambert: Right.

[00:27:12] Julie Van Dyke: But already you're looking for your stick. Okay?

[00:27:15] Susan Lambert: Yeah.

[00:27:16] Julie Van Dyke: So now compare that with the idea of, the sentence now is find. So if the word is find, you're not going to look for the stick, because you don't find things with something. Find the cat with spots, for example. Right? So you're not using an instrument to do the finding, you're finding something that has some other kind of property. Maybe it's not the best example.

[00:27:39] Susan Lambert: I get it though. Yep.

[00:27:40] Julie Van Dyke: In the reading context, the good example is when you have an object relative clause. So the man that the girl liked.

[00:27:49] Susan Lambert: Mm-hmm.

[00:27:49] Julie Van Dyke: Right? So that's hard because the man that the girl, we're not used to having two nouns next to each other, right? You're not predicting that.

[00:27:58] Susan Lambert: Mm-hmm.

[00:27:58] Julie Van Dyke: It's unusual to see that. If it was the man the girl liked, right? You're more accustomed to seeing a verb that comes after the subject, right?

[00:28:07] Susan Lambert: Okay.

[00:28:08] Julie Van Dyke: So if I'm measuring your eye movements, when you have the man the girl liked, you're going to have difficulty there and we can measure it.

[00:28:15] Susan Lambert: I see. Yep.

[00:28:15] Julie Van Dyke: Because it's not what you're predicting. So all this is to try to help you to understand that at a word level, we have an expectation about what's going to come next. And it's reflected in a syntax, right? So once I have a certain noun, I'm going to be looking for a certain type of a sentence structure. I'm going to look for a certain type of thing to follow it.

[00:28:38] Susan Lambert: Mm-hmm.

[00:28:39] Julie Van Dyke: And I'm already making those expectations.

[00:28:41] Susan Lambert: Mm.

[00:28:42] Julie Van Dyke: And we can measure this. So it really is happening on a word-by-word base.

[00:28:46] Susan Lambert: Yeah.

[00:28:46] Julie Van Dyke: So if you're jumping already to the sentence, you've missed all of that processing and potentially all that confusion that could have happened at the sentence level.

[00:28:57] Susan Lambert: Yeah, that makes sense. And so as a reader, when I am reading a paragraph and it's like, "Wait, that doesn't make sense." Partly it could be those complicated sentence structures that make me have to go back and reread and say, "Hey, wait, wait, what? I wasn't expecting that. I've got to go back and reread the sentence to make it make sense."

[00:29:19] Julie Van Dyke: Mm-hmm. Mm-hmm. That's what we do. But interestingly, there are ambiguities in the language, right? And so often there are instances where our brains will commit to a certain meaning and then revise it without you even realizing.

[00:29:35] Susan Lambert: Wow.

[00:29:35] Julie Van Dyke: So when we measure eye movements, we see regressions all the time. You don't even know.

[00:29:40] Susan Lambert: And regressions, can you explain for listeners?

[00:29:43] Julie Van Dyke: Meaning, meaning a look back, A look back. A look back.

[00:29:45] Susan Lambert: Right. Yep.

[00:29:45] Julie Van Dyke: So, um: I believe the man was late. Simple sentence, right? I believe the man was late. Now, if we think about the syntax of that, well first let's think about the meaning of that. Do you believe the man?

[00:29:58] Susan Lambert: No, I believe the man was late.

[00:29:59] Julie Van Dyke: Exactly. So what you're believing is the whole clause. The man was late. But you don't know that until you get to the period. But your brain has already made the connections. There's a moment in time where your brain says, "Oh, usually when I see, I believe the man," like that's a common thing in language, right?

[00:30:22] Susan Lambert: Right. Yeah.

[00:30:22] Julie Van Dyke: It could be that. That's what the sentence is. So the initial understanding is, I believe the man. And then the sentence continues and your brain has to revise that. We call it a re-analysis. because you've actually made the wrong connection. You've said, "Oh, I believe the man." But really, that's not what the sentence is. I believe the man was late. You see?

[00:30:44] Susan Lambert: Hmm. Yep.

[00:30:44] Julie Van Dyke: And we see that in young children, they have a harder time making those kinds of reanalysis. So young children might actually continue to believe that it's the man that you're believing. It's not, they don't know what to do with the "was late" part.

[00:30:56] Susan Lambert: Oh. Yeah.

[00:30:58] Julie Van Dyke: All of that is syntactic.

[00:30:59] Susan Lambert: Yeah.

[00:30:59] Julie Van Dyke: Right? Because it's, you, what are you believing? You're believing just that noun phrase? The man? Or you're believing the whole clause. The man was late. The only way to really refer to that as a unit, I mean, we say, okay, it's an embedded clause, right? But that's a syntactic thing. That's a whole idea.

And we do that, like if I measure your eye movements on those kinds of sentences, you do have a little bit of a delay and a looking back. So, I believe the man. And then when you get, was late, you can measure the reanalysis. You see eye movements that go back. You see slowing them down. But you're not conscious of that. Right?

[00:31:36] Susan Lambert: Right.

[00:31:36] Julie Van Dyke: Because our brains know how to do that. That's the statistical linguistic processing that we do.

[00:31:41] Susan Lambert: Yeah.

[00:31:41] Julie Van Dyke: But I'll say, the brains know how to do it fast when they have expert skilled language ability, right? So if you have a child who's struggling because English is not their language, they might have a lot harder time doing that.

[00:31:57] Susan Lambert: Yeah. Right.

[00:31:57] Julie Van Dyke: So this is again, just sort of bringing out these aspects of why syntax matters,. Because if you don't have the basic language ability to do these things that are automatic, then you're going to have a whole lot of problem when you're trying to actually measure, "oh, well, so what did they understand?" Like that's the product part.

[00:32:14] Susan Lambert: Yep, yep. And it's a great example of the differences between oral language and written language then.

[00:32:20] Julie Van Dyke: Yes, yes.

[00:32:20] Susan Lambert: And the complexities of written language, which is why you are rallying for explicit instruction in syntax.

[00:32:29] Julie Van Dyke: Yeah. Yeah, I mean, from my perspective it's just so basic. Because we are explicit about phonology because that's the part of the linguistic system that matters for decoding.

[00:32:43] Susan Lambert: Yep.

[00:32:43] Julie Van Dyke: If you want to increase comprehension, you need to be explicit in syntax because that's the part of the language system that matters for comprehension. So to me, it's just a simple equation. And I think that the, like I said, the reason that we aren't doing that is because people aren't, they don't feel comfortable in it.

[00:33:02] Susan Lambert: Yeah.

[00:33:03] Julie Van Dyke: And perhaps they don't see the importance of it.

[00:33:05] Susan Lambert: And so maybe given all that and all that we talked about in, in all its glory and complexity...

[00:33:12] Julie Van Dyke: yeah. I feel bad. I gave some like really difficult, uh, examples that are just going to make people want to flee. I feel like we're not meeting our goal of getting people to wake up and want to study syntax in the morning. But, um.

[00:33:23] Susan Lambert: But if they're still with us, which I believe they are with, still with us...

[00:33:26] Julie Van Dyke: The thing is, is that people should know that it's systematic, it's rule-based. Compare how you felt before you started to learn how to teach phonics. Right? You know, it's like it was all over the place. Right?

[00:33:37] Susan Lambert: Yep.

[00:33:37] Julie Van Dyke: But now we just need the same kind of professional development and support, so that people can see the beauty and the structure in it.

[00:33:44] Susan Lambert: Yeah.

[00:33:45] Julie Van Dyke: So I hope we'll get there.

[00:33:46] Susan Lambert: I agree with that. Do you have any specific recommendations for any educators that are listening right now of things they can do outside of, "wake up every morning and get excited to learn about syntax?"

[00:33:58] Julie Van Dyke: Yeah, I mean, well, there's a lot of recommendations that I've been making. You know, I've been speaking quite a lot in different venues, so a lot of your listeners may have heard me. Even, I also spoke during your comprehension week, and so there were some recommendations that I made there. I will say that I developed a system called Cascade Reading, which is an AI-based artificial intelligence-based system, which will take any sentence and format it into a visual format that gives you cues to what the syntactic structure is.

[00:34:31] Susan Lambert: Okay?

[00:34:31] Julie Van Dyke: And this is something that can really support teachers because as I said, teachers may have a really hard time feeling comfortable with syntax, and this helps them to know where the syntactic units are. Because one of the things that's really important for syntax is to help students to be able to sort of hear and feel the rhythm in the language.

[00:34:51] Susan Lambert: Mm.

[00:34:51] Julie Van Dyke: Because the rhythm in the language is the syntactic structure.

[00:34:54] Susan Lambert: Yep.

[00:34:55] Julie Van Dyke: And so, you know, you may have heard people talk about scooping, right?

[00:34:59] Susan Lambert: Yep. Showing the phrasing within a sentence, right? Yeah.

[00:35:01] Julie Van Dyke: That's right. Yeah, that's right. So the problem there is that it's sometimes not so simple where you put those scoops.

How do you know where the end of the phrase is?

[00:35:09] Susan Lambert: Right.

[00:35:10] Julie Van Dyke: That really is being able to pick out the syntactic unit. And so Cascade will do that automatically because the syntactic units are on each line separately. So that's one thing that's really very helpful.

[00:35:21] Susan Lambert: And what's the relationship between that syntactical phrasing or that phrasing both in reading and then, I would imagine—this probably sounds like a dumb connection to you. You're like, of course it is—that's related to fluency then, right?

[00:35:36] Julie Van Dyke: That's right. Yeah.

[00:35:36] Susan Lambert: That's how we help develop fluency is syntax.

[00:35:39] Julie Van Dyke: Right. You just sort of closed the loop, right? So in order to read fluently, you need to have syntactic knowledge. You need to know where those constituents are.

[00:35:47] Susan Lambert: Yep.

[00:35:47] Julie Van Dyke: Fluency is marked by natural phrasings, which includes pauses at constituent boundaries. And if you don't know what they are, you're going to be hesitating.

[00:35:57] Susan Lambert: Right.

[00:35:57] Julie Van Dyke: If you have trouble with language in any way, you're going to be hesitating. So I did a study with a colleague of mine, Mara Green, which actually knocked my socks off because I just didn't think it was going to work.

So we did this with kids with developmental language disability, DLD, so Tiffany Hogan has spoken a lot about that. I know that you've had her on your podcast.

[00:36:15] Susan Lambert: Yep.

[00:36:16] Julie Van Dyke: These kids have a real difficult time with syntactic structures. And so we did this study where we asked students to read a sentence, but we gave them a sentence. We read it to them. We gave them the actual phrasing. They had a model of what the sentence was. All they had to do was parrot it back.

[00:36:34] Susan Lambert: Hmm.

[00:36:35] Julie Van Dyke: And they could not do it in a way that sounded fluent.

[00:36:37] Susan Lambert: Hmm.

[00:36:38] Julie Van Dyke: Right. Because they just don't access that syntactic structure to know where the pauses are.

[00:36:45] Susan Lambert: Wow.

[00:36:45] Julie Van Dyke: So bringing out those constituent boundaries, which is where the phrasing changes when you, when you're reading fluently, is that knowledge of syntax. So we have a study with Cascade where just reading in the cascade, it both improves comprehension, but it helps students to hear how the sentence should sound. Because seeing that format is training them to know where the constituent boundaries are.

[00:37:14] Susan Lambert: Mm. I mean, another just sort of easy connection here is the importance of teacher modeling then.

[00:37:22] Julie Van Dyke: Mm-hmm. Yeah.

[00:37:22] Susan Lambert: At the sentence level.

[00:37:23] Julie Van Dyke: Yeah, that's also right.

[00:37:24] Susan Lambert: Hmm. I want to mention the winter edition of IDA;s Perspectives, and it was actually a two-part. This would've been, is it the winter edition?

[00:37:37] Julie Van Dyke: We have winter and spring, yes.

[00:37:38] Susan Lambert: Okay. Winter and spring. Okay.

[00:37:40] Julie Van Dyke: Yeah.

[00:37:40] Susan Lambert: Two-part series called "Syntax Comes First."

[00:37:43] Julie Van Dyke: Yes.

[00:37:44] Susan Lambert: Which I got super excited about when I saw the first one, and that's when I talked to you and then you're like, wait, there's another one coming!

[00:37:50] Julie Van Dyke: Yeah. Yeah.

[00:37:52] Susan Lambert: How did that happen and why did you think it was important to bring those two editions?

[00:37:57] Julie Van Dyke: Yeah. Well, how that it happened is, is a great question. I want to just give a real shout out to Margie Gillis and Nancy Eberhardt. So, Nancy Eberhardt is the co-editor of Perspectives. She sits on the editorial board.

[00:38:11] Susan Lambert: Mm-hmm.

[00:38:12] Julie Van Dyke: And, uh, Margie Gillis is my longtime colleague from Haskins Laboratories and she runs Literacy How, which is, uh, she does coaching and she just won a career award at IDA. We just came back from IDA. I've known her for a long time, and I've expressed to her for many, many years my desire to get all of this language knowledge, language science knowledge into the literacy conversation.

[00:38:38] Susan Lambert: Yeah.

[00:38:39] Julie Van Dyke: And Margie and Nancy were, I think, leaders in the idea of bringing the importance of syntax for early education out. They have a book that they wrote as part of the Literacy House series on syntax. So, yeah, you know, it's really this idea that there is so much knowledge about how the brain processes language, and it's not getting into the so-called Science of Reading discussion. So I mentioned this to Margie and she introduced me to Nancy and over here in Connecticut, we had a nice cup of tea, uh, in a cafe in Darian.

Nancy went home with lots of new information in her brain and she happens to sit as, as I mentioned, on the editorial board and the co-editor in chief of Perspectives, and she really made this happen. So she, thank you to Nancy and to Margie ultimately, but she gave me a platform to try to bring some of this information into the community. And I took advantage of it by inviting many of my colleagues who do language science, but are not part of the literacy community to write articles that are very accessible, you know, targeted for teachers.

[00:39:46] Susan Lambert: Yeah.

[00:39:47] Julie Van Dyke: But also talking about their science and how it's relevant for reading and language comprehension. So we have a collection of 17 articles in there. There are articles by, I would say, five language scientists who you won't see writing anywhere else about their language, language research.

But we also wanted it to be very relevant and accessible for teachers, you know, for their classroom practice. So we also have classroom related articles written by Nancy Eberhardt and Margie Gillis, and also Nancy Hennessy and Julie Solomon. And then my co theme editor is Kelly Powell Smith. She did a tour, the force, um, summarizing the various ways of assessing syntax.

She has a two-part article there, including a table that summarizes all of the available assessments. So, and then the other thing I'm really excited about is that there is a six-page resource section at the end of the spring edition that includes both practical recommendations for what teachers should know in terms of professional development, and then suggestions for what they should do in the classroom.

And then we also have two pages of resources. And the resources section we're continuing to develop and I'm hoping to have that available online on my personal website, sometime in between the semesters. I'm hoping it's a little project, but I think I'm just really trying to use it as a vehicle to get more knowledge about why it's important to teach syntax, how to teach syntax and really bring together all the resources that are available because there are people who are doing really great work in terms of bringing syntax into classroom practice.

[00:41:24] Susan Lambert: That's awesome. And for listeners, check out the show notes, because we're going to provide you some links.

[00:41:28] Julie Van Dyke: Yeah and can I just say about that? So I, yeah. I've now sort of unintentionally be, become an advertisement for IDA, but there is an option for a free six-month trial membership that teachers can get.

So I know that teachers, you know, they're asked to, to spend their own money on a lot of things, and IDA is a, you know, an important organization, it's worth your support.

[00:41:48] Susan Lambert: Yeah.

[00:41:49] Julie Van Dyke: But you can do this thing for a free six-month membership. That's also an option.

[00:41:53] Susan Lambert: Oh, I love that.

[00:41:53] Julie Van Dyke: It's on their membership website.

[00:41:55] Susan Lambert: That's awesome. Thank you for that. And I mean, I understand now since that has come out, and maybe even before, you are a very busy lady. Speaking, doing webinars, podcasts, and doing all the things to highlight the importance of syntax. Isn't that right?

[00:42:11] Julie Van Dyke: Yeah, yeah. You can find me online in a lot of different places.

You know, I'm, I'm really interested in supporting teachers in bringing this into their classroom practice, and I'm very aware that it's difficult and burdensome to think about doing something new.

[00:42:27] Susan Lambert: Yeah.

[00:42:27] Julie Van Dyke: Especially if you don't feel well-trained in it.

[00:42:31] Susan Lambert: Yep.

[00:42:31] Julie Van Dyke: And so, you know, I'm trying to really point to resources, um, that can help make that an easier transition.

[00:42:37] Susan Lambert: And I wonder how this makes you feel as a former researcher to finally see, what did you say, 20 years or so?

[00:42:44] Julie Van Dyke: Yeah.

[00:42:44] Susan Lambert: To finally see your work come to light and starting to make a difference for both awareness and, you know, soon, impact in the classroom. That's kind of cool.

[00:42:55] Julie Van Dyke: Yeah. I mean, we've got a lot, a, a long way to go, you know. I mean, it, it, the thing that I keep coming back to is we've made so much progress in terms of knowing how to teach reading. And doing it in a way that is aligned to the way the brain expects to learn, right?

[00:43:12] Susan Lambert: Yeah.

[00:43:12] Julie Van Dyke: Like, you know, we know that brains care a lot about phonological information, right? And now we're using phonological information to leverage learning on the word side, right?

[00:43:22] Susan Lambert: Mm. Mm-hmm.

[00:43:23] Julie Van Dyke: So now we need, the next phase, is to learn, learn how to use syntactic information to leverage that learning for comprehension. And I'm a little bit, um, you know, we need to move the needle on the Nation's Report Card. Right? We still have two-thirds of students who are unable to read what we call proficient.

[00:43:46] Susan Lambert: Mm-hmm.

[00:43:46] Julie Van Dyke: And I think the only way that we can really get that number to be closer to the 95% of what we really want to see is to do something new. And syntax is the new thing. That is the part that we need to really develop because it's the whole other part of learning that we just haven't really been focusing on so far. So, so this is a beginning, you know. I think we, we've still got a ways to go, but I think we're headed in the right direction.

[00:44:11] Susan Lambert: That's awesome. Well, as we wrap up, do you have any other closing thoughts you'd like to share?

[00:44:18] Julie Van Dyke: Uh, I don't know. I think the, the main thing is to just not be afraid of bringing this new perspective into your classroom practice.

[00:44:25] Susan Lambert: Hmm.

[00:44:26] Julie Van Dyke: I completely get it that nobody wants to learn a whole new, a whole new thing. And syntax is, you know, I and, and I'm sorry if I did a lousy job of ...

[00:44:35] Susan Lambert: Oh, you did great.

[00:44:35] Julie Van Dyke: ...Portraying how un-difficult it can be. But the thing to take home is that it just really is rule-governed and systematic. The same systematicity and rule governance that you find in teaching phonics, it exists in syntax. It's a little bit different because we don't have this as a whole topic that we could discuss. We don't have the same kind of laid out scope and sequence because you don't have an end point to syntax. You're just trying to build exposures to the language.

[00:45:06] Susan Lambert: Right.

[00:45:06] Julie Van Dyke: And build sort of the cognitive gymnastics of handling novel sentences, but that's the beauty of the language. So anything I can do to your listeners to try to make it easier and to provide resources to help students really see the beauty and the flexibility of language, people can be in touch with me.

[00:45:27] Susan Lambert: That's awesome. And you know, just a little tease is that we have asked Nancy Eberhardt to join us on a session, and so she will also be following up with an episode of her own. So we're really excited.

[00:45:39] Julie Van Dyke: That's great. Yeah.

[00:45:40] Susan Lambert: Well, Julie, thank you again for joining. This has been really, really fun. I learned some new things and I appreciate your time and congratulations again.

[00:45:48] Julie Van Dyke: Thank you. Thank you.

[00:45:51] Susan Lambert: That was Dr. Julie Van Dyke. She holds a joint appointment as an associate research professor at the Institute for Brain and Cognitive Sciences, University of Connecticut, clinical assistant professor at the Yale University Child Study Center, and research scientist at the Yale UConn Haskins Global Literacy Hub.

She's also the inventor and chief scientist at Cascade Reading. Take your learning to the next level by checking out the show notes where you'll find more resources, including Dr. Van Dyke's recent Science of Reading webinar, Finding the Missing link in Reading Comprehension. Remember next time we're sticking with Syntax, I'll be joined by Nancy Chapel Eberhardt for some analysis of this episode, as well as some strategies for bringing these ideas into the classroom.

[00:46:41] Nancy Chapel Eberhardt: So I think one of the biggest ways that syntactic knowledge can help us with our reading comprehension is through this parsing of the language, which is basically helping us with prosody and we know that prosody will help us with our fluency with reading.

[00:46:58] Susan Lambert: That's coming up. Meantime, check out all that's been happening on the Beyond My Years podcast. After recent episodes on cell phone policies and teaching styles, host, Ana Torres is joined by leading math educator, Keri Brown, for a conversation about getting students to share their thinking.

[00:47:17] Keri Brown: I run my reading centers a lot like how I run my math centers. It's very routine. I have them talking in sentences. I have them, you know, very independent.

[00:47:29] Susan Lambert: That's available now in the Beyond My Years feed. We'll have a link in the show notes.

Finally, if there's something you'd like to know from Dr. Van Dyke or a question you have about syntax or comprehension, please visit amplify.com/SORmailbag. We'd love to tackle it on an upcoming episode. Science of Reading: The Podcast is brought to you by Amplify. I'm Susan Lambert.

Thank you so much for listening.