Listen or read the following collection of Scott’s interviews with leading experts and learn from their studies.

Dr. Timothy Shanahan

Dr. Timothy Shanahan is Distinguished Professor Emeritus at the University of Illinois at Chicago. He is past president of the International Literacy Association and a leading member of the National Reading Panel.

What is your background?

My background is as a school teacher (I taught first and third grades and remedial reading). From there I went on and got a Ph.D. in Education, largely working with psychologists (experimental, developmental, clinical) and linguists (including working especially closely with Dr. Richard Venezky who first proved the morphemic aspects of the English spelling system) in that training. In the doctoral program that I was in, most of the students had master’s degrees in psychology and I was the only one with a teaching degree or background so I had to pedal especially hard to try to keep up.

(Followup) My own understanding of the English system dictates that words are spelled according to meaning first, regardless of pronunciation. Analyzing disyllabic words according to phonics rules gives you: ac/ tion rather than act + ion. The ac/ tion obscures the morphology of the word. What is your understanding of phonics as it relates to morphology?

You are correct on the linguistics of the matter.

What [is missing] is the fact that no one learns to read (or to speak or to listen) by being taught the linguistic system. If we took the phonics that we teach children and programmed it into a computer, the computer wouldn’t be able to read.

What we teach are clues to a system to facilitate the learner’s abstracting of those patterns and relationships from their experience with text. Thus, your “action” example is correct linguistically, but what the youngster who is learning to read is trying to do is to come up with an approximate pronunciation of a word that is in his mental lexicon.

Even if that were not the case, part of phonics instruction usually includes a simple admonition… to try it the other way if the pronunciation that you come up with doesn’t work. That’s why Dr. Venezky (he was the one that showed that our spelling system was morphophonemic) was a big supporter of phonics instruction—he recognized it was a primitive system used for teaching not for programming computers (our instruction gives kids clues to some of the most helpful patterns and relationships and they have to learn to apply those patterns based on their statistical experiences with language). It is more likely the child will come up with action by saying ak-shun than saying act/ion…

(Followup) Are the clues we teach linguistically accurate? Can you be both linguistically accurate and still teach patterns and relationships that already exist within the system?

Generally, you would like them to be accurate phonetically and morphemically, but given the example (action) that you gave me, I would say that there are very few words a six-year-old is likely to be reading that would turn on that issue. However, by the time kids are seven and can read about 700-800 words (of the approximately 12,000 words in their oral vocabularies) they are taught to recognize units like <sion>  and <tion>  when decoding. I would say if accuracy means teaching kids things that they can’t use now, because the vocabulary demand is not there, or if teaching those things would block them from being able to decode, like the action example, then I would not strive for linguistic accuracy. In any event, I don’t think those distinctions arise very often or that they ultimately matter much in kids’ learning.

(Followup) Is reading pronouncing the words?

No, that isn’t what reading is, but it is an essential part of reading initially. In order to read, one has to figure out how to translate the print into pronunciation so that the pronunciation can be matched with the students oral language. Phonics is pointless if one doesn’t have an oral language version of the language. Separation of meaning and pronunciation are important when kids are learning to read, and frankly, when they are learning to speak. One of the things that you stated is that meaning comes first in language learning. That [is the viewpoint of] a linguist, not a psychologist or psycholinguist. My eight-month-old grandson is getting a lot out of parroting “I’m bad” right now. He is figuring out the phonemic system at the moment. He’ll get to the meaning of that in a few months.

(Followup) Is this [the child will come up with action by saying ak-shun than saying act/ion] reading or pronouncing? Is “reading out loud” a construct that adults created in order to understand what students understand when they read? In other words, is the pronouncing of words out loud just a measurement tool?

What a skilled adult reader does is a bit different from what a beginning reading six-year-old does. In fact, Chuck Perfetti has pretty much proven that beginning readers really cannot read silently for instance. They’ll get there, but that isn’t where they start. Pronouncing words out loud during beginning reading is more than a measurement tool. It is an important aspect of the learning process.

The reason that I say that phonics programs don’t have to be perfect representations of all the linguistic systems is because 1) teaching such programs makes a measurable and consistent difference in kids’ learning (if they are so imperfect why would their teaching lead to better reading?) and 2) there are not big differences in learning derived from programs that teach very different sequences or that may be a mess linguistically (Dr. Venezky used to be pretty critical of the Open Court Program because it made so little sense morphologically – and, yet, research shows the program improves reading anyway). I’m sure we could spoil some program badly enough that it wouldn’t work or work as well as another program, but generally the point is that these programs are primitive representations of the linguistic system and they require a good deal of construction on the part of the learner (and this construction seems to come about through statistically weighing one’s experience – which is why English readers overwhelmingly read nonsense words beginning with <c>  as if they began with the letter <k>  and not with the letter <s>.

What is your profession?

I am a college professor. I do research on literacy instruction and train literacy teachers and researchers.

What are your personal and/or professional experiences with language?

My original research focus was in investigating how language knowledge is shared across reading and writing. For example, how related are students ability to use phonology and morphology to decode words and to spell words. Though I am not actively collecting data on those issues, that area of research has burgeoned since my first forays into it, and we are coming to understand better these connections.

I was also part of a research team that looked at measures of the language ability of preschoolers and its relationship with later reading comprehension (we found that if language was measured more fully—vocabulary, grammar, listening comprehension, etc.—you had a strong connection with later reading, but if your early measures focused only on a single dimension of language like receptive vocabulary, the connections were almost zero.

Finally, I’m not doing research in the area of quantitative linguistics, but I am carefully monitoring it for purposes of curriculum design, etc. What I am referring to here are studies of the language features of texts in various disciplines. The grammar features that are evident in science, history, and literature provide readers with very different language demands that I suspect teaching can facilitate.

Where can we find out more?