In the context of early years reading instruction, one of the most helpful entry-points for understanding the difference between oral language and written language is the distinction between biologically primary and biologically secondary skills, as articulated in the work of US (University of Missouri) evolutionary psychologist, Professor David Geary.
Geary has collaborated with Dr Dan Berch on what is known as “an evolutionarily informed theory” which holds that ‘Folk domains represent universal forms of knowledge and competencies that emerge from a combination of inherent cognitive biases and evolutionarily expectant experiences’ (Geary & Berch, 2016, p. 219).
By folk domains, Geary and Berch mean those knowledge and skill sets that humans have an evolved, and hence biologically primary advantage for developing. These include social domains such as recognition of the meaning behind each other’s facial expressions (e.g., friendly or hostile), the development of oral language (expressive and receptive verbal skills), the emergence of theory of mind*, and ecological domains such as folk biology and folk physics, i.e., our “commonsense” understandings, acquired early in life, about how the world around us works. An example of folk biology is the inherent understanding that parents are older than their children. Folk physics on the other hand, takes in concepts such as gravity, and the inability of humans to fly unassisted.
Geary and Berch explained that biologically primary skills have assisted humans to hone capacities that support vital physical and social survival in groups – hence the nod to human evolution in the name of the theory.
So - this is a theory that asks us to think about what knowledge and skills we might expect children to absorb or “intuit” through everyday life experiences “in the village”, as opposed to the knowledge and skills that they can learn, if appropriate instructional experiences are provided. Here we might think of playing a musical instrument, learning how to play chess, and of course, the icing on the life-trajectory cake…..learning to read, and then to write.
In his commentary on Geary’s work and its applications to education, David Didau (2017) summed up its significance for educators this way:
If it’s a primary adaptation, then maybe we don’t need to teach it at all as children will have an innate ability to pick it up from their environments. That said, maybe we do need to make sure that children’s environments are conducive to acquiring the folk knowledge we all take for granted.
As can be seen, a close read of Didau’s words highlights that there may need to be some qualifications to this conceptual framework. While I think the biologically primary-biologically secondary distinction is profoundly important as a macro lens on education, I also think there are some notable caveats to bear in mind.
I have written previously that “. . . although oral language development is biologically primary and ‘natural’, it is by no means ‘set and forget’”. This is because a range of biological, ecological and circumstantial factors come into play in children’s lives that result in them experiencing different home language and literacy environments in the preschool years, and different instructional environments in the school years. We have known for decades that there is a social gradient, for example, that impacts the quality and quantity of language exposure children experience in the preschool years (see here for references). Language development is not configured to succeed without adult input. Adults (parents and teachers in particular) invest considerable time into encouraging and supporting the development of connected discourse skills (especially conversation and narrative genres in the early years) through repeated serve-and-return interactions across the day, real-time modelling, and expansions and elaborations as part of everyday feedback and encouragement. Oral language is clever, but it doesn't get there on its own.
I often make the observation when I present professional learning, that oral language is a paradox. It is both biologically primary and highly fragile. Certain neurodevelopmental disorders (otherwise known as forms of neurodiversity - sometimes diagnosed, sometimes not) impact on children’s oral language development, notably developmental language disorder, autism spectrum disorder, and various forms of syndromal and non-syndromal intellectual disability, to name a few.
Epidemiological research carried out by Professor Courtenay Norbury and her team in the UK tells us that we should expect two children in every class of 30 to meet diagnostic criteria for Language Disorder/Developmental Language Disorder (DLD) on formal assessment at school entry. What we should not assume, however, is that the remaining 28 are flying, unencumbered with respect to their oral language skills. A significant, but unknown proportion of children will have weak language skills that may go undetected and/or may be mistaken for disinterest/poor motivation and/or lack of social skills. Some children’s language disorder is not identified until they struggle with reading comprehension, in spite of adequate decoding skills. This 2018 open-access paper by Adlof and Hogan is a recommended read on this and on DLD.
So, this capacity for which humans have evolved a special evolutionary advantage is not quite as self-sufficient as it is sometimes made out to be.
Nowhere is this lack of self-sufficiency more evident, in fact, than in the expansion of children’s vocabulary beyond everyday so-called “Tier 1” words, to the other side of what the late Dr David Corson described as the “lexical bar”, to the higher-order, more literate language that is needed for academic success. Corson observed that (my emphasis):
To look at the impact that students’ life histories have on their learning and use of academic English words, one starting point is the commonplace fact that the vocabulary of English falls into two very different categories. (There is a)….. striking incompatibility between Anglo-Saxon and Graeco-Latin elements in English: ‘the familiar homely-sounding and typically very short words’ that we learn very early in life and use for most everyday purposes and ‘the more learned, foreign sounding and characteristically rather long words’ (p. 138)” [that are used for academic purposes].
Corson was, of course, referring to the categories of words that we now commonly describe as Tier 2 (relatively common but less so than Tier 1) literate language used by educated citizens, and Tier 3 words, which are subject-specific and only infrequently arise in decontextualised ways, for example words such as enzyme, catalyse, rhomboid, and isthmus. Readers are referred to the work of Isabel Beck, Margaret McKeown and Linda Kucan for a more detailed consideration of tiers of vocabulary and how to teach across them.
Corson’s reflection provides an important “yes, but” to the cut and dried idea of oral language being biologically primary. This maxim holds to the extent that under typical circumstances, children will acquire a generous store of Tier 1 words, with which they can navigate the business of everyday life, in the here and now with those around them. Without exposure to impactful classroom instruction, however, most children are not going to experience enough of the literate language that authors use, to get across Corson’s lexical bar, as users and consumers of more sophisticated language. This exposure needs to occur both through classroom-based vocabulary teaching and through children’s own reading. The latter of course forms a compelling argument for impactful early reading instruction, so this is not left to chance and a never-ending game of catch-up for all concerned.
Interestingly, there are also some “yes but” scenarios in play on the biologically secondary side of the ledger, with respect to children learning to read. These are borne out in Nancy Young’s Ladder of Reading and Writing infographic (see below), which represents the fact that a small percentage of students (5-10% in the darker green shaded area) are successful with seemingly little to no formal reading instruction. They appear to skip across the metaphorical bridge from biologically primary (oral language) to biologically secondary (reading) skills on the strength of a positive alignment between genetic endowment and preschool language and literacy experiences, followed by a compounding Matthew Effect on school entry. A further 35-40% (those in the paler green zone) seem to succeed at school, regardless of where the reading pedagogy they experience sits on a continuum of explicitness (see here for an open access paper contrasting balanced literacy and structured literacy teaching).
Reproduced with permission from https://nancyyoung.ca/the-ladder-of-reading-writing/ The existence of these “early adopters” of biologically secondary skills should not overly surprise us, given that reading skills, like intelligence and language skills are spread across a distribution. Reading is probably most accurately conceptualised as a bio-psycho-social skill – one that draws on genetic influences, individual temperament/motivation factors, and the nature of the child’s environment (including the instruction to which they have been exposed). Some children will be very high performers, some will be very low, and most will clump around the mid-point. This pattern is strongest where a so-called normal distribution occurs, but of course a core aim of education is to disrupt the normal distribution by pushing the entire curve to the right, in a more peaked (less spread out) pattern. They key thing when considering this bio-psycho-social framework, though, is the fact that schools cannot influence children’s genetic endowment. They can, however, influence the psycho-social domains by teaching reading sub-skills and promoting a strong sense of self-efficacy as a reader.
What is perhaps not discussed enough, is the potential value-add for these early adopters, of explicit teaching of spelling and the incorporation of morphological awareness and knowledge to (a) promote vocabulary development (especially on the high side of Corson’s lexical bar, as discussed above) and (b) provide explicit understandings of the ways in which morphology and spelling interact with each other. Consider for example, the convention that we double a final consonant in CVC words when we add the suffix “-ing” (e.g., cup becomes cupping) and we drop word-final “e” in the same situation (e.g., cable becomes cabling). We could hope that all strong readers notice and intuit these conventions, which of course some do, or we could explicitly teach them, as a way of fast-tracking improved writing. In the same way, we can hope that strong readers intuitively “notice” that the words construct, destruct, instruct, structure, and obstruct all have the same (Latin) base “struct”, and make the conceptual leap that it means “build”. Alternatively, we could invest classroom (instructional) time into teaching the meanings of the Latin and Greek bases that make up so much of the literate language over which successful students achieve mastery. In so doing, we can teach them the generalisable habit of being curious about word meanings, origins, and connections. In a language with the large and rich vocabulary we have to work with in English, this would seem like a solid investment.
Moving down into the red-shaded area of Nancy Young’s infographic, we see how the other half lives – the 50% of students for whom the bridge from biologically primary to biologically secondary skills is experienced as a flimsy rope assembly suspended over rapidly running waters a long way below. These are the students who truly embody Geary and Berch’s evolutionarily-informed theory with respect to reading skills. They are also the 50% who remind us that we cannot build entire education systems around the good fortunes of half the population. Doing so is to wilfully sentence half of the populace to under-achievement and its associated social and economic baggage across the life-span.
While reading scientists are rightly at pains to emphasise that, in the words (and indeed book tile) of Harvard psychologist, Steven Pinker, we have a language instinct, (but not a reading instinct), there are certain adaptations that our brains seem to make once we begin to learn to read, and these should be understood and capitalised on in early reading instruction. Right across the proficiency spectrum, we want children to be cashing in on what Professor David Share described back in the 1990s as the self-teaching hypothesis. This is especially important in an orthography such as the one we use in English, where there is not a reliable two-way 1:1 correspondence between grapheme-phoneme correspondences (GPCs), and some GPCs occur relatively infrequently. Those low-frequency, less regular GPCs occur more and more in texts children are reading beyond their mastery of basic GPCs and fluent reading of decodable (phonically controlled) texts in the early stages of reading success. Share cited evidence that the average fifth grader encounters around 10,000 new words per year, a phenomenon he aptly described as an “orthographic avalanche”. He argued that only efficient, transferable decoding skills afford children opportunities for high rates of incidental success in decoding novel GPCs in printed words. Such success of course contributes to reading fluency, which is a strong driver of reading comprehension.
You can learn more about Share’s self-teaching hypothesis at this Five from Five page. A closely related construct, statistical learning, is outlined by Joanne Arciuli in this 2018 open-access paper. She notes (p. 634) that:
Children are given explicit instruction regarding some of these regularities to assist them when they are beginning to “crack the code.” Such instruction is vital in the early years of decoding. However, in a language such as English, which has a deep orthography with many-to-many mappings between orthography and phonology rather than one-to-one mappings and complex mappings that depend upon positional and other contextual regularities, it is not possible to convey explicitly all of the regularities that children need to know in order to become skilled readers. Thus, it must be assumed that some regularities are acquired implicitly.
Self-teaching and statistical learning, in turn, are related to a third contemporary construct in reading science research and practice: set for variability, otherwise known as mispronunciation correction. I touched on this in my most recent blog-post and provided a range of links there to readings and podcasts for more information. Mispronunciation correction means that we are guiding students to form hypotheses about possible pronunciations of unfamiliar GPCs and then supporting them to use this information to test a spoken (pronounced) approximation against their oral language lexicon and assess its likely accuracy.
These lines of research and classroom/clinical practice remind us that reading, while a biologically secondary skill, does not have to flail around like an upturned tortoise every time an unfamiliar word is encountered. Under optimal circumstances, it can be resourceful and like a bank account, earn compounding-interest on its returns over time.
The bottom line then, is that oral language skills
are biologically primary, yet can be vulnerable, and reading skills are biologically
secondary, yet can be resourceful. The challenge (and opportunity) is to
titrate instruction to ensure that a biologically primary skill like oral
language is not taken for granted and is supported to develop beyond the
proximal horizon of Tier 1 words and simple sentence structures. On the flip
side, its biologically secondary cousin, reading skill, should be supported
through high-quality initial instruction to become autonomous and a secret
weapon for ongoing oral language development. This requires a nuanced understanding
on both sides of the biologically primary Vs secondary ledger by teachers and curriculum
leaders.
[*Theory of mind (ToM) is sometimes mistakenly treated as being synonymous empathy, but that is not correct. The two are related, but ToM is the understanding that another person my hold a false belief, by virtue of their application of the information they have available to them. You can watch a video demonstrating its features here].
(C) Pamela Snow (2024)
