An interview with David C Geary – Part 1

David C. Geary is a Curators’ Distinguished Professor in the Department of Psychological Sciences (and a former departmental chair) and the Interdisciplinary Neuroscience Program at the University of Missouri. His many publications include The Origin of Mind and Evolution of Vulnerability. I wanted to interview him because his ideas on biologically primary and biologically secondary abilities have been discussed at length on this blog.

Question 1: Regular readers of this blog will be familiar with the idea of biologically primary abilities that we have evolved to acquire, and biologically secondary abilities that are culturally specific and that we have therefore not evolved to acquire. Is this an accurate summary of the theory and, if so, what empirical evidence supports it and what potential empirical evidence could prove it wrong? 

Yes, this is accurate. Primary abilities have not only evolved, they are human universals that emerge developmentally through a combination of skeletal, inherent constraints and a bias to engage in the types of activities that will flesh them out and adapt them to local social and ecological conditions.  Language is one example, but there are many others in the broad domain of folk psychology – an intuitive understanding of the self, others, and the ability to engage in social relationships. Other primary domains include folk biology (e.g., learning which foods can be eaten and when) and folk physics (e.g., the ability to navigate from one place to another). The research of Liz Spelke, Rochel Gelman and others have provided evidence for very early (in infancy, toddlerhood) attentional and behavioral biases in these domains, and evidence that common childhood activities are needed for their full expression.

The motivational and behavioral component is expressed through children’s engagement in self-initiated activities, such as social play or exploring the environment. These activities are typically considered fun, consistent with a built-in reward mechanism that reinforces children’s activities in these areas. Children need to engage in these types of activities because it takes many years for primary abilities to become fully developed. The plasticity of these abilities and a bias to engage in activities that flesh them out allows them to become adapted to local conditions (e.g., the local language).

In other words, the combination of skeletal constraints and built-in motivational and behavioral biases ensures that certain competencies emerge (e.g., various social skills) but at the same time are adjusted to better fit the nuances of the contexts in which the children are situated. The development of these competences occurs with little if any explicit instruction or mental effort.

Biologically secondary abilities, such as reading, writing, and arithmetic, do not have this deep evolutionary history, are not human universals, and in fact the cultural-goal (at least in developed nations) of universal education is less than 200 years old. There are many implications to my basic argument but the key ones are:

a. The cognitive, attentional, motivational, and behavioral biases that ensure the acquisition of primary abilities will not be sufficient for the learning of secondary abilities.

b. I’ve laid out why secondary abilities will require the engagement of domain-general abilities, such as working memory, at least during the early phases of learning elsewhere (e.g., 2008 article in Educational Psychologist, and 2007 monograph, Educating the Evolved Mind). As you know, John Sweller and others have repeatedly demonstrated the importance of working memory for school learning.

c. We cannot expect a universal motivation for secondary learning. Some people will of course like to learn new things, especially once they have the basics mastered, but the goal is universal education, not just the education of the curious and cognitively able. For most children, the goal to learn secondary competencies in school is competing against a strong motivation to engage in activities that flesh out primary abilities, such as social play or other social activities.

d. These points suggest that for most children to become proficient in secondary domains, well-developed curricula and competent teachers must provide students the structure and scaffolding provided by evolution to support primary learning. This will involve explicit curricular goals and for many children explicit instruction (see below).

e. Making school more fun essentially involves making the associated activities more similar to the activities that promote primary learning. This might work, with teachers’ help, during the initial (preschool) phases of learning secondary abilities, but won’t be effective as material becomes more abstract.

If my thesis is incorrect, then it can be easily falsified. As one example, whole language should be more effective in helping children learn how to read than instruction that includes an explicit focus on phonology, word decoding, reading fluency, and understanding the structure of written stories. The National Reading Panel in the U.S. concluded that children need to learn these things, and many experimental studies funded by the National Institutes of Health show that explicit instruction works better than other approaches.

It is not as well known, but this has also been assessed for mathematics with Project Follow Through. The study, conducted in the U.S., was a quasi-experimental evaluation of the basic tenents of broad educational approaches and was the largest educational intervention that has ever been conducted. The project was conducted from 1968 to 1977 and included about 200,000 at-risk children from kindergarten through third grade. Each participating school partnered with a university, research institute or some other group to develop and implement the four-year intervention grounded in differing educational principles and academic foci.

The principles reflected the basic teaching philosophy and broadly included direct instruction, cognitive approaches (e.g., focus on problem solving), and social/affective approaches focusing on, for instance, self-efficacy. The educational goals ranged from a focus on mastery of basic skills to more complex problem solving and conceptual understanding. Relative to Title I (a U.S. program for children at-risk for poor long-term academic outcomes) comparison groups and with respect to performance on standard achievement tests, children in the direct instruction programs with a focus on basic skills and mastery had better outcomes on basic skills, problem solving, and had a better academic self-concept than did students in the other programs. There was some fade out over time, but many of these gains were maintained throughout the elementary school years. Of course, the results were roundly criticized, and eventually suppressed in favor of more Romantic (child centered) approaches to education, eventually igniting the ‘math wars’ in the U.S.

One response to the math wars was the creation of the National Mathematics Advisory Panel (U.S.). One of many things covered by the panel included reviews of intervention effectiveness. Especially for at-risk students a rigorous curriculum with explicit instruction results in the most gains.

There are obviously people who, as I noted, are curious and cognitively capable and learn quite a bit on their own.  This learning, however, typically occurs after they have learned the basics of reading, writing, and arithmetic.  And, does not work for most children or adults; if it did, we wouldn’t need organized schools.

The interview continues in Part 2


27 thoughts on “An interview with David C Geary – Part 1

  1. Tom Burkard says:

    I was with Prof Geary until he commented “The development of these [primary] competences occurs with little if any explicit instruction or mental effort.”

    This came up in my linguistics prelims at UEA Norwich in 1989 when we were discussing children’s acquisition of language. There was a fair sprinkling of mature students in our seminar, including a couple of youngish mothers, one of whom hotly disputed the notion that children acquire language ‘naturally’. She contended that mothers–and fathers–in fact spend one hell of a lot of time explaining what words mean, teaching children how to say things and correcting their fractured syntax.

    Now, most infant teachers will agree that some children come to school incapable of conveying their most basic needs with even the most primitive speech. I’ve long since stopped following linguistics and I’m not sure if Chomsky still dominates the field, but I think it’s fair to suggest that all cultures employ a fair amount of explicit instruction even in regard to the development of primary competences. This includes social competences–all societies go to great lengths to discourage certain behaviours and encourage others, and this process starts in infancy. They’d collapse into anarchy if they didn’t.

  2. I think that what he says is that by being in an environment where speech is present is what is needed. The more rich and complex and supportive that environment the faster and so deeper after a time the language of the child will be. Providing the opportunity to learn rather than teaching speech.

  3. thanks for the interview, looking fwd to part 2;

    in the meantime i was wondering how Geary views language?

    it seems from this part 1 interview that language is a primary ability yet writing is a secondary ability – does this mean Geary thinks writing is not language?

    or that he takes writing to just be the notation/document/inscription and not the script/sign/semiology?


  4. Hello,

    Regarding parental teaching of primary competencies. This certainly does occur in many upper-middle class and middle class households, but is not a human universal. In fact, David Lancy’s work on parent-child interactions in traditional contexts reveals very little direct instruction for anything.

    Reading and writing are built upon the evolved language system and are supported by many of the same brain regions. But, written communications systems only emerged in a few cultures, unlike the universality of language, and learning to read and write requires some explicit instruction for most children.



    • John Pierry says:

      written communications systems only emerged in a few cultures, unlike the universality of language

      And what is it about those few? Do you see it as just a random accident, a reflection of the superiority of the cultures that developed those writing systems, or developing out of a need that was particular to that society? For example, cultures that conglomerated and depended on large flowing rivers (like the Nile) that made transport and trade a viable option, and therefore needed records of that trade?

      My argument would be that writing skills are innate and used only in cultures where those situations call upon them. A bit like skiing – someone from the Amazon would be able to learn how to ski (a physically and neurologically demanding activity), even though their environment would have made such an activity inconceivable in their culture.

      • How would evolution know to evolve an innate writing skill prior to a time when anyone actually wrote anything? Does evolution have some kind of spooky foresight?

      • John Pierry says:

        Does evolution have some kind of spooky foresight?

        Is that question meant to be serious or rhetorical?

        Quite simply – we write because we can. If we couldn’t, we would do something else – like those aliens in that movie that communicated with inky clouds.

        Evolution is clearly more nuanced than you seem to give it credit for. Start thinking in terms of increments rather than great leaps and you might get the idea. I haven’t read this entire article but it seems to capture what I’m talking about:

      • Stan says:

        I still think you are missing the point here. Sure the genes that determine the mental and other capabilities for written language evolved gradually over time. What other theory is there on that score?

        But that doesn’t mean they evolved so that we can write without transmission of cultural information, that is non-genetic information. Nor does it mean that reading and writing skills can be picked up as easily as speaking a language.

        To see what is the case we don’t have to explore genetics. We can look at cases of illiterate adults in western countries over the time written texts became fairly universal. You could go back to the introduction of printed books and pamphlets in English from about 1604 onwards. If your theory is correct and learning to read and write is no different than learning speech then we would expect a large number of people to pick up written language skills without training from that time on. I suggest we don’t. Instead we see a massive effort to introduce education in part so that workers could understand written instructions and everyone had access to important texts of the day such as the Christian bible.

        Or an analogy. Touch typing is far superior to picking out letters with two fingers. Most people who use a keyboard on a daily basis are aware of this. Clearly the genetic basis for the skills to touch type are common in both people who can and can’t. Yet people do not naturally become touch typists. Unless they take steps that amount to explicit training and teaching of touch typing they will tend to type inefficiently with two fingers. An exception – someone who without training naturally took to touch typing wouldn’t prove anything except exceptions exist – the norm is that touch typing requires training that two finger typing does not.

      • a small but important point about effort needed to learn to read – syllabic scripts can be learned relatively easily (compared to alphabetic scripts) – Cree syllabic script in a few days, Vai (western Liberia) script in a few weeks;

  5. Ryan says:

    This is excellent, looking forward to part 2. Well done Greg, In particular, I’m curious as to whether he feels evolved biologically primary ever carries deficits that need to be taught with biologically secondary instruction, I’m thinking specifically of things like cognitive biases in decision making and negative tendencies in collaboration that have an evolutionary basis eg: tendencies to defer to the highest status person in the room etc

    He mentions one possible route to falsification, I suppose another way might be evidence that it is biological secondary that influences biologically primary and not the other way around.

    • David C Geary says:


      Absolutely, there are many primary biases that can interfere with our understanding of reality. These include many social biases, as mentioned, as well as biases about how the biological and physical worlds work.This is why education is so important.

      We know that secondary learning can influence awareness of primary systems. Learning to read makes students more aware of language sounds and learning symbolic math makes people better at intuitive (primary) estimations of quantity (e.g., relative size of two collections of objects). I’ve argued that the explicit awareness of the operation of primary systems may be one step in secondary learning in a related domain (e.g., language and reading; or intuitive sensitivity to quantity and children’s understanding of the quantity of number symbols, such as 3).

      Some of my early discussion of this is here;


      • Thank you very much for that reply and the link. In large part after reading the introduction of your book and Robert Trivers I’ve argued for the teaching of collaboration under certain circumstances where our evolved nature interferes with efficient group outcomes in modern world situations. And especially where those evolved tendencies eg: self deception can be actively harmful to individuals and groups. I see you’ve discussed this further in part 2. Thanks again.

  6. Pingback: An interview with David C Geary – Part 2 – Filling the pail

  7. Pingback: Fair schooling — take-off | Educational test item design

  8. Pingback: What is stopping you from being your best? – Filling the pail

  9. Pingback: Survival of the fittest… ideas in your head – Filling the pail

  10. Pingback: Should cognitive load theory be taught to trainee teachers – Filling the pail

  11. Pingback: Teaching methods still matter – Filling the pail

  12. Pingback: Nicholas Christakis and the universal act of teaching – Filling the pail

  13. Pingback: Matthew Oldridge: Playing dice with our children’s futures – Filling the pail

  14. Pingback: Ashman’s Taxonomy – Filling the pail

  15. Pingback: Rousseauian Nonsense – Filling the pail

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.