Kieran Egan’s “The educated mind” 2

The second post in a two-part review of Kieran Egan’s book The Educated Mind: How Cognitive Tools Shape our Understanding.

For Egan, a key point in the historical development of understanding was the introduction by the Greeks of a fully alphabetic representation of language – it included symbols for vowels as well as consonants. He points out that being able to represent speech accurately in writing gives people a better understanding of how they use language and therefore of the concepts that language represents. Egan attributes the flowering of Greek reasoning and knowledge to their alphabet “from which all alphabetic systems are derived” (p.75).

This claim would be persuasive if it were accurate. But it isn’t. As far as we know, the Phoenicians – renowned traders – invented the first alphabetic representation of language. It was a consonantal alphabet that reflected the structure of Semitic languages and it spread through the Middle East. The Greeks adapted it, introducing symbols for vowels. This wasn’t a stroke of genius on their part – Semitic writing systems also used symbols for vowels where required for disambiguation – but a necessary addition because Greek is an Indo-European language with a syllabic structure. The script used by the Mycenaean civilisation that preceded the Greeks was a syllabic one.

“a distinctive kind of literate thinking”

Egan argues that this alphabet enabled the Greeks to develop “extended discursive writing” that “is not an external copy of a kind of thinking that goes on in the head; it represents a distinctive kind of literate thinking” (p.76). I agree that extended discursive writing changes thinking, but I’m not convinced that it’s distinctive nor that it results from literacy.

There’s been some discussion amongst teachers recently about the claim that committing facts to long-term memory mitigates the limitations of working memory. Thorough memorisation of information certainly helps – we can recall it quickly and easily when we need it – but we can still only juggle half-a-dozen items at a time in working memory. The pre-literate and semi-literate civilisations that preceded the Greeks relied on long-term memory for the storage and transmission of information because they didn’t have an alternative. But long-term memory has its own limitations in the form of errors, biases and decay. Even people who had memorisation down to a fine art were obliged to develop writing in order to have an accurate record of things that long-term memory isn’t good at handling, such as what’s in sealed sacks and jars and how old it is. Being able to represent spoken language in writing takes things a step further. Written language not only circumvents the weaknesses of long-term memory, it helps with the limitations of working memory too. Extended discursive writing can encompass thousands of facts, ideas and arguments that a speaker and a listener would find it impossible to keep track of in conversation. So extended discursive writing doesn’t represent “a distinctive kind of literate thinking” so much as significantly extending pre-literate thinking.

the Greek miracle

It’s true that the sudden arrival in Greece of “democracy, logic, philosophy, history, drama [and] reflective introspection… were explainable in large part as an implication of the development and spread of alphabetic literacy” (p.76). But although alphabetic literacy might be a necessary condition for the “Greek miracle”, it isn’t a sufficient one.

Like all the civilisations that had preceded it, the economy of the Greek city states was predominantly agricultural, although it also supported thriving industries in mining, metalwork, leatherwork and pottery. Over time agricultural communities had figured out more efficient ways of producing, storing and trading food. Communities learn from each other, so sooner or later, one of them would produce enough surplus food to free up some of its members to focus on thinking and problem-solving, and would have the means to make a permanent record of the thoughts and solutions that emerged. The Greeks used agricultural methods employed across the Middle East, adapted the Phoenician alphabet and slavery fuelled the Greek economy as it had previous civilisations. The literate Greeks were standing on the shoulders of pre-literate Middle Eastern giants.

The ability to make a permanent record of thoughts and solutions gave the next generation of thinkers and problem-solvers a head start and created the virtuous cycle of understanding that’s continued almost unabated to the present day. I say almost unabated, because there have been periods during which it’s been impossible for communities to support thinkers and problem-solvers; earthquakes, volcanic eruptions, drought, flood, disease, war and invasion have all had a devastating and long-term impact on food production and on the infrastructure that relies on it.

language, knowledge and understanding

Egan’s types of understanding – Somatic, Mythic, Romantic, Philosophic and Ironic – have descriptive validity; they do reflect the way understanding has developed historically, and the way it develops in children. But from a causal perspective, although those phases correlate with literacy they also correlate with the complexity of knowledge. As complexity of knowledge increases, so understanding shifts from binary to scalar to systematic to the exceptions to systems; binary classifications, for example, are characteristic of the way people, however literate they are, tend to categorise knowledge in a domain that’s new to them (e.g. Lewandowski et al, 2005).

Egan doesn’t just see literacy as an important factor in the development of understanding, he frames understanding in terms of literacy. What this means is that in Egan’s framework, knowledge (notably pre-verbal and non-verbal knowledge) has to get in line behind literacy when it comes to the development of understanding. It also means that Egan overlooks the key role of agriculture and trade in the development of writing systems and of the cultures that invented them. And that apprenticeship, for millennia widely used as a means of passing on knowledge, is considered only in relation to ‘aboriginal’ cultures (p.49). And that Somatic understanding is relegated to a few pages at the end of the chapter on the Ironic.

non-verbal knowledge

These are significant oversights. Non-verbal knowledge is a sine qua non for designers, artisans, architects, builders, farmers, engineers, mariners, surgeons, physiotherapists, artists, chefs, parfumiers, musicians – the list goes on and on. It’s true that much of the knowledge associated with these occupations is transmitted verbally, but much of it can’t be transmitted through language, but acquired only by looking, listening or doing. Jenny Uglow in The Lunar Men attributes the speed at which the industrial revolution took place not to literacy, but to the development of a way to reproduce technical drawings accurately.

Egan appears sceptical about practical people and practical things because when

those who see themselves as practical people engaging in practical things [who] tend not to place any value on acquiring the abstract languages framed to deal with an order than underlies surface diversity” are “powerful in government, education departments and legislatures, pressures mount for an increasingly down-to-earth, real-world curriculum. Abstractions and theories are seen as idle, ivory-tower indulgences removed from the gritty reality of sensible life.” (p.228)

We’re all familiar with the type of people Egan refers to, and I’d agree that the purpose of education isn’t simply to produce a workforce for industry. But there are other practical people engaging in practical things who are noticeable by their absence from this book; farmers, craftspeople, traders and engineers who are very interested in abstractions, theories and the order that underlies surface diversity. The importance of knowledge that’s difficult to verbalise has significant implications for the curriculum and for the traditional academic/vocational divide. Although there is clearly a difference between ‘abstractions and theories’ and their application, theory and application are interdependent; neither is more important than the other, something that policy-makers often find difficult to grasp.

Egan acknowledges that there’s a problem with emphasising the importance of non-verbal knowledge in circles that assume that language underpins understanding. As he points out “Much modernist and postmodernist theory is built on the assumption that human understanding is essentially languaged understanding” (p.166). Egan’s framework elbows aside language to make room for non-verbal knowledge, but it’s a vague, incoherent “ineffable” sort of non-verbal knowledge that’s best expressed linguistically through irony (p.170). It doesn’t appear to include the very coherent, concrete kind of non-verbal knowledge that enables us to grow food, build bridges or carry out heart-transplants.

the internal coherence of what’s out there

Clearly, bodies of knowledge transmitted from person to person via language will be shaped by language and by the thought-processes that produce it, so the knowledge transmitted won’t be 100% complete, objective or error-free. But a vast amount of knowledge refers to what’s out there, and what’s out there has an existence independent of our thought-processes and language. What’s out there also has an internally coherent structure that becomes clearer the more we learn about it, so over time our collective bodies of knowledge more accurately reflect what’s out there and become more internally coherent despite their incompleteness, subjectivity and errors.

The implication is that in education, the internal coherence of knowledge itself should play at least some part in shaping the curriculum. But because the driving force behind Egan’s framework is literacy rather than knowledge, the internal coherence of knowledge can’t get a word in edgeways. During the Romantic phase of children’s thinking, for example, Egan recommends introducing topics randomly to induce ‘wonder and awe’ (p.218), rather than introducing them systematically to help children make sense of the world. To me this doesn’t look very different from the “gradual extension from what is already familiar” (p.86) approach of which Egan is pretty critical. I thought the chapter on Philosophic understanding might have something to say about this but it’s about how people think about knowledge rather than the internal coherence of knowledge itself – not quite the same thing.

the cherries on the straw hat of society

The sociologist Jacques Ellul once described hippies as the cherries on the straw hat of society* meaning that they were in a position to be critical of society only because of the nature of the society of which they were critical. I think this also serves as an analogy for Egan’s educational framework. He’s free to construct an educational theory framed solely in terms of literacy only because of the non-literate knowledge of practical people like farmers, craftspeople, traders and engineers. That brings me back to my original agricultural analogy; wonder and awe, like apple blossom and the aroma of hops, might make might make our experience of education and of agriculture transcendent, but if it wasn’t for coherent bodies of non-verbal knowledge and potatoes, swedes and Brussels sprouts, we wouldn’t be in a position to appreciate transcendence at all.


Lewandowski G, Gutschow A, McCartney R, Sanders K, Shinners-Kennedy D (2005). What novice programmers don’t know. Proceedings of the first international workshop on computing education research, 1-12. ACM New York, NY.

Uglow, J (2003). The Lunar Men: The Friends who made the Future. Faber & Faber.

*I can’t remember which of Ellul’s books this reference is from and can’t find it quoted anywhere. If anyone knows, I’d be grateful for the source.

Kieran Egan’s “The educated mind” 1

I grew up in a small hamlet on the edge of the English Fens. The clay soil it was built on retains nutrients and moisture, so, well-drained, it provides an ideal medium for arable farming. Arable crops aren’t very romantic. The backdrop to my childhood wasn’t acres of lacy apple blossom in spring or aromatic hops in summer, although there were a few fields of waving golden wheat. I grew up amongst potatoes, swedes and Brussels sprouts. Not romantic at all, but the produce of East Anglia has long contributed to the UK population getting through the winter.

A few weeks ago on Twitter Tim Taylor (@imagineinquiry) asked me what I thought about Kieran Egan’s book The Educated Mind: How Cognitive Tools Shape our Understanding. This book is widely cited by teachers, so I read it. It reminded me of the sticky clay and root vegetables of my childhood – because sticky clay, root vegetables and other mundane essentials are noticeable by their absence from Egan’s educational and cultural framework. For Egan, minds aren’t grounded in the earth, but in language. To me the educational model he proposes is the equivalent of clouds of apple blossom and heady hops; breathtakingly beautiful and dizzying, but only if you’ve managed to get through the winter living on swedes and potatoes. My agricultural allusion isn’t just a simile.


Egan begins by claiming there’s a crisis in mass education systems in the West due to their being shaped by three fundamentally incompatible ideas; socialisation, Plato’s concept of reason and knowledge, and Rousseau’s focus on the fulfilment of individual potential. To resolve this inherent conflict, Egan proposes an alternative educational framework based on the concept of recapitulation. Recapitulation was a popular idea in the 19th century, fuelled by the theory of evolution and the discovery that during gestation human embryos go through phases that look remarkably like transformations from simple life forms to more complex ones. As Ernst Haeckel put it ‘ontogeny recapitulates phylogeny’.

The theory of recapitulation has been largely abandoned by biologists, but is still influential in other domains. Egan applies it to the intellectual tools – the sign systems that children first encounter in others and then internalise – that Vygotsky claimed shape our understanding of the world. Egan maps the historical ‘culturally accumulated complexity in language’ onto the ways that children’s understanding changes as they get older and proposes that what children are taught and the way they are taught should be shaped by five distinct, though not always separate, phases of understanding:

Somatic; pre-linguistic understanding
Mythic; binary opposites – good/bad, strong/weak, right/wrong
Romantic; transcendent qualities – heroism, bravery, wickedness
Philosophic; the principles underlying patterns in information
Ironic; being able to challenge philosophic principles – seeing alternatives.

At first glance Egan’s arguments appear persuasive but I think they have several fundamental weaknesses, all originating in flawed implicit assumptions. First, the crisis in education.

crisis? what crisis?

I can see why a fundamentally incoherent education system might run into difficulties, but Egan observes:

“…today we are puzzled by the schools’ difficulty in providing even the most rudimentary education to students”… “the costs of…social alienation, psychological rootlessness and ignorance of the world and possibilities of human experience within it, are incalculable and heartbreaking.” (p.1)

Wait a minute. There’s no doubt that Western education systems fail to provide even the most rudimentary education for some students, but those form a tiny minority. And although some school pupils could be described as socially alienated, psychologically rootless or ignorant of the world and possibilities of human experience within it, that description wouldn’t apply to many others. So what exactly is the crisis Egan refers to? The only clue I could find was on page 2 where he describes ‘the educational ineffectiveness of our schools’ as a ‘modern social puzzle’ and defines ‘modern’ as beginning with the ‘late nineteenth century development of mass schooling’.

To claim an educational system is in crisis, you have to compare it to something. Critics often make comparisons with other nations, with the best schools (depending on how you define ‘best’) or with what they think the education system should be like. Egan appears to fall into the last category, but to overlook the fact that prior to mass schooling children did well if they manage to learn to read and write at all, and that girls and children with disabilities often didn’t get any education at all.

Critics often miss a crucial point. Mass education systems, unlike specific schools, cater for entire populations, with all their genetic variation, socio-economic fluctuations, dysfunctional families, unexpected illnesses and disruptive life events. In a recent radio interview, Tony Little headmaster of Eton College was asked if he thought the very successful Eton model could be rolled out elsewhere. He pointed out, dryly, that Eton is a highly selective school, which might just be a factor in its academic success. One obvious reason for the perceived success of schools outside state systems is that those schools are not obliged to teach whichever children happen to live nearby. Even the best education system won’t be problem-free because life is complex and problems are inextricably woven into the fabric of life itself. I’m not suggesting that we tolerate bad schools or have low aspirations. What I am suggesting is that our expectations for mass education systems need to be realistic, not based on idealised speculation.

incompatible ideas

Speculation also comes into play with regard to the incompatibility of the three ideas Egan claims shape mass education in the West. They have certainly shaped education historically and you could see them as in tension. But the ideas are incompatible only if you believe that one idea should predominate or that the aims inherent in each idea can be perfectly met. There’s no reason why schools shouldn’t inculcate social values, teach reason and knowledge and develop individual potential. Indeed, it would be difficult for any school that taught reasoning and knowledge to avoid socialisation because of the nature of schools, and in developing reasoning and knowledge children would move towards realising their potential anyway.

If, as Egan argues, Western mass education systems have been ineffective since they started, his complaint appears to be rooted in assumptions about what the system should be like rather than in evidence about its actual potential. And as long as different constituencies have different opinions about the aims of the education system, someone somewhere will be calling ‘Crisis!’. That doesn’t mean there is one. But Egan believes there is, hence his new framework. The framework is based on the development of written language and its impact on thinking and understanding. For Egan, written language marked a crucial turning point in human history.

why write?

There’s no doubt that written language is an important factor in knowledge and understanding. Spoken language enables us to communicate ideas about things that aren’t right here right now. Written language enables us to communicate with people who aren’t right here right now. The increasing sophistication of written language as it developed from pictograms to syllabaries to alphabets enabled increasingly sophisticated ideas to be communicated. But the widely held belief that language is the determining factor when it comes to knowledge and understanding is open to question.

The earliest known examples of writing were not representations of language as such but records of agricultural products; noting whether it was wheat or barley in the sacks, wine or oil in the jars, when the produce was harvested and how many sacks and jars were stored where. Early writing consisted of pictograms (images of what the symbols represent) and ideograms (symbols for ideas). It was centuries before these were to develop into the alphabetic representations of language we’re familiar with today. To understand why it took so long, we need to put ourselves in the shoes (or sandals) of the early adopters of agriculture.

food is wealth

Farming provides a more reliable food supply than hunting and gathering. Farming allows food that’s surplus to requirements to be stored in case the next harvest is a bad one, or to be traded. Surplus food enables a community to support people who aren’t directly involved in food production; rulers, administrators, artisans, traders, scribes, teachers, a militia to defend its territory. The militia has other uses too. Conquering and enslaving neighbouring peoples has for millennia been a popular way of increasing food production in order to support a complex infrastructure.

But for surplus food to be turned into wealth, storage and trade are required. Storage and trade require written records and writing is labour-intensive. While scribes are being trained and are maintaining records they can’t do much farming; writing is costly. So communities that can’t predict when a series of bad harvests will next result in them living hand-to-mouth, will focus on writing about things that are difficult to remember – what’s in a sealed container, when it was harvested etc. They won’t need to keep records of how to grow food, look after animals, histories, myths, poems or general knowledge if that information can be transmitted reliably from person to person orally. It’s only when oral transmission stops being reliable that written language as distinct from record-keeping, starts to look like a good idea. And the more you trade, the more oral transmission gets to be a problem. Travellers might need detailed written descriptions of people, places and things. Builders and engineers using imported designs or materials might need precise instructions.

Spoken language wasn’t the only driving force behind the development of written language – economic and technical factors played a significant role. I don’t think Egan gives these factors sufficient weight in his account of the development of human understanding nor in his model for education, as I explain in the next post.