“Teacher, Tutor, Scholar, I”

Remarks to: the Worshipful Company Of Educators, on the occasion of the annual Franklin Lecture, Mansion House, 18:00 to 19:00, Thursday, 9 February 2023.

“Teacher, Tutor, Scholar, I” – A Metaverse Of Education Or Conundra[1] Of Confusion

My Lord Mayor, Master, Aldermen, Sheriff, visiting Masters, Distinguished Guests, Ladies & Gentlemen,

Just two minutes ago, before coming up here, Stephen Bernhard asked me if I felt nervous. “Of course”, I replied. Stephen said, “I too always got nervous giving our company’s seminars, and I knew what I was talking about.” Thanks, Stephen.

Our Master [Caroline Haines] asked me to consider the future of education.  F Scott Fitzgerald said that “the test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function.”[2]  My first idea is that something absolutely must be done to make education far more relevant to the modern world, while the second is that we have little idea what exactly to do. 

Annika Small of Futurelab put it well, “We are now at a point where we need to teach what no one knew yesterday, and prepare our students for what no one yet knows.”

“Teacher, Tutor, Scholar, I”
[With thanks to OpenAI’s DALL-E for collaboration on the image]

I’m not an education expert, but I have breadth, attending some 18 schools in four countries, three universities, four professional qualifications, a very short stint teaching primary school, many secondary school lectures, school Almoner for nine years, and abundant tertiary university work.  I’ve long been at the boundary of science and economics, including a long stint as a director of Ministry of Defence research.  With a PhD in complex & chaotic systems I may dwell on systems theory a bit more than I should, but my fear tonight is acting like a proverbial eighth grader.

After a day listening to eighth graders exchange gossip, a teacher quoted Abe Lincoln, “Better to remain silent and be thought a fool than to speak out and remove all doubt.”  After consideration, one student asked, “What does it mean to remove all doubt?”

Though I have been active in machine-learning research for over 45 years, no I’m not tricking you with the latest media gimmick reading this lecture from OpenAI’s ChatGPT (though, on the other hand, how would you know?).   ChatGPT is not the enemy, rather a content creator and plagiarists’ friend.  This lecture is not going to say that technology is simply whizzo good stuff.  At the end I hope you think that the exciting ‘action’ in technology is with teachers and the management of lifelong learning, not blockchain, big data, AI, and the Metaverse. 

Mansion House

This lecture is held at an interesting time, with industrial action by NASUWT The Teachers’ Union, but I’m not going to regale you with dire statistics about UK education and then plead for money. 

Last June, Peter Hyman, gave a lecture to the Management Consultants on his “Head, Heart & Hand Curriculum” as co-founder of School 21 in East London, saying our education system is not fit for purpose.  It’s linear and progressive from 4 to 16, but then diversifies into technical, vocational, and academic. The majority of advanced learning takes place from ages 16 to 21, then employers are supposed to pick up professional development.  Yet today’s primary school student will undertake multiple careers with different advanced skills, so lifelong learning is key. An economic assessment is that UK education is about average, mediocre perhaps, in the OECD peer group, with highs and lows in universities and schools, but could do better with the resources it has.  Vocational training needs a bit more work.  Globally, we need bigger and bolder education systems where everyone can be excellent at something, be happy, and give back to society. 

Three years ago I shared some musings on The Frontiers of Education to the Educators and Actuaries, concluding with four questions.

  • How do we properly feed back to students the real job market, the real labour market information, a range of incomes not a specific salary, the likelihoods of unemployment, the volatility of employment, and the tools to do something about it?
  • Should we be prepared to push all education through simulators?
  • How do we build a lifetime funding model for self-improvement?
  • How can an educational system, in fact any process system, genuinely encourage mental diversity, and if so, how can we measure it?

I want to move those four questions further along by invoking systems theory, examining “I”, “Scholar”, “Tutor”, and “Teacher”, then end on five discussion points: focus on time, speed up scientific experimentation, increase variation tolerance, emphasise technology for teachers, develop lifelong education communities.

Systems Theory

Let’s start with systems theory.  May I ask everyone to take a very deep breath and hold it.  You may now breathe out.  Now may I ask everyone to take a very very very deep breath and hold it tightly. 

While you’re busy turning red, many systems are working.  Oxygen levels decrease and carbon dioxide accumulates.  Free hydrogen ion concentration rises changing your blood pH.  Accumulating soluble carbon dioxide crosses your blood-brain barrier to a trillion energy dependent neurons. Parts of the brain are inordinately sensitive to pH and seek more oxygen, pushing you to exhale carbon dioxide in an attempt to restore the normal balance between oxygen and carbon dioxide. If you continue long enough you’ll faint, but instead you exhale and breathe.

That’s a system with a feedback loop; breathe in some oxygen, when the pH level gets too high, exhale.  In turn, systems have sub-systems.  We could explore respiration sub-systems, the muscular system and its use of ATP, or the digestive system and Krebs cycle, and so on.

As a sailor, I’ll remind you that the entire basis of systems theory is navigation and helming boats.  The origins of cybernetics during and after WWII used Greek words for navigators and helmsmen.  Kubernaein [kubernáo], governance, meaning ‘to steer’ is the root word for us today.  Systems boil down to feedforward and feedback.  Feedforward is deciding where to go.  Feedback is deciding how well you’re doing getting there.  Feedforward and feedback often come carrots and sticks, gains and losses.

Systems theory also encompasses machine-learning, artificial intelligence (AI) if you will.  Here, the most basic concept is to distinguish patterns and anomalies from historic data.  In many ways, teaching scientific enquiry may be the most important and basic skill, how to deal with anomalies.

The purpose of education isn’t hard to fathom.  Though I’m inordinately fond of theologian Timothy Radcliffe’s definition of universities, as places where we “learn to talk to strangers”, a Wellcome Trust submission to a 2016 government consultation says, quite reasonably, “The purpose of education is to prepare people for life, equipping them with the knowledge and skills to contribute to a thriving society”.[3] Education trains us to reinforce good patterns, remove bad patterns, and spot anomalies. 

Systems theorists represent systems as nodes and flows.  Complexity theory, building on systems theory points to surprising ways in which many systems are alike, from financial systems, environmental systems, and governmental systems, to tonight’s topic, global educational systems.  Here are four:

  • Emergent behaviour – however simple the components, the system can behave in complex ways, seemingly with a personality. 
  • Self-organising – often resulting in unintended consequences. 
  • Criticality – absorbing large amounts of change up to a threshold but collapsing suddenly in a shock.  Making a system more efficient or synchronised or standardised often increases fragility by creating single points of failure.

Last year, in this Egyptian Hall, a government minister managed to insert 46 mentions of ‘innovation’ in a ten minute address.  I was counting.  Impressive, but he might as well have said ‘detergent’ or ‘lace knitting’ for all that 10% of his verbiage meant.  The world is changing fast, yet large systems are particularly poor at innovation.  Systems are designed to drive out variety and redundancy.  In the UK we’ll put 100 accountants to work to find ways of ensuring two people aren’t caught performing the same job.  W B Yeats believed that “Education is not the filling of a pail, but the lighting of a fire”.  We are born polymaths, interested in everything.  Then we are killed off, often by the education system itself reducing variation.  We talk about pupil-centred education but hear a group chant from Monty Python’s Life of Brian, “we are all individuals”.

The bigger the system the more sub-systems we tend to give it to manage.  We’re guilty of concluding that efficiency always results from banging two organisations together; what was the question?  Think Inland Revenue and Revenue & Customs, military regiments, or DFID and FCO.  Our education system is responsible for many sub-systems:

  • Academic learning;
  • Pastoral care;
  • Status and prestige;
  • Credentials and degrees;
  • Employability;
  • Citizenship and pride;
  • National productivity;
  • and so on…
  • And as we saw during the pandemic, an implicit system of childcare and babysitting, in brief, state orphanages.

This is way way too much, to the point we can’t answer the question, “when would we know our education system is working?”  I learned long ago that before you automate or integrate, you first simplify.  Instead, we complicate.  Education is now being roped into recent government plans to make the UK a technology superpower.  Further, systems conflict.  If you believe that a small percentage of any classroom would thrive in a corner teaching themselves, and that a small percentage of any classroom verges on hopeless, then the economics of productivity argue to leave some children behind.  Another economic conflict is hearing on radio talkshows that graduates, particularly doctors, nurses, and STEM graduate, are ‘owned’ by the country and should pay back for the mandatory education they had to take if they emigrate.  Another systems conflict is that if you’re going to be inclusive you can’t be elitist, so I sometimes wonder if ‘good’ schools impede social mobility.

Yet another example of conflict among systems is between breadth and depth, the tension between generalism and specialism.  Globally, this tends to be a conflict between the liberal arts of natural sciences, social sciences, arts, and humanities pitted against jobs and professional courses.  In the UK, the liberal arts university end, philosophy, politics, and economics (PPE), a descendant of the “Modern Greats”, faces off against vocational colleges and comes with a lot of class baggage, but the same debates occur between breadth and applicability around the world, even if with a bit less vehemence.  Locally, I’d point to an attempt by the Livery Companies Apprenticeship Scheme to pilot post-nominals for apprentices.

So turning to the title of the lecture, let’s begin with:

“I” – Solo learning until recently consisted of reading in the corner.  For a minority of the best students, this often worked well but was unsuitable for the masses.  Most faddish educational technology is solo, direct person-to-computer learning.  Solo learning applications are characterised by ‘gamification’, such as mathematical or language apps, ‘mind gyms’, and versions of augmented and virtual reality, such as SkyMap helping children and adults with navigating the skies.  I myself am an avid DuoLingo user.  Frankly, if you’ve tried these, admit it, they can be fun, at first.  They can also become repetitive and tedious swiftly.  Other solo applications have taken on some classroom rote tasks – the ones we chanted when we were little, such as the alphabet – where repetition is helpful.  Traditional large lecture formats are also going solo.  On my own I can watch the-best-of-the-best from Oxbridge or Gresham to the Ivy League to TedX to Grand Écoles to Nobel Laureates, and increasingly Asian.

These applications continue to develop in a highly competitive environment. Deeper science is being, based on genuine brain activity, as witnessed at Birkbeck’s ToddlerLab, where infra-red skull caps track toddlers real-time at home, in classrooms, at play, and in virtual reality.  Research is questioning fundamentals; does dopamine function by reinforcing forward from cause to effect, or backward from effect to cause.  We’re beginning to see that glial cells may be as important as neurons in thinking.

Giving students today’s solo technology is a bit like giving them a quill pen centuries ago, confident that if better pencils and pens come along they will only enhance the time spent using them.  Though we do have to limit apps that “detonate children’s powers of attention”.[4]    As an aside, I recall the story of a student who when asked where her homework was, replied, “It’s still in my pencil.”

Xun Kuang (Chinese: 荀況; c. 310 – c. after 238 BCE, often known as Xunzi) said in the third century BC something that has rattled down the ages: “Not having learned it is not as good as having learned it; having learned it is not as good as having seen it carried out; having seen it is not as good as understanding it; understanding it is not as good as doing it.” Solo technology automates imparting knowledge, demonstrating knowledge, understanding knowledge, and assessing knowledge acquisition, up to a point.

“Scholar” – This role has seen perhaps the most change to date.  Note that entire disciplines are now just technology disciplines.  My daughter reminded me that her MSc was in Geomatics, a discipline that would almost disappear if technology disappeared.  The same applies to the hot job market in data visualisation.  Large-scale databases and algorithms have transformed direct research, with connections to wider and deeper resources than ever before.  Preprint services, such as Arxiv.org, share draft research papers publicly before peer review, speeding up global research tremendously. Many fields have been transformed by these enormously larger networks, ranging from early English studies to mental health. 

Despite this frenzy of connectivity, there are questions about the significance of progress, quantity over quality, and a reduction in innovation ‘efficiency’.[5]  Research is one area where machine-learning may help us enormously, as we truly have little idea how good scholars genuinely conduct research.  We need to rethink how research is done making it more practical earlier. We need to work out how to combine more research students and workers better in a single institution.

“Tutor” – Of all four roles, tutor is most problematic.  Expensive one-on-one interaction with little ability to gauge its effectiveness.  Thus, tutorial is likely to be the area most affected by new management technologies.  For solo learning we can set a fixed amount of time, say two hours out of six per day per student.   We can evaluate scholars on research advances.  The Tutor role is where we need better technology to help assess students, target specific interventions, and increase the amount of interaction.  This technology is likely to range from appointments and assessments to biological tracking things like eye movements or brain activity.

“Teacher” – Teacher is an emotive word encompassing a profession of inordinate responsibility and complexity.  Yet, solo learning is lessening the need for rote lessons and for lectures.  New technology, such as white boards in every classroom have largely been a placebo to suit ministers who announce ‘one for every classroom’.  They do no harm, but haven’t done much good either.  Technology can and should focus on helping teachers.

“Worker” – We were bound to get to productivity.  Lifelong learning brings us to earnings.

National productivity problems are widely ascribed to three things – education & skills gaps, lack of investment, and a failure to make knowledge connections, Late Lord Mayor Vincent Keaveny worked on skills as People & Purpose.  Lord Mayor Lyons is working on Financing Our Future.  Subject to election I hope to work on connecting knowledge networks.  Still, despite our Master’s efforts and the City’s intense involvement in education, with our history of guild training, Christ’s Hospital, Gresham, Birkbeck, London Metropolitan, City, Imperial, City & Guilds, and numerous other schools, apprenticeships, and support, UK educational system performance has been flat as assessed by the Programme for International Student Assessment (PISA), a triennial survey by the OECD of 15-year-old students around the world, from 2000 till the last release in 2018.  And before you blame money, UK funding has been remarkably flat and average within the OECD.

In some senses productivity gone backwards. If you’ve done A, B, and Q, perhaps you are a qualified project manager, even though you don’t have the full certificate.  Because we are poor at assessing inferred skills, we rely too much on education qualifications that require much more time to acquire.  Society’s productivity has gone down to some degree because we require far more qualifications than are needed for many roles.  We have a skills problem indeed, but we also have a skills recognition problem.  The future lies in a solid 3R grounding with people piecing together many more mini courses with their real-life experiences to validate inferred skills.  This is happening widely.

A $1.7 billion UK unicorn, Multiverse, focuses on human-to-human coaching and in two years  has gone from 1,000 apprentices to over 10,000, now granting applied degrees that measure what you can do, not just what you know.  A lot more scientific rigour needs to be applied though to a lot more such experiments around the world.

Let’s return to innovation.  In 1968, the Finnish parliament introduced legislation to reform the education system. Finland spends 30 percent less on primary education per child than many countries, money spent on learning rather than administration or classroom technology.  At age seven, Finns start school later, with few tests on progress, but the few they have matter.  First, the tests are random samples, not universal. Second, scores are used to assess students not the school. Third, test results don’t affect funding or ranking.  By 2006, Finland was first out of 57 countries in science and mathematics and second in reading.  There are some basic tenets, but I would emphasise two, ‘equal opportunities for all’ and ‘in teachers we trust’.  I might also note that the Finns have fewer, stronger, and shorter feedback loops on progress.

Let’s pause here for a moment.  We’ve quickly explored what systems theory infers and then surveyed worker, teacher, tutor, scholar, I.  Now for two commentaries, on changing systems and what it means to be human.

Changing Systems

A systems engineer would look at the mess of systems in education and immediately think of cutting them down to a more manageable structure.  Such an engineer would streamline processes, and seek checks and balances among accreditation, testing, certification, and training.  Such an engineer would align risks and rewards, as well as strengthening feedback and feedforward loops.  Along the way, the bulk of information technology would be aimed at recording, analysing, and reporting to ensure that people within the system had the information they need to make decisions.  This is all good, but there is a complementary approach, to recombine systems.

In the Netherlands, there is a much higher degree of student involvement in running institutions.  Community governance is seen as training.  One anonymous Dutch university graduate said to me, “Dad, I learned so much helping to run the school, such as every meeting needs an agenda.”  Note that I spend about five years getting my UK graduate employees to grasp this basic skill.  Note too so many systems are wrapped up in one package, organisational theory, teamwork, management.  I was pleased to hear at Merton Court, a primary school in Sidcup, that the council chairman was a student, and the student held the gavel, not the headteacher.

“What we need is a good economic argument for giving hungry children free meals.”
[With thanks to OpenAI’s DALL-E for collaboration on the image]

During the World Cup in November, Japanese fans and players garnered kudos by exhibiting their normal habits of cleanliness.  So here’s a question – who cleans Japanese schools?  The students, of course.  As one head remarked, 700 secondary school or 300 elementary school students can do wonders in fifteen minutes every day.  Just after the war, the Japanese seemed to take a recent Private Eye cartoon caption seriously, “What we need is a good economic argument for giving hungry children free meals.”  So, who cooks in Japanese secondary schools?  The students, of course.  Trick question, who cooks in Japanese elementary schools?  The students, of course.  Yes, there is support and supervision, but since the war the students do the cooking, serving, and cleaning every day.  So the Japanese have re-engineered a mess of feedback into one system – citizenship, teamwork, chemistry, biology, planning, scheduling, logistics, etc.  This is real systems innovation. 

What It Means To Be Human

Korea is trialling robots in 300 kindergartens to prepare children for the future.[6]  Robert Hercock at BT Labs was asked how humans might find working with AI. He pointed out that we already have experience, for example working with dogs and horses.  Personal productivity in the future will both need strong basics in reading, writing, and arithmetic, combined with the ability to use technology that takes up our ‘cognitive load’.  Speculative fiction author Neal Stephenson has two characters discussing hiking in rough rock terrain.  ““Even walking is hard—every footstep requires you formulate a plan … in such terrain. It’s exhausting. Mentally exhausting.  The cognitive load of having to think about each step.” “That’s why people who knew this land used horses. The horse handles the cognitive load. You just tell it where to go and how fast.”“[7]

In an earlier work, The Diamond Age, Stephenson wrote about an adaptive book, the Young Lady’s Illustrated Primer: A Propædeutic Enchiridion.[8]  The Primer steers its reader intellectually towards a more interesting life. “In your Primer you have a resource that will make you highly educated, but it will never make you intelligent. That comes from life.”

Machine-learning is helping us to understand what is irreducibly human.  This has two implications, first we are beginning to see more clearly where people can add value to society, in human-to-human interaction and support.  If you think something is likely to be automated in the near future, then put less emphasis on teaching it.  The premium is on people.  Like the Barbra Streisand number, “People, People who need people, Are the luckiest people in the world”.

Advances in machine-learning imply that we should be spending much more time ensuring the soundness of deep foundations in reading, writing, and arithmetic, the three Rs, and tolerate wide shallows elsewhere.  Do we move from the 3Rs to skills too rapidly in early years? 

Secondly, empathy with other people will become the crucial skill.  You will be less able to hide in a non-social specialist skill set. What can’t be automated is being human.  Automation polarises my view of education in the future – more hard basics and a focus on people skills, with a light dusting of many broad subject areas.

One day a teacher asks her student Johnny, ‘Johnny, if there are two birds on a wire and I fire two barrels from a shotgun, how many birds will I hit?’.  ‘One, Miss’.  ‘Johnny, please listen, if there are two birds on a wire and I fire two barrels from a shotgun, how many birds will I hit?’.  ‘One, Miss’.  ‘Why Johnny?’.  ‘Well Miss, after you fire the first barrel the second bird will fly away.’  ‘Johnny, that’s the wrong answer, but I like the way you think.’

The next day Johnny comes into the classroom.  ‘Miss, my Dad says that I must save my allowance.  One bank offers me an educational booklet.  The other bank has a very pretty teller.  Which bank should get my account?’  The teacher blushes, and says, ‘Well, perhaps the one with the very pretty teller?’  Johnny replies, ‘No Miss, the one with the biggest government guarantee, but I like the way you think!’.

Rather than seeing education as an effort to try and get the universe into your skull, try to learn how other people think.  Think like a scientist, an accountant, an electrician, a lawyer, an administrator.  It’s why I love the Raspberry Pi, not end-user computer skills you can pick up on the job, but learning how programmers think.  In Delft they have a Studieverzameling (study collection) where thousands of old engineering exhibits help today’s students see how other engineers solved similar problems in earlier times with different technologies.

And old tech, books, teaches empathy.  Over a lifetime’s reading you can empathise with thousands and thousands of people.  Carmen Callil, the Founder of Virago, said, “Reading is a way of becoming the person you’re interested in being.”

I shall suggest five areas for us to discuss tonight, focus on time, speed up scientific experimentation, increase our variation tolerance, emphasise technology for teachers, develop lifelong communities,:

One: focus on time

Why in an era of lifelong learning do so many recent UK graduates state, “I’ve had my last exam; I’m never taking a test again”?  Longanesi said, “Everything that I don’t know I learned in school.”[9]  Why are children bored with education?  Because we waste their time.  While a certain amount of student time should be set aside for high-quality solo application, the rest of their time must be spent on interesting human engagement.  That means teachers, and each other.

William James, said, “My experience is what I agree to attend to”.[10]  Time expenditure is everything.  Herbert A Simon took this further – “What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.”  Tim Wu, professor of law at Columbia University, adds, “… we must reflect that, when we reach the end of our days, our life experience will equal what we have paid attention to, whether by choice or default.”  Perhaps a teacher’s core responsibility is to help students learn where and how to spend their attention.

A good example of helping students with time is touch typing.  The time spent learning touch typing early in life provides enormous payback during the rest of their lives.  I would encourage constant time mapping, tracking, and analysis of students.  Where is Jonny spending his time exploring how others think?  All his time as a computer game general, or getting a good breadth of roles from soldier to ballerina to forest ranger.

Two: speed up scientific experimentation

When I was ten, we were allowed two films a year at one of the schools I attended. I remember one film was a fairy tale about a cat, the other, ‘Hemo the Magnificent’, about our circulatory system.  You really remembered things when you only had two.  How much things have changed.  There is a lot of experimentation around.  Kids’ Lit Quiz turns reading children’s literature into sport; we have the Harmony Project, Heritage School Cambridge, London Interdisciplinary School, Black Mountains College, University of Chicago Laboratory Schools, (42) coding school in Paris, King Abdullah University of Science and Technology in Saudi, NewGlobe in Nigeria.

We have seen huge changes in just a few years.  Yet it’s not enough to keep up.  We need much more experimentation.  Covid helped show how slow change was.  Almost instantly, video-conferencing abolished snow days, ill children attend class, teacher training is online, alumni feed back their experiences directly to students, socially reluctant students participate in collaborative software, parent evenings are smoother and faster and fairer, and on Zoom everyone is in the front row of class. 

It’s not just tech, it’s environments and processes.  We’re learning from studies about myopia due to young students not having sufficient direct sunlight or why China has stopped tutor cramming.  Due to covid, flipped classrooms, where pupils complete readings at home and work on live problem-solving during class time, have burgeoned. 

More competition will help increase experimentation.  As we over-certify our educational organisations in so many respects, we create barriers to entry, hampering start-ups and challengers.  I would challenge independent schools and charitable schools to take more risks, not fewer, to prove that they are helping the advance of knowledge while the state sector lags.  I would also insist we pull our socks up on scientific evaluation.  One of my daughters participated in a school iPad rollout.  She pointed to students, chastised for reading books in class, now able to read books on the iPad but hide it from teachers.  It was apparent to her at 15 that it wasn’t an experiment, just trying to look trendy.

Managing scientific research years ago, when budget time came my favourite question was “if we’re a learning organisation and your department has learned so much this past year, then why is your budget identical to last year?”  I would discourage too many bulk central purchases.  I would let schools pick and choose different portfolios of new applications so we see what works, where, in what combinations.

Three: increase variation tolerance

The UK exhibits Postcode Lottery Terror.  Many UK fact-finding missions head to the Nordics or Asia to see exemplary systems.  They ask what has been scientifically proven with about twenty years of evidence and, converted to some initiative, fly back to the UK.  I remember a Finn incredulously asking, “why don’t they ever ask us what we’re doing now?”.  There is no time for cultural adaptation or piloting; the initiative has to pay off not just in one electoral cycle, but one minister’s term of office.  Certainly no time for progressive rollout.  If the initiative were successful then it would unfairly disadvantage those to whom it hadn’t been applied, wouldn’t it?  The infamous postcode lottery.  Not surprisingly without adaptation, piloting, and progressive rollout, many UK government initiatives fail; there is no learning loop, no feedback.  Britain has a problem accepting variation.  Some countries may tolerate excessive variation, the US perhaps, others seem less tolerant than the UK, though this is often confused with their small size, for example Singapore. Good ideas such as national curricula and league tables sound great, but come with a cost of reducing variety.

We need to increase our tolerance of variability.  Education systems are not about just educating the ‘middle’ or average, but also the smart and the disadvantaged. Recognising that there are ethical issues by experimenting on students, we should work to make safeguards stronger, simpler, and faster to speed up experimentation, rather than use ethics as reasons to hamper learning.

Four: emphasise technology for teachers

Jacques Barzun said, “Teaching is not a lost art, but the regard for it is a lost tradition.”  If student solo time is fixed within a closed competitive framework, then the real productivity gains will come from making teachers much more effective.  More effectiveness means de-emphasising lecture skills and putting more emphasis on social interactions and group learning, helping the disadvantaged regain their potential, and providing the carrots and sticks of motivation.

I remember providing semi-automated bookkeeping systems for headteachers around 1990.  Things are only a bit better.  Current teacher aid software is clunky and seamfull, not integrated and seamless.  Proper class, teacher, and student dashboards are coming but not yet here. 

Still, we can already see that keystroke to keystroke recording will enable diagnostics helping teachers work out in minute detail their personal performance, their class performance, and each pupils’ performance with specific guidance points on what interventions will improve things.  Teachers should be able to gain time to engage much more socially with students from tutorials to Socratic lessons.  Teachers should have marketplaces for lesson plans, buying and selling popular playlists of lessons that work.  Heads should be able to benchmark things like their school’s culture.  Support software should be used to reduce teacher attrition rates, particularly reducing stress for initiates.

Teaching can be terrifying, facing up to student cliques and groups.  I remember delivering a lecture on economics to an unruly East London secondary school where the teachers had lost control of discipline.  Groping for an idea I shouted, “Who wants to sit down and learn how to make lots of money?”  It worked, but what about other situations. A prison school I visited focused on discipline by highlighting choices and consequences, and it seems to work well.  Helping teachers learn how to discipline is one gap in the software world we ought to address. 

The overall technology lesson is don’t get overexcited about any specific technology; it’s evolution not revolution.  Yet do get excited about the end goal, freeing teachers for more and more relational time with students.  And just before we turn to our fifth discussion point, community, why isn’t educating parents in their role as primary educators more serious, not just cyber-risk, but assessing parental strengths and weaknesses, how to help with homework, provide counselling, simple advice about getting teenagers out of bed.

“Lifelong Education Communities”
[With thanks to OpenAI’s DALL-E for collaboration on the image]

Five: develop lifelong education communities

Eight years ago the world sat down and hammered out 17 Sustainable Development Goals.  Number 4 reads – “Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all.”  Shouldn’t a dedication to lifelong education change the entire way we live?

Over the past two centuries, our lives have moved from learn, work then die; to learn, work, retire then die; to now learn, work, relearn, keep contributing, then die.  Education is looking a lot more like work, experiential and team-based.  The navy recognises this: Rear Admiral Jude Terry recites, “Join well, train well, live well, work well, leave well, but most of all think well”.

Z/Yen’s Long Finance initiative studied long-lived institutions and concluded that two forms dominated, ecclesiastical organisations and educational establishments, often combined in the same institution.  Guilds began as a combination of ecclesiastical, educational, and employment. It was community that kept them going through time – anthropologists have a harsh definition of community, it’s not touchy-feely, it’s about debt.  Communities are groups of people with shared indebtedness to one another, and to previous and subsequent generations.

Education is a long-lived product with feedback taking a long time to get back to inform improvements.  What is education’s value added?  Who is accountable for education?  Should education be free?  How can education be less of a pig-in-a-poke purchase?  Can we ethically answer the cost/benefit questions students have?  These conundra, if solvable, may result in radical new societal, not just educational, models.

One future model might be to auto-enrol in a cohort that will take you from nursery through to death.  Sure you can change cohorts, records should follow the learner not the provider, as we want choice and competition, but a good cohort takes on all of your education and employment needs.  Rather than a school guidance counsellor sending you off to work at the end of your education without much feedback for them about your career success or failure, the guidance counsellor takes you into work and has the necessary information to guide your combination of career and lifelong learning going forward. 

Some newer education models do link themselves to employment guarantees already.  We could go much further, augmented education bonds perhaps. Imagine if your university reached back to secondary, then primary, then nursery, while also drawing on its investment in people who pass through its doors via tithes on employed community members, while taking back those who need retraining. All on one-site demographics eliminating the academic-vocational distinction.  Or educational cohorts might resemble souped up trades unions or pension schemes for education and life-time employment.  Or cohorts may emerge from massive open online courses.  Or from large employer combines, as with the Handelskammer in Germany or our guilds, might step up to succour cohorts.  Or perhaps cohorts take the form of lifetime general practitioners, trusted gateways to education and employment services. 

It’s trendy to talk of 15 minute cities, where work, shopping, education, health, and leisure, are all within an easy 15-minute walk.  Think of 15 minute educational cohorts where everything from the 3Rs to counselling to laboratories to tutorials are all within 15 minutes of your home, until you go out into employment.  We should be talking about super-communities rather than super-heads.  School provides children with a fresh start every year, a chance to reinvent themselves.  From sports child to nerd, from shy to gregarious, from irresponsible to responsible.  This can happen within communities where people support one another’s development.

Before this lifespan education sounds too crazy, remember the craziness of a system that asks twenty year olds to make forty year student loan decisions.  We only asked seven years of indentured servants.  And this from governments that can’t make sensible short or medium-term financial decisions, let alone sensible pension decisions.  Governments can be unreliable allies for long-term-thinking institutions.  Educational cohorts would be mutual cooperatives with people looking after one another, and relying on one another.  A community.  By sharing our futures we solve our problems together. 

If “Education is what remains when we have forgotten all that we have been taught”.[11]  Wouldn’t it be nice if what remained was community?


Roy Amara’s Law has gained widespread currency, “We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.” My corollary to Amara’s Law might be, “We tend to overreact to problems in the short run and underreact to problems in the long run.”  I would encourage us Educators to react to today’s problems. 

We’ve explored changing systems and what it means to be human, as well as set out five suggestions, focus on time, speed up scientific experimentation, increase variation tolerance, emphasise technology for teachers, and develop lifelong education communities.

I will close by sharing with you Jaron Lanier’s concept of ‘karmic vertigo’ and computer systems.  Lanier observes, “The computer code we are offhandedly writing today could become the deeply embedded standards for centuries to come. Any programmer or system designer who takes that realization on and feels the full karmic burden, gets vertigo.” Stewart Brand explains: “The karmic view of the future can be as distorting as the discounted view. Instead of the reduced responsibility of discounting, karma can impose crushing responsibility, paralyzing to contemplate.”

The pace of educational change needs to match the revolutionary scale of change in society.  Changing education systems involves crushing responsibility, but we must overcome the dizziness and paralysis of that karmic vertigo and change our systems to meet the responsibility we owe to those whom we educate.  This change can come from within with the resources we have today. 

As they say in the mental health field, you don’t have to be ill to get better.  Whereas, if you don’t want to improve, nobody can help you.  I implore Educators to overcome karmic vertigo and get going on improving our systems.

Thank you.


Preparing for this lecture involved visiting primary schools, secondary schools, and numerous universities in several countries over the past twelve months.  In no way could I have written this without the help of many people, and while in no way are they responsible for the direction I’ve taken things, I would like to thank the following I can remember with apologies to any I’ve overlooked. You were all helpful in so many ways, though probably unaware how much – Ian Angell, Elisabeth Barratt, Nick Barratt, Brian Blake, Philip Bond, Nicholas Botfield, D’Maris Coffman, Tim Connell, Sarah Cook, Jon Davis, Brennen Dicker, Katie Duff, Martin Elliott, Richard Evans, John Franklin, Clare Fraser, Marian Gamble, Malcolm Gill, Malcolm Grant, Catherine Griffiths, Anthony Finkelstein, Caroline Haines, Andy Haldane, Ian Harris, Robert Hercock, Ted Hoefling, Simon Hughes, Kevin Ibeh, Greg Jones, Wendy Joseph, David Latchman, John Lloyd, Maxine Mainelli, Xenia Mainelli, Nicholas Mainelli-Barre, Denis Mareschal and the team at Birkbeck’s ToddlerLab, Jeremy Mayhew, Maura McGowan, Luke Miller, Ajit Mishra, William Orr-Ewing, Robert Pay, Dom, Sarah, and Henry Price and the team at Merton Court, Linyun Qiu, Richard Russell and the team at Colfe’s School, Anne Punter, Simon Reid, Janet Reynolds, Wolfgang Schlör, Naresh Sonpar, Susan Steele, Neal Stephenson, Jude Terry, Emily Thomas and the team at HMP Isis, Nick Timothy, Leslie Tucker, Charles Vermont, Mike Wardle, Edward Wild, Sophie Windsor, Keri Wong, and Stephen Wordsworth.



Further Browsing:

[1] Conundra or conundrums – https://en.wiktionary.org/wiki/conundrum,  https://asktheleagueofnerds.com/the-conundra-conundrum/, https://www.theguardian.com/notesandqueries/query/0,5753,-5253,00.html

[2] https://quoteinvestigator.com/2020/01/05/intelligence/

[3] https://wellcome.org/sites/default/files/wtp060177.pdf

[4] https://thecritic.co.uk/rage-against-the-machine/

[5] and a joke I couldn’t resist sharing – Student: “How did I do on my research paper?”  Teacher: “Actually, you didn’t turn in a research paper. You turned in a random assemblage of sentences. In fact, the sentences you apparently kidnapped in the dead of night and forced into this violent and arbitrary plan of yours clearly seemed to be placed on the pages against their will. Reading your paper was like watching unfamiliar, uncomfortable people interacting at a cocktail party that no one wanted to attend in the first place. You didn’t submit a research paper. You submitted a hostage situation.”


[7] Neal Stephenson, Termination Shock, Harper Collins (2021).

[8] Neal Stephenson, The Diamond Age: Or, A Young Lady’s Illustrated Primer, Bantam Spectra (1995).

[9] “Tutto quello che non so, l’ho imparato a scuola”, Leo Longanesi (1905-1957), editor and journalist.

[10] William James, The Principles of Psychology, Vol.1 (1890).

[11] https://quoteinvestigator.com/2014/09/07/forgotten/