Democratech: reflections on the human nature of blockchain

Posted April 30, 2025 by Jon Dron

At short notice I was invited to be guest of honour and keynote at Bennett University’s International Conference on Blockchain for Inclusive and Representative Democracy yesterday. I was not able to attend the entire conference – my opening keynote was at 9:30pm here in Vancouver and I eventually needed to sleep – but I made it for a few hours. I was impressed with the diversity and breadth of the work going on, mainly in India, and the passionate, smart people in attendance. It was a particular pleasure to hear from Ramesh Sharma, who I have known for many years in an online learning context, here speaking of very different things, and I really loved the ceremonial lighting of the lantern – the sharing of the light – with which the conference began. It is a powerful and connecting metaphor.

Like most geeks I do have the occasional thought about blockchain and democracy but I can’t describe myself as an expert or even an enthusiastic amateur in either field. So, rather than speaking about things the delegates knew far more about than I, and given the compressed time-frame for preparing the keynote, I chose to ground the talk in familiar territory, taking a broad-brush view of how to think of the technological ecosystem into which the technologies must fit. It led to some new thoughts here and there: in particular, I rather like the idea of technologies in general acting as a kind of distributed ledger of human cognition. The result was these slides – Democratech: reflections on the human nature of blockchain.

In rough note form (not a polished academic work and not particularly coherent!), the text below is approximately what I spoke about for each of the slides:

1 In this talk I will be using ideas from my most recent book: here it is. You can download it for free or buy it in paper or electronic form if you wish. See http://teachingcrowds.ca. It is at least as much about the nature of technology as it is about the nature of education, and that’s what I want to talk about today: what kind of a technology is blockchain, and why does it matter?

2 “Technology” is a fuzzy term that can mean many things to different people. I spend a whole chapter in the book exploring many definitions of what “technology” means. To, save time, I am going to use what I conclude to be the best definition, from Brian Arthur, “orchestrating phenomena to our use”.

3 I prefer to think of this as “organizing stuff to do stuff”, because it makes it clearer that the stuff that it organizes nearly always includes stuff already organized to do stuff: as Arthur observes, almost all if not all technologies are assemblies of other technologies, at least when they are put to use.

Technologies are made of technologies, at every scale, and they are parts of webs of technologies that stretch far into time and space. Kevin Kelly calls this massively interconnected network the technium. And, as he puts it, technology can be thought of as both a thing and a verb or, as Ursula Franklin puts it fish and water – a slippery thing to pin down. It is something we do and something we have done. In fact it is typically both.

4 By this definition, democracies are technologies too – in fact, hugely complex assemblies of technologies. They orchestrate phenomena using systems, physical objects, and assemblies of them, to approximate a fair voice for all in the governance of where we dwell. So are words, and language, and, as Franklin notes, there are technologies of prayer.

5 If you take nothing else from this speech, take this: only the whole assembly matters. The parts are very important to the designer and make a big difference to how a technology works and is experienced, but it is how the parts are assembled and act together that makes the technology as it is experienced, as it is instantiated. That includes what we do with them – more on that in a moment.

If you are not convinced, think about some of the parts of the computer you are looking at now: some are sharp, some contain harmful chemicals, and there’s a good chance that there is a deadly amount of electricity flowing through them, and yet we gain benefit from them, not loss of life, because we assemble them in ways that (at least normally) eliminate the harm by adding technologies to prevent it: counter technologies. Often, a large part of what we recognize as a technology is in fact a counter technology to other parts of it – think of cars, for example, where many of the components are simply there to stop other components blowing up, seizing, or killing people.

6 Technologies create what Stuart Kauffman calls “adjacent possibles” – empty niches that further technologies can fill, individually or in conjunction with others, including others that already exist. Every new technology makes further technologies possible, adding new parts to new assemblies. This accounts for the exponential growth in technologies over the past 10000 years or so: technologies evolve from and with other technologies, almost never out of nothing.

Those adjacent possible empty niches are fundamentally unprestatable, as Kauffman puts it: no one can imagine all the possible assemblies into which we might put something as simple as a screwdriver. A stirrer of paint, a back scratcher, a scribe, a pointer, a stabbing weapon, a weight, a missile, a crow bar… And this is true of every technology. All can be assembled differently, in indefinitely many assemblies, to make indefinitely many wholes. This is true at the finest of scales. Though there may be some very close resemblances between instances, you have never written your own signature, nor washed your clothes, nor eaten your food the same way twice. Only machines can do that, but they are part of our technologies as much as we are part of them: the machine may behave consistently but the technology through which we use it – the instantiation in which we participate – most likely does not.

Technologies also come with path dependencies that can harden and distort assemblies, because the soft must shape itself around the hard. What exists shapes what can exist.

7, 8 When instantiated, we are participants in, not just users of, the technology. Using a technology is also a technology: whether organizing it or being part of the organization

9 , 10 We are coparticipants in a largely self-organizing web of technology that is part organic, part process, part physical object, part conceptual, part structural. Technologies democratize cognition though they also embed and harden values of the powerful, and the uses to which they are put are too often to subdue, constrain, or abuse our fellow humans. It is always important to remember that the technology that matters is seldom its most obvious components: it is the assembly they are in. As they are used, they are different technologies to everyone that uses them, because they are parts of different assemblies: the production line is a very different technology for its boss, its workers, its shareholders, the consumers of what it produces, orchestrating different phenomena to different users. This means that technologies – as instantiated – are never neutral. They have histories, contexts, and propensities.

11 And our input matters: it is not just the method but the way things are done that matters. Every assembly can be a creative assembly, and it is possible to do it well or badly. And so we all create new adjacent possibles for one another. Through technologies we participate in the collective cognition of the human race: in effect, technologies form the distributed ledger of our shared cognition. But all of us assemble and interpret in the ways we use technology, whether we form part of it (hard technique) or are the organizers (soft technique).

12 Blockchain is a technology capable of achieving great good: potentially accountable but equally interesting in ways it can support anonymity, free from central control but also interesting in the context of an existing system of trust, good for both privacy and transparency, etc. It has indefinitely many adjacent possibles, from the exchange of property to the assertion of identity, from enabling reliable voting to making supply chains accountable.

13 But all technologies are what Neil Postman called Faustian bargains. When you invent the ship you invent the shipwreck as Paul Virilio put it. The story of the Monkeys Paw, by W.W. Jacobs is a tale of horror in which a monkey’s paw grants three wishes to a modest couple, who ask only to pay off their mortgage with their first wish. Moments later, they learn their son has died in a horrible accident at a factory in which he works and the company will pay compensation: the exact cost of the outstanding mortgage. And so the story goes on. Technologies are like that.

Blockchain can be subverted by organized crowds (botnets and human), malware, cracking, etc, and quantum computing means all bets are off about reliability an security. It is possible to lose votes as easily as it is to lose millions in bitcoin. Blockchain can conceal criminal activity, and, conversely, enable a level of surveillance never seen before. Remember, this is all about the assembly, and blockchain is a very versatile component. It’s a super-soft technology that connects many others. Blockchain makes new forms of democracy possible, but it also enables new forms of tyranny.

To understand blockchain we must understand the technologies of which it forms only part of the assembly. Never forget that it is only ever the assembly that matters, not the parts. This is and has always been true of all the technologies of democracy. Paper voting, say, in its raw form is incredibly and fundamentally unreliable, prone to loss, error, abuse, corruption, coercion, loss of privacy, etc and it is terribly, terribly inefficient and insecure. However, we throw in a lot of counter technologies – systems to assure reliability, safes, multiple counts, policing procedures, surveillance, electronic counts, , observers, etc – and so the process is now so well evolved that it often enough works. Paper is not the technology of interest: it is the whole system that surrounds it. Same for blockchain.

14 Understanding technologies mean we we must know the adjacent possibles but, remember, we we can only ever see the most brightly lit of these from where we currently stand. The creative potential, for both good and evil, is barely visible at all. Someone, somehow, somewhere, will find new assemblies that achieve their ends, whether it benefits all of us or not. Sadly, those most able are typically those least trustworthy thanks to the fundamental inequalities of our societies that reward greed and that give most to those who already have most. Anything is weaponizable, including democracy, as (here in Canada) our neighbours south of the border are discovering to their cost. And it means understand what happens at scale: the environmental impacts and counter technologies to that: but, as Reneé Dubos put it, fixing problems with counter technologies is a philosophy of despair, because every counter technology we create is another Faustian bargain that creates new problems to solve, and new adjacent possibles we never foresaw.

15 We must understand where blockchain fits in the massive web of the collective technium – the Ricardian contracts, the oracles, the legal frameworks that surround them, the ZKP techniques, the privacy laws, the voting practices, the laws of ownership, and so on. It is unwise to simply drop it in as a replacement for what we already do because it will harden what should not be hardened – when we automate we tend to simplify – and create new relationships that may be incompatible or positively dangerous to existing technologies of democracy. But, as we reinvent it, we must always remember the unprestatable adjacent possibles we create, the things we reinforce, the things we lose. And we must remember that someone, somewhere is seeing adjacent possibles we did not imagine, assemblies we have yet to conceive, and they may not be friendly to democratic ideals.

16 To understand this means we must look far beyond the bits and bytes and flashing lights; we must make empathetic leaps into the hearts and minds of our coparticipants in the technium. We are technologies, as much a part of blockchain as it is part of the broader web of the technium.

What kind of technologies do we want to be?

Video and slides from my webinar, How to Be an Educational Technology: An Entangled Perspective on Teaching

Posted September 24, 2024, updated onSeptember 25, 2024 by Jon Dron

For those with an interest, here are the slides from my webinar for Contact North | Contact Nord that I gave today: How to be an educational technology (warning: large download, about 32MB).

Here is a link to the video of the session.

I was invited to do this webinar because my book (How Education Works: Teaching, Technology, and Technique, briefly reviewed on the Contact North | Contact Nord site last year) was among the top 5 most viewed books of the year, so that was what the talk was about. Among the most central messages of the book and the ones that I was trying to get across in this presentation were:

that how we do teaching matters more than what we do (“T’ain’t what you do, it’s the way that you do it”) and
that we can only understand the process if we examine the whole complex assembly of teaching (very much including the technique of all who contribute to it, including learners, textbooks, and room designers) not just the individual parts.

Along the way I had a few other things to say about why that must be the case, the nature of teaching, the nature of collective cognition, and some of the profound consequences of seeing the world this way. I had fun persuading ChatGPT to illustrate the slides in a style that was not that of Richard Scarry (ChatGPT would not do that, for copyright reasons) but that was reminiscent of it, so there are lots of cute animals doing stuff with technologies on the slides.

I rushed and rambled, I sang, I fumbled and stumbled, but I think it sparked some interest and critical thinking. Even if it didn’t, some learning happened, and that is always a good thing. The conversations in the chat went too fast for me to follow but I think there were some good ones. If nothing else, though I was very nervous, I had fun, and it was lovely to notice a fair number of friends, colleagues, and even the odd relative among the audience. Thank you all who were there, and thank you anyone who catches the recording later.

And now in Chinese: 在线学习环境：隐喻问题与系统改进. And some thoughts on the value of printed texts.

Posted April 13, 2024 by Jon Dron

Warm off the press, and with copious thanks and admiration to Junhong Xiao for the invitation to submit and the translation, here is my paper “The problematic metaphor of the environment in online learning” in Chinese, in the Journal of Open Learning. It is based on my OTESSA Journal paper, originally published as “On the Misappropriation of Spatial Metaphors in Online Learning” at the end of 2022 (a paper I am quite pleased with, though it has yet to receive any citations that I am aware of).

Many thanks, too, to Junhong for sending me the printed version that arrived today, smelling deliciously of ink. I hardly ever read anything longer than a shopping bill on paper any more but there is something rather special about paper that digital versions entirely lack. The particular beauty of a book or journal written in a language and script that I don’t even slightly understand is that, notwithstanding the ease with which I can translate it using my phone, it largely divorces the medium from the message. Even with translation tools my name is unrecognizable to me in this: Google Lens translates it as “Jon Delong”. Although I know it contains a translation of my own words, it is really just a thing: the signs it contains mean nothing to me, in and of themselves. And it is a thing that I like, much as I like the books on my bookshelf.

I am not alone in loving paper books, a fact that owners of physical copies of my most recent book (which can be read online for free but that costs about $CAD40 on paper) have had the kindness to mention, e.g. here and here. There is something generational in this, perhaps. For those of us who grew up knowing no other reading medium than ink on paper, there is comfort in the familiar, and we have thousands (perhaps millions) of deeply associated memories in our muscles and brains connected with it, made more precious by the increasing rarity with which those memories are reinforced by actually reading them that way. But, for the most part, I doubt that my grandchildren, at least, will lack that. While they do enjoy and enthusiastically interact with text on screens, from before they have been able to accurately grasp them they have been exposed to printed books, and have loved some of them as much as I did at the same ages.

It is tempting to think that our love of paper might simply be because we don’t have decent e-readers, but I think there is more to it than that. I have some great e-readers in many sizes and types, and I do prefer some of them to read from, for sure: backlighting when I need it, robustness, flexibility, the means to see it in any size or font that works for me, the simple and precise search, the shareable highlights, the lightness of (some) devices, the different ways I can hold them, and so on, make them far more accessible. But paper has its charms, too. Most obviously, something printed on a paper is a thing to own whereas, on the whole, a digital copy tends to just be a licence to read, and ownership matters. I won’t be leaving my e-books to my children. The thingness really matters in other ways, too. Paper is something to handle, something to smell. Pages and book covers have textures – I can recognize some books I know well by touch alone. It affects many senses, and is more salient as a result. It takes up room in an environment so it’s a commitment, and so it has to matter, simply because it is there; a rivalrous object competing with other rivalrous objects for limited space. Paper comes in fixed sizes that may wear down but will never change: it thus keeps its shape in our memories, too. My wife has framed occasional pages from my previously translated work, elevating them to art works, decoupled from their original context, displayed with the same lofty reverence as pages from old atlases. Interestingly, she won’t do that if it is just a printed PDF: it has to come from a published paper journal, so the provenance matters. Paper has a history and a context of its own, beyond what it contains. And paper creates its own context, filled with physical signals and landmarks that make words relative to the medium, not abstractions that can be reflowed, translated into other languages, or converted into other media (notably speech). The result is something that is far more memorable than a reflowable e-text. Over the years I’ve written a little about this here and there, and elsewhere, including a paper on the subject (ironically, a paper that is not available on paper, as it happens), describing an approach to making e-texts more memorable.

After reaching a slightly ridiculous peak in the mid-2000s, and largely as a result of a brutal culling that occurred when I came to Canada nearly 17 years ago, my paper book collection has now diminished to easily fit in a single and not particularly large free-standing IKEA shelving unit. The survivors are mostly ones I might want to refer to or read again, and losing some of them would sadden me a great deal, but I would only (perhaps) run into a burning building to save just a few, including, for instance:

A dictionary from 1936, bound in leather by my father and used in countless games of Scrabble and spelling disputes when I was a boy, and that was used by my whole family to look up words at one time or another.
My original hardback copy of the Phantom Tollbooth (I have a paperback copy for lending), that remains my favourite book of all time, that was first read to me by my father, and that I have read myself many times at many ages, including to my own children.
A boxed set of the complete works of Narnia, that I chose as my school art prize when I was 18 because the family copies had become threadbare (read and abused by me and my four siblings), and that I later read to my own children. How someone with very limited artistic skill came to win the school art prize is a story for another time.
A well-worn original hardback copy of Harold and the Purple Crayon (I have a paperback copy for lending) that my father once displayed for children in his school to read, with the admonition “This is Mr Dron’s book. Please handle with care” (it was not – it was mine).
A scribble-filled, bookmark-laden copy of Kevin Kelly’s Out of Control that strongly influenced my thinking when I was researching my PhD and that still inspires me today. I can remember exactly where I sat when I made some of the margin notes.
A disintegrating copy of Storyland, given to me by my godmother in 1963 and read to me and by me for many years thereafter. There is a double value to this one because we once had two copies of this in our home: the other belonged to my wife, and was also a huge influence on her at similar ages.

These books proudly wear their history and their relationships with me and my loved ones in all their creases, coffee stains, scuffs, and tattered pages. To a greater or lesser extent, the same is true of almost all of the other physical books I have kept. They sit there as a constant reminder of their presence – their physical presence, their emotional presence, their social presence and their cognitive presence – flitting by in my peripheral vision many times a day, connecting me to thoughts and inspirations I had when I read them and, often, with people and places connected with them. None of this is true of my e-books. Nor is it quite the same for other objects of sentimental value, except perhaps (and for very similar reasons) the occasional sculpture or picture, or some musical instruments. Much as I am fond of (say) baby clothes worn by my kids or a battered teddy bear, they are little more than aides memoires for other times and other activities, whereas the books (and a few other objects) latently embody the experiences themselves. If I opened them again (and I sometimes do) it would not be the same experience, but it would enrich and connect with those that I already had.

I have hundreds of e-books that are available on many devices, one of which I carry with me at all times, not to mention an Everand (formerly Scribd) account with a long history, not to mention a long and mostly lost history of library borrowing, and I have at least a dozen devices on which to read them, from a 4 inch e-ink reader to a 32 inch monitor and much in between, but my connection with those is far more limited and transient. It is still more limited for books that are locked to a certain duration through DRM (which is one reason they are the scum of the earth). When I look at my devices and open the various reading apps on them I do see a handful of book covers, usually those that I have most recently read, but that is too fleeting and volatile to have much value. And when I open them they don’t fall open on well-thumbed pages. The text is not tangibly connected with the object at all.

As well as smarter landmarks within them, better ways to make e-books more visible would help, which brings me to the real point of this post. For many years I have wanted to paper a wall or two with e-paper (preferably in colour) on which to display e-book covers, but the costs are still prohibitive. It would be fun if the covers would become battered with increasing use, showing the ones that really mattered, and maybe dust could settle on those that were never opened, though it would not have to be so skeuomorphic – fading would work, or glyphs. They could be ordered manually or by (say) reading date, title, author, or subject. Perhaps touching them or scanning a QR code could open them. I would love to get a research grant to do this but I don’t think asking for electronic wallpaper in my office would fly with most funding sources, even if I prettied it up with words like “autoethnography”, and I don’t have a strong enough case, nor can I think of a rigorous enough research methodology to try it in a larger study with other people. Well. Maybe I will try some time. Until the costs of e-paper come down much further, it is not going to be a commercially viable product, either, though prices are now low enough that it might be possible to do it in a limited way with a poster-sized display for a (very) few thousand dollars. It could certainly be done with a large screen TV for well under $1000 but I don’t think a power-hungry glowing screen would be at all the way to go: the value would not be enough to warrant the environmental harm or energy costs, and something that emitted light would be too distracting. I do have a big monitor on my desk, though, which is already doing that so it wouldn’t be any worse, to which I could add a background showing e-book covers or spines. I could easily do this as a static image or slideshow, but I’d rather have something dynamic. It shouldn’t be too hard to extract the metadata from my list of books, swipe the images from the Web or the e-book files, and show them as a backdrop (a screensaver would be trivial). It might even be worth extending this to papers and articles I have read. I already have Pocket open most of the time, displaying web pages that I have recently read or want to read (serving a similar purpose for short-term recollection), and that could be incorporated in this. I think it would be useful, and it would not be too much work to do it – most of the important development could be done in a day or two. If anyone has done this already or feels like coding it, do get in touch!

Preprint – The human nature of generative AIs and the technological nature of humanity: implications for education

Posted October 9, 2023, updated onOctober 14, 2023 by Jon Dron

Here is a preprint of a paper I just submitted to MDPI’s Digital journal that applies the co-participation model that underpins How Education Works (and a number of my papers over the last few years) to generative AIs (GAIs). I don’t know whether it will be accepted and, even if it is, it is very likely that some changes will be required. This is a warts-and-all raw first submission. It’s fairly long (around 10,000 words).

The central observation around which the paper revolves is that, for the first time in the history of technology, recent generations of GAIs automate (or at least appear to automate) the soft technique that has, till now, been the sole domain of humans. Up until now, every technology we have ever created, be it physically instantiated, cognitive, organizational, structural, or conceptual, has left all of the soft part of the orchestration to human beings.

The fact that GAIs replicate the soft stuff is a matter for some concern when they start to play a role in education, mainly because:

the skills they replace may atrophy or never be learned in the first place. This is not even slightly like replacing hard skills of handwriting or arithmetic: we are talking about skills like creativity, problem-solving, critical inquiry, design, and so on. We’re talking about the stuff that GAIs are trained with.
the AIs themselves are an amalgam, an embodiment of our collective intelligence, not actual people. You can spin up any kind of persona you like and discard it just as easily. Much of the crucially important hidden/tacit curriculum of education is concerned with relationships, identity, ways of thinking, ways of being, ways of working and playing with others. It’s about learning to be human in a human society. It is therefore quite problematic to delegate how we learn to be human to a machine with (literally and figuratively) no skin in the game, trained on a bunch of signals signifying nothing but more signals.

On the other hand, to not use them in educational systems would be as stupid as to not use writing. These technologies are now parts of our extended cognition, intertwingled with our collective intelligence as much as any other technology, so of course they must be integrated in our educational systems. The big questions are not about whether we should embrace them but how, and what soft skills they might replace that we wish to preserve or develop. I hope that we will value real humans and their inventions more, rather than less, though I fear that, as long as we retain the main structural features of our education systems without significant adjustments to how they work, we will no longer care, and we may lose some of our capacity for caring.

I suggest a few ways we might avert some of the greatest risks by, for instance, treating them as partners/contractors/team members rather than tools, by avoiding methods of “personalization” that simply reinforce existing power imbalances and pedagogies designed for better indoctrination, by using them to help connect us and support human relationships, by doing what we can to reduce extrinsic drivers, by decoupling learning and credentials, and by doubling down on the social aspects of learning. There is also an undeniable explosion in adjacent possibles, leading to new skills to learn, new ways to be creative, and new possibilities for opening up education to more people. The potential paths we might take from now on are unprestatable and multifarious but, once we start down them, resulting path dependencies may lead us into great calamity at least as easily as they may expand our potential. We need to make wise decisions now, while we still have the wisdom to make them.

MDPI invited me to submit this article free of their normal article processing charge (APC). The fact that I accepted is therefore very much not an endorsement of APCs, though I respect MDPI’s willingness to accommodate those who find payment difficult, the good editorial services they provide, and the fact that all they publish is open. I was not previously familiar with the Digital journal itself. It has been publishing 4 articles a year since 2021, mostly offering a mix of reports on application designs and literature reviews. The quality seems good.

Abstract

This paper applies a theoretical model to analyze the ways that widespread use of generative AIs (GAIs) in education and, more broadly, in contributing to and reflecting the collective intelligence of our species, can and will change us. The model extends Brian Arthur’s insights into the nature of technologies as the orchestration of phenomena to our use by explaining the nature of humans participation in their enactment, whether as part of the orchestration (hard technique, where our roles must be performed correctly) or as orchestrators of phenomena (soft technique performed creatively or idiosyncratically). Education may be seen as a technological process for developing the soft and hard techniques of humans to participate in the technologies and thus the collective intelligence of our cultures. Unlike all earlier technologies, by embodying that collective intelligence themselves, GAIs can closely emulate and implement not only the hard technique but also the soft that, until now, was humanity’s sole domain: the very things that technologies enabled us to do can now be done by the technologies themselves. The consequences for what, how, and even whether we learn are profound. The paper explores some of these consequences and concludes with theoretically informed approaches that may help us to avert some dangers while benefiting from the strengths of generative AIs.

Originally posted at: https://landing.athabascau.ca/bookmarks/view/20512771/preprint-the-human-nature-of-generative-ais-and-the-technological-nature-of-humanity-implications-for-education

Cognitive prostheses and the future of the human race

Posted August 18, 2023, updated onOctober 14, 2023 by Jon Dron

I have lived in Canada for over 16 years so I was surprised when, a few months ago, a conference-friend whom I had not seen for many years contacted me to ask whereabouts in Malaysia I lived. I believe that they were misremembering a connection from a conversation long ago in which I’d mentioned a recent visit (it was in fact a day-trip from Singapore) and combining that with the accurate recollection that I no longer lived in the UK.

Not long after, I was on a panel discussing the impact of ChatGPT during which I prompted ChatGPT to introduce me. Here was its first attempt in response to the prompt “tell me about Jon Dron”. I’ve highlighted things that are true in green, and things that are false in red (emphasized for those unable to see the colours):

Jon Dron is a Canadian educator, researcher, and writer. He is known for his work in the field of e-learning and collaborative technologies. Dron has a PhD in Educational Technology from the University of Calgary, and has held academic positions at several universities in Canada, including Athabasca University, where he is currently a professor.

Dron has written extensively on topics such as online learning, social media, and the use of technology in education. He is the author of several books, including “Teaching Crowds: Learning and Social Media” and “E-Learning Theory and Practice.”

In addition to his academic work, Dron is also a musician and has released several albums of ambient and experimental music.

I’d say that there is not much difference between the human and machine recollections. I would almost certainly make at least as many mistakes if I were to asked to confidently describe a person I don’t know particularly well. In fact, I might make similar mistakes (not, please note, hallucinations) about quite close friends. Most of us don’t have eidetic memories: we reinvent recollections as much as we recall them. While there are surely many profound differences between how humans and large language models (LLMs) like ChatGPT process information, this is at least circumstantial evidence that some of the basic principles underlying artificial neural networks and biological neural networks are probably pretty similar. True, AIs do not know when they are making things up (or telling the truth, for that matter) but, in fairness, much of the time, neither do we. With a lot of intentional training we may be able to remember lines in a play or how to do long division but, usually, our recollections are like blurry JPEGs rather than RAW images.

Even for things we have intentionally learned to do or recall well, it is unusual for that training to stick without continual reinforcement, and mistakes are easily made. A few days ago I performed a set of around 30 songs (neither ambient nor experimental), most of which I had known for decades, all of which I had carefully practiced in the days leading up to the event to be sure I could play them as I intended. Here is a picture of me singing at that gig, drawn by my 6-year-old grandchild who was in attendance:

Despite my precautions and ample experience, in perhaps a majority of songs, I variously forgot words, chords, notes, and, in a couple of cases, whole verses. Combined with errors of execution (my fingers are not robotic, my voice gets husky) there was, I think, only one song in the whole set that came out more or less exactly as I intended. I have made such mistakes in almost every gig I have ever played. In fact, in well over 40 years as a performer, I have never played the same song in exactly the same way twice, though I have played some of them well over 10,000 times. Most of the variations are a feature, not a bug: they are where the expression lies. A performance is a conversation between performer, instruments, setting, and audience, not a mechanical copy of a perfect original. Nonetheless, my goal is usually to at least play the right notes and sing the right words, and I frequently fail to do that. Significantly, I generally know when I have done it wrong (typically a little before in a dread realization that just makes things worse) and adapt fairly seamlessly on the fly so, on the whole, you probably wouldn’t even notice it has happened, but I play much like ChatGPT responds to prompts: I fill in the things I don’t know with something more or less plausible. These creative adaptations are no more hallucinations than the false outputs of LLMs.

The fact that perfect recall is so difficult to achieve is why we need physical prostheses, to write things down, to look things up, or to automate them. Given LLMs’ weaknesses in accurate recall, it is slightly ironic that we often rely on computers for that. It is, though, considerably more difficult for an LLM to do this because they have no big pictures, no purposes, no plans, not even broad intentions. They don’t know whether what they are churning out is right or wrong, so they don’t know to correct it. In fact, they don’t even know what they are saying, period. There’s no reflection, no metacognition, no layers of introspection, no sense of self, nothing to connect concepts together, no reason for them to correct errors that they cannot perceive.

Things that make us smart

How difficult can it be to fix this? I think we will soon be seeing a lot more solutions to this problem because if we can look stuff up then so can machines, and more reliable information from other systems can be used to feed the input or improve the output of the LLM (Bing, for instance, has been doing so for a while now, to an extent). A much more intriguing possibility is that an LLM itself or subsystem of it might not only look things up but also write and/or sequester code it needs to do things it is currently incapable of doing, extending its own capacity by assembling and remixing higher-level cognitive structures. Add a bit of layering then throw in an evolutionary algorithm to kill of the less viable or effective, and you’ve got a machine that can almost intentionally learn, and know when it has made a mistake.

Such abilities are a critical part of what makes humans smart, too. When discussing neural networks it is a bit too easy to focus on the underlying neural correlates of learning without paying much (if any) heed to the complex emergent structures that result from them – the “stuff” of thought – but those structures are the main things that make it work for humans. Like the training sets for large language models, the intelligence of humans is largely built from the knowledge gained from other humans through language, pedagogies, writing, drawing, music, computers, and other mediating technologies. Like an LLM, the cognitive technologies that result from this (including songs) are parts that we assemble and remix to in order to analyze, synthesize, and create. Unlike most if not all existing LLMs, though, the ways we assemble them – the methods of analysis, the rules of logic, the pedagogies, the algorithms, the principles, and so on (that we have also learned from others) – are cognitive prostheses that play an active role in the assembly, allowing us to build, invent, and use further cognitive prostheses and so to recursively extend our capabilities far beyond the training set, as well as to diagnose our own shortfalls.

Like an LLM, our intelligence is also fundamentally collective, not just in what happens inside brains, but because our minds are extended, through tools, gadgets, rules, language, writing, structures, and systems that we enlist from the world as part of (not only adjuncts to) our thinking processes. Through technologies, from language to screwdrivers, we literally share our minds with others. For those of us who use them, LLMs are now as much parts of us as our own creative outputs are parts of them.

All of this means that human minds are part-technology (largely but not wholly instantiated in biological neural nets) and so our cognition is about as artificial as that of AIs. We could barely even think without cognitive prostheses like language, symbols, logic, and all the countless ways of doing and using technologies that we have devised, from guitars to cars. Education, in part, is a process of building and enlisting those cognitive prostheses in learners’ minds, and of enabling learners to build and enlist their own, in a massively complex, recursive, iterative, and distributed process, rich in feedback loops and self-organizing subsystems.

Choosing what we give up to the machine

There are many good ways to use LLMs in the learning process, as part of what students do. Just as it would be absurd to deny students the use of pens, books, computers, the Internet, and so on, it is absurd to deny them the use of AIs, including in summative assessments. These are now part of our cognitive apparatus, so we should learn how to participate in them wisely. But I think we need to be extremely cautious in choosing what we delegate to them, above all when using them to replace or augment some or all of the teaching role.

What makes AIs different from technologies of the past is that they perform a broadly similar process of cognitive assembly as we do ourselves, allowing us to offload much more of our cognition to an embodied collective intelligence created from the combined output of countless millions of people. Only months after the launch of ChatGPT, this is already profoundly changing how we learn and how we teach. It is disturbing and disruptive in an educational context for a number of reasons, such as that:

it may make it unnecessary for us to learn its skills ourselves, and so important aspects of our own cognition, not just things we don’t need (but which are they?), may atrophy;
if it teaches, it may embed biases from its training set and design (whose?) that we will inherit;
it may be a bland amalgam of what others have written, lacking originality or human quirks, and that is what we, too, will learn to do;
if we use it to teach, it may lead students towards an average or norm, not a peak;
it renders traditional forms of credentialling learning largely useless.

We need solutions to these problems or, at least, to understand how we will successfully adapt to the changes they bring, or whether we even want to do so. Right now, an LLM is not a mind at all, but it can be a functioning part of one, much as an artificial limb is a functioning part of a body or a cyborg prosthesis extends what a body can do. Whether we feel any particular limb that it (partly) replicates needs replacing, which system we should replace it with, and whether it is a a good idea in the first place are among the biggest questions we have to answer. But I think there’s an even bigger problem we need to solve: the nature of education itself.

AI teachers

There are no value-free technologies, at least insofar as they are enacted and brought into being through our participation in them, and the technologies that contribute to our cognition, such as teaching, are the most value-laden of all, communicating not just the knowledge and skills they purport to provide but also the ways of thinking and being that they embody. It is not just what they teach or how effectively they do so, but how they teach, and how we learn to think and behave as a result, that matters.

While AI teachers might well make it easier to learn to do and remember stuff, building hard cognitive technologies (technique, if you prefer) is not the only thing that education does. Through education, we learn values, ways of connecting, ways of thinking, and ways of being with others in the world. In the past this has come for free when we learn the other stuff, because real human teachers (including textbook authors, other students, etc) can’t help but model and transmit the tacit knowledge, values, and attitudes that go along with what they teach. This is largely why in-person lectures work. They are hopeless for learning the stuff being taught but the fact that students physically attend them makes them great for sharing attitudes, enthusiasm, bringing people together, letting us see how other people think through problems, how they react to ideas, etc. It is also why recordings of online lectures are much less successful because they don’t, albeit that the benefits of being able to repeat and rewind somewhat compensate for the losses.

What happens, though, when we all learn how to be human from something that is not (quite) human? The tacit curriculum – the stuff through which we learn ways of being, not just ways of doing – for me looms largest among the problems we have to solve if we are to embed AIs in our educational systems, as indeed we must. Do we want our children to learn to be human from machines that haven’t quite figured out what that means and almost certainly never will?

Many AI-Ed acolytes tell the comforting story that we are just offloading some of our teaching to the machine, making teaching more personal, more responsive, cheaper, and more accessible to more people, freeing human teachers to do more of the human stuff. I get that: there is much to be said for making the acquisition of hard skills and knowledge easier, cheaper, and more efficient. However, it is local thinking writ large. It solves the problems that we have to solve today that are caused by how we have chosen to teach, with all the centuries-long path dependencies and counter technologies that entails, replacing technologies without wondering why they exist in the first place.

Perhaps the biggest of the problems that the entangled technologies of education systems cause are the devastating effects of tightly coupled credentials (and their cousins, grades) on intrinsic motivation. Much of the process of good teaching is one of reigniting that intrinsic motivation or, at least, supporting the development of internally regulated extrinsic motivation, and much of the process of bad teaching is about going with the flow and using threats and rewards to drive the process. As long as credentials remain the primary reason for learning, and as long as they remain based on proof of easily measured learning outcomes provided through end-products like assignments and inauthentic tests, then an AI that offers a faster, more efficient, and better tailored way of achieving them will crowd out the rest. Human teaching will be treated as a minor and largely irrelevant interruption or, at best, a feel-good ritual with motivational perks for those who can afford it. And, as we are already seeing, students coerced to meet deadlines and goals imposed on them will use AIs to take shortcuts. Why do it yourself when a machine can do it for you?

The future

As we start to build AIs more like us, with metacognitive traits, self-set purposes, and the capacity for independent learning, the problem is just going to get bigger. Whether they are better or worse (they will be both), AIs will not be the same as us, yet they will increasingly seem so, and increasingly play human roles in the system. If the purpose of education is seen as nothing but short-term achievement of explicit learning outcomes and getting the credentials arising from that, then it would be better to let the machines achieve them so that we can get on with our lives. But of course that is not the purpose. Education is for preparing people to live better lives in better societies. It is why the picture of me singing above delights me more than anything ever created by an AI. It is why education is and must remain a fundamentally human process. Almost any human activity can be replaced by an AI, including teaching, but education is fundamental to how we become who we are. That’s not the kind of thing that I think we want to replace.

Our minds are already changing as they extend into the collective intelligence of LLMs – they must – and we are only at the very beginning of this story. Most of the changes that are about to occur will be mundane, complex, and the process will be punctuated but gradual, so we won’t really notice what has been happening until it has happened, by which time it may be too late. It is probably not an exaggeration to say that, unless environmental or other disasters don’t bring it all to a halt, this is a pivotal moment in our history.

It is much easier to think locally, to think about what AIs can do to support or extend what we do now, than it is to imagine how everything will change as a result of everyone doing that at scale. It requires us to think in systems, which is not something most of us are educated or prepared to do. But we must do that, now, while we still can. We should not leave it to AIs to do it for us.

There’s much more on many of the underpinning ideas mentioned in this post, including references and arguments supporting them, in my freely downloadable or cheap-to-purchase latest book (of three, as it happens), How Education Works.

A decade of unwriting: the life history of "How Education Works"

Posted June 24, 2023, updated onOctober 14, 2023 by Jon Dron

About 10 years ago I submitted the first draft of a book called “How Learning Technologies Work” to AU Press. The title was a nod to David Byrne’s wonderful book, “How Music Works” which is about much more than just music, just as mine was about much more than learning technologies.

Pulling together ideas I had been thinking about for a few years, the book had taken me only a few months to write, mostly at the tail end of my sabbatical. I was quite pleased with it. The internal reviewers were positive too, though they suggested a number of sensible revisions, including clarifying some confusing arguments and a bit of restructuring. Also, in the interests of marketing, they recommended a change to the title because, though accurately describing the book’s contents, I was not using “learning technologies” in its mainstream sense at all (for me, poetry, pedagogies, and prayer are as much technologies as pots, potentiometers and practices), so it would appeal to only a small subset of its intended audience. They were also a bit concerned that it would be hard to find an audience for it even if it had a better title because it was at least as much a book about the nature of technology as it was a book about learning, so it would fall between two possible markets, potentially appealing to neither.

A few months later, I had written a new revision that addressed most of the reviewers’ recommendations and concerns, though it still lacked a good title. I could have submitted it then. However, in the process of disentangling those confusing arguments, I had realized that the soft/hard technology distinction on which much of the book rested was far less well-defined than I had imagined, and that some of the conclusions that I had drawn from it were just plain wrong. The more I thought about it, the less happy I felt.

And so began the first of a series of substantial rewrites. However, my teaching load was very high, and I had lots of other stuff to do, so progress was slow. I was still rewriting it when I unwisely became Chair of my department in 2016, which almost brought the whole project to a halt for another 3 years. Despite that, by the time my tenure as Chair ended, the book had grown to around double its original (not insubstantial) length, and the theory was starting to look coherent, though I had yet to make the final leap that made sense of it all.

By 2019, as I started another sabbatical, I had decided to split the book into two. I put the stuff that seemed useful for practitioners into a new book, “Education: an owner’s manual”, leaving the explanatory and predictive theory in its own book, now grandiosely titled “How Education Works”, and worked on both simultaneously. Each grew to a few hundred pages.

Neither worked particularly well. It was really difficult to keep the theory out of the practical book, and the theoretical work was horribly dry without the stories and examples to make sense of it. The theory, though, at last made sense, albeit that I struggled (and failed) to give it a catchy name. The solution was infuriatingly obvious. In all my talks on the subject my catchphrase from the start had been “’tain’t what you do, it’s the way that you do it, that’s what gets results” (it’s the epigraph for the book), so it was always implicit that softness and hardness are not characteristics of all technologies, as such, nor even of their assemblies, but of the ways that we participate in their orchestration. Essentially, what matters is technique: the roles we play as parts of the orchestration or orchestrators of it. That’s where the magic happens.

But now I had two mediocre books that were going nowhere. Fearing I was about to wind up with two unfinished and/or unsellable books, about half way through my sabbatical I brutally slashed over half the chapters from both, pasted the remains together, and spent much of the time I had left filling in the cracks in the resulting bricolage.

I finally submitted “How Education Works: Teaching, Technology, and Technique” in the closing hours of 2020, accompanied by a new proposal because, though it shared a theme and a few words with the original, it was a very different book.

Along the way I had written over a million words, only around a tenth of which made it into what I sent to AU Press. I had spent the vast majority of my authoring time unwriting rather than writing the book and, with each word I wrote or unwrote, the book had written me, as much as I had written it. The book is as much a part of my cognition as a product of it.

And now, at last, it can be part of yours.

30 months after it was submitted – I won’t go into the reasons apart from to say it has been very frustrating – the book is finally available as a free PDF download or to read on the Web. If all goes to plan, the paper and e-book versions should arrive June 27th, 2023, and can be pre-ordered now.

It is still a book about technology at least as much as it is about education (very broadly defined), albeit that it is now firmly situated in the latter. It has to be both because among the central points I’m making are that we are part-technology and technology is part-us, that cognition is (in part) technology and technology is (in part) cognition, and that education is a fundamentally technological and thus fundamentally human activity. It’s all one complex, hugely distributed, recursive intertwingularity in which we and our technological creations are all co-participants in the cognition and learning of ourselves and one another.

During the 30 months AU Press has had the book I have noticed a thousand different ways the book could be improved, and I don’t love all of the edits made to it along the way (by me and others), but I reckon it does what I want it to do, and 10 years is long enough.

It’s time to start another.

A few places you can buy the book

AU Press (CA)

Barnes & Noble (US)

Blackwells (UK)

Amazon (CA)

Amazon (JP)

University of Chicago Press (US)

Indigo (CA)

Booktopia (AU)

Mediaeval Teaching in the Digital Age (slides from my keynote at Oxford Brookes University, May 26, 2021)

Posted May 26, 2021, updated onMay 26, 2021 by Jon Dron

These are the slides from my keynote today at the Oxford Brookes “Theorizing the Virtual” School of Education Research Conference. As theorizing the virtual is pretty much my thing, I was keen to be a part of this! It was an ungodly hour of the day for me (2am kickoff) but it was worth staying up for. It was a great bunch of attendees who really got into the spirit of the thing and kept me wide awake. I wish I could hang around for the rest of it but, on the bright side, at least I’m up at the right time to see the Super Flower Blood Moon (though it’s looking cloudy, darn it). In this talk I dwelt on a few of the notable differences between online and in-person teaching. This is the abstract…

Pedagogical methods (ways of teaching) are solutions to problems of helping people to learn, in a context filled with economic, physical, temporal, legal, moral, social, political, technological, and organizational constraints. In mediaeval times books were rare and unaffordable, and experts’ time was precious and limited, so lectures were a pragmatic solution, but they in turn created more problems. Counter-technologies such as classes, classrooms, behavioural rules and norms, courses, terms, curricula, timetables and assignment deadlines were were devised to solve those problems, then methods of teaching (pedagogies) were in turn invented to solve problems these counter-technologies caused, notably including:
· people who might not want (or be able) to be there at that time,
· people who were bored and
· people who were confused.
Better pedagogies supported learner needs for autonomy and competence, or helped learners find relevance to their own goals, values, and interests. They exploited physical closeness for support, role-modelling, inspiration, belongingness and so on. However, increasingly many relied on extrinsic motivators, like classroom discipline, grades and credentials to coerce students to learn. Extrinsic motivation achieves compliance, but it makes the reward or avoidance of the punishment the goal, persistently and often permanently crowding out intrinsic motivation. Intelligent students respond with instrumental approaches, satisficing, or cheating. Learning seldom persists; love of the subject is subdued; learners learn to learn in ineffective ways. More layers of counter-technologies are needed to limit the damage, and so it goes on.
Online, the constraints are very different, and its native forms are the motivational inverse of in-person learning. An online teacher cannot control every moment of a learner’s time, and learners can use the freedoms they gain to take the time they need, when they need it, to learn and to reflect, without the constraints of scheduled classroom hours and deadlines. However, more effort is usually needed to support their needs for relatedness. Unfortunately, many online teachers try (or are required) to re-establish the control they had in the classroom through grading or the promise of credentials, recreating the mediaeval problems that would otherwise not exist, using tools like learning management systems that were designed (poorly) to replicate in-person teaching functions. These are solutions to the problems caused by counter-technologies, not to problems of learning.
There are better ways, and that’s what this session is about.