The description of the following ideas as "non universals" originates from Alan Kay. The purpose of this page is to expand and elaborate further on the meanings and educational implications of the list of non universals

What Alan Kay said about his Universals / Non Universals slide at the EuroPython 2006 keynote (transcribed by me from source)

  • social
  • language
  • communication
  • culture
  • fantasies
  • stories
  • tools and art
  • superstition
  • religion and magic
  • case based learning
  • theatre
  • play and games
  • differences over similarities
  • quick reactions to patterns
  • loud noises and snakes
  • supernormal responses
  • vendetta, and more (about 300 of these have been identified across cultures)

"In effect anthropologists have been studying humans for about a Century now and firstly 3000 human cultures seem to be very very different. Then they start realising that they seemed surprisingly parametric. Every culture had a language, every culture told stories ... (goes through some of the items on the Universals list)

If you look at these you can see our modern internet culture - it's basically social, it enables us to communicate in various ways and so forth, basically a story based culture"


"What's interesting is to look for things that are not universal, that seems to have some importance as well. Most people have lived and died on this Earth for 100,000 years without reading and writing, without having deductive maths and model based science .... (goes through non universals list)

These are a little harder to learn than the ones on the left because we are not directly wired to learn them. These things are actually inventions which are difficult to invent. And the rise of Schools going all the way back to the Sumerian and Egyptian times came about to start helping children learn some of these things that aren't easy to learn. It can be argued that if you are trying to be utopian about education what we should be doing is helping the children of the world learn these hard to learn things. Equal rights is a really good one to help children learn. No culture in the world is particularly good at it."


(listed here, go here for actual discussion)
  • Aren't the non universals already in the curriculum?
  • What is blocking a more widespread uptake of the importance of teaching the non universals?
  • If School is failing then how and why is it failing?
  • Should there be other items on the non universal and universal lists? Are the lists complete?
  • How should the non universals be taught?
  • How do the non universals interact with the universals (the relationship b/w them)?
  • What are the implications for the curriculum wars (back to basics versus learning through discovery)?
  • Are there some exemplars of teachers who do teach the non universals successfully?
  • Are we asking the right questions?


Thoughts About Teaching Science and Mathematics to Young Children by alan kay

Our Human Condition "From Space" by alan kay

I like his idea of the unsane, the mental state where our ideas don't fit reality, the map doesn't represent the territory. We like to think of ourselves as mostly "sane" and contrast that with a few "insane" personal moments or the more permanent state of a few unfortunates. But the "unsane" idea makes room for a different self perception. What if more often than not we are unsane?

When Alan Kay said in his writings from time to time that computers were not all that important, I couldn't really believe that he meant that. Computers have been his career, so I couldn't take him seriously. Even now I feel a strong pressure to qualify this observation.

But in this article he explains it clearly. Most of modern science (400 year tradition) can be done with simple tools - to grasp it requires point of view, effort, time but not money or computers

One of the first quotes I heard from him years ago was "point of view is worth 80 IQ points". Initially, that sounded quaint and elitist, "how old hat to be talking about IQ", I thought. But having read this article and some other of his writings I now see it as an important insight. Point of view changes everything. So what happens? From the outside you look the same but from the inside everything looks different, you see the world through different eyes. My feeling is that the important thing is to pursue the non universal powerful ideas more rigorously.
(16 Aug 08 - Bill Kerr)


alan kay:
We should look a bit at three different kinds of understanding: rote understanding, operational understanding, and meta-understanding. If we leave out the majority of teachers who don't really understand math in any strong way, we still find that the kinds of understandings that are left are not up to the task of being able to see the meaning and value of a new perspective on mathematics. For example, it is possible to understand calculus a little in the narrow form in which it was learned, and still not be able to see "calculus" in a different form (even if the new way is a stronger way to look at it). Real fluency in a subject allows many of the most powerful ideas
in the subject to be somewhat detached from specific forms. This is meta-understanding.

For example, the school version of calculus is based on a numeric continuum and algebraic manipulations. But the idea of calculus is not really strongly tied to this.

The idea has to do with separating out the similarities and differences of change to produce and allow much simpler and easier to understand relationships to be created. This can be done so that the connection between one state and the next one of interest is a simple addition. Actual continuity can be replaced by a notion of "you pick and then I pick" so that non-continuities don't get seen. This other view of calculus as a form of calculation was used by Babbage in his
first "difference engines" because a computing machine that can do lots of additions for you can make this other way to look at calculus very practical and worthwhile. The side benefit is that it is much easier to understand than the algebraic rubrics. If we then add to this the idea of using vectors (as "supernumbers") instead of regular numbers, we are able to dispense with coordinate systems except when convenient, and are able to operate in multiple dimensions.

All of this was worked out in the 19th century and quite a bit was adopted enthusiastically by science and is in main use today.

To cut to the chase, Seymour Papert (who was a very good mathematician) was one of the first to realize that this kind of math (called "vector differential geometry") fit very well into young
children's thinking patterns, and that the new personal computers would be able to manifest Babbage's dream to be able to compute and think in terms of an incremental calculus for complex change.

Any one fluent in mathematics can recognize this (but it took a Papert to first point it out). But, virtually no one without fluency in mathematics can recognize this. And surveys have shown that less than 5% of Americans are fluent in math or science. Many of the 95% were able to go through 16 years of schooling and successfully get a college degree without attaining any fluency in math or science.

- extract from a longer discussion at


alan kay:
"Theory of Harmony" is kind of like "Deductive Abstract Mathematics" in that most traditional cultures have some form of counting, adding
and subtracting -- and some make music with multiple pitches at once (as did Western Culture before 1600). But the notion of harmony
before 1600 was essentially as a byproduct of melodies and voice leading rather than a thing in itself in which chords have the same
first class status as melodic lines. How and why this appeared is fascinating and is well known in music history.

Some of the most interesting composers in the Baroque period (especially Bach) tried to make both the old and the new schemes work
completely together. Bach's harmonic language in particular was an amazing blend of harmonies and bass lines with voice leading and
other contrapuntal techniques (quite a bit of his vocabulary is revealed in his harmonized chorales (some 371 or 372 of them)). That
these two worlds are very different ways of looking at things is attested to by a wonderful piece by Purcell "The Contest Between
Melodie and Harmonie".

As with "Greek Math", history doesn't seem to have any record of a separate and as rich invention of a harmonic theory. So it is really rare.


alan kay:
"Similarities over Differences" was to contrast with the standard processes of most nervous systems of most species to be more interested in "differences over similarities" (which is listed on the universal side). At most levels from reflexes to quite a bit of cognition, most similarities are accommodated and normalized while differences to the normalizations have a heightened significance (of "danger" or "pay attention").

Paying attention to differences is good for simple survival but makes it hard to think in many ways because it leads to so many cases, categories and distinctions -- and because some of the most important things may have disappeared into "normal" (in particular, things about oneself and one's own culture). So we unfortunately are much more interested in even superficial differences between humans and cultures and have a very hard time thinking of "the other" as being in the same value space as we are....

Part of the invention of modern math by the Greeks was their desire to get rid of the huge codexes of cases for geometry and arithmetic. This led to many useful abstractions which could be used as lenses to see things which looked different to normal minds as actually the same. For example, the Greek idea that there is only one triangle of each shape (because you can divide the two short sides by the long one to make a standard triangle of a given shape). This gets rid of lots of confusion and leaves room to start thinking more powerful thoughts. (The Greeks accomplished the interesting and amazing feat of using normalization to separate similarities and differences but paid attention to the similarities.) Calculus is a more subtle and tremendously useful example of separating similarities and differences. Convolution theory is yet more subtle ...

One way to think of my chart is that a lot of things we correlate with "enlightenment" and "civilization" are rather un-natural and rare inventions whose skills require us to learn how to go against many of our built in thought patterns. I think this is one of the main reasons to have an organized education (to learn the skills of being better thinkers than our nervous systems are directly set up for).

History suggests that we not lose these powerful ideas. They are not easy to get back.

The non-built-in nature of the powerful ideas on the right hand list implies they are generally more difficult to learn -- and this seems to be the case. This difficulty makes educational reform very hard because a very large number of the gatekeepers in education do not realize these simple ideas and tend to perceive and react (not think) using the universal left hand list .....

  • Aren't the non universals already in the curriculum?

  • What is blocking a more widespread uptake of the importance of teaching the non universals?
  • If School is failing then how and why is it failing?
  • Should there be other items on the non universal and universal lists? Are the lists complete?

Question (Bill Kerr):
I could think of some non universals / powerful ideas that are not on your
list, eg. Darwinian evolution, computer-human symbiosis for starters ...

Reply (Alan Kay)
Sure. There are lots (and they should be paid attention to). But certainly Darwinian Evolution (and a lot of other things fall under Science), etc. Computer-human symbiosis falls under the larger topics of how human thinking can be changed by the use of media (for better or for worse), etc, For a short list, it's best to use the biggies. Similarly, if we listed every built-in human trait (especially the zillions of bad ones), the list would be too long for any discussion purposes.

Neither list is complete. But the important property of the universals list is that most the items are well vetted. The importance of my non-universal list is just that 5 or 7 items are the most important changes that humans have made in their 200,000 years on the planet (and most of these came very recently (even agriculture)). What more do people need to start thinking with? What more arguments about modern science need to be made? (And if they do need to be made, then what new kind of argument would work?)

In other words, if the items on my list are ignored then it really doesn't matter much what else could be on the list. For example, the notion that there are "powerful ideas" could be the number one powerful idea, since it should lead to trying to understand powerful ideas, and to trying to find more of them.

John Maxwell:
I think the non-universals are by their very nature disputed and disputable, and incomplete. They originate with people, and are thereby politically situated and charged. They require an active cultural committment. The dispute over evolution is such an interesting case: here is an extremely powerful idea, the implications of which are still being worked out, and this is fundamentally threatening to all sorts of people, on levels which are deeply and historically embedded and which take centuries to sort themselves out.

  • How should the non universals be taught?

Alan Kay:
I like Bruner's term "scaffolded learning" because real discoveries are rare -- we've learned how to teach 10 year olds a good and mathematical version of calculus but no child has ever discovered calculus without guidance (and it took 200,000 years for two smart adults to do it with hints). Much of the "discovery and inquiry learning" curricula I've seen is pretty soft.

But learning and teaching would be easy if it could be transmitted by words or actions. Instead, some changes have to happen in the learner's mind/brain through some actions on their part (which could involve doing something or just sitting in a chair pondering). Things are sometimes not obvious because they are literally invisible, or because the explanations fall outside of existing commonsense thinking patterns. Or some new set of coordinations have to be learned/built that were not there before.

These have many of the trappings of creativity and the having of ideas that are not simple increments from the ideas of the surrounding context. The phrase I use for this is "Learning a powerful idea requires a lot of the same kinds of creativity as it took to invent it in the first place". This is because it has to be invented anew by the learner.

The good news is that learners for already invented ideas almost never have to be as smart and unusual as the original inventors (calculus can be learned by pretty much everybody, but Newton and Leibniz were unusual). On the other side, some real work has to be done to "cross the barriers".

Tim Gallwey (the incredible tennis teacher) use to say: you have to hit thousands of balls to learn to play tennis -- my method gets you to hit those thousands of balls, but feeling and thinking differently. A good method in mathematics (like Mary Laycock's or Seymours) still requires you to do lots of things (to get your mind/brain fluent) but can be and feel mathematical for most of the journey rather than painful in many ways. This is what we've called "Hard fun", and it is a process that is shared by any set of arts/sports/skills that have been developed.

Another way to look at it is "If you don't read for fun, you will never get fluent enough to read for purpose".

The big problem with the "standard algebraic route" is not so much algebra, but that the standard route requires lots of work but doesn't deliver "real math" very well. It's not situated in mathematical thinking, but much more in rule learning and following. People have turned Logo (and other computing) into rule learning and following, etc. It can be done to any initially terrific subject.


Some of the most important "powerful ideas" can be drawn from Anthropology, Bio-behavior, Neuroethology, etc., (how History can be interpreted in the light of these, etc.) and have to do with insights about ourselves that are critical and have remained hidden for 10s of centuries. Our research project is ultimately about getting children to start learning these, but we decided that we needed to learn how to teach math and physical science (and what kinds of each of these) to children first. Jerome Bruner saw this earlier than anyone and pioneered one of the greatest curriculum designs for elementary school children in "Man A Course Of Study" (MACOS), an intellectually honest presentation of Anthropology to 5th graders. This was implemented in more than 10,000 schools in the US in the late 60s, was a masterpiece, and ultimately was destroyed by religious fundamentalists in Congress.

more about Bruner and MACOS:
"Jerome S. Bruner also became involved in the design and implementation of the influential MACOS project (which sought to produce a comprehensive curriculum drawing upon the behavioural sciences). The curriculum famously aimed to address three questions:

What is uniquely human about human beings?
How did they get that way?
How could they be made more so? (Bruner 1976: 74)

MACOS was attacked by conservatives (especially the cross-cultural nature of the materials). It was also difficult to implement - requiring a degree of sophistication and learning on the part of teachers, and ability and motivation on the part of students. The educational tide had begun to move away from more liberal and progressive thinkers like Jerome Bruner"
- jerome bruner and the process of education

  • How do the non universals interact with the universals (the relationship b/w them)?
  • What are the implications for the curriculum wars (back to basics versus learning through discovery)?

John Maxwell:
Where I end up is that all the stuff about "how people learn" is kind of beside the point compared with the curricular importance of powerful ideas. This is the real value of education. But it's politically safer to do work on learning styles and assessment methodologies than to focus on the importance of something like evolution in the curriculum.

  • Are there some exemplars of teachers who do teach the non universals successfully?
  • Are we asking the right questions?

EuroPython keynote by alan kay (part 2)

Thoughts About Teaching Science and Mathematics to Young Children by alan kay

Squeakland list discussion about the non universals (August 2007 archive, sorted by subject, scroll down to start)

Tracing the Dynabook: A Study of Technocultural Transformations (PhD Dissertation) by John W. Maxwell
Ch. 2 How do we know a good idea when we see one? Answer: Draw attention to the division of labour that we take for granted, the reified discourse of "experts", "engineers", "designers", "end-users", "miracle workers", "plain folks" ... perhaps these boundaries don't have to be there
Ch 6. What is a powerful idea anyway? Discusses what powerful ideas are (eg. algebra, liberal democracy, atomic theory of matter, evolution) and the decline in powerful idea discourse.