Mindstorms by Papert

I’m finally getting around to reading Mindstorms by Seymour Papert. I’ve enjoyed his writing before so was looking forward to this seminal work on computing education. I’m surprised by the first chapter though.

In a typical experiment in combinatorial thinking, children are asked to form all the possible combinations (or “families”) of beads of assorted colors. It really is quite remarkable that most children are unable to do this systematically and accurately until
they are in the fifth or sixth grades. Why should this be? Why does this task seem to be so much more difficult than the intellectual feats accomplished by seven and eight year old children? Is its logical structure essentially more complex? Can it possibly require a neurological mechanism that does not mature until the approach of puberty? I think that a more likely explanation is provided by looking at the nature of the culture. The task of making the families of beads can be looked at as constructing and executing a program, a very common sort of program, in which two loops are nested: Fix a first color and run through all the possible second colors, then repeat until all possible first colors have been run through. For someone who is thoroughly used to computers and programming there is nothing “formal” or abstract about this task. For a child in a computer culture it would be as concrete as matching up knives and forks at the dinner table. Even the common “bug” of including some families twice (for example, red-blue and blue-red) would be well-known. Our culture is rich in pairs, couples, and one-to-one correspondences of all sorts, and it is rich in language for talking about such things. This richness provides both the incentive and a supply of models and tools for children to build ways to think about such issues as whether three large pieces of candy are more or less than four much smaller pieces. For such problems our children acquire an excellent intuitive sense of quantity. But our culture is relatively poor in models of systematic procedures. Until recently there was not even a name in popular language for programming, let alone for the ideas needed to do so successfully. There is no word for “nested loops” and no word for the double-counting bug. Indeed, there are no words for the powerful ideas computerists refer to as “bug” and “debugging.”

This just seems really a strange point of view. People have been making combinations of beads for a very long time, surely. Clearly “nested loops” are extremely common in textiles, and the double-counting bug simply doesn’t arise when sorting beads physically, unless you treat them as ordered pairs… and surely this kind of mathematics is very old. The text continues to describe all this as a new style of thinking, arguing that this is the first time we have the tools to think about thinking… Just what??

I’ll carry on reading but would love to hear what others think about the book, or join folks for a reading group discussion around the book if some of you aren’t completely zoomed out by now…


Fascinating topic, glad to zoom in to discuss this! Always excited to talk about nested loops…

I don’t know the author, and at first glance I couldn’t detect any “social darwinist” overtones. On the contrary, i find it quite healthy to address the leap in culture brought about by the rise of code speech, which provides us with more intuitive ways of performing complect tasks. I’ve made the point in the past (on the tidal forum) that recursion could be tracked down in time up to Levallois débitage, i.e. Neanderthal Times, so I not only it agree with you that this type of mental pattern is old, it must be really old. I’ll make the link with what prehistorians identifybas traces of “children flint knappers” or better even " apprentice or beginner level" knappers. I always wondered, how they could determine why the less well executed pieces were made by inexperienced individuals, less even by way of age. I can easily imagine children getting it right super easily. So for me the problem is really: how can Papert claim that ALL children are unable to classify beads before they’re x year old? I’m eager to know what you think.

1 Like

Sorry for the very slow reply! I don’t detect social darwinist overtones either, I just see a strange blindspot, in that in a bead sorting task, Papert doesn’t consider bead-sorting cultures. I think this blindspot runs through computer science culture, the idea that there ideas travel along a one-way street from computer science to culture. So there’s the assumption that by learning computer programming we are able to think about bead sorting in a new, formal way. Presumably all the patterned necklaces in museums were made without thought. It’d be interesting to see if children are able to do such bead sorting earlier in communities where bead sorting is an everyday activity.

It would be good to get back to organising some zoom calls, I am finishing off a project at the moment but will hopefully have full attention on algorithmic patterns from December and be much more organised from then.

1 Like

Takes me back, I remember reading (bits of) Mindstorms by Seymour Papert when it came out in the ‘80s. I was setting up a lab of BBC B micros mainly for use by the large Primary School Teacher Training department of Nottingham Trent Polytechnic, complete with Logo Turtles. Using computers in schools was a new idea back then. The DTI (Department of Trade and Industry) established its Micros in Schools scheme in 1981/2 and every school found themselves with a BBC B or a Research Machine 380Z and encouraged by the PM herself to accept the challenge afforded by the new technology (Speech on microcomputers in schools | Margaret Thatcher Foundation).

Two things I recall from those days. Firstly the initiative was driven by the DTI and not the Department of Education ~ allegedly the government was wary that if the DoE instructed teachers what to do and what to teach the teacher’s unions would object, but that an incentive from Trade and Industry might be accepted. And secondly, the change in schools. Some (perhaps most) teachers took little or no interest or notice of the new micros (“How am I going to use one micro to teach my history class of 30?”) and a few struggled to integrate the micro into the lessons they had been teaching, without a computer, for ages, a fairly difficult task as there was little or nothing to help them. Few teachers had any experience of computers or programming (or coding as it’s now called) and if you admitted to knowing what a computer was the likelihood was that you would be given the box containing the new micro that had been delivered and told to get on with it (and if you were lucky you may get a new job title: Head of Computing). Rumour had it that many schools just put the box in a cupboard and carried on as before! Time for learning new tricks was scare even then.

Papert (Seymour Papert - Wikipedia ) was one of a few who thought computers could make a difference. He worked as a researcher in a variety of places, Cambridge, NPL and MIT where he was co-director of the AI Lab and he worked with Piaget (Jean Piaget - Wikipedia). His take on pedagogy developed into constructionism (Constructionism (learning theory) - Wikipedia) ~ hence, I guess, his interest in sorting coloured beads.

1 Like

@yaxu Sorry for my slow replies too! This topic is of particular interest, but I realize I also missed out on another thread. I’d be glad to join a jitsi or zoom call when the time comes.

@SimonR it’s great to read your reply because it really helps contextualizing. However, despite Papert’s great career highlights and his (sort of) activism toward student-centered education, it’s possible to think about this particular example and challenge it somewhat. After all, Piaget’s model of what children are, although innovative and open-ended, can be questioned. As I remember what I learned form him, hekind of view children as miniature/unfinished adults, which nowadays seems to fit European / 18th century children better than many other groups of kids the same age. On top of that, Piaget engaged in an experimental feedback loop by setting up his model by analyzing his own children, whose behaviour we can guess was influenced schrödinger-style by being the subjects of constant “scientific” observation. Despite these ideas that I keep from way-back when I was studying psychology for some time, I appreciate Piaget a lot.

At this stage, and without diving too deep, @yaxu 's point seems crucial in that different children in different societies with different social models would have different abilities. It would be great to have time to dig deeper into this. What I sense is that there would be a point in demonstrating what computer culture "does to " people whom it influences deeply. Whether patterned necklaces can be crafted “without though” or not depends on how we define conscious thought ; presumably programming a patterned necklace on a computer involves another form of involvement than sorting beads and patterning them. However, as I might have already pointed out too often already, this brings up topics such as animal thinking and the development of conscious, sentient inner discourse : what kind of subjectivity is the bower bird gifted with and how much intention do these creatures put into their works? How much can we say individual bees reflect on their hierarchical position?

It would also be quite nice to discuss biases in computer culture. The idea that "ideas travel from computer science to culture is very deep-rooted. It’s related to the way our culture thinks of science, the circulation of knowledge and the way culture interacts with knowledge. Could we say that popular culture has an impact on scientific culture? If so, the model would be quite different, because it potentially involves several feedback loops : computer science does influence popular culture, but popular culture also influences computer science, so the way these elements interact is varying over time, perhaps even following some kind of cyclic pattern…

I’m going much too fast here, but gathering and processing more topic-related literature would be a good way to keep thinking about this.

1 Like

Thanks for putting things in context @SimonR, I’ll get to reading more of mindstorms before commenting further, but I have to comment on this bugbear:

I remember the single BBC micro arriving in my primary school. It was a very rare 10 minutes that I got near it, enough time to load and play a game for a short while. It was stolen a couple of times. This experience was so far from the narrative of the golden BBC micro era painted by the raspberry pi people that I can only assume that they all went to private school…

I haven’t done much bead sorting but have been introduced to weaving, tablet weaving and braiding, and this definitely requires conscious calculations, although they are so embedded in physical movement and relations that they are quite hard to talk about. Also there’s no real motivation to talk about the calculations, it’s enough to see the structure of the resulting textile, as a kind of writing.

I’d say computer programming is the same, it involves conscious calculations but it’s not too easy to talk about them. This is obscured though because the process of programming produces a written program - which is explicit and it’s easy to talk about how it works as its own process. But talking about writing a program is very different from talking about running one…

The way the topics get woven here is very nice. Memories of primary school computers have a distinct feel from those of crafting objects, probably because the machines were rare and the teachers might have at times conveyed a sense of anxiety toward what was happening to these computers. My memories relate to a turtle, which I found was related to the Logo language. A few hours were enough to realize that some kids were faster than others, and then the whole apparatus disappears. We must have had a couple machines in a dedicated room, which different classes had scarce access to.

@yaxu I agree, running a program and writing one are hardly comparable tasks. I really had live coding in mind, which involves having a clue about quite a lot of concepts before starting to “create a necklace”. Let’s assume one knows about loops and variables, about printing colored pixels on a screen. Then, writing a program does involve conscious calculations, but maybe also (quite importantly) some sort of abstraction mechanism in order to define rules relating to what beads looks like, how to design those in different sizes and colors, chain them, etc.

Do you think the conscious calculations involved in creating a physical patterned necklace share the same pre-requisites? My impression is that the rules pertain to the symbolic, conceptual and visual result, as opposed to the syntax of the meta-rules which enable the very possibility of writing a valid program. In an archaeological perspective, patterning seems directly related to programming, but I wouldn’t say the two are the same. Would you agree that the notion of “creating a stencil” is relevant in order to define what it could mean to “program a necklace”… focusing on a set of rules rather than on the execution of an individual work?

The question of physical movement is a very interesting one, too, when it comes to creating patterns! Mastering minute finger motion can be quite a challenge depending on an individual’s background and experience. Zhuangzi (3rd century BC?) has some magnificent paragraphs about the bodily intelligence of craftsmen who manage “thinking with their bodies”, thus transcending a trial-and-error based empirical approach. Although programming also involves precise finger motion (and gathering relevant visual cues!) , maybe what makes a real difference here is that in order to write code, one has to be able to read it simultaneously (a feedback loop?) and access its abstract implications. In a way, “programming a necklace” would mean defining rules which can output many different patterned necklaces by changing some variables. Creating a physical necklace might involve more focus on an individual pattern, which also involves a clear and conscious relation to patterning, but perhaps has a specific impact/fingerprint on the way individuals thinks of the possibility of creating variations?