NEW SCIENTIST REVIEW

Down with School! Up with Logoland!

The Children's Machine: Rethinking in the classroom, with mixed results. large and small, designed to implement his School in the Age of the Computer I was one of them. About ten years ago, ideas, and has received a wealth of feed­ by Seymour Papert, Basic Books, New York, I was part of a team that developed and back, much of it deeply discouraging. But HarperCollins in Britain, pp 241, £22·50 taught an introductory course in computer one can learn even more from "mistakes" science aimed at universirv students who than from a string of successes-that is a Daniel Dennett hated and feared computers but whose central tenet of Papert's vision of learning, parents, in many cases, had said "You must and he practices what he preaches. So this IN 1956, the mathematician John McCarthy learn about computers before you gradu­ sequel engagingly recounts what he has coined the term "artificial intelligence" ate." These students were seasoned veter­ learned, and especially the mistakes he for a new discipline that was emerging ans of what Paperr calls School-experts at made along the way. His own thinking from some of the more has undergone a transforma­ imaginative and playful tion; he is still an infectiously explorations of that new optimistic visionary, but a mind-tool, the computer. A wiser one. few years later he devel­ Logo has nowjoined forces oped a radically new sort of with Lego, the plastic build­ programming language, ing blocks, and a new wave Lisp, which became the of delectable settings for lingua franca of AI. Unlike learning has been created and the sturdier, stodgier com­ explored. Papert favours puter languages created by parables; in these pages you and for business and indus­ will find no statistical surveys try, Lisp was remarkably of the effects wrought by open-ended and freewheel­ Logo-Lege projects and no ing. Instead of concentrat­ data on control groups with ing on numbers, it was which to compare the inspir­ designed to take any sym­ ing tales he tells. Indeed, at bols or strings of symbols several points he says that he (lists) as its objects, and is quite sure that no test because its own machinery could be counted on to reveal consisted of just such lists by before-and-after compari­ (and lists of lists), Lisp son the benefits he has anec­ creations easily inhabited dotally conveyed. As you read the very world they acted further you come to realise upon, and hence could that this is not a dodge on his reflect upon themselves and part, but a fundamental their own reflections indefi­ implication of his message: nitely, revising and rein- the sort of testing required by venting themselves, breaking down the piling on the facts, drilling for the big test­ such attempts to measure success destroys artificial barrier between program and data. and they were pathologically uncomfort­ the very conditions for learning that the Seymour Papert was one of the most able in any setting where they had to think. computer, at its best, can create. playful of the AI pioneers, and more than (My impression then was that many of There is a recurrent pattern that has any of the others, his own reflections them, given a choice between solving a bedevilled many-in the end, almost all­ turned to the nature of that playfulness simple puzzle and memorising two pages of attempts to use computers in education; no and its role in learning and discovery. In the phone book, would gratefully choose sooner are the classrooms equipped than 1980, he published Mindstorms, in which the latter project.) I had read Papert's book the dead hand of educational bureaucracy he presented his Utopian vision of comput­ and discussed its ideas with him at length. begins to impose conditions that systemati­ ers in the classroom, of which the centre­ Logo was designed for five-year-olds, so it cally squander the power the equipment piece was Logo, a dialect of Lisp that he might, I thought, be just right for my uni­ promised to deliver. Of all the teachers who and others had developed specifically for versity students. In fact, it was spectacu­ enthusiastically took up the themes of very young children. larly successful. The students forgot their Mindstonns, "many felt seduced and aban­ The key element of his design was Turtle phobias and inhibitions and took flight, doned by the talk of a computer revolution graphics, an inspired interface that made creating a trove of idiosyncratic projects, as the use of the computer became children's interactions with Logo not just effortlessly learning the fundamentals of routinised". Instead of railing at the system visible, but instantly comprehensible­ programming, and building a robust base or blaming the teachers or administrators, feelabie, you might say. The tales he told on which we could then help them to con­ Papert looks at the broader problem, and )f those early encounters were compelling. struct a more "adult" set of edifices. So I am sees that it springs from deeply held-and Ihese became an important ingredient in one of the many who can personally attest of course ill-examined-assumptions about the barrage of persuasive literature that that Logo, in the right circumstances, does the point of School. led teachers and schools all over the US, work wonders. Consider the awkward confrontation you and indeed the world, to invest huge sums The right circumstances are hard to come would inevitably create were you to imple­ in "computerising the classroom". Thou­ by, alas. In the years since Mindstorms, ment one of Papert's fundamental princi­ sands of teachers tried their hand at Logo Papert has participated in a host of projects, ples of learning: take your time. Learning

6 November 1993 45 REVIEW NEW SCIENTIST

happens best when you can browse around that has been shaped by the pressures of a be crowned with success, but saps our self­ in a problem space, savouring the shapes, market economy,he encourages us to make confidence. With no fund of shared experi­ fiddling with the bits and pieces, twiddling a leap of faith, and trust in the distributed ence of screwing up, we are apt to harbour the knobs-but always, always, taking your wisdom of many local experiments under wildly unrealistic fantasies about the intel­ time. So there you are, noodling away minimal control: "The RigorousResearcher lectual prowess, the clarity and rigour, of contentedly and constructively when the will object to the populist tone of this our teachers and peers. This Victorian bell rings, and you have not finished your argument. It is appropriate to buy a food prudishness about our own cognitive dis­ assignment. Well? What sort of payoff does processor or garlic press on the basis of abilities is built right into the fundamental society want? And how soon? It is always a individual whim, but education is more structure of School, and it spawns a host of nagging problem. serious. Every child deserves the best. Sci- secondary effects, all debilitating. There are Lying behind this and other confronta- ence should be used to find out what is the some wonderfully liberating passages in tions is an assumption that Papert calls the best, and then everyone should adopt the which Papert, as usual practising what he Gothic cathedral model of learning. Sup- proven methods." preaches, describes his own confusions, pose educating a child were like "building Papert has a good retort. What makes the false starts and insecurities, and then a Gothic cathedral out of 40 000 blocks Rigorous Researcher think there is a best recounts the childlike solutions he discov­ . of stone. Clearly, strict way? Herbert Simon, ers. We should all be such children. organisation is needed to another of the founders I am a believer, but the adult in me says perform such a task. One of AI, has built his dis- that responsible scepticism has still not cannot have individual tinguished career around been met halfway. Take a few thousand workers deciding that showing the importance children and feed them a lovingly prepared they want to put a block of Voltaire's maxim: diet of Logo-Lege, and it is not surprising here or there just because "The best is the enemy that you can harvest a few-or even a few they are inspired to do of the good", and Papert hundred-inspiring success stories. What so." Instead of conceding would replace the hyper- happened to the other children? I don't that education must, after rationalist quest for the think anybody knows, at least not in detail. all, be a building process best-which has given Presumably they were at.least not harmed conducted under the wis- us a series of stultifying, in any way, but were there enough benefits est set of controls we can straitjacketed systems of to justify expanding these programmes? muster, Papert takes on education-with" a free- When the thrill of pioneering has to be the task of persuading us roaming, "cybernetic", traded in for year-in, year-our practices that that education is not that feedback-guided track- can be reliably adopted by thousands of less sort of task at all. We ing of the good. Many inspired teachers, will the same wonderful must dare to let go-not will be tempted to dis- effects be in evidence? Papert is cautiously always, but as much as miss Papert's proposed optimistic, and he suggests that in any case we can bear. We must adopt a more "sys- revolution as a trendy amalgam of politi- the current systems are so toxic for so many temic" view of learning as a variegated cally correct themes, unabashedly fes- children that we risk little by taking family of "emergent" phenomena. This is a tooned with the buzzwords of feminism, the leap, but he offers no blueprints and truly radical idea, which Papert cannily empowerment, multiculturalism and anti- no guarantees-it would fly in the face of makes more palatable by an analogy. His elitism-Dionysus si! Apollo no! and all his radical message to offer either. For the revolutionary proposals would create a that. But in fact he has given these themes time being he is content to make it clear market economy of educational experi- a much more unified, circumscribed and that there must be feedback if these mentation, in contrast to today's traditional cogently reasoned justification than one experiments are to sort themselves out in a educational hierarchy, which he likens to generally encounters. desirable direction, but he has little to say the Soviet Union's disastrous planned" And then there are the details. There is about how to design the appropriate feed­ economy: "... while our economic system, just no escape from his shrewd analyses of back mechanisms. We can hope that his with all its faults, is above the threshold of the crushing obstacles to learning that next book will have more to say on this functionality and theirs was below it, our stand untouched by traditional pedagogy. important topic. 0 education system falls on the same side of To my mind, the most important of these the line as the Soviet economy". The timing analyses are his observations on the crip- Daniel Dennett is the cofounder of the of Papert's proposal could not be more pIing taboos against self-exposure.We learn Curricular Software Studio at Tufts University in ticklish: just as the US is finally coming to in Schoolto conceal our own ignorance and Massachusetts, where he is Distinguished grips with the spiralling cost and diminish- confusion, and this not only inhibits us Professor ofArts and Science. He is the author of ing effectiveness of a health-care system from exploring the very moves that would Consciousness Explained (1992).

46 6 November 1993