How do we move from getting kids into school to actually educating them? Provocative new book by Lant Pritchett

I approached Lant Pritchett’s new book ‘The Rebirth of Education’ with glee and trepidation. Glee because Lant is one of the smartest, lant pritchettwittiest and best writers and thinkers on development. Trepidation because this issue is an intellectual minefield of Somme-like proportions (remember the epic Kevin Watkins v Justin Sandefur battle?). And sure enough, Lant took me into all kinds of uncomfortable places. Allow me to share my confusion.

First the book. Based on a data-tastic summary of a lot of research and case studies, Lant argues, in the words of the book’s subtitle, that ‘Schooling Ain’t Learning’:

  • In India less than half of children surveyed in grade 5 could read a story for second graders (and over 1 in 4 could not read a simple sentence), and only slightly more than half could do subtraction. Results over several years were getting worse, not better. See graphic for more examples.
  • In Tanzania over 65 percent of students who sat the 2012 examination for secondary school (Form IV) completers failed, with the worst possible results.
  • A majority of 15 year-olds in low- and middle-income countries have only learned enough to reach the bottom 5 percent of their peers in high-income countries.

UP 5th gradersThe evidence is piled high that there is a crisis in education – kids are now in school, thanks to a huge effort by governments, publics and aid donors – but they aren’t learning anywhere near enough. In most countries, kids dropping out through boredom and futility far exceed the lack of enrolment as a cause for under-achievement. What a horrendous waste.

How did it get like this and what could turn it around? The big proposition here is that the system required to deliver mass enrolment is fundamentally different from that needed to deliver genuine education. Lant is adamant that ‘more of the same’ – more schools, more teachers, more books, more teacher training – cannot deliver if it takes place within the current system, and cites a lot of cases where such increases in inputs have not delivered. He calculates that even if Mexico increased spending on its current education system fivefold, it would still lag way behind the OECD average.

Instead, he argues that this is a system problem. (His writing on systems is superb – more on that tomorrow). He draws a distinction between ‘spider’ and ‘starfish’ systems:

‘A spider uses its web to expand its reach, but all information created by the vibrations of the web must be processed, decisions made, and actions taken by one spider brain at the center of the web. The starfish, in contrast, is a very different kind of organism. Many species of starfish actually have no brain. The starfish is a radically decentralized organism with only a loosely connected nervous system. The starfish moves not because the brain processes information and decides to move but because the local actions of its loosely connected parts add up to movement.’

i.e. education systems have to move from a state-led command and control model to a looser network that encourages evolution and emergent change. He likens spider systems to an eggshell that has allowed embryonic education systems to reach full enrolment, but must now break to allow the move from enrolment to actual education.

He argues that the switch from spider to starfish ‘is not the usual battleground of ‘markets’ v ‘governments’’, but one of accountability, rebirth-education-lant-pritchettwhere state-run systems in Germany, France and the Netherlands have produced effective starfish systems. I’m not entirely convinced by that. Although he does have examples of starfish states, he then doesn’t explore them, or draw any lessons for today’s education reformers. This seems an extraordinary omission.

Instead the tone of the book is overwhelmingly hostile to state education and evil bureaucrats and teachers’ unions seeking only to defend their ‘vested interests’ and snuff out innovation. And the positive examples he cites are community-controlled schools, private providers, decentralized ‘small government’ and charter schools – hardly a statist shopping list.

Where he gets really interesting (and where I particularly need a steer from readers who know more about educational history than me) is on the political economy of education. He argues that the dominant driver in the introduction of mass public education has never been a desire to educate, but ‘everywhere and always a contest for the control of socialization’, a process which the state refuses to outsource. He cites a wonderfully frank 19th Century Japanese education minister: ‘In administration of all schools, it must be kept in mind, what is to be done is not for the sake of the pupils, but for the sake of the country.’

He sees two other drivers of mass education:

  • The need to prepare workers for a modernizing economy, but that doesn’t explain why both slow and fast-growing economies have moved so fast to introduce mass public systems
  • A desire to ‘keep up with the Joneses’ by e.g. meeting the MDGs, which leads to a  focus on enrolment, rather than education

His conclusion?

‘The spider systems we have today were designed in the nineteenth century and adapted and adopted in the twentieth century to meet a certain set of demands: to prepare workers for a transition out of agriculture, to build nations to support states, and to legitimate the regimes that controlled those states. It would be extraordinary indeed if those spiders just so happened to be systems designed for the learning and educational challenges the youth of the twenty-first century will face.’

True to his understanding of emergent change in complex systems, he then refuses to provide a complete blueprint for an alternative, and instead suggests six characteristics of effective starfish systems:

Open: How is the entry and exit of providers of schooling structured?

Locally operated: Do those who manage schools and teach in schools (and the local coalitions of parents and citizens they are accountable to) have autonomy over how their school is operated?

Performance pressured: Are there clear, measured, achievable outcome metrics against which the performance of schools can be assessed?

Professionally networked: Do teachers feel a common professional ethos and linkages among themselves as professional educators?

Technically supported: Are the schools, principals, and teachers given access to the technical support they need to expand their own capacities?

Flexibly financed: Can resources flow naturally (without topdown decisionmaking) into those schools and activities within schools that have proved to be effective?

As examples he cites the US/UK university system, the International Baccalaureate and Brazil’s recent education reforms.

Fascinating, but I have some concerns:

  • What are the trade-offs in moving from spider to starfish? What will be the impact on equity of decentralizing control or finance, let alone of a much greater role for private providers?
  • And can we really assume that non-state providers will be any freer of suspect motives than governments? Religious schools?
  • Can a real education reform process really be this jaundiced about the motivation of teachers and their organizations? Surely they have to be far more central?
  • 100% enrolment is most definitely not a done deal, especially for the most marginalized. Watch out for next week’s Global Monitoring Report on that.
  • Lant’s claim that ‘more’ (teachers, textbooks, classrooms) doesn’t help is, to put it mildly, contested

Which (even though I, like everyone else, am convinced of the crisis in education) all leaves me floundering between two apparently contradictory intellectual poles. I am convinced (as is Oxfam) that state provision (not just funding) has historically played a fundamental role in producing equitable and effective essential services. I am also convinced that systems thinking, applying the ideas of emergent change and evolution etc, can provide rich ideas on how change happens. To what extent are the two compatible/in conflict? Off to my padded cell to agonize……..

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


18 Responses to “How do we move from getting kids into school to actually educating them? Provocative new book by Lant Pritchett”
  1. Nicholas Colloff

    Historically I think the motivator was how do we supply a sufficiently educated workforce to a newly industrialising economy whilst maintaining/guiding socialisation. Education that is good enough but not socially destabilising (or liberating)!

    I cannot see why you cannot have a state system that provides the infrastructure and standards (both for inputs -qualified teachers/curriculum and outputs – educational performance) but then trusts teachers/parents to self-organise the educational experience – except we appear incapable of trusting professional judgement (and recognise that the more we seek to channel and control it, the less we get, we disempower it).

  2. Alex Gwyther

    It sounds like a really fascinating book. We at UKCDS are hosting an event this Thursday on education systems research that should be especially relevant.

    We are hosting a (free) Town Hall event this Thursday on funding for education systems research. There is imminent new UK funding into education research for international development and we have organised the event, with four major funders coming to speak, as an opportunity for academics to learn the ins and outs of how funders decide what research to commission and to hopefully improve the research tenders and procurement process.
    There are still spaces available, but for those unable to attend we will also be live streaming the event.

    For how to attend and more details follow this link:

  3. Rui

    As you point, without the state intervention the equity is compromised and the marginalized will remain marginalized. Who has greater economic power will always have “powerful knowledge” and the others will have the “knowledge of the powerful”. (about this see eg. Michael Young (2013) Overcoming the crisis in curriculum theory : a knowledge- based approach , Journal of Curriculum Studies, 45:2 , 101-118 , DOI : 10.1080/00220272.2013.764505 )
    A community school in the rural area can provide a “powerful knowledge” similar to a community school in the city? I don’t think so with a completely decentralized system.
    Historically, for instance, there are different curriculum traditions in Germany and in USA. While in USA the Tyler ‘s basic principles of curriculum prevails (a model that transnational organizations promote in developing countries) in Germany prevails the concept of “Bildung”. This too traditions are very different.
    There are several case studies that show that when the state system were dismantled, especially in African countries in the 80 result of structural adjustment plans, educational systems have become more unequal and learning problems was not improved.
    I think should be a balance between decentralization and state intervention.

    (I hope you can understand my English because is not my mother tongue)

  4. Caroline Sweetman

    There’s no clash between believing the state should provide education and understanding how change happens in a complex way – they are two utterly different concerns – not opposite poles of the same spectrum as you imply. The first is a politically-informed conclusion about who should provide and the second gives us insights into how best to provide – and challenges states to develop ways of working which are more starfishlike… so you can come out of your padded cell. Does sound extraordinary that Lant doesn’t provide examples of state which have attempted this, but presumably that’s because he’s not interested as he’s a private sector boy himself? Anyway there’s a research project there for someone to do.

  5. aldo matteucci

    While pondering the second installment, I’d suggest you have a look at Steven JOHNSON; Where good ideas come from. I’ll will gently wean you off dichotomies, as in spider/star.

    Johnson concludes his book with the metaphor of the reef – akin to Darwin’s “crowded bank”. On a reef there is room for both spiders and stars. It all depends on the ever-changing context.

  6. Jed Stevenson

    For a clear, concise, and well-referenced account of the origins of the Western education system, and its globalization in the 20th century, see the first chapter of “Literacy and mothering: How women’s schooling changes the lives of the world’s children” by Robert A LeVine and colleagues (2011).

    (The whole book is well worth reading; I’ve reviewed it here:

  7. Sonia Languille

    In a context of jobs scarcity (or job creation in the unskilled informal sector) and in the absence of a job-intensive developmental path, the expansion of African education systems indeed fulfills a traditional role of integration of the youth; it responds to a social order imperative. This is certainly not specific to developing countries: in continental Europe, the expansion of senior secondary education happened quite recently – in the end of the 1970s/1980s: a period that exactly coincided with the dramatic increase in unemployment. And in Western countries, education systems have built in mechanisms that allow achieve a balance between the integration of the mass and the reproduction of social differentiations (in France, ‘grandes ecoles’, in the UK, the Oxbridge system…).
    The issue of quality education in sub-Saharan Africa should not be delinked from broader political economy considerations and from the economic prescriptions imposed on sub-Saharan African countries by the international community, since the structural adjustment programme era: fiscal discipline, trade liberalisation (that decreased African states’ tax base), a shrinking public service (traditionally purveyor of formal jobs and therefore of demands for skilled labour) or the delegitimation of state led industrial/development policies have profoundly shaped African education systems. In contrast, the ‘human capital’ component of the South Korean ‘miracle’ was one of a tight relationship between the state’s industrial policy and education and training policy, which evolved in time and was associated with a specific political settlement.
    Lant Pritchett’s work (I have not yet read his new book), which largely puts the blame on ‘rent seeking’ bureaucrats, teachers’ unions and teachers’ behaviour, denies a key political economy aspect of the question, which is the fiscal space (and therefore global and domestic choices over resources allocation) to sustain quality education for all: even if, without doubt, delivering quality education cannot be subsumed into a simple matter of resources, it remains a matter of fact that African schools are resource starved. And African teachers’ attempts to adopt more learner-centred pedagogical methods are largely constrained by the lack of resources. Besides, restructuring teachers’ incentives – for instance through performance-based salaries – may well go against one of Pritchet’s features of effective starfish systems (‘Professionally networked: Do teachers feel a common professional ethos and linkages among themselves as professional educators?’). Tying salaries to students’ performances may indeed induce a competitive, individualistic ethos against the traditional collective values and norms of the teaching profession, committed to a socially valued objective.
    Moreover, it remains difficult to understand why private providers (the market) would care about students’ performances more than the state. Evidence that would shed light on the mechanisms by which private ownership status would deliver, by itself, better learning results remains actually very thin (a similar absence of analytical and empirical demonstration stands for other public services like water or electricity). On the other hand a growing literature has shown the detrimental effects of charters schools or voucher mechanisms in terms of growing educational inequalities.
    Finally, one may suggest that quality education fundamentally remains a political issue. As such, more than a localised reengineering of schools’ management, it primarily calls for a national as well as localised democratic deliberation on its meaning, its features and on the national resources to be allocated to it: in that context, the state (rather than ‘nature’) remains certainly the least bad instrument to ensure an equalisation of resources among local territories, schools and citizens (of course subject to conflicts).

  8. Peter Chowla

    I find this approach to anti-statism very unconvincing. The popular press is full of stories about how Asians are passing us in education. PISA records massive dominance of Singapore, Shanghai and Korea n rankings. How do you think these are not “spider” systems with massively domineering central-state bureaucracies?

    I am not advocating the Asian model – I have enough family members who led from it because of its rote learning and lack of teaching of critical thinking. But it does a damn good job in getting 99-100% on the fifth graders’ maths questions in your graphic. If that is your test of learning, we need more spider systems if you look at some of the evidence…

  9. Steve Klees

    While I will agree with you that Lant Pritchett is smart, he is very far from one of the “best writers and thinkers on development.” Lant is an ideologue. This book and previous works all are written from a commitment to a very narrow, neoliberal version of economics. I don’t use the “ideologue” label as name-calling but as descriptive. Perhaps all we have are ideologues, myself included. The ubiquitous calls for evidence-based policies are supposed to get us away from ideologies but they don’t. Evidence always can be marshalled and interpreted differently so we are always left with debate. What Lant offers is his biased view of evidence and all I can offer is my alternative.
    You begin with Lant’s lament over the sorry state of education in developing countries. That is correct but how to understand why that is so is complex. While Lant’s focus is on test scores, we have not done very well on access either. There have been international commitments to Universal Primary Education since the 1960s. Although we have made some progress, once again we are about to postpone UPE until 2030. Yet UPE is relatively easy to accomplish. There is an important argument to be made that the Education for All accord and the MDGs, despite good intentions, are more cosmetic than serious efforts.
    We are unwilling to try the “more” policies that Lant criticized – more and better teachers, learning materials, facilities. In fact, the last 30 years of the neoliberal-cut-and criticize-government-era has gone in the opposite direction. In most developing countries, class sizes have become unconscionable, teachers are less trained, learning materials are scarce, and facilities are poor. It is no wonder that little learning is going on and that parents everywhere search for any possible improvements in charters or private schools. But this is not because they are inherently better but because we have abandoned our public school systems. This is the true crisis in education.
    Moreover, most education problems are social problems that cannot be fixed by education remedies alone. A recent World Bank report said that 3 billion people on this planet are living at the margins of society. Given problems with poverty, health, lack of parent’s education, lack of jobs, a sharp educational pyramid that limits educational opportunities, it is again little wonder that education is in crisis.
    This is not a problem of developing countries alone. The huge amount of disadvantaged children in the U.S. have severe learning problems in school and often drop out of high school. While we have always done a decent job of educating middle class kids, we have never done a good job with disadvantaged kids because we have been unwilling to devote the education resources it takes and because of the similar-to-the-above huge disadvantages children face outside of school (see Richard Rothstein’s book, Class and Schools).
    All we get from Lant is a tired old re-write of the ideological solutions pushed for the past 30 years under the new guise of spiders and starfish (perhaps to give it a so-so-biology basis). The tired solution for the neoliberal era has been changes in school management – for thirty years we have heard talk of restructuring, re-engineering, reorganization, knowledge management, civil service reform, decentralization, privatization, and more. Even if we only look at narrow measures of achievement – which we shouldn’t – none of these has made a difference. In this era of hostility to government, nobody has been willing to try what we know can work – better trained teachers, smaller classes, adequate learning materials, facilities conducive to an education along with the concomitant social reforms necessary to improve students’ lives out of school.
    There are as many innovative public schools as there are private school in developing countries and elsewhere. Autonomy is not some magic wand, and, as you said, decentralization always increases inequality. Moreover, we are part of larger groups – regions, nations, globes. These larger pictures must be taken into account in running schools. In the U.S. we are moving towards more centralized influence (e.g., standards). It is ideological blinders, not evidence, that leads Lant to reject government and teacher unions as the lead players in improving education.

  10. Daniela Lloyd-Williams

    When you have poorly paid, badly trained teachers who live in the classroom (because they were posted to a rural school), when you have more than 100 children in class, when you don’t have textbooks or other teaching materials, when the children arrive at school hungry and leave hungry, is it any wonder that the majority don’t achieve their expected learning outcomes? In some countries, such as Burkina Faso, what is needed is simply more schools, more teachers, better trained teachers and more learning resources. Sure, the education system could be restructured but first invest some money at the coal face.

  11. Joern Geisselmann

    To my knowledge one of the most innovative and effective approaches to (secondary) education comes out of Columbia:

    The “Sistema de Aprendizaje Tutorial”, SAT, is a formal—but flexible—educational program that arose from FUNDAEC’s efforts initiated in 1974, aimed at contributing to the progress of rural communities. The System has developed a methodology which makes it possible for any individual—youth or adult—from the most remote rural region to have access to a secondary-level education par excellence.

  12. Jane Carter

    If you want an example of a starfish type education system, I think Switzerland is a case in point. Here education is the responsibility of each canton (of which there are 26 in a country of just under 8 million people), and takng State = canton, education is almost entirely State funded. Private schools exist, but are few and really for specialised teaching – the vast majority of children attend State schools which are well funded, and teachers well paid. I found it mind-blowing to realise that curricula are (or at least were) designed at cantonal level, leading to text books with local examples and, more significantly, no clear national standard at school leaving. There was a referendum a few years ago on introducing a federal curriculum, which passed, so that is a project now in progress – teacher friends of mine find it a bit of a nightmare even if they believe it’s the right way to go. There are of course 3 national languages so that’s an obvious challenge to begin with. Overall, I’m a big fan of the Swiss system; certainly my children have had a truly excellent education – better (it seems to me) than their English cousins attending expensive private schools, but that perception may be baised by my politics!
    Lessons to be drawn
    – States can run decentralised education systems that respond to local needs (eg. the town in which I live is biligual French/German, so bilingual studies are offered, especially at secondary school level)
    – Private schools are not a necessary part of the system to ensure an offer of quality
    – Funding is crucial – excellent facilities and well paid teachers who have respect in the community of course make the Swiss education system what it is. Here I guess an important issue is the taxation base; if too local, poorer areas would get poorer schools and vice versa. This is evened out through subsidises across the canton.
    I totally agree with previous comments about underfunding of education leading to poor results; if teachers are inadequately trained, poorly paid and have to work in crumbling buildings without adequate materials, it’s no wonder that children fail to learn – especially if they themselves have to walk miles to school on less than full stomachs. But I guess Lant does not dispute that!

  13. Lant Pritchett

    Loved the review. Hope the book is as good as its review.

    One minor quibble is that I go out of my way to not say “inputs don’t matter” or that more budget/training/textbooks “doesn’t help”. My point is that an input based approach doesn’t appear from any of the available evidence to be enough for anything like ambitious goals. Many developing countries are at 1 to 2 standard deviations behind the OECD (something like 4 to 5 “grade levels” behind at grade 9) and there is no articulated, evidence based plan in which inputs make up that difference. So I tried to position as “more and better”–but have to admit I put more emphasis on the “better” than the “more”–in part because there is so much organizational and political dynamic for “more and let’s worry about better later” out there already–but still never claimed the strong version (that many economists do) that more “doesn’t help.”

  14. Sam Gardner

    I fear pedagogues and educations specialists. I fear for them theoretical systems are more important than children. We remember the reading pedagogy debacle, where only after a decade or so of using the “whole word”system, someone took the effort to see wether it worked. It didn’t. Were these lessons learned?

    I agree, it is fantastic to be a kid in a starfish system with competent teachers, headmasters, parents, ministers of education. My kids were in an international school, and I wish everybody could.

    I fear we live in a system where teacher pay is not enough to get that quality. In the US, the starfish is rather crippled. Recentralising seems to be necessary. In France and most eastern systems, decent education outcomes are reached with central control, accepting the fact that teachers didn’t get their Phd yet, and they give them some good guidelines, to beat the insecurity when they face a full class.

    In Vietnam the PISA outcomes for the non-dropouts are on par with the results from the western countries. We try to introduce “escuela nueva” methodology. Is Columbia really so much better? I guess the children are happier.

    Should we move to a more starfish approach? probably, but with caution. Step by step. Probably faster for the investment in toilets than for the curriculum.

  15. Audrey

    as someone who works not only for Oxfam but also in a primary school, I must say that all this sounds way too complicated. Only two people mention the welfare of the kids. Surely that’s what education is about? Just like Oxfam’s awesome catchphrase, children should be at the heart of education. What we are aiming with educating children is giving them a chance to fulfill their potential, learn, socialize, dream, aspire and be themselves. This is a space for them to express and realize, not fit into some “educational system”. Sure, policies, curriculum etc are important, but one of my friends’ kid is home-schooled and is one of the happiest, brightest kid I’ve ever met. Go back to basics: support children, teachers and teaching staff and the community will follow.

  16. In Brazil the time goes very quickly and Government Propositions to Improve Education are imcompatible with reality of the country. A great deal of problems involving our educacional system of education as keep fake goals to reach without a serious political of resources to finance searchings in many areas of knowledge, the illiterate pupils and also lot of functionally illiterate that had already left the school to get job. Moreover we cannot see aims at showing to reafirm the importance of reading of literary books in the beginning of the first stages from elementar school, this activity,is the extremely importance to Children’s cognitive development , involving criativity, thought,language, perception, memory, reasoning, besides that, expanding the universe of knowledge of the Peoples , concepts formation, sociability, geographical world, above all, as supporting accting like preparation of student to acquire basic literacy and literacy. This preparation, made by specific activities inherent reading, provide to child, through the different themes and elements that make up the numerous works, building their impressions about the narratives which were submitted to it, his foreknowledge about the themes, their their previous experiences, your creating your own story based on your understanding and assimilation who can build on these fictions, based on the unreal and confrontation these, with facts and livings of its reality, representations and theatricality and their experiences. In 2014 the Brazilian Government announced a set of aggressive measurements that will be put until 2014 in order to reafirm a positive position to lead the education problems but at least until 2016 a few steps were given towards this aim. There’s no boost in the Brazilian educational system that can justify any government worry to change our performances, results or skills to reflect educational programs as PNE (Plano Nacional de Educação). Our poor system of education is the equivocated result of many influences proposed by many goverments and pedagogues and educations specialists that are so far our reality and necessities in this área. On the other hand teachers are constantly forgiven as part of the educational process, they need to improve their own qualifications and skills with the low salary they earn. While the government do not want to combat the true roots of problems that involve Brazilian education system we will continue lead with with class overbooking classes of pupils, day by day scenes of violence, profissionals without time to best prepare themselves because they need to work in three schools or more, without motivation program, no careear plan to get evolution and so on, what meaning that is “long and widen road” to go if Brazilian society (mainly the government) want change the ways of our education and future of million Brazilian that are coming.