Good quality education for all children is denied when ghettoes are allowed to flourish. They may live cheek-by-jowl, but a huge and widening chasm divides Britain’s city dwellers - the “haves” and the “have-nots”.

This division between rich and poor caused Disraeli in the 19th century to observe “dwellers in different zones or inhabitants of different planets”. Today it is most apparent in urban schools where the children of the two tribes mix less and less. Truly comprehensive urban schools are getting harder to find.

What is the optimal kind of school we should be aiming for in the next millenium and why is the ideal so difficult, if not impossible, to sustain especially in our cities?

The optimal city school has to be “part of the fine city” which Tim Brighouse spoke of in his 1994 GreenwichTES lecture. “Make the global local,” is a convenient way of summing it up; the intercultural potential is one of the most exciting aspects of such schools. Already, we have new hybrid identities and cultures emerging; Glaswegian Chinese, Afro-Caribbean scouser, Sylheti cockney and so on. The DEMOS report, The Creative City, urged us to “build bridges beyond the ethnic fragments and produce something new out of the multicultural patchwork of our cities”.

In the same report, the powerful idea of the “learning city” is advanced with the reminder that,“in the inter-urban competition game, being a knowledge-intensive organisationIhas acquired a new strategic importance”. The bedrock on which such an urban vision rests is a school system which gives children and young people opportunities to contribute to and enjoy their city. Such a system needs to include a diverse range of learning strategies (for example, distance learning, project and fieldwork) with built-in guarantees of access and progression.

The optimal school also has to be a place where most human beings, child or adult, would willingly spend time- a decent physical environment. The comparisons might be a fast food outlet or supermarket, for these are the places which most families have come to expect as normal public environments. In the National Commission on Education’s study of 11 effective schools - published last Autumn as Success Against the Odds - the importance of the school’s physical environment was constantly emphasised. The quality of working conditions, for teachers and learners, is critically important, materially and symbolically.

The pathetic poverty of our aspirations is made clear when we consider a recent report on American elementary schools, commissioned by the business sector in California. It points out that: * Classroooms have inadequate telecommunications, electrical connections, lighting and space which seriously hinders the introduction of new curricula and teaching methods that require computers, television screens and flexible groupings of children. “Only one school has a ‘phone in every classroom. Without a telephone line into the classroom, children are not going to be able to receive instruction or curriculum that moves on these lines.”

* The absence of adequate storage space affects the security and safety of computers, videos and other equipment. But children’s learning is also affected because new methods of assessment require the maintenance of extensive portfolios for each child and space in which to keep student projects.

* No physical space has been planned for staff development: “None of the spaces teachers now gather in - the cafeteria, student library, or teachers’ lounge - is designed to enable teachers to work on computers, view televised instruction, or watch a video. Few are large enough for whole staff instruction and most lack the privacy for small-group instruction.”

These, then, are fundamental inadequacies in schools observed by the business community in California. Compare its aspirations with the meagre, shoddy conditions of most British schools.

The optimal urban school should also be what the Americans call a “full service” school or what we might call a community school. In Florida, the state authorities argue that schools are appropriate locations for providing a wide range of services. People are more likely to know where schools are and they often have surplus space; “a ‘one-stop shop’ fosters communication sbetween agencies, can reduce duplication, and saves parents’ time”.

This surely makes sense. The Prince’s Trust, in its successful Millennium bid for homework and study centres, should take note. Extending and refurbishing existing local centres - whether schools, colleges or, perhaps, public libraries - must be better, in most cases, than building new?

The optimal school, whether at primary or secondary level, should also be socially mixed for two reasons. First, we must take seriously the call for greater social cohesion and avoid a Blade Runner scenario of alienation and violence in our cities. It makes sense if our children and young people learn to live and work together while at school. Tony Blair has emphasised the importance of achieving “a sense of communal belonging”, saying: “It is only by working together in a community of people that the individual’s interests can be advanced.” Our seeming addiction to a class system certainly gets in the way of anything that seeks to be inclusive (seen as low status), rather than exclusive (high status).

Second, a true social mix helps improve pupil performance, particularly of the less able and less highly motivated. Research from Edinburgh University suggests that where there is a significant corps of more motivated and demanding pupils (and parents), teachers have higher expectations and peer pressure has a direct, positive influence. This confirms what some of us have sensed and observed in practice.

Finally, the optimal school should, by implication, be a local resource. In some cases, this might mean that it lacks diversity because there are urban neighbourhoods which are homogeneous, but a policy decision is then needed which pays regard to the practical possibilities and to what the users of the service think is desirable.

The educational benefits of local schools include easier contact between home and school, and better liaison between primary and secondary; both OFSTED and the Teachers Training Agency have said liaison needs to be improved.

What we must curb are the long, expensive and sometimes dangerous journeys to school which children make these days. Not simply journeys which are chosen by parents, but increasingly those which are forced on children who can’t get a school place nearer to home. We need guaranteed, local school places -something which should give the diocesan authorities food for thought given their “gatekeeper” role in school admissions.

Of course, there are many optimal comprehensive schools in England, Wales and Scotland, well supported by their local communities and successfully promoting high levels of achievement for children from all social classes - whether academically, socially, on the sports field, in the orchestra or through work experience in local businesses. What most of these schools lack, though, is a rich cultural and ethnic mix. Nor do they have the city’s concentration of art galleries, theatres, hiistoric buildings, universities and leading edge businesses on their doorsteps.

The best teachers will want to work in such schools. Indeed, many already do. But currently there are few optimal schools in Britain’s major cities - and fewer now than 10 or 20 years ago. Before we make pronouncements about who is to blame, the city’s demographic and economic context needs to be properly understood.

The increased division between rich and poor over the past 15 to 20 years, is well documented and accepted. What seems to be less well understood or debated is that the most extreme manifestation of this division is found in cities - and that it affects children, especially. One in four children now lives in a household on income support and poor families are increasingly concentrated in particular urban areas.

Between 1981 and 1991 poverty increased markedly in inner London and other large metropolitan areas. Meanwhile, the better off neighbourhoods moved sharply in the other direction, with increases in owner-occupation, two-car ownership, high economic activity and low levels of unemployment.

In inner London the extremes are growing; most boroughs now have markedly higher proportions of social classes I and II (professional and managerial) and of unemployed people than the average for either Greater London or the UK. This means that increasing numbers of rich, well-educated residents live cheek-by-jowl with the poor and ill-educated.

Simultaneously, the proportion of households with no economically active adult and with dependent children has risen, having more than doubled in all inner London boroughs. In Greenwich, for example, the proportion has risen from 6 to 17 per cent in a decade. All 10 boroughs are in the list of England’s 30 most deprived areas, alongside Birmingham (fifth most deprived), Liverpool (sixth), and Sandwell (ninth).

How, then, does all this relate to education and to schools, in particular? Sir Peter Newsam has described how secondary schools in the inner-London area have increasingly moved beyond a simple first and second-class division to a much more hierarchical and polarised system (TES, March 22). In his analysis, more than three-quarters of children in the top 25 per cent of the ability range are now in half the available schools, including selective independent schools, which represent a third of this favoured half. The other half comprises schools which contain very few and often no pupils who are in the top 25 per cent of the ability range.

As Sir Peter comments: “such schools are no more failed comprehensives than a Brussels sprout is a failed cabbage.”

At one end of a finely-graded schools’ spectrum is the boys’ “comprehensive” which enrols pupils at 11, few of whom have a reading age near their chronological age, many of whom have been unable to gain admittance to the schools nearest to their home and most of whom are black. Because such a school is the witless victim of its own successful accommodation of such pupils, it is highly probable that throughout the school year there are further enrolments - of the recently-arrived homeless, of immigrants and of those excluded from other schools. At 16, more than 20 per cent of the pupils have achieved no GCSE grades at all, compared to a national average of 8 per cent.

At the other end of the spectrum is the mixed comprehensive which has opted out of its local education authority and made use of its grant-maintained status so it can select up to 50 per cent of its intake at 11. On the basis of a series of tests (in verbal reasoning, non-verbal reasoning and maths), which are skewed towards identifying the top 10 per cent of the age cohort (not simply the top 25 per cent), the school is already “super-selective”. The other half of the intake is drawn from the school’s mainly middle-class catchment area, with priority given to siblings of present pupils and to the children of staff. In the case I am thinking of, this category was also over-subscribed; overall, there were 1,200 applications for 250 places.

It is not surprising that in this school, even before the full impact of its super-selectivity has hit Year 11, its 1995 GCSE results were more than eight times better than those of the other school I described. Yet both are called “comprehensive”.

Whether by virtue of selection on the basis of religion, musical ability or more conventional measures of academic attainment, an increasing proportion of inner London’s comprehensive elite is becoming more elitist.

Most of this increasing selectivity has the effect of keeping the (mainly white) middle-class together and corralling the rest into schools which are confronted with a range of educational challenges, most of which arise directly from intense social and economic disadvantage. School improvement strategies are certainly needed in these schools, and no one should scoff at the impact or relevance of these, even if the initial challenge is to raise pupil attendance from an awful 75 to a less awful 80 per cent and to increase the percentage of Year 11 students gaining five GCSEs at grades A to C from 6 to 15 per cent, compared to a national average of 43 per cent and a national target of 85 per cent by age 19.

But my main point is that the optimal urban school is not represented in either of the two extremes I have described. In neither case are the pupils being offered the best, fairest or most challenging kind of education - and certainly not one designed for the 21st Century.

What is to be done?

* Let’s not mess with success. Where there are schools which meet my criteria for optimal schooling, leave them alone. But any moves towards greater exclusiveness need to be resisted and stopped.

* Let’s consider giving a financial incentive to schools which primarily serve their local communities. Not a large amount but, simply at the margins, an additional element in each school’s budget which acknowledges that it is successfully serving the families who live closest to the school.

* Invite local consortia to put forward proposals for co-ordinated 14 to 19 provision. Such consortia could include secondary schools - local authority, voluntary-aided, grant-maintained and independent - further education and city technology colleges. Local businesses, charities and universities might want to associate themselves with such consortia; local authorities and diocesan bodies might well be leading players.

Each consortium would work out what kind of arrangements would best serve the needs of all the young people in their locality. “Junior colleges” might well emerge, but the Secretary of State would need to ensure that certain basic criteria were met, such as increased post-16 access and participation. Successful proposals would be awarded specific capital and revenue grants.

* We should see if the most accomplished and skilled of our teachers and heads - especially those in their 30s and 40s, often with their own children - would be attracted to either stay or come to work in urban schools on five or eight-year contracts, which also included the provision of good quality housing.

* We need a well articulated family policy which links several government departments, at both national and local authority level: education, health, social services, environment, heritage, to name but five. A Cabinet minister is needed at senior level and likewise, at local or regional level, a co-ordinating manager at deputy chief executive level, reporting to an elected board.

* Local government itself needs to be strengthened, especially in London where strategic vision, co-ordination and powers are needed - expertise and goodwill already exist by the bucketful. Which government will be brave enough to tackle this?

Beyond the structures are individuals who make personal decisions. We live in what has been called the “decency decade”, or the “nicer Nineties”. If this is so, then perhaps the larger critical mass of more advantaged, middle-class parents will do as many of their predecessors did in the 1970s and “hold hands” with each other and others who are not so privileged? We need them to participate in our schools. We also need the participation of the fee-paying, independent schools.

None of our children will thank us if we insist on splitting and wrecking our cities. The emergence of what Tony Blair has described as a “growing anxious class” will hasten and deepen unless we develop urban schools which are inclusive, not exclusive, and which celebrate rather than seek to deny the realities and immense potential of our cities.

Margaret Maden is Director of the Centre for Successful Schools at Keele University and former chief education officer for Warwickshire. A full copy of the lecture titled “Divided cities: dwellers in different zones, inhabitants of different planets” is available from TES Promotions, Admiral House, 66-68 East Smithfield, London E1 9XY. Please enclose a self-addressed label and a cheque for Pounds 5 made payable to The TES