ºÃÉ«ÏÈÉúTV

The government's policy on housing, by Richard Rogers via John Prescott, is to promote high-density development on brownfield land. Assuming we don't want to waste half our lives in traffic jams or on station platforms, this is fairly uncontroversial stuff. But anybody can have a policy. The devil's question is: can you implement it?

CABE's expert design review panel has just published a report examining this very question, based on some 20 large high-density urban housing projects that it has reviewed. The findings confirm that high density is difficult to get right, and suggest that the British housebuilding industry is getting it wrong.

The problem is that it requires meticulous, site-specific design to preserve the privacy of the cheek-by-jowl neighbours, sufficient sunlight for their herb window boxes, and enough – but not too much – tarmac to store their car. "Standardised solutions will often not be appropriate, and a high level of design skill is needed if such schemes are to become and to remain places where people want to live," writes Paul Finch, the design review panel's outgoing chairman.

In other words, if housebuilders deal with urban sites by packing their stock-in-trade suburban boxes and curly road layouts a bit more tightly together, then the resulting neighbourhoods will not work in the long run. To avoid this, they should bite the bullet and give freer rein to architects to purpose-design dwellings and layouts.

The CABE publication is not intended as a design guide, as it argues that "there is no simple recipe" that could be put in such a thing – and in any case, it is anxious not to constrain architectural ingenuity. Instead it presents a discussion of "the complex array of ingredients" involved.

Standardised solutions WILL not often result in schemes becoming places where people want to live

Paul Finch, chairman of CABE design review

One of the ingredients that is much discussed is the use of courtyard or part-courtyard forms, which the review notes are becoming commonplace in residential projects. It warns that "this form of building is not well-enough established nor sufficiently understood by designers in England to guarantee success as a matter of course". It emphasises this with the observation that "a grid pattern of courtyards, while providing enclosed private space, produces a series of buildings turned in on themselves, leaving streets without surveillance or pedestrian activity". It also warns that "a courtyard configuration can result in some habitable rooms receiving very little natural daylight and having little aspect, as well as being overlooked."

As to the subject of roads and car parking, the review makes its attitude clear from the start: the section is headed "Taming the car". And here, courtyards come in for yet more stick. "Proposals often contain courtyards dominated by parking for cars when streets might be a better place to keep them, allowing courtyards to become more private spaces." It adds: "One way of easing the pressure of car ownership is to ensure that residents can feel safe walking or cycling and have easy access to the public transport system."

The review rounds off by looking at wider social and management issues. It notes that high-density housing often appeals to working households, who abandon the neighbourhood during daylight hours. To avoid this, it recommends that a range of other uses should be included in larger developments, such as live–work units, offices, shops, restaurants and other community facilities, "not only for residents but also for those living in the surrounding neighbourhood."

Finally, and realistically, the review notes that good neighbourhoods do not end with good design. "In the past, many housing schemes have failed not because of poor design but through ineffective management." Accordingly, it recommends the establishment of a "robust management company".

All in all, the CABE review presents a well-informed and timely discussion of all the design challenges involved in high-density urban housing. Although it is hard to pick holes in any of its arguments, there is something incestuous about the dry, theoretical discussion, which will only be intelligible to architects and urban designers. It skirts daintily around the pitfalls that it acknowledges all too many housebuilders blunder into. There is little in the document that will appeal to recalcitrant housebuilders, little to open their eyes to new resident-friendly approaches to high-density urban development, and little that will encourage them to appoint architects who can deal with its challenges.