Type a keyword and hit enter to start searching. Press Esc to cancel.
Ross Atkin Associates
Research, design and development

Let’s stop talking about ‘shared space’

Let’s stop talking about ‘shared space’

Note: I wrote this at the start of February this year and circulated it to some people for review but didn’t get around to publishing it. Since then John Dales has referred to it in a piece he wrote more recently so I thought I’d better publish.

Last week I went to one of Urban Design London’s ‘challenging practice’ session to help out with a discussion on ‘shared space’ being run by John Dales. At the end of the discussion John put a question to the forty or so assembled street designers and engineers – ‘Do you feel that shared space is a useful term?’. Remarkably not one of them raised their hand.

A month earlier I had sat in the Houses of Parliament at a meeting convened by Lord Holmes of Richmond to discuss the report he had published in July and his call for a moratorium on new shared space schemes in the UK. In that meeting I made the point that we might make more progress on dealing with the issues caused by inconsiderate urban design for blind and partially sighted people if instead of focusing on shared space we broke it down into a set of design features and explained how they impact different people (I’ll come onto these later).

Designers and campaigners mean different things when they say ‘shared space’

Between the two meetings (as well as the events we ran as part of the DfTs consultation on changes to tactile paving guidance over the autumn) I realised that the sight loss and urban design communities mean different things when they say ‘shared space’.

You can start to see this by comparing the definition of shared space preferred Stuart Reid the author of the DfT’s guidance on the subject:

‘Shared Space: a street or place accessible to both pedestrians and vehicles that is designed to enable pedestrians to move more freely by reducing traffic management features that tend to encourage users of vehicles to assume priority.’

and that used by Lord Holmes in the open Survey Monkey data collection that formed the evidence base for his report:

‘A shared space is a street or area where people and traffic are not clearly separated. The area might have level surfaces, no pavements and kerbs and road crossings without traffic signals or zebra markings.’

The definitions may sound similar but their differences are significant.

The former definition by using ‘street or place’ and focusing on the objective implies a scheme sized intervention. When designers talk about shared space schemes this is what they usually mean – an entire street, or at least a major junction. ‘Place’ is important because it refers to ‘placemaking’ which is often a major objective for a shared space scheme (re-balancing a street in favour of the people spending time in it at the expense of those moving through it).

Lord Holmes’s definition by exchanging ‘place’ for ‘area’ and focusing on specific design features (or their absence) can be interpreted to mean something much smaller – a specific crossing point or a side road entry treatment for example, or just somewhere where a particular feature (like a controlled crossing or guard rail) is not present.

Inspection of the list of locations to which respondents to Lord Holmes’s survey confirms that this is indeed how it has been interpreted. As would be expected, it includes major shared space schemes like New Road Brighton, Hackbridge and Exhibition Road, but also sites like Judd St, and Rivington Street in London which are conventional streets that happen not to have controlled crossings; or Earls Court Road where guardrail has been removed but otherwise a conventional street layout with signal controlled crossings and kerbs remains (sorry these are all in London but I’m picking streets I know). It also includes some 1980s level-surface pedestrianised or partially pedestrianised areas like Queen Street Oxford, Fitzroy Street Cambridge and York town centre.

This distinction is important because we could have a complete moratorium on shared space tomorrow and problematic features would continue to pop up all over the place. Designers would continue to implement features, like undelineated level surfaces at raised side road entry treatments or the removal of signal controlled crossings, that cause difficulties for blind and partially sighted people, because they don’t identify them as being anything to do with shared space.

One feature at a time

As an alternative to talking about shared space I would recommend we talk instead about the following four design features (which are often combined in a shared space scheme) and how they impact different disabled people. By doing this we will be able to constructively address the problematic parts of shared space schemes, as well as dealing with these features when they crop up not as part of a shared space.

Removal of Signal Controlled Crossings

Of the features associated with shared space this has by far the greatest impact on the greatest number of disabled people. In theory removal of a signal controlled crossing in a shared space that is ‘working properly’ should advantage the ‘average pedestrian’ as they should be able to cross whenever they want rather than having to wait for the green man. Even where this seems to be happening (at Poynton for example) we must discriminate between the average pedestrian and more vulnerable pedestrians who are likely to feel less confident establishing priority over vehicles, for all kinds of reasons. The two million blind and partially sighted people in the UK will be particularly affected because of their reduced ability both to detect approaching traffic and to read the intentions of drivers and cyclists. For many of these people confidence that the traffic is being held at a red light can be extremely valuable.

If a moratorium is to be called on something it would be most usefully be called on the conversion of signal controlled to uncontrolled or ‘courtesy’ crossings. This feature seems to correlate well with the less successful shared space schemes, like at Ashford or Exhibition Road where attempts have been made to ‘tame’ very traffic dominated environments but where vehicle volumes and speeds remain very high (presumably the reason they had signal controlled crossings in the first place). The more successful shared space schemes like Leonard Circus or Seven Dials did not involve the removal of signal controlled crossings (presumably because they were already somewhat pedestrian dominated spaces before they became shared spaces).

As well as putting a brake on shared space schemes whose ambition in terms of vehicle speed reduction is likely to be higher than they will deliver, a moratorium on signal controlled crossing removal would prevent these being removed as part of non shared space schemes.

Some clarity from campaigners that ‘signal controlled crossings are good for disabled people’ would help to counteract the libertarian argument against traffic signals which runs much wider than shared space.

Level Surfaces

To date most of the campaigning against shared space from the vision impairment sector has focused on level surfaces. Guide Dogs first campaign was called ‘Stop shared surfaces, keep our pavements’, which was then followed by ‘Say NO to Shared Streets; If people can’t tell where the road begins and the pavement ends- how can you feel safe?’. The RNIB’s current ‘On My Street’ campaign begins ‘Shared space or shared surfaces is a new concept in urban street design in which street signs, road marking and pavements are eliminated in favour of a space which is shared by pedestrians, cyclists and vehicles’.

There are a few issues with this focus. Most obviously it diminishes the the much more important issue of signal controlled crossings. Additionally, by tying the level surface issue to shared spaces, it removes attention from all of the problematic level surfaces (like those raised side road entry treatments) that are not part of shared space schemes. It also fails to recognise that some shared space schemes, like Leonard Circus, retain kerbs to demarcate a safe space where vulnerable pedestrians are not forced to share with vehicles.

Additionally it prevents sensible discussion about what can and should be done about all of these other level surfaces. Some participants in the tactile consultation workshops seemed absolutely convinced that the introduction an evidence-based requirement for dangerous undemarcated level surfaces to be properly marked with tactile paving would somehow act to encourage more shared space, when in fact the reverse was expected by design professionals.

Finally, unlike the signal controlled crossing situation, the level surface one is not a clear cut ‘bad for disabled people’ or even ‘bad for blind and partially sighted people’ issue. The vast majority of the two million blind and partially sighted people in the UK do not use a mobility aid (like a long cane or guide dog) relying instead primarily on their residual vision to navigate. For these people a kerb is not a particularly helpful design feature. They are much more likely to be using the tonal contrast between road and pavement, or the yellow lines to determine where the road is. Indeed for some of them the kerb functions only as something else to trip over.

This is absolutely not to understate the negative impact that level surfaces (particularly ones without delineation) have on blind people who do rely primarily on guide dogs and long canes (4,500 guide dog users plus probably something of the order of 100,000 long cane users). Just to point out that the argument ‘level surfaces are bad for two million blind and partially sighted people’ would be unlikely to turn out to be true, were rigorous research conducted.

Removal of tonal contrast between road and pavement

As explained above, for the majority of blind and partially sighted pedestrians who are relying primarily on their residual vision, removal of the the tonal contrast between the pavement and road can have a much more detrimental impact than removal of the kerb. Additionally the introduction of meaningless tonal contrast (such as the criss-cross bands of darker paving on Exhibition Road or the concentric rings in Bexley) can cause them to have to continually check for a level change making navigating such spaces slow and confusing.

More awareness amongst designers of how people with low vision use tonal contrast to navigate could make a big impact on public space accessibility generally as well as shared spaces specifically. The freer use of tone afforded by shared space schemes could provide opportunities to provide more useful tonal contrast to pedestrians with low vision rather than less.


The final important feature associated with shared space that affects blind and partially sighted people is the removal of street furniture, particularly guard rail around crossing points. Like level surfaces this is a change that is common at many sites that the designers would not consider shared space. In many local authorities it is now standard practice.

For all of those blind and partially sighted pedestrians who are relying primarily on their residual vision decluttering is usually considered a good thing, removing sources of trips and collisions. However for some long cane users items of street furniture can serve as vital navigational waypoints, with runs of guardrail surrounding crossing points serving as very useful ways of locating the crossing point. Where guardrail is removed from a junction that has been built to Figure 17 of the tactile paving guidance this exposes an undelineated level surface that can easily allow a long cane user to unwittingly enter the road. Because of this some attendees at the tactile consultation workshops felt very strongly that guard rail should not be removed from streets they regularly use.

The effect of decluttering on long cane users is not well understood by designers which may be why steps are rarely taken to mitigate its effects. Within the current climate campaigning to reinstate guardrail on relatively quiet streets is unlikely to prove successful. A strategy that formalises the mitigation strategy (such as a standard for the use of tactile paving or maintenance of a detectable level change – as was proposed in last year’s DfT tactile paving consultation) would be likely to have a more positive impact.