Wednesday, 22 May 2013

Lessons from Copenhagen

As I mentioned last post, I was invited to go on a study tour of the cycling infrastructure of Copenhagen, representing the Cycling Embassy of Great Britain. The tour took place on 8–9 May, and was very well-organised by Phil Jones and our Danish partners, Henriette Lund of Transport for London, Niels Jensen of Copenhagen Municipality, consultant Niels Hoe, and Lene Hartman, of Furesø Commune. Their contributions made it a really informative trip, though of course, you learn most on a trip like this from not hearing what people say, but from trying and testing the infrastructure yourself, on a bike, and observing how everybody else uses the infrastructure, and how everything works. When you do this you gain a physical appreciation of how a system of traffic flows, of how all the environmental elements, the design and habits and practices of the users, interact.

There is really no substitute for this individual, personal experience. The same appreciation cannot be gained from looking at photos, reading statistics or graphs, or even watching videos. As I commented after the Dutch study tour with David Hembrow, cycling infrastructure, and how it creates the experience of cycling in a particular place, is not an "underrstood" thing, fundamentally, it is a thing that is felt. This is what Hembrow means when he talks so much about subjective safety. Subjective safety is difficult, if not impossible, to quantify, and therefore it is a problematic factor for engineers to incorporate into their designs. But the differences in the cycling experience in various environments very largely come down to this concept of subjective safety, and the quality of the experience determines how many people actually cycle.

Of course through writing a blog I can't allow you to feel how it feels to cycle in Copenhagen, Groningen or Münster, you have to go there yourself. I'm glad that such a good representative selection of transport professionals from the UK attended this trip and were able to experiencve what it feels like to cycle in Copenhagen, and were able to observe all the features that go together to make up the Danish cycling experience, and to draw their own conclusions. We had officers from TfL, from the Boroughs of Lambeth and Waltham Forest, from Bournemouth, and the cities of Cardiff and Birmingham, as well as independent consultants. One engineer from Birmingham told me he had not cycled for many years. but he gamely got on his hire bike, with all the rest, and cycled, in the end, maybe 40km around the Copenhagen area, over two days.

So I can't really explain cycling in Copenhagen fully to you in a blog, but I can show you my pictures and videos, and explain some of the features of the environment that appeared to me to be important, and relate other facts about the history, development and culture of cycling in Copenhagen, as they were related to us by our Danish guides.

There's certainly always been a strong cycling culture in Denmark (as long as there have been bikes), cycling certainly never became so marginalised as it did in the UK, and progress towards the current infrastructure situation in Copenhagen has been continuous over a very long period, for more than a century, in fact.
Km of cycle track over time in Copenhagen, courtesy Niels Jensen, Copenhagen Municipality
But this graph, showing a pretty linear trend in the mileage of cycle tracks provided in the city over time, does not represent the history very well, according to our Danish hosts. Up until the 1970s, Denmark suffered from the same history as other western nations, of trying to accommodate more and more cars on city streets, allowing cycling to be squeezed out, and providing an increasingly poor environment for cyclists, particularly on major roads. This triggered a protest movement, as it had slightly earlier in the Netherlands, with the Dutch Stop de Kindermoord campaign, with huge demonstrations, such as this one, below, shown in this photo of Copenhagen City Hall, around 1980. These initially influencved politicians, and later engineers and planners. The extent to which cycling as a mode of transport is actually prioritised on the streets of Copenhagen today, with cycle tracks in all the places they are needed, traffic exclusion and reduction where it is appropriate, and all the surrounding measures, is directly attributable to this movement.

Cyclists' demo at Copenhagen Rathaus, about 1980
So what we see today in Copenhagen and environs is a pretty comprehensive network of cycle tracks and paths that create a high degree of subjective safety. There tends to be quite a lot of national chauvinism around cycle facility provision, with some wishing to denigrate the progress made in other places. Without entering into that, I can still attempt to make an assessment of the quality of the Danish cycle infrastructure as objectively as I can. It is clearly more advanced than seen in any country outside the Netherlands. Its strength is its comprehensive character and consistency of design, whereby road users of all types know what to expect, and know what is expected of them (contrasting so strongly with the confusion and inconsistency that surrounds providing for cycling in the UK). It is more similar to the infrastructure I wrote about in Münster, Germany, than it is to the Dutch model, but it shares much with the Dutch model. Build and maintenance quality on average tend to be slightly lower than that seen in the Netherlands, the absolute proportion of space that has been allocated exclusively to cycling is lower, and the treatment of junctions, relying more on road users' respect for priority rules, and less on full separation of flows through signalling and segregation, is like the German model rather than the Dutch. But the sense of subjective safety, and, another important and related term I have found I have need for a lot in talking about Danish cycling, fun, is very high.

If I need to enumerate the main elements that go towards this achievement of the sense of cycling safety and fun on the streets of Copenhagen, they are:
  1. Segregation on almost all busy roads (I estimate 90–95%)
  2. Design of car parking to protect segregated cycle tracks
  3. Simple junctions, nearly always simple signalised cross-roads, with consistent methods of working which are expected and understood
  4. Roundabouts with cycle tracks on which cyclists are given priority, in the Dutch manner, with surfacing that supports correct behaviour
  5. General separation of cyclists from buses (again at least 90% of the time)
  6. Invariable provision for cyclists to protect them from the effects of road works (even if it means considerable space compromise for all categories of road-user).
The standard Copenhagen one-way cycle track: Tarmac, 1.5–2m wide, drop from pavement about 5cm, drop to road about 5cm, protected by car parking: the "roast beef" (or Danish bacon) of Copenhagen cycle provision
Quite a famous view, the crossroads leading on to Nørrobrogade bridge: simple, standard Danish junction design, the blue is only marked across junctions, not on the segregated cycle tracks
Roadies enjoying the priority they are given on the cycle track on a roundabout in the suburban commune of Furesø. Note how the smooth surface for bikes and rough surface for cars, zebra, and sharp geometry all reinforce priority.
Bus stop with cycle track running behind, Vesterbrogade, in the city centre. Note use of the track by a mobility scooter.
Near the central station, works for construction of a metro line have restricted space, but cyclists still have a protected track, though it's not working perfectly, with pedestrians straying into it. But users seemed to get along with this kind of thing in a good-natured fashion.
On top of these factors, there's a certain amount of "fluff", if I may disrespectfully put it like that, which tends to attract a lot of attention from foreigners, often abetted by the Danes themselves, with their love of novelty design, which I don't see as very significant. In this category may come the green wave: little lights installed in a few cycle tracks that are supposed to help you to ride at a speed where you will sail through all the junctions on a route on green signals; train carriages with huge bikes painted on them; and the famous counter on Nørrebrogade bridge that thanks you for being the 2541st cyclist (or whatever it is) on the bridge that day.

I think it's fairly important to disentangle what is the meat of Danish cycle provision from the Danish pastry, as not to do can lead to confusion abroad. It was indeed peculiar that the recent, highly marginal, innovation of the green wave was mentioned in some of Transport for London's early concept material relating to the Barclays Cycle Superhighways as being a possible intervention for London, when they were not proposing that the Superhighways were to be Danish-style segregated cycle tracks, but merely areas of bus lanes painted blue! This was really getting stuff crazily upside-down, as the green wave depends on the existence of the protected cycle tracks, as does every other aspect of the Copenhagen cycling experience, at bottom. This was like discussing what colour you are going to paint a house that you have no intention of building.

The green wave in operation: it's those tiny lights in the left of the track
You win the raffle! The famous cycle counter on the Nørrebrogade bridge. 
In the photo above, the real thing to consider is not the much-discussed counter, but the width of the cycle tracks. The space on the bridge has recently been reconstructed, like that on London's Blackfriars Bridge. But of course the Danes have done things very differently to TfL. They had cycle tracks already, and they had two lanes of motor traffic in both directions. Recognising the modal shift that their policies had supported, they realised more pace was needed for bikes, and less for cars. They reduced the carriageway to one lane in each direction, and expanded the cycle tracks to 5m width each. Both are mono-directional (as is the Danish norm, except for cycle paths disconnected from roads), and the small chevrons are supposed to indicate a two-lane arrangement, to allow faster cyclists to overtake slower ones.

I've not mentioned filtered permeability in the list above. This is because, though there is quite a bit of filtered permeability in the centre of Copenhagen, particularly in the mediaeval street layout between the Tivoli gardens and the castle, we, the UK group, did not feel this was particularly critical to the operation of the cycle network, though it does make some contribution. Most of the time, the shortest and most convenient routes to places one needs to get to one finds are on main roads with cycle tracks on them. This seems to me to give us a very powerful clue as to what we will need in London if we are to ever reproduce anything like the Danish cycling system here. As I say, at least 90% of the roads with significant traffic on had cycle tracks. The tracks could sometimes have been wider, ideally, but they were nearly always there, as expected, where they were needed. And the city is planning on installing them on the few main roads where they do not currently exist.

The motor traffic restriction measures that were in place in the densest, oldest areas of the municipality seemed to be more opportunist than part of strategic cycle network planning, and more to do with civilising the place generally, and making space for pedestrians. I did cycle through these streets, and I did find that they were generally not the most efficient routes on which to navigate the city.

Vestergade by the City Hall is pedestrianised 4am–11pm, and that means no cycling, for obvious reasons – it's too busy. There are good parallel cycle routes however.
Street off Vester Volgarde, with a form of filtered permeability: no-entry to motors from this direction, but two-way for cyclists, with painted lanes
Nice expensive shared-space treatment in Landemærket, by the famous Round Tower, part of a one-way arrangement that eliminates through motor traffic. But this isn't so much for the benefit of cyclists. The area is full of pedestrians, and fast cycle journeys are best made on the tracks on the bigger roads.
The Danish planners, as is pretty well-known, have had a policy since the 1980s of quite consistently putting in cycle facilities on the routes that the discovered cyclists were actually already using. This means they put them on the main, direct routes. Experiments with trying to direct cyclists on to the back-strteet network, using filtered permeability solutions, failed to work, for the same reason they would fail to work in London, or in most other old cities: those are not the direct, efficient routes to the places that people actually need to go, in the majority of cases. Following this discovery, the modern, successful policy continued, which is still underway.

Comparison with Dutch cycling is inevitable. Perhaps the most obvious difference in the infrastructure concerns junctions. Typically, at junctions where all movements are allowed, the semi-raised Danish cycle tracks drop to carriageway level shortly before a junction and merge with the right-turn lane for motor vehicles. This means that if you get to the junction when traffic is flowing, you will be interacting with moving motor vehicles potentially crossing your path. The rule is that all turning traffic must give way to all non-turning traffic, and this includes pedestrians. So if there are pedestrains crossing on a green phase which is simultaneous with the green for bikes and cars going forwards, turning bikes and cars must give way. Turning cars must also give way to bikes going straight across. Left-turning bikes (doing the equivalent of a UK right turn) do not pull out of the cycle track/lane and do not change lane. The proceed as if going straight across the junction, and at the far corner they come to a standstill ahead of the crosswise traffic being held at a red signal, and do a sort of turn on the spot (sometimes known as the "jug-handle" move, though it is often not executed in a jug-handle shape), so as to face the red signal in the direction in which they need to go. They then wait for the green, whereupon they move off in front of the motor traffic going in that direction, and join the cycle track on the road into which they have turned. This is known as doing the "Copenhagen left" ("Copenhagen right" in UK terms).

The merging of the cycle track with the right-turn lane in H C Andersens Boulevard in the city centre.
If cyclists arrive at the junction when their direction of traffic is held at red, and there are cars in the combined lane waiting to turn right, the cyclists will probably position themselves to the left of those cars, to make it easier to go straight on, or do the two-stage left turn. If they cannot get to the left of the cars, and must stay behind them, then they may get obstructed by the cars being unable to turn, because of pedestrians crossing. Right-turning cyclists can generally go at any stage of the lights, provided they give way to pedestrians, as they just continue on the tracks round the corner.

This junction system has considerable merit in terms of simplicity and comprehensibility. I did not see any close-calls at junctions (except at one junction which clearly had not been designed to acommodate the moves that cyclists wished to make), and I saw total compliance by cyclists with the two-stage left procedure. On the other hand, these junctions are not so subjectively safe as typical Dutch junctions. Dutch junctions use the two-stage left as well, but they build-in a protected waiting area. The standard Dutch junction resembles more a signalised roundabout for cyclists superimposed on a crossroads for cars, whereas the Danish junction just has the crossroads, usually marked in blue paint on the road like a big hopscotch grid. There is no formal waiting area and no protection for left-turning cyclists, they just accumulate on the corner, as the video below shows. The mixing of straight-on cyclists with right-turning motors reduces the sense of subjective safety, a situation which the Dutch avoid with separate signal phases. On the other hand, the relative lack of HGVs moving about during the daytime, compared to British cities, mean that dangerous interactions on corners with vehicles with poor driver visibility are unlikely.

Large Copenhagen junction with many cyclists from the right doing a two-stage left. Most junctions have cycle lanes marked across them in blue, but this one does not.

Dutch junction design is explained in this excellent video by BicycleDutch, which you may well have seen before.

It is seen the the Dutch employ an absolutely consistent principle of keeping cycling to the right of the motor flow, whereas the Danes do not. That being said, there are less potential conflicts at Danish junctions than you might expect. One reason must be the generally much lower level of motor traffic, in comparable urban circumstances, as compared with a British city, as so many people are on bikes, and so many goods are moved around by cargo bike. Another reason is that many turns across cycle routes are banned to motor vehicles, increasingly so. Another is that the Danes, at very busy junctions, are now increasingly adopting the Dutch principle, and providing a green cycle signal phase (using small, eye-level traffic lights), separate from the motor green phase. One thing the Danes do not do is to use UK/US-style advanced stop boxes, except on very low-traffic streets. They do not aim to put cyclists gratuitously in the way of motor vehicles, and the observance of the two-stage left turn is part of this culture.

This suburban junction, though still lacking Dutch-style protecting islands, does not merge cyclists with right-turning motors. I am not sure if the cycle phase is separated or not (there are small cycle signals).
Another difference between Copenhagen and certainly the better Dutch cities, I would say, is that overall, the amount of space allocated to cyclists is rather lower. Since the provision is still very comprehensive, this manifests itself mainly as cycle tracks that are on the narrow side: often only 1.5m wide. Whereas cycling in the UK you are put constantly on-edge by close passes on your right by motor vehicles, in Copenhagen, if you are cycling slowly, you are rather put on-edge by the constant passes of faster cyclists on your left. They are experts at overtaking right on the edge of the narrow raised tracks, leaving literally millimeters to spare. Having said that, I prefer the Danish problem to the British. Not all the tracks are too narrow, some are fantastically generous, but it is an issue. It arises from the fact that the municipality has fitted the tracks in even in quite narrow two-way motor streets, combining them also with bus routes (the cycle track always on the inside of the bus route, with no crossing over), where the Dutch would certainly have adopted a different solution: either removing one direction of motor traffic entirely, or creating an autolowe or nearly car-free street, and giving the whole carriageway, effectively, to bikes. In other words, the Danes seem to practice separation of cycle and car traffic at a route level (as opposed to at a street level) much less than the Dutch. This reflects in turn some unwillingness, I would assert, on the part of Copenhagen Municipality to really tackle the volume of cars allowed into the city centre. The city centre certainly felt moderately polluted to me, though nothing like London.

One feature of the narrow cycle tracks is that they can get congested when a cyclist wishes to turn left (the equivalent of our right) into a minor road. If this is at an unsignalised junction with the minor road, the latter having no cycle tracks, the two-stage left does not really apply. Instead, there will be a bit of tarmac at the edge of the raised cycle track, opposite the turning, to create a ramp down to the carriageway, and the truning cyclist must wait in this area for a gap in both directions of motor traffic in order to make the turn. They could move onto the carriageway to wait, but this might cause problems on a narrow road with a cargo bike. On the other hand, if the cargo bike remains on the track, it will be in the way of cyclists behind. However, it seemed that a deal of good-natured give-and-take typically smoothed out such situations.

The standard half pavement height cycle track is not the only type seen in Copenhagen. I quite liked this street in Fredericksberg, below, where both the edge of the track, and the meridian strip of the road (which prevents some potential turns across the track), have been executed using lines of low, rough bricks.

Gamel Kongevj in Fredericksberg: track at carriageway level with minimal segregation
In this case, when it passes a minor road entrance (across which the pavement is continuous, emphasising pedestrian priority), the track does not change level, but becomes a cycle lane marked with a dotted white line.

The same route passing a minor road junction
Bus disembarkation on this road was not on to an island in front of the cycle track, as there was no space for that, but onto the cycle track itself. I show this example mainly because it strikes me as a good parallel to the sort of not too-wide main road in London that we are often faced with, where people say there is no room for cycle tracks, and certainly no room for floating bus stops. The Danes really do put protected space for bikes first. If there is a major flow of bikes on a road, they give them their own space. There's no such thing for them as a motor through-road that is "too narrow for cycle tracks". There's a lesson there, if we really wish to prioritise cycling as the Danes do. As some representative from Denmark (I forget who) once said, in response British politicians', or planners', claims that our roads were "too narrow for cycle tracks", "The lack of space is in your heads". Of course, some roads really are too narrow for cycle tracks. The solution in these cases is simple: remove the motor traffic, except for essential access.

Bus disembarkation/embarcation on Gamel Kongevj
A consequence of this prioritisation of bike space is that the pavements are often quite narrow, narrower than we might expect on comparable British streets. But this works because the kind of distances that are done as longish walks, or as very short car or bus journeys, by people in the UK, journeys of the order of 0.5–1.0 km, that most UK cyclists probably wouldn't bother unlocking and re-locking their bikes for, tend to be done by bike in Copenhagen. There's actually less need for pavement space, and transport for children and the disabled is well catered-for on the cycle tracks one way or another, either utilising electric, or human-powered, or hybrid vehicles.

I've not mentioned speed limits. The limit mostly seems to be 40km/h (25mph) on roads having segregated cycle tracks in the city centre. As on Dutch roads, what the limit actually is may be said to be less important to subjective safety than aspects of street design which both tame traffic and also keep motor traffic and bikes separated. Features to physically calm the traffic do exist, such as this pedestrian island road narrowing, with centre hatching, below. This could be a British design  – except for the critical detail of the omnipresent one-way cycle tracks. In the UK, such islands, designed to help pedestrians, allow motorists to squeeze, endanger and intimidate cyclists on the carriageway. In Denmark, cycling is separate, in its own space. The island is irrelevant to cyclists.

Traffic island to help pedestrians: irrelevant to cycling
At the recent Cycle City Expo conference in Birmingham (I may seem to be going off at a tangent now, but bear with me), there was a very interesting session chaired by John Dales of Urban Movement, in which Johan Diepens of the Dutch Cycling Embassy gave a talk. More or less the last thing he said was, I felt, a particularly significant statement, that I tweeted at the time. He showed some maps of the cycle and road networks in Dutch cities, said that in planning for cycling, the critical thing is to design your network correctly. Everything else, he said, was trivial. I think this is quite a stunning statement, from the point of view of both engineers and campaigners in the UK, who spend so much time agonising over design solutions for individual roads and junctions. I think the point is this, and it's strongly reflected in the success of the Copenhagen model: once you decide on your basic repertoire of designs, your units of the network, your standardised way of treating main roads and junctions, all you have to think about is an optimised network. You don't have to keep reinventing design for every location, you use the standards, as a bricklayer building a house from a plan works with standard bricks and blocks. The Danes do sometimes do special designs for special places, as in this new, nicely-paved cycle track at the side of City Hall, below (which actually has not met with universal approval), but the basic pattern of the tracks is pretty utilitarian and standard. It's continuity, and a basic level of quality and protection, which is key (as it is in the Netherlands), and it works because the network is big enough, and the routes are sensible and direct enough.

Refurbished cycle track (on the right hand side of the studs) in Vester Volgarde, by the Rathaus
Here's Copenhagen's cycle track plan, below. This covers only the central municipality, not the other parts of the greater Copenhagen urban area. Those outer areas are shown in grey, including Fredericksberg, which is the "island" within Copenhagen Municipality. It will be seen that those areas have cycle tracks as well, and that the network links up.

Copenhagen Municipality's plan of priority cycle tracks, courtesy Copenhagen Municipality
This is not all there is to it. These are the priority routes in and out, on main roads, but there is also a "green route" network under development. The green routes tend to be more orbital, more involved with parks and waterfronts and less connected to the main roads, and they tend to consist of wide two-way tracks, with a pavement along one side. The routes chosen are such that the two-way nature of the tracks does not cause conflicts at junctions; for example, when putting a route along a waterfront, all the junctions are on one side. There are also major new cycle/pedestrian bridges involved in this network, over motorways, railways and waterways.

The green route plan, courtesy Copenhagen Municipality
Cycling on a typical green route: not shared space; the yellow path is for pedestrians
Junction where the green route crosses a main road
New bridge on another green route, over the railway tracks at Osterport Station
Not part of the green network, but this track on Havenegade by the waterfront is an example of a two-way track by a road.
Then, additional to all this, is the "superhighway" network, the routes for longer-distance commuting, out to about 25 km from the city centre. These superhighways can be alongside motorways, where they are necessarily two-way on both sides of the road.

On cycle superhighway 2 from the northern suburbs to the city
The demographic that you see cycling in Copenhagen and surrounds is incredibly mixed. It is somewhere inbetween the UK cycling demographic and the Dutch, though much closer to the latter. This, of course, reflects the infrastructure that is provided. We didn't see many primary-age children cycling on their own, as I had done in the Netherlands, but this may have been because we were in the wrong places at the wrong times. Parents carrying such children using child seats, trailers and Christiana-type machines seemed far more common. Apart from than that, we saw a total mixture, of fit, sporty cyclists, workers on cheap mountain bikes, women and men of all ages on reliable, fully-equipped roadsters, and people of all ages, up to the very elderly, using cargo bike to transport goods and shopping. About 75% of the bikes in use seemed to be fully-equipped roadsters, of the type normal in the Netherlands, often of very high quality, with baskets, chain cases, integral locks, hub gears, coaster brakes, and hub dynamo lighting.

Just some cyclists on a track in central Copenhagen
Just some more
I think you'll have got the picture by now. Some of my tweets and pictures from the trip were collected by Ely Cycling Campaign and made into this Storify page. Yet more photos taken by me and other people on the trip can be found in this collection on Flickr. I came back tweeting the following:
  1. It's the end of my #cphbiketrip. Final thoughts: cycling in DK is fantastic FUN. In UK cities it is NOT FUN. That's the issue to be overcome

    Got to understand this: you need all to stick your Hierarchies of Provision, Quietways, Greenways, 20mph etc in the bin... #cphbiketrip

    ..Cause the solution is segregated cycle tracks on *all* main roads. That's the only thing that gives you fun cycling for all.#cphbiketrip

    UK politicians, don't waste time, don't bother with cycling at all if you are not interested in doing this. Over and out. #cphbiketrip
Now these tweets attracted some criticism, getting labeled kerb nerd zealotry by Bill Chidley, writer of Buffalo Bill's Bicycle Blog. Now, I can take the insults. I've been advocating networks of segregated cycle tracks on main roads as the principal solution to raising cycling levels in the UK for 20 years, and insults have always been directed at me, and those who think like me, by others in the UK cycling community, because of this. The popular slogan used to be "cycle tracks are designed by idiots for idiots", and now we're "kerb nerds" and "zealots".

But, despite the limitations of subtlety and completeness inherent in Twitter as a medium, I have to say I meant what I said in those tweets, and stand by their message. It's what I have thought for a long time; every visit to a high-cycling city in Europe rams the truth home to me again, and visiting Copenhagen did, again. I say as I find it, I am certain of what I say, and I think someone has to say this very clearly. David Hembrow's message from the Netherlands I think is much the same.

There's a more moderate, nuanced critique of (I think) my point of view than Chidley's given by Mr Happy Cyclist in his piece Extreme straw men and reductionist thinking (though not mentioning me), where he points out that
A lot of the arguments [between cycle campaigners] seem to centre around a question of what is the single intervention that would get more people on to their bikes. The problem is that there is not a single intervention that is likely to succeed in achieving such an end.
....The message that seems to come across is that all you need to do is to create a completely separate infrastructure for cyclists in which there is absolutely no contact between cyclists and motor vehicles, and cyclists will come flocking.
Now I'll accept that no single type of intervention is in itself sufficient. The trouble with saying this, and pointing out the range of interventions that are used in Denmark and the Netherlands, is that UK policy-makers get hold of the wrong things, implement measures in the wrong order, do only the easiest things, fail to tackle the real problems, and, in short, just fail again to mainstream cycling. They concentrate on trying to train cyclists to cycle on hostile roads, or on providing cycle parking where there are no cyclists who want to use it, or on trying to route cyclists away from "difficult" junctions to places nobody wants to go, or on creating 20 mph limits on roads which are unsuitable as cycling through-routes, or on installing air pumps on the pavement, or creating incomprehensible "cycle hubs", or some such stuff. Transport for London getting hold of the Copenhagen green wave idea but ignoring that fact that everything there depends on the existence of the segregated cycle tracks is a classic example of this.

What I am doing is pointing out what is important, what is key, what needs to be done first. Cycle tracks with a high degree of efficiency and subjective safety, that are direct and on the main roads that cyclists already predominantly use, are the keystone of the arch, without which the cathedral of mass cycling, with all its other components, cannot stand. That's what I'm pointing out most of the time in this blog.

A pump installed on Copenhagen's cycle superhighway 2: nice to have, but it's in the category of "fluff" again. The important thing is the thorough separation from the motor traffic on a direct arterial route.
Cycling just works in Copenhagen, because the cycle tracks give adequate subjective safety, and are there in nearly all the places they are needed, and form a coherent network that is practical and direct, because the space has been taken out of the main roads. The reliable, predictable sense of subjective safety takes the stress out of cycling and makes it fun, as well as convenient. This is why the people cycle in such large numbers, I am certain of it. Yes, as Bill Chidley commented, "fun" is a subjective quantity, and you can have fun cycling on UK roads – of course you can. But though fun is subjective, you can measure some related things. For instance, I did not see one helmet camera on an cyclist in Denmark (there were plenty of helmets, following promotion campaigns, that have caused a fall in cycling according to  Mikael Colville-Andersen, but no cameras). This is an indication of the lack of stress and worry in the environment. The actual statistical safety levels are very good as well – Lene Hartman of Furesø Commune told the group that, in her patch, with a population of 30,000, there was an average one one cycling casualty reported per year: a fact which stunned some of the British local authority officers. But I feel this is a slightly separate issue.

I saw a contribution yesterday on the ever-excellent As Easy As Riding A Bike blog which I thought put this extremely well:
The biggest drawbacks to cycling [in the UK] at present, I find, is the exhaustion of having to be constantly in ‘defensive’ mode, looking out for idiot motorists – plus the damaging effect on one’s faith in human nature when continuously presented with the reality that when no outside force forces good behaviour, most human beings unthinkingly obey the rule of ‘might is right’ and ‘its fine if I can get away with it’.
It's the lack of this sense of "the exhaustion of constantly having to be in defensive mode" that marks a huge difference between the feel of cycling in the UK, and cycling in a place like Copenhagen. For most people, being in this perpetual defensive state will always be antithetical to having "fun". That's why the UK style of vehicular cycling on the roads can never develop into mass cycling, Netherlands or Denmark-style. Put simply, most people are not prepared to cycle on roads with cars.

My worry is that British politicians and other opinion-formers, those who are favourable to cycling, still don't "get" the scale of the change that is required. I worry when I hear talk of single, big, high-profile projects, such as the Crossrail for the Bike or the Barclays Cycle Superhighways, as a few isolated routes, as if these can make a much difference on their own; I worry that there's still a tremendous under-estimating going on here of what we are actually need. I felt this very strongly after seeing the highly-developed system in Copenhagen. Of course, we have to start somwhere, and a complete network cannot be built in a year or two, but I also worry about Andrew Gilligan's emphasis on the "Quietways", which sound like they might repeat the failures of the past by putting the emphasis on trying to cater for cyclists on indirect, impractical routes, rather than by putting cycling centre-stage, on the main routes, as they have shown works so well in Copenhagen. David Hembrow expresses the same thought, I think, with his "Dutch Infrastructure is now xx ahead" widget, pointing to the way that neither politicians nor most campaigners in the UK seem to have grasped the scale of the infrastructural change that is required to achieve mass cycling.

I'll leave you with a strange picture. This hangs in the Statens Museum for Kunst in Copenhagen. It's by Lucas Cranach the Elder (I think). It shows some children trying to get a ball through a hoop that is probably just not quite big enough for it to go through. Is this an allegory for cycling in the UK, that there's just "something slightly wrong" which we'll never fix, despite our best intentions? In more pessimistic moments I think it might be. I leave you to ponder.


  1. Great post, you covered it all!

    I found the footways in Copenhagen often quite narrow, some bus stops not accessible and dropped kerbs for pedestrians poor - I think cyclists have been prioritised over pedestrians. But, I found drivers very accommodating as they would stop on quieter streets when they thought you wanted to cross and gritting/ sweeping cycle tracks after a bit of snow was a sight to behold. Finally, bikes parked everywhere! I will be going back!

  2. One of the issues facing us in the UK is the lack of political will to upset anyone by improving conditions for pedestrians and cyclists.

    I'm in conversation (if you can call it that) with a member of my local residents group who both complains about rat running traffic on our own road yet is outraged that the council have addressed rat running on another local street, thereby removing his shortcut.

    Another blog recently recounted the comments of a traffic engineer "You touch parking you die".

    There is an, understandable, fear from local politicians that they'll be forced out of office, either when next up for election by a well organised lobby group of motorists or, more by the press (witness Evening Standard hounding out of office the Westminster politician who tried to implement parking charges in the evenings. In Lambeth at the moment there has been press coverage, and MP Kate Hoey support, for a campaign AGAINST a cycle lane in Clapham Old Town.

    How did this not happen in Copenhagen or Amsterdam - whilst there were many campaigning for change surely there must have been many who perceived themselves as being disadvantaged and would have been vocal about it?

    1. How it was done in the past, I don't know, but I think there are two factors that help in the Netherlands today. We've just had a street redesign here (residential street in Amsterdam,, where 20% of parking was lost for more cycling provisions and space for pedestrians, so I've some first-hand experience:

      First, there is a culture of consensus based decision making (Poldermodel), which is felt throughout the process. As a resident, you're invited for every step in the design process, from gathering the first ideas to the final design. If you don't take part in these, it's 'your own fault' if your interests aren't served. If you do, you'll find that all groups are represented: parents with kids will ask for wider pavements and slower speeds, car owners will ask for more parking, everybody will complain about bicycle parking and rat running traffic, and most will ask for more trees, better lighting, et cetera. When presented with the final design, you can always feel that every aspect is taken into account, and it's much harder to be angry at your neighbors asking for something than at an anonymous office.

      Second, there's no direct link between politicians and traffic regulations. Most of it is hidden in excellent benchmark designs (CROW et al) and committees that don't have a direct link to a party. City council would of course be involved in accepting a new circulation plan, but I wouldn't have a clue who to blame when my favourite parking spot would be removed.

    2. Jan, a couple of thoughts regarding the consultation of local residents. Of course this is a good idea, and it is done in many cases in the UK, although you could argue how well, and whether the feedback obtained is always fed into the decision making process. But I have heard people complain about a couple of problems with this.

      Firstly, residents in the immediate vicinity of any proposed work will rarely be interested in their own street being a useful thoroughfare for anyone else. For them it is a destination, and they often have the casting vote. But even so, of the two types of example citizen you suggest, parents and car owners, the parents will generally also be car owners, and they often have no experience of their local area being pleasant for walking or cycling (or god forbid, playing!). They've given up on the idea. They are used to just getting the kids straight in the car so they want as much parking as possible as close to the house as possible - because of course the streets are dangerous.

  3. "the critical thing is to design your network correctly": thankfully we already have in London a network of wide roads where cycle tracks could be installed, that connect all the important areas of the city, and take people to the places they actually want to go; it's the TfL network. Have TfL implement Dutch tracks on those routes, starting with the ironically named "Cycling Superhighways" (compare them with the example above!). Some boroughs (Westminster, I'm looking at you) will moan, but ignore them, they'll have to follow eventually anyway.

    On that London-centric note, was there any talk from the TfL representatives to go to The Netherlands? Although CPH looks like bike Valhalla compared to London, it is still inferior to the Dutch treatment (that you say they are starting to implement anyway). The Dutch total separation would be much better suited to the high volumes at busy junctions in London anyway, imagine 50+ riders waiting at a corner to turn right.

    Across the channel Anne Hidalgo (Socialist Party), the most likely successor to the current Mayor of Paris, mentioned her desire to continue reducing the amount of cars in Paris by installing trams and cycle tracks (in French:, as well as reducing the amount of diesel vehicles to combat particulate pollution. I don't see Boris doing that.

    1. I have indeed heard talk from several sources (including this) about TfL representatives visiting the Netherlands to look at the infrastructure soon, though not necessarily these same ones.

      I tend to agree about the superiority of Dutch junction designs. Whether we can achieve the best designs in one jump, in all cases, starting from our present motor-centric infrastructural and political situation, I think is the moot point, and the difficult judgement call. The Dutch did not get there in one jump, and the political difficulty we will have will be over the space reallocation and signal time reallocation. It's still of general educational value to see the different approaches in different countries.

  4. First of all, thank you very much for this blogpost and the time/effort/travel you took to do it.

    Second, three nitpicks (can't help it, it's part of my job and I happen to think I'm quite good at it): (1) Under "you've won the raffle" (very funny!) the word "pace" should be "space" although the latter will in this instance beget the former; (2) the "pedestrian island" or "pedestrian refuge" as we call them in Australia, is primarily intended to provide a place for pedestrians to cross an entire, usually busy-with-traffic road in two stages if need be, and is installed in crossing-the-road demand areas where a signalised pedestrian crossing isn't "warranted", i.e. it isn't intended to be a primary "traffic calming" measure although this can be a consequence (but usually isn't when you've seen numerous ones at which the signs/handrails in the refuge have been completely run over by motor vehicles); (3) I reckon the ball does fit through the hoop.

    Lastly, many well-intended Anglospherian cycling advocates suggest that we should aim for a "Copenhagen" cycling network. I think this is misguided. Although I would gladly accept what CPH has tomorrow if it were able to be put in instantaneously, that is a non-existent option. We have basically nothing in terms of a cycling network. We know what the best in the world is. Why weaken our negotiating position by saying upfront that we are prepared to accept second-best, especially when second-best is showing clear signs of failing (i.e. reduced mode share, helmets, hardly any kids and elderly cycling). If we are serious advocates then we should be asking for the best and accepting nothing less. Life is too short, only 75 years on average. Infrastructure lasts for at least 15 years and usually much more. I don't want to put up with crap infra for any longer and I don't want anyone else to have to either.

  5. Who makes the cycle pump anyone help please

  6. Hi David. I came back to read your (comprehensive) take on cycling in Copenhagen after my recent visit there. Two small points...
    - Your video of the 2-stage turn is no longer available to view.
    - I wonder if planners/engineers in Copenhagen made/make a conscious effort with regard to "Design of car parking to protect segregated cycle tracks", or whether, once the stepped cycle track became the building block of choice, it's a given. By building a stepped track out from the pavement car parking naturally moves out with the kerb line.

    What came first - the stepped track or the principle of keeping bikes to the right of motor vehicles at all times on busy/main roads? Academic perhaps, but interesting.