Cyclenation and Europe: one from the archives

My post entitled Audacity, on the subject of a report on cycling in the town of Raalte in The Netherlands, and how that report was transmitted to a UK audience via Cyclenation, generated quite a lot of interest. The issue is now entirely cleared up. David Henbrow wrote a piece on A View from the cycle path which followed this up (though in fact he had written, but not posted, his article before I posted mine) and showed us exactly what cycling conditions are like in Raalte and why they have such a high cycling rate. The issue with the summary of the report that was published in English was, as I said I suspected in my piece, just a critical bit of mis-translation. As David's friend Mark Wagenbuur expained in a comment on this blog:

This is a fine example of getting lost in translation! There is but one clear mistake in the article, but it turns the whole meaning around: Where it says: 

"It has a high percentage of bicycle use without any clearly visible reasons."

It should have read:

"It has a high percentage of bicycle use without ONE clearly visible reason."

Because there is a whole lot of factors that contributed to this success story. The most important ones were the enthusiasm of the council and the guts they had to implement a daring plan!

The reason I was at some pains to go into this, down to the fiddly detail of exactly how Dutch words have been translated, and which sentences have been quoted by Cyclenation, is that Cyclenation, formerly known as CCN, the Cycle Campaign Network, has a history of, how shall I put it, creatively re-interpreting studies, data and documents, particularly those that relate to cycling in continental Europe, for its audience of UK cycle activists. The webmaster of the Cyclanation website is John Franklin, though the site's web news is edited by Jayne Rodgers. The Cyclenation newsletter is now edited by Graham Hankins, but going back into the past, when it was CCN News and a paper document, it was edited by John Franklin.

The business of Cyclenation and the report into Raalte reminded me of an affair from the distant past, the details of which I have exhumed from the archives, and which I now serve up again here. So, dear reader, pick up your pipe, don your deerstalker, and hop into a Hansom, for a journey through the swirling mists of turn-of-the (21st)-century London, for this is The Strange Case of the (So-called) Finnish (or Helsinki) Study.


Our tale begins back in 2001, in Helsinki City Planning Department. There, a Dr Eero Pasanen publishes a document called The risks of cycling. Strangely, he publishes it in English. You can read this original document, from which so much fuss later came, still, on the web – it is not very long.

The document came to the attention of the UK cycle campaign community through the pages of CCN News edition 56, January 2002, edited by John Franklin (and he was also the Chair of CCN at the time). This also can still be read on the web, but, to save you the trouble, I have extracted the relevant text from that newsletter here:

Finnish study cautions against road-side paths

A study based on Helsinki has shown that it is much safer to cycle on the roads with traffic than on the twoway cycle paths that form much of the cyclists' network. 45% of cycling kilometres in Helsinki take place on the cycle paths, but they are the location of 56% of injury crashes to cyclists. 

The situation is worst at junctions. The risk of a crossing crash is 3 times higher coming along a cycle path than along the road. Cyclists riding on the pavements are also more at risk. 

According to the author, the main problem is that car drivers have nothing to fear from cyclists. At crossings, drivers focus their attention on other cars and not on cyclists. The two-way nature of most of the cycle paths exacerbates the problems as drivers expect even less to see cyclists arriving from the 'wrong' direction. The most common type of crash, by a factor of ten, is a car turning left hitting a 'wrong-side' cyclist approaching from the driver's right. 

The Finnish two-way cycle network was based on a Nordic traffic planning guide from the 1960s which considered cyclists and pedestrians to be a homogenous group of vulnerable road users, to be separated from motor traffic. Though an appealing principle at the time, it has led to a considerable number of unnecessary car/cycle crashes on at-grade crossings and pedestrian/cycle conflicts on paths connected to sidewalks. 

Helsinki has 800 km of cycle paths, half of which are especially dangerous. It is hard to imagine that this system could be rebuilt, but in those countries and cities that are just beginning to build cycle networks, roadside paths and particularly two-way types should be avoided in an urban environment. 

Two other findings of the report are that cycling is usually safer the more cyclists there are about, and that cycling leads to more police reports of pedestrian injuries per kilometre travelled than private motor vehicle use.

[End of report] 

The February 2002 edition of London Cyclist, magazine of the London Cycling Campaign, carried a short news item based on this report in CCN News, stating the same conclusions, that two-way cycle paths alongside roads are dangerous.

Now, the significance of this to me, and to others in Camden Cycling Campaign at that time, was that we were promoting the construction of two-way segregated on-road cycle tracks at that time. We had designed one, and got that built, on Royal College Street, Camden Town, London. At the time this report came out, we were lobbying to get another built, the east-west track through Bloomsbury, on Torrington Place, Tavistock Place, and other roads (known as "The Seven Stations Link", or as LCN Route 0). These tracks were the idea of Paul Gannon, who had lived and cycled in The Netherlands, and the designs were based on Dutch models, specifically, the main strategic cycle tracks constructed in The Hague. I will tell in more detail the story of the construction of these tracks in another post. They remain to this day the longest and most successful segregated cycle tracks ever constructed in London.

Now this report from Finland was very worrying to us, since it suggested that, despite the apparent success of these designs in The Netherlands, such designs were actually dangerous. It suggested that we were doing completely the wrong thing in promoting them in London. Certainly many cyclists in London and members of other branches of the London Cycling Campaign already believed, for a variety of reasons, that Dutch-style segregated cycle tracks were the wrong way to go, and this report provided evidence for them.

Well, we discussed this at length, looked at the original document on which the CCN report was based, critically examined it, and decided we were not wrong after all. The problem lay with the "Finnish Study", not with our designs, which had, after all, already been tried and tested in The Netherlands for decades. I forthwith wrote two letters (emails) rebutting the interpretation of the "Finnish Study" that had been disseminated in the UK press: one, to the editor of London Cyclist, which was published, and the other to John Franklin, as editor of CCN News, which he did not publish. They were similar, so I quote here the letter to John Franklin:

Dear John,

The danger of cycling in Helsinki

Your report in CCN news 56, “Finnish study cautions against road-side paths” is rather misleading in terms of its reporting of the research in question. The record needs to be set straight in view of London Borough of Camden’s current work on creating two-way segregated on-street cycle cycle tracks both in Camden Town and central London (the Seven Stations Link, or London Cycle Network route 0), campaigned for by Camden Cycling Campaign, and because Camden Council is now the lead borough of the London Cycle Network, responsible for co-ordinating and advising on its implementation London-wide.

The Finnish paper can be read on the web at http://www.bikexprt.com/research/pasanen/helsinki.htm.

The implication of your reporting of it, that cycling in Helsinki is unsafe, and that the style of engineering that has been employed there in order to promote cycling, very similar to that used in Sweden, Netherlands, and Denmark, has not been a success, is very wide of the mark.

The two most important points arising from the statistics in this report are as follows. Firstly, the casualty rate for cycling in Helsinki is 20 per thousand million km cycled, only one-third of the overall British level. This safety record is broadly similar to the other countries I mentioned, all of which have constructed extensive networks of on-street segregated cycle tracks in cities over the past 40 years. Secondly, the level of cycling in Finland overall in terms of km per person per day is 7 times that of the UK: an astonishing fact in view of the often sub-zero climate. (In the Netherlands the level is 30 times ours.)

It is clear from these facts that, rather than being “cautioned” by the Finnish experience, we should be seeking to emulate what they and the other Nordic countries have done right in cycling policy.

The report tells us further that 56% of cycling miles in Helsinki are conducted on the network of on-street segregated cycle tracks, proof of the high level of popularity of these facilities (our experience in Camden also confirms this). One point your report have focuses on is that there are more crashes at junctions on this network than there are on ordinary roads, relative to the numbers of cyclists using them. However, this “discovery” needs to be treated with caution. The vital data that the report omits are the average levels of motor traffic at the different junction types. It seems most likely that the apparent safety difference is due to the segregated tracks being implemented in the busier city-centre locations, where the higher levels of traffic will inevitably result in more incidents than on suburban roads, irrespective of facility design. The vast bulk of experience of the decades of implementing these designs in the countries mentioned, the fact that they continue to be demanded by cyclists and continue to be built there, and the fact that these are regarded as the leading cycling nations in the developed world, indicates that two-way segregated cycle tracks, when well-built, are safe for cyclists, much safer than British roads, as the figures show, and that they have a major effect in expanding cycling. One incompletely-researched paper does not change this.

It should also be noted that the overall thrust of this paper, authored by one individual engineer, is anti-cycling, proposing compulsory helmet legislation (to which UK cycling organisations generally are opposed), and arguing that encouraging cycling goes against the wider interests of the population in making the provision of public transport less economic, another idea with which we would doubtless take issue.

David Arditti

Camden Cycling Campaign (A local group of the London Cycling Campaign)

I worded this quite carefully and politely. But I never heard from Franklin about it. However, discussion within LCC rumbled on, with opponents of segregated infrastructure continuing to claim that the "Finnish Study" proved that segregated cycle tracks, particularly two-way ones, were "dangerous". This kept coming back with people continuing to cite the same flimsy evidence against segregated tracks, and  correspondence on this subject in early 2005 on the LCC Yahoo! groups culminated in a very long post from Paul Gannon, where, in reply to points made by Charlie Lloyd and others, he wrote the following, which I can only term a thorough and forensic demolition of the "Finnish Study", along with a condemnation of the "thoroughly dishonest" way in which Franklin had used it to push his own anti-cycle facility agenda. This is long (it also goes into technical details of the Camden segregated tracks), but worth reading in full:

When I questioned the existence of statistical evidence demonstrating the assertion that segregated tracks are less safe than cycling on the road, Charlie pointed to the ‘Helsinki study’ and the junctions on the Royal College Street track as examples of how that study was applicable.

The full text of the Helsinki paper is in the attachments (doc and txt - that latter might lose the statistics). The following text was circulated on the LCC Issues list and attributed to the Cyclists Network some time ago and illustrates the way in which paper has been used as evidence.

[Gannon here reproduces the report published in CCN News]

Those who have cited one factoid from the Helsinki Document may feel they have ‘put another nail in the coffin’ hoping to bury the arguments for continental-style cycle tracks in the UK, but is that really so?

Let’s look at this Finnish study. To start with, it’s not the “Finnish Study” to which status the CCN web report has elevated it. It is not a ‘study’, but a synthesizing policy/political document that cites statistics from other studies and/or synthesizing documents. Citing it as the ‘Finnish study’ gives it an undeserved status and authority. Citing it as the Finnish study on just one of the document’s highly biased conclusions is, to say the least, disingenuous.

The report was written to support the view that cycling should be constrained rather than encouraged, and the reason for that is that cycling is dangerous and therefore undesirable.

Let’s have a look at some of the things the CCN report happened to overlook.

The report is very short, assembled from a few paragraphs with highly contentious assertions and claims: “cycling is dangerous”; “cycling competes with public transport”; “Cycling is attractive and healthy for cyclists, but public transport is essential for many and perhaps the most manageable way towards sustainable traffic.”; “Cycling is not a harmless traffic mode for pedestrians.”; “One could claim that many cyclists feel themselves as ‘saviours of the world’. With their non-polluting, silent and relatively harmless vehicles, they may imagine that they have more rights than other road users.”; “An obligatory helmet may decrease the attraction of occasional cycling, but would this be a serious problem?”; “At present, popular arguments for sustainable traffic and the freedom of cycling often seem to take precedence over questions about safety”.

It is worth here re-reading the CCN report of the Helsinki document to remind ourselves that we are talking about the same report. The choice of material is so biased as to give an entirely misleading impression of that report. It is a thoroughly dishonest account in my opinion for pulling out one bit of information that suits the writer of the CCN report and ignoring all the rest. Then down the line the CCN spin on the report becomes the wisdom among the British cycling activists’ establishment – another bit of evidence damning these childish longings for segregated tracks.

It is important to assess the purpose of the Helsinki report. It is clear that the author is assembling a case for not spending money on cycling facilities on two grounds: first, the assertion that safety issues associated with cycling are insurmountable, or require severe restrictions on cyclists’ “freedom” (“It is clear that we must radically improve the safety of cycling. But how can this be done? So far, most of the important successful steps in traffic safety work have been based on restrictions of the freedom of car driving. When trying to improve the safety of cycling, the starting point is different. Popular arguments for sustainable traffic and the freedom of cycling often seem to neglect safety problems.”); second, it conflicts with the role of public transport (“Still, we want to increase cycling for environmental and health reasons. But are these reasons strong enough to compensate the serious safety problems of cycling, especially when cycling seems to compete mainly with the very safe public transport? The important question is: Does increased cycling weaken the level of public transport service?”).

We must keep in mind this context in order to understand the purpose of the report and its ‘evidence’ when citing the report itself as ‘evidence’ in other arguments. The author has had to go out of his way to assemble his ‘evidence’ and has had to present and interpret it. It should not be accepted or cited uncritically and to report it as statistical evidence is the push at the limit of credibility. The report is a political one, with a political aim and should be viewed – and cited - as such.

Before dealing with the issue of segregated tracks, let’s look (very briefly) at the general assertion of the author. The author’s ‘cycling is dangerous’ assertion is based on one ‘statistic’, a simple comparison of injury causing bicycle accidents per km travelled’, compared (presumably, though the report’s wording is not clear on this) with ‘injury causing car accidents per km travelled’. But are these statistics comparable? The author’s figures show that a much higher proportion of injuries in accidents involving cars are inflicted on ‘other people’ than is the case with bicycles. Surely this fact is of considerable importance in evaluating safety measures and policy for influencing mode of transport usage? But, of course, it depends on the case you want to make how you present and interpret the statistical evidence.

It raises in particular the question of whether the number (and severity) of injuries can be reduced by reducing car use, something that the author doesn’t consider. The author’s table shows that cars accidents cause twice as many ‘other victims’ as did cycling in Helsinki.

The question that really needs to asked is whether cyclist and pedestrian injuries can be reduced by a reduction in levels of motor vehicle traffic – a conclusion I would draw from a rounded assessment of all the statistics presented, rather than just citing simplistic arguments based on one dubious figure. Cite one or two statistics and you can easily get the ‘cycling is so dangerous it must be curbed’ argument; carefully assess what the statistics might indicate and you can get a more sophisticated result.

The key point is that these are qualitative matters at least as much as they are purely quantitative. A headline-grabbing statistic may help one feel that one has landed a telling argument, but it adds nothing of substance to an important discussion if you make no effort to understand what is being measured and why.

We need to understand that all traffic statistics are of doubtful quality, easily misunderstood, and frequently abused. Statistics can be no more than a guide, critically assessed and carefully used for drawing conclusions. And it is important not to alight on one statistic, but to use in reaching judgements an array of statistics.

I expect that a lot of people will agree with me that the Helsinki Document must be properly assessed when dealing with its assertions about cycling being dangerous, doing down public transport, about cyclists being self-righteous jerks, and so forth. However, plenty of the same people are happy to cite this highly tendentious document on another matter.

Which brings me to the misuse of statistics about the alleged lack of safety on cycle tracks. First, the author. The foregoing means that the author’s evidence and arguments are suspect; also he appears to be anti-cycling and is amassing evidence to argue against cycle track construction and/or for measures to limit cyclists’ freedom. None of us has seen the original studies which he cites so we are dealing with his selection and presentation.

Second, the statistics. He says, “45% of the cycling kilometres in Helsinki are on cycle paths along streets, but 56% of injury accidents happen to these cyclists”. This is cited to justify the conclusion “we cannot reduce bicycle accidents by building new cycle paths.”

But we simply don’t know whether these are comparable categories or not as insufficient information is given for a judgment to be made on the usefulness of the numbers and the categories. The roads used for cycling in a city such as Helsinki are quite probably quieter residential roads with comparatively lower levels of cycle and motor vehicle traffic, while cycling on tracks will be in the city centre and on major commuting routes with comparatively heavier traffic. This statistic is meaningless.

The author’s other relevant statistic refers specifically to two-way cycle tracks and points to what is widely known to be the case – which is why CCC has argued for specific design measures to deal with this potential problem.

First the statistic must be looked at closely. Only one interaction shows much higher crash rate than all the other interactions, namely motor vehicles turning left (for UK this is turning right).

Any road design system has its weak points – in the UK there is a higher proportion of accidents among ‘cyclists turning right’ than for other maneouvres. The report’s absolute figures don’t tell us anything we didn’t already know and they certainly cannot support the conclusion that all cycle tracks are more dangerous. As these are absolute figures all they show is that one motor vehicle maneouvre is more dangerous than others.

The answer may be to restrict that movement rather than to restrict the building of cycletracks. Just as with the overall argument of the author where he interprets the high rate of injuries caused by cars to cyclists as evidence that cycling should be curtailed, so some people are arguing, here in the UK, that this particular statistic should be generalized and applied to all interactions at junctions with segregated tracks.

The interaction ONLY applies to two-tracks and not to one-way ones and can be removed completely be preventing the particular manoeuvre.

Now Charlie said that this situation applied to the junctions on Royal College St. But this is not true. Royal College Street is one way and there is no equivalent of the maneouvre (ie motor vehicles turning right across the track). Therefore his claim is wrong.

However, if he had said the junction of the ‘Seven Stations Link’ east of Byng Place (near ULU) with Gordon Square, he would have been right. This junction does indeed allow for this interaction.

However there is another, and I think important, factor. When Camden Cycling Campaign proposed the tracks, it was part of the plan that all these three junctions should be closed to all motor vehicle crossings and motor vehicles routed via (current) traffic-signalled junctions (giving a substantial junction-free run on the Royal College Street route in particular). The decision about the appropriate junction treatment depends on overall motor vehicle movements per hour and the direction of the flows. I am surprised that Charlie appears not to know that we wanted these junctions closed to motor vehicles and he repeatedly circulates the impression that there is a flaw in the design rather than the implementation of these junctions. For, this important part of the proposal was rejected by the council’s traffic engineers. They are now looking at redirecting motor traffic on the Gordon Square junction as originally proposed.

What the statistics from Helsinki show is not that segregated tracks are unsafe, but that in particular situations there can be raised risks (ie, only one interaction and only on two-way tracks). The decision as to whether you see that as an argument against such tracks or against such interactions is a political decision, not a statistical one. As mentioned above there are far more cycle crashes in the UK while a cyclist is turning right than any other interaction. Does anyone go on to conclude from this one statistic that all road cycling is unsafe? Or is it a case of double-standards? The statistics cited do not offer any evidence at all that ‘segregated tracks’ are unsafe, only that poor design is unsafe, but then we know that anyway and it applies to BOTH ‘track’ and ‘non-track’solutions. To see this as an inherent weakness of tracks, but not of non-track approaches’ is again to apply double-standards.

The Helsinki report does not show what Charlie asserted it showed, namely that segregated cycle tracks are unsafe and it is misleading to suggest that it does.

Paul Gannon

Phew. Well that should have been the end of the matter, but was not of course, and down the years various British opponents of segregated cycle engineering have continued to drag out the bedraggled corpse of the notorious "Helsinki Study", or its distorted Franklinesque reflection, to attempt to add credence to their views. For example, another voluble opponent of segregation, Tony Raven, wrote this on the Camden Cycling Campaign Yahoo! group in 2007:

Here is what the Helsinki Traffic Planning Division says about their  massive investment (~€800m) in 800km of two way cycle lanes (some of which I have cycled): 

"A recent study in Helsinki showed that it is safer to cycle on streets amongst cars than on our two-way cycle paths along streets. It is hard to imagine that our present two-way cycling network could be rebuilt. But in those countries and cities which are just beginning to build their cycling facilities, two-way cycle paths should be avoided in urban street networks."

Now by 2007 none of this was "recent", and Tony Raven perpetrated here a further distortion to the record by claiming that the "Finnish Study" was the work of "Helsinki Traffic Planning Division" when it was not, it was the work of its author, Eero Pasanen, and there is no evidence that the weird views contained in it were ever anybody's but his own. However, this incident goes to show how things like this tend just to go round in a series of "Chinese Whispers" in the cycle campaign community, an original misinterpretation, or deliberately distorted interpretation, of a study becoming relayed, embellished and altered again, turning over time into the "folk wisdom" of those who seem to be determined to try to prove, against all the real, proper statistical evidence, deriving from the overall casualty figures for cyclists in various places (some of which I covered in this post) that "segregated cycle paths (or tracks) are dangerous". (If you are interested, there is a further discussion of the evidence on the safety of segregated cycle facilities by Mr C on MCR Cycling.)

The survival of the "Finnish Study" continues to be ensured by the fact that it features to this day on Franklin's webpage Cycle path safety: A summary of research – so desperate is Franklin to continue to publicise any work that could, by any twisting, be seen to support his point of view, no matter how discredited it may be. And earlier this year someone else on the LCC "Issues" Yahoo! group, I forget who, dragged it out again, someone who was not aware of all the previous LCC correspondence surrounding it.


John Franklin has been a central figure, perhaps the central figure, in CCN, now Cyclenation, for many years. Through that organisation he has had huge opportunity to disseminate his well-known anti-segregated cycle facility views to the whole of the UK cycle campaign community, sometimes explicitly, but sometimes more covertly, through the selection of material for reporting, choosing what to publish and, equally importantly, what not to publish, and in the way he has produced interpretations and summaries of domestic and foreign research work. This activity by Franklin has, I believe, been very damaging to cycling in the UK. His selection and his interpretation has invariably been done to prop up his own ideas on how cyclists should be catered for on the roads. To quote Mr C:

Looking through the literature on his website, I see a homeopathy-like penchant for cherry-picking research which agrees with his message on the alleged safety issues of separate cycle facilities, whilst ignoring the wider body of work showing they improve cyclists' safety and promote higher cycling rates when implemented well. Reading through his published work, it seems disconnected from reality. 

Another issue has been that, up until perhaps the last five years, we actually had very little good information in English on the cycle infrastructure experience in Europe, and particularly on infrastructure standards. We tended to get more information from the United States: but the relevance of this to UK urban utility cycling was often very dubious. Another trick Franklin has perpetrated has been to represent at least one study of cycle casualties emanating from the US as evidence for the "danger" of segregated cycle facilities, when the base data in question actually related to American cyclists falling off their bikes on mountain trails – totally irrelevant to what he was claiming evidence for. But to actually uncover such facts is very hard, as most of the links on his website point to documents that are not really the base studies being quoted, but summaries and interpretations, again. They are mere "heresay", not "evidence".

The lack on information in English on cycle standards on the continent has changed with the appearance of several invaluable websites filling this gap, such as the Fietsberaad website in English, the Cycling Embassy of Denmark, David Hembrow's superb View from the Cycle Path and now, Great Britain's own Cycling Embassy. But this is all very recent. Before this, we were dependent on only one or documents that happened to be translated, and there was no guarantee that these would be mainstream, or genuinely representative of the continental infrastructure experience, rather than something loopy and unrepresentative, such as the "Finnish Study", or, I may add, the ideas of the late Hans Monderman on "Shared Space", which have gained some currency in the UK.

For the "Finnish Study" was always an obviously mad document if you read it all. No-one should ever have regarded it as a serious contribution to the cycle infrastructure debate. It was only through the distorted reporting of CCN (now Cyclenation), "thoroughly dishonest", as Gannon called it, that, once misrepresented in a way that was attractive to those who wanted to find evidence to support their preconceived ideas about the "dangerousness" of segregated cycle engineering, it became canonised as "evidence" of what was wrong with the European approach, for British cyclists of a certain frame of mind, who looked more towards the American approach of Forrester than the European one. It is a cautionary tale.

The recent Cyclenation report on Raalte may have nothing whatever to to with Franklin, but that incident brought all this back to mind again. I think it is an interesting history lesson, and, if you have borne with me this far, I hope you have found it instructive. The lesson is the simple one that is taught (or should be taught) to every young scientist: don't believe what you are told about "what the research proves", always look yourself at the original work, and make up your own mind if it is true.