Thanks to Pete for the comments.  In case of misunderstanding, I would
certainly say cycle lanes are not always appropriate (e.g.  rarely in
20mph zones), and that they should be well designed in relation to the
particular road, traffic speeds, etc.  I fully accept that in some
situations traffic may overtake faster or closer if a cycle lane is
present.  I agree that many of the cycle lane designs at roundabouts
were bad.

I am only familiar with the situation in Edinburgh, where the evidence
is very strong that any negative effects of prominent onroad cycle lanes
(and ASLs) are outweighed by the benefits.  The evidence that the
widespread introduction of coloured onroad facilities in Edinburgh in
the last few years has led to a big increase in cycle use (and hence in
safety) is strong.  The many letters we have seen from individual
cyclists in the recent council consultation show that widespread onroad
facilities have made cycling on the roads seem safer and seem more
legitimate (and NOT just in the cycle lanes).  And perception of safety
is the biggest negative for potential cyclists.

Comments on some of Pete's specific points now follow.  I'm trying not
to make this too long so won't comment on everything. [it's still long!]

1 > It is important not to conflate the issues of ASLs with cycle lanes.

Certainly, but even more so not to conflate cycle lanes with off-road
paths as you did later! Albeit that there are many very popular off-road
paths, some of which are even ok for and extensively used for commuting.

2 > The only way to asses the desirability of a cycle "facility" is whether
> it makes conditions better or worse for cyclists.

Not sure what you mean by better or worse, but if the combined effect of
a lot of visible onroad cycle facilities is to bring out more cyclists,
and significantly reduce casualty rates at the same time, as has
happened in Edinburgh, that would seem desirable.

3 > > Surely the number one question is 'Does widespread provision of cycle
> > lanes/ASLs encourage more people to cycle?'
>
> The answer to this is no.

Pete's first counterexample was Milton Keynes - but that is an offroad
network, and also poorly designed, according to John Franklin.

Pete's second counterexample ...
> The highest levels of cycling in Cheshire (and also the safest place to
> cycle) is Crewe - a place that is fairly devoid of facilities.

Whilst cycling may be high in Crewe, how has it changed in the last few
years? In Edinburgh and London cycle commuting has doubled, in parallel
with the new visibility of cycling.  If onroad cycle facilities are
significantly dangerous one might even expect a decline.

4 > > The reason for my interest in this is that, as far as I am aware, the
> > only places in UK to have seen substantial increases in cycle use are
> > London and Edinburgh

London: Pete is obviously right that the congestion charge is a major
factor; also the bombings.  However that the onroad visible facilities
are also very significant is the opinion of Transport for London - and
of (initially sceptical) independent transport commentators such as
Christian Wolmar.  What does London Cycling Campaign think?

> Edinburgh: A council that is genuinely trying to reduce motor traffic
> through a whole host of measures - and also has a history of cycle
> friendlyness through less visible measures eg sinusoidal road humps.

This has been ongoing for many years. However it is just in the last few
years, parallel to the appearance of the visible onroad facilities, that
there has been a big increase in onroad cycling numbers.  In earlier
years, most cycle expenditure was on the offroad network which - whilst
hugely appreciated by the significant numbers who use it - only saw a
fairly slow increase, and with little effect on onroad cycle numbers.

5 > We should also look at the Gloucester "safer city" project in which
> a large amount of money was spent on safety measures for all road
> users. Gloucester became a safer city for all categories of road users
> with one exception - cyclists. And what did they do for cyclists - cycle
> lanes.

I don't know this case, so can't usefully comment.  Maybe the nature of
the facilities is different, and there will of course be many
differences between Gloucester and Edinburgh.

6 > > [There will be more
> > details of our campaign, together with many quotes, in the next Spokes
> > bulletin, early March].
>
> And how about a counter example.

Of the 40 or so replies that were copied to us, not a single respondent
to the recent Edinburgh council consultation asked for cycle facilities
to be removed, whereas many asked for more and for more prominent.

Further to Pete's European examples of slow-speed areas, we are very
happy with Edinburgh's plans for many new residential 20mph zones,
including some shopping areas, which will need very few cycle
facilities.  This seems unlikely to be the solution for the
longer-distance through-routes however.

7 > > Of course it is much harder, if not impossible, to research this effect
> > than to research details such as the car-distance question.  Hence the
> > imperative to home in on the latter - it is easy to study!
>
> The advocates of homeopothy make the same arguments.

I would be delighted for research to be done into why Edinburgh and
London have seen doubling of cycle use in a relatively short period, and
whether this can be replicated elsewhere.  I am just saying it is much
harder to do objectively than the assessement of the localised safety
effects of specific cycling facilities.  I am also saying that the
results of the latter, and any lessons from it, should only be
generalised with great care.

8 > Both the advocates of lids and lanes are starting from the proposition
> that cycling on the road is a risky activity that requires special
> protection on the form of segregated space or protective equipment.

Certainly not.  I see well designed and prominent onroad facilities as a
very effective promotional tool far more than as a safety measure.  I
also believe cycling to be a safe activity and a safe means of
transport.

9 > > Finally, as regards the specific question of cycle lanes and
> > car-distance, that question should not be 'how much distance?' but 'is
> > the distance adequate (and in relation to the car speeds) ?'
>
> Take a look at
> http://www.warringtoncyclecampaign.co.uk/report/cycle-lanes.pdf
> Fig 3b  That is too close for my liking.

Ok, take a look at www.spokes.org.uk - downloads - bulletin 89 - page 1.
That is even closer.  However there would be a riot if the council
suggested removing that cycle lane.  Prior to that it was impossible to
cycle up The Mound except in the stream of cars (which is often either
static, so used to be very frustrating and fumey, or moving faster than
an uphill cyclist, so used to feel under pressure).  Now it's nearly
always a clear path up.  The benefits to *pedestrians* from the vehicle
separation are also very obvious in the photo, if you imagine how it was
with the traffic up against the pavement.

It just shows there are many different situations, and we can all find
example photos to highlight our case!