Side effects

My thoughts on this document by Paul Smith, still cited by his acolytes even now.

Most of Smith's document is, of course, simple assertion, not backed by any kind of evidence. I disagree with pretty much every one of these assertions and state why. I've also stated a few items of evidence and further reading. For the most part, though, it amounts to "assuming that speed enforcement is evil, then speed enforcement is evil".

A philosophical question: Do you think that one of the things that would characterise a genuinely safe driver, is the willingness to compromise safety in order to exceed the speed limit without detection? Drivers who do the things Smith describes, are not, I would argue, safe drivers. So the question is whether we should compensate for their poor driving by allowing them to drive faster with impunity. If you assume, as Smith does, that speeding is victimless, or at least more so than speed enforcement, then you will agree with him. If, on the other hand, you accept that speeding has some societal cost (for example increased noise and risk to residents of the road) then perhaps you won't.

Text in italics is from the document, comments below are mine. The headings are Smith's.

Road safety culture damage

 * Although it is not much talked about in road safety, cultural values at very much at the heart of any safety system. However, in industrial safety, developing and maintaining a safety culture is considered central to reducing risk. Unfortunately nothing in modern policy is helping to build a sound road safety culture. The side effects listed here are actually causing severe damage to our once good road safety culture. Our previous good safety culture was a central component in achieving the safest roads in the world.

If there has been an effect on road safety culture, it is more likely the result of militant libertarian motorist activism attacking every conceivable form of regulation of motoring. This has, of course, been the case since the first motorised wheel turned in Britain. The same strident voices opposed registration, licensing, speed limits, compulsory insurance - and in virtually every case using similar arguments about the salt-of-the-earth majority being let down by the unruly minority. In no case has this ever been found to stand up to independent scrutiny, but it has become markedly louder and more aggressive in the last decade or so.

For a comprehensive treatment of this phenomenon, read Plowden's The Motor Car And Politics, a political history of motoring in the UK. It is a most interesting read.

Drivers' general attitude to driving is worsened.

 * One of the key factors that identifies a low risk driver is having a good attitude. A good attitude comes from taking responsibilities seriously and goes towards allowing safe margins for error. Drivers with a good attitude learn from their mistakes and don't take safety for granted. There's a significant risk that excessive speed enforcement is having a general negative effect on drivers' attitudes.

Leaving aside for a moment the fact that this appears to be arguing for people to be allowed to make more mistakes, the consequences of which are directly in opposition to the supposed improvement in safety culture, the idea that drivers with "good attitude" are likely to be the same drivers who are unwilling to simply keep within the limit, is questionable at best. In reality, of course, drivers with multiple convictions are also more likely to crash (e.g., which finds a significant positive correlation between points and crash risk).

Reduced roads traffic policing

 * Road traffic policing has been in steady decline in the speed camera era. It is sometimes argued that speed cameras have nothing to do with the accepted decline. We blame policy and point out that the same policy makers have caused the rise in cameras and the decline in roads traffic policing.

The decline in traffic policing is the result of traffic policing being taken out of the core duties of police forces by the last Conservative government. There is no direct link between this and the timeline of camera growth. Of course the camera partnerships set up by the same government erected cameras, that was what they were there for, but they were not directly controlled by the police and it's unclear what benefit the police are supposed to have attempted to derive by pressing something on which they were, by common consent, no longer being measured.

But here's a curious thing: even with this de-emphasis on traffic policing and all the appalling carnage Smith asserts it released, still the road traffic fatality and injury figures continued to fall, year on year, both absolutely and per billion passenger kilometres, as the number of cameras spiralled.

The best Smith was ever able to come up with here was an assertion of "loss in trend" that turned out to be down to comparing two years which had different data collection bases, affected only motorcyclists, was visible only on the types of roads least likely to have cameras, and was evident only in fatalities, not in the much larger (and therefore statistically dependable) KSI figures. Odd, isn't it?

Speed management replaces road user quality management

 * The advent of speed cameras steadily taking the place of traffic police has had the inevitable effect of changing the balance refined over the years by competent police forces to place greater emphasis on speed limit compliance.

This "replacement" has no basis in fact. The emphasis on traffic policing was an effect of the change in core duties, not the growth in camera numbers, though both may have had a coincident cause, and in fact the use of cameras has had the exact opposite of the effect Smith asserts: it is now much more likely that police will be looking for offences that require human judgement (tailgating, erratic driving and so on) as the policing of absolute speed can safely be left to dumb machines.

Ironically a camera is now in early trials which can detect a large number of dangerous driving behaviours including erratic driving, tailgating and so on.

Speed enforcement replaces sound road safety engineering.

 * Slapping a camera up was almost the Pavlovian response to a couple of accidents in the same stretch of road. What used to happen pre-camera era – and what must happen again as we recover – is that road engineers examine the root cause of the accident or hazard and orchestrate a means to reduce/eliminate the hazard or make it much more visible.

There is simply no evidence at all that traffic engineers have stopped trying to make roads safer, and precious little evidence that what they did actually had that effect in the first place. In a well-known experiment (one cited by Smith in other places, I think), a garden gnome was placed on some corners which were identified as accident blackspots. The sites with gnomes exhibited a consistent improvement in accident rates - something generally known as ; Smith is well aware of this effect as he (correctly, for once) identifies it as a factor in overestimating the lives saved by cameras.

Aside from gross effects, such as grade-separated junctions and segregated traffic lanes, most of what traffic engineers have done in the name of road safety is highly disputable and indeed occasionally disputed by Smith (but only when it involves reducing traffic speeds). There is good evidence that drivers respond to "road safety" measures such as edge markings, signage and so on, by driving faster and relying on the signage rather than nreading the road. It's also very clearly the case that drivers accustomed to not finding a tractor round a given bend no a given road will, over time, come to assume that no tractor will ever be there, regardless of the warning sign to the contrary. A large part of traffic engineering work centres on signage telling people what they should have recognised and allowed for anyway.

Risk Compensation 1 - motorists

 * Drivers may follow closer or drive more aggressively to preserve personal subjective risk levels when forced to travel at a speed significantly lower than their optimum safe speed of progress for the conditions.

This begs the question of whether their "optimum safe speed" is any such thing. Patently in many cases it is not; there is no informed dissent from the view that drivers generally overestimate their own skill and underestimate the extent to which documented risks apply to them. In short, drivers think they are better and more able to handle risk than they actually are.

Of course slowing them down does not fix this, and does not, as Smith implies, reduce the speed imperative, but neither does it make it provably worse, and it is unquestionably the case that, all other things being equal, a collision at 30mph is less likely to cause death than one at 40mph. In fact it has been shown that on a perfectly dry road a car braking from 40mph at its maximum effect is still travelling at 30mph where the same car on the same road would have stopped had it been doing 30mph. This is just classical Newtonian mechanics, of course, but the tests bear it out.

Risk Compensation 2 - motorists

 * Drivers may pay ‘just enough’ attention to preserve subjective risk levels at any speed. If speeds are lower, then attention is lower. This is very dangerous if it is punctuated by periods of complete inattention.

This is a statement of the obvious. Drivers only ever pay just enough attention - or, rather often, just not enough, with around five million recorded motor claims per year (of which, incidentally, over 100,000 are fraudulent).

Of course it would be better if drivers were better educated as to the risk they pose, but this is hard to do - a basic need would be a programme of continuing driver education, most likely including periodic re-tests. This is politically sensitive because although most drivers would happily concur with the view that driving standards are lower than they should be and it would be better if some proportion of the worst drivers were taken off the road, the drivers affected, which may well include some who falsely perceive themselves to be very good drivers, are unlikely to be any happier about it than those convicted of speeding offences.

Ultimately the application of road traffic law is a key part of allowing as many people to drive as possible by keeping the parameters, as much as possible, within the competence of the average individual. Drivers would probably not welcome the subjective methods used in some jurisdictions to measure intoxication and other impairments, and they probably would not relish a return to the old days when excessive speed was down to the say-so of the police constable on the ground, because in the end most drivers perceive bad and dangerous driving to be a problem of other drivers, not themselves.

As Bernard Shaw put it: "What is dangerous driving? I have a tendency to believe that everyone's driving is dangerous, except my own"

Risk compensation 3 – motorists and pedestrians

 * Slower traffic may create an illusion of safety. This may result in lower levels of care from drivers and especially from pedestrians.

The plain fact is that it's not an illusion. In 2000 the government set itself the following targets:
 * 40% reduction in the number of people killed or seriously injured in road accidents compared with the average for 1994-98;
 * 50% reduction in the number of children killed or seriously injured;
 * 10% reduction in the slight casualty rate, expressed as the number of people slightly injured per 100 million vehicle kilometres.

By 2008 the figures show the number of people killed or seriously injured was 40 per cent below the 1994-98 average; the number of children killed or seriously injured was 59 per cent below the 1994-98 average; and the slight casualty rate was over thirty per cent below the 1994-98 average.

The reduction in urban traffic speeds and the much wider public profile of the dangers of excessive speed cannot be asserted to be increasing risk, when at the same time the measured level of risk has reduced not just slightly but by over 50% in the case of child casualties. says that one should not multiply hypotheses. If Smith's assertion that reduced traffic speeds increase pedestrian risk-taking is true, then another hypothesis must be advanced which accounts for the substantial measured decrease in pedestrian injuries over this period.

Stimulation effect

 * Less stimulation for drivers (lower work rates / lower information rates) leads to more sleepiness and poorer concentration.

This is a recurrent theme of Smith's. I begs the question of whether a driver who is incapable of concentrating at lower speeds is a fit person to be in charge of a motor vehicle, it is also at odds with my own personal experience, which is that the slower you drive the more hazards you notice. Paul Ripley also says much the same, recommending keeping to within the speed at which you can process all the available information.

Longer exposure to accident risk due to longer journey times.

 * Some accident risk on the roads is time-based. Where journeys take longer, the time exposed to danger is increased. This effect must be quantified and allowed for. It is especially relevant for "fell asleep at the wheel" type accidents, which are likely to be more prevalent due to reduced stimulation.

This assertion has no evident basis in fact. Quite aside from the fact that a driver who is on the point of falling asleep should stop and sleep, the number of junctions, vehicle interactions and so on is essentially independent of the speed - all you achieve by speeding up is increasing the relative speed at which a collision can happen.

If you genuinely believe that you can't stay awake without driving at excessive speed then you urgently need to stop driving before you kill someone. The need for constant panic responses - which is what Smith is referring to here - to maintain a base level of concentration is a sure sign that you should not be on the road.

Reduced rate of driver skills acquisition

 * Higher driving speeds are a ‘stressor’ that promotes experience learning. Without the stressor it is highly likely that vital experience based skills are acquired more slowly or not at all. Trained Police drivers are frequently heard to say: “You never really learn to drive until you are travelling well over the speed limit.” Of course, that’s not to say that people should be encouraged to driver well over the speed limit. However the effect is certain to be present to a degree in less extreme circumstances. Any reduction in the rate or extent of experience learning reflects directly in reduced average driver skills.

The public roads are not a race track. This claim, which is based entirely on assertion and not on documented fact, ignores the very real truth that most drivers invest nothing financially in their driver skills development over time. To suggest that speeding on public roads represents an investment in skills, rather than simply selfishness and a failure to perceive the act as having any significant negative consequences, is a bold assertion and one which requires genuine evidence, not arm-waving.

If it is genuinely the case that fast driving is a necessary part of skills acquisition then we had better all get off the roads until we have completed off-road performance training, especially on a skid pan, in our current car, because after all every car has different dynamics.

You may think it more likely that this claim is simply self-serving bollocks. I might be tempted to agree.

Dangerous overtaking of lorries due to ‘HGV40’ enforcement

 * Having different speed limits for different classes of vehicles on the same stretch of roads, particularly single carriageway roads, and rigorously enforcing the speed limit of the slower vehicle types, leads to “trains” of the slower vehicles and frustration of the faster ones. This can lead to riskier overtakes than would otherwise occur, exacerbated further by the overtaking vehicle also having to be wary of blipping over the limit in the act of passing in case of a camera.

The cause of dangerous overtaking is bad risk assessment and excess impatience - the speed imperative - not the vehicles being overtaken. It is unarguably the case that anyone who overtakes unsafely simply because the vehicle in front is not going quite fast enough for them, is not a safe driver. No safe driving manual in the world would advocate this behaviour, all advocate the precise opposite. To blame the slower vehicles for the behaviour of unsafe drivers is intellectually dishonest.

Responsibility effect 1

 * Reduced individual driver responsibility in general and for choice of speed in particular, leads to a reduced tendency for drivers to reduce speed when necessary.

No evidence is offered to support the existence of this purported effect, and the substantial reductions in traffic fatalities that coincide with the reductions in urban speeds in the first decade of the 21st Century directly contradict this assertion.

Priority Distortion 1

 * Drivers' priorities are distorted. (i.e. speeds are set to legal limits rather than for safe driving reasons). We believe that millions of drivers have come to regard the speedometer as a barometer of safety; it is no such thing.

Some people believe that the universe was created by an invisible and undetectable, that does not make it true. No evidence is produced to back this assertion, and it is directly contradicted by the fact that in order to pass a driving test - itself widely recognised as a woefully inadequate standard on which to base what amounts to a lifetime's driving - you must first demonstrate the ability to drive within the speed limit. Smith's claim is basically that this ability to drive within the limit without excessive concentration on the speedometer is, apparently alone among driving skills, reduced with experience. According to Smith, then, only fast driving increases necessary skills, and by driving fast we apparently lose the ability to drive slowly. It will be interesting to see if any research evidence ever backs this up.

Priority Distortion 2

 * Maintaining a legal speed may sometimes instantaneously be more important to a driver than observing or dealing with road hazards.

Maintaining a legal speed will never under any circumstances be more important to a safe driver than observing or dealing with road hazards. So Smith's assertion covers only dangerous drivers.

The case that we should not enforce rules because dangerous drivers make bad decisions is not, to me, a compelling one. Apart from anything else, it lacks any obvious evidential basis.

False safety beliefs

 * Messages that imply that exceeding the speed limit is dangerous come with a misleading counterpart – the implication that if exceeding the speed limit is dangerous, then observing the speed limit must be safe. It certainly isn’t true, but we believe that most of us are affected from time to time and millions are affected continuously.

I am not aware of any campaign that has ever stated that as long as you remain within the limit, you are safe. I am aware of campaigns that, entirely accurately, inform drivers of the consequences to victims of higher collision speeds. I am not aware of any research that shows safety to be improved in general by exceeding the speed limit. I am aware of research which shows that, for the same types of road and traffic conditions, both probability and severity of crashes rises with speed.

The question once again is: is speed ever more important to a safe driver than safety? I think not. A safe driver can drive safely at any speed. Unsafe drivers need to be slowed down to mitigate the risk they pose to themselves and others. Smith's entire thesis here is that he, as a self-declared safe driver, finds that obeying the speed limit is burdensome, and therefore the speed limit and speed enforcement is wrong. Even if he were the safest driver on the road, this would not be a valid inference.

Oversimplified messages

 * Modern road safety messages tend to focus on very simple ‘dos’ and ‘don’ts’, Stick to the speed limit. Wear your seatbelt. Don’t use your mobile phone. But road safety needs sophisticated messages such as. Develop your skills. Learn to better manage risk. Take responsibility for your actions. To some extent speed camera policy is responsible for this oversimplification of road safety messages.

Um, no. Safety campaigns have pretty much always been single-issue. it is not clear how, for example, speed camera campaigns influenced the campaign which started in 1971. As with so many of Smith's claims against enforcement, the two supposedly related factors probably have a coincident cause (in this case the impossibility of conveying complex messages in very short films or on billboard signs) but that does not make the relationship between the factors themselves causal.

Driver distraction

 * Drivers pay less attention to the road ahead because they pay much more attention to speed limit signs, the speedometer, speed cameras and indeed anywhere where there might be a speed camera. While this effect is small in any individual driver, it probably amounts to several percent of total national driver attention. This is a huge hidden danger and, given the size of the population, it is certain that there will be cases where this low level but widespread distraction coincides with incident development.

This is a bold claim but no evidential basis is cited. It may be the fact that drivers who are determined not to slow down do these things, but the idea that a safe driver who is comfortable with the idea of complying with the law would be so distracted is unproven. Apart from anything else, inability to detect changes in speed limit should in most cases lead to a fail on the driving test, certainly both failure to observe the limit and failure to make adequate progress are fail criteria. As ever, Smith fails to explain how the ability to drive both safely and within the limit mysteriously declines with experience when everything else supposedly improves.

Traffic displacement

 * Traffic diverts to less safe roads due to enforcement on busy routes. Of particular concern here are dangerous ‘thrill seeking’ groups of road users. It’s known that such use of the roads carries a high risk and it is only reasonable to assume that anyone intending to use our roads in their own vehicles in such a way will actively seek out camera free routes to get their thrills. So the traffic displaced by cameras includes highly risky groups. This may contribute very strongly to an illusion of speed camera benefit and a significant escalation of risk elsewhere.

The phenomenon of drivers diverting to less safe routes is known as "rat-running" and has been documented in the road safety literature since at least the 1950s. It is a theme in Buchanan's  report of 1963. Smith provides no evidence of any link with cameras, and no evidence that the roads to which the drivers divert are any more or less likely to have automated enforcement.

As of 2010, average speed cameras are present on the. Is this more likely to increase or decrease high speed motorcycling on that road, which has featured on top ten of most dangerous roads in the country ever since such records began?

Thrill-seeking is normally associated with particular roads with particular characteristics - like the Cat & Fiddle, legendary among motorcyclists since time immemorial as a place to get your knee down and occasionally die. Smith offers no evidence that enforcement on such roads leads the thrill-seekers to take their fun somewhere more dangerous - they were pretty much in the most dangerous place already, that is rather the point.

‘Race-away’ behaviour

 * It has been admitted at the highest level that crashes beyond speed camera sites sometimes increase as drivers or motorcyclists ‘race away’ from the speed camera site and ‘crash on the next bend’. It remains to be determined if this effect results primarily from anger or over-excitement.

Once again the behaviour of dangerous drivers is being used as a justification for not enforcing the law. No safe driver would do this, and it seems quite likely that the drivers who speed away from the camera site would be just as likely to speed through it were the camera not there.

The solution is simple: average speed cameras. Presumably Smith would have no problem with that.

The risk of accidents directly caused by enforcement.

 * It is well known that, irrespective of speed of free travel, some drivers automatically brake or slow down significantly for a camera. If a speedo check reveals a speed within the limit, but is carried out at the same time as panic braking by the vehicle in front, the gap between vehicles can reduce to dangerous levels requiring severe braking and can result in loss of control and collisions.

This is a slight variation on Smith's recurrent claim that if you happen to be looking at the speedo at the time a child runs out, you are more likely to hit the child. Aside from the fact that no published research makes any such claim, it rather misses the point that if you are observing the speed limit (i.e. you know what it is and you are within it) then you won't need to do this. It also blames crashes which are actually caused by a combination of dangerous driving (failure to observe the speed limit followed by late and therefore excessive reaction to cameras) and following too close, on the cameras.

A driver who has no idea of the limit, doesn't know how fast he's going, can't spot the warning signs and the big yellow box in time, and brakes sharply to a foolishly low speed. Is this an indictment of the driver or the enforcement camera?

Injuries caused by enforcement hardware – fixed cameras.

 * Any street furniture is vulnerable to collision with any road user. Often we see bent traffic light support poles and street lights. Keep left signs are more vulnerable. The extra poles to support speed cameras are not exempt from such incidents. Given the additional reinforcement to avoid vibration and to render them vandal-resistant they are actually particularly dangerous.

The number of traffic signs, street lamps, bollards, trees, bridge abutments and other impedimenta cannot be less than the tens of millions. There are only a few hundred fixed camera sites. Yes, I have seen people calling for the removal of "dangerous trees". They are rightly ridiculed.

Average speed cameras promote speedo obsession

 * Average speed cameras were promoted to overcome the argument of the blip overtaking being more risky thanks to fixed “instantaneous” enforcement. The side effects are far worse though. People are more obsessed with their speedo in such zones, leading to dangerously low attention spans to the real hazards.

Another evidence-free assertion based on the idea that experienced drivers are less able to keep safely within the speed limits than learners. The quietest car I ever owned was a Vauxhall Senator. At 50mph it was still possible to tell whether you were speeding up or slowing down, it was not necessary to spend your whole life looking at the speedo in order to confirm this. Of course if you are in the grip of the speed imperative and convinced that if your average speed is only 49.9 instead of 50.0mph then your dick will shrivel, you may indeed suffer from this supposed obsession, but I don't think safe drivers fit that mould. A safe driver will easily be able to keep their speed between, say, 48 and 51mph through an average speed enforced 50 zone based on the usual combination of sensory cues, including engine note, road noise, wind noise and of course the occasional glance at the speedo. And guess what? That still leaves a comfortable margin for error before you hit the prosecution guidelines.

So, as with so many of Smith's assertions, this is based on the idea that fundamentally dangerous drivers - those to whom absolute speed is more important than safety - should be allowed to dictate that we do not control their illegal behaviour. Put in those terms it doesn't sound quite so reasonable does it?

Average speed cameras promote close proximity driving

 * It is highly visible in motorway road works sections overseen by SPECS average speed cameras that vehicle drive in close proximity to one another, both close following and long periods of side-by-side driving are extremely commonplace as vehicles match their speeds to the exact speed limit. Side-by-side driving increases risk because in the event of an incident there’s no escape space.

As drivers slow down and drive through roadworks, they close up. Headway distances always decrease with speed, it's not just a fundamental rule it's also the reason why the M25 variable speed limits have achieved a significant increase in flow when speeds are reduced - the optimum throughput of a multi-lane highway is at something like 50mph, and modelling of this is one of the most complex areas of traffic engineering. Turbulent flow analysis is the field, and it is very much more advanced than rocket science - described it as "the most important unsolved problem of classical physics".

So while it may be the case that drivers follow more closely in speed-limited roadworks, and speed-limited roadworks also have average speed enforcement, no evidence is provided of a causal link, and it is much more likely that they are simply part of a much more complex and larger cause and effect system.

Poorer public / police relationship.

 * Road traffic policing has already had a clear effect on the public's perception of the police. Many law-abiding people only come into contact with the police over road traffic issues. The blunt nature of the law, with eroded presumptions of innocence, eroded right to silence and absolute offences frequently leads to the Police being seen in a bad light. This degrades the Police’s ability to deal with all crime, and especially means that Police road safety messages are regarded with less respect. One particular strong contributor are the frequent cases reported in press where it is perfectly clear that the Police cannot obey the speed limit laws either (when off duty, or otherwise unable to avail themselves of their speed limit exemption).

Let's start by getting one thing perfectly clear: there is no such thing as a law-abiding speeder (see the Road Traffic Regulation Act sections 81, 86, 89 & schedule 6).

Nor is there any eroded presumption of innocence. Evidential speed cameras, like evidential breath testers, mechanistically detect absolute limits. It's an imperfect system but it's the one we have.

Reduced respect for law.

 * The vast majority of UK drivers regularly exceed the speed limit, and this includes the legislators, Police (without an emergency need) and court officials. When a citizen faces conviction for a speeding offence he knows full well that those responsible for convicting him are regularly guilty of the same offence. The hypocritical application of the law brings it into serious disrepute and the ultimate consequences can only be guessed at. There are many ways that reduced respect for the law can bring new dangers to our roads.

This is in fact two arguments in one.

First, Smith implies that enforcement of the law against a majority of drivers who freely admit breaking it, reduces respect for the law. There's a small problem here: if they respected the law, they'd not have broken it in the first place - As JS Dean put it in 1947, "the position has long since passed far beyond the limits of ordinary law-breaking and become an exhibition of national degeneracy." Speeding is not, as some would have you believe, the safe driver occasionally being nabbed for doing a tiny bit over the limit; the drive to enforce limits is a response to the documented fact that a significant majority of drivers deliberately and consciously break the law as a matter of course, and by a significant margin. Few people genuinely trying to keep within the limit will be in excess of the ACPO prosecution guideline of limit + 10% + 2mph; people drive to the (well-known) prosecution guideline, and the complain when they are found to exceed it. It is hard to argue that enforcement represents a lessening in respect for the law.

Second, Smith asserts that those prosecuting the offence are held to different standards. They aren't. Yes, Police officers, even senior officers, have been convicted of speeding offences. Not many, and given the immense noise generated by the anti camera lobby you can be sure that every single one gets full prominence. It does not look as if serving police officers are anything like proportionally represented among speeding convicts, and I would be surprised if they were - any senior officer is likely to have stern words with members of his force who are caught breaking the very laws they are paid to enforce.

Reduced confidence in official road safety messages

 * The false safety messages surrounding the Department for Transport’s speed camera campaign result in substantial public disbelief and loss of confidence. The opportunity to communicate valuable road safety messages is being eroded.

Or is it that the assertion by the likes of Smith that the message is false is responsible for this purported effect?

There was similar noise about the drink-drive laws and attendant campaigns, but nobody now seriously believes that their driving performance is enhanced by a drink or two.

Road user groups are set against one another

 * Some cyclists and pedestrians are noticeably angry with car drivers because they have been persuaded that ‘exceeding the speed limit is dangerous’ and they know that exceeding the speed limit is commonplace. This growing animosity is very bad for road safety which depends on co-operation and consideration.

I quoted briefly above the words of JS Dean of the Pedestrians' Association, written in 1947: "The trouble with the facts about the law-breaking of the motorists and the motor interests is that there are too many: it is difficult even to grasp them. In fact, the position has long since passed far beyond the limits of ordinary law-breaking and become an exhibition of national degeneracy." Where is the evidence to support Smith's assertion that this attitude is the result of speed enforcement? For sure, pedestrians' advocates point out that urban speeding is dangerous to pedestrians. It is. Provably and unequivocally so. As noted above, a car travelling at 40mph will still be travelling at 30mph at the point where it would have stopped had it been doing 30mph in the first place. Smith was fond of coming up with unlikely scenarios somehow attributable to speed enforcement that made the 30mph driver brake later, but no evidence of such scenarios has ever been independently published. Pedestrians have always criticised motorists for driving too fast ever since the dawn of the motor car. Britain's first motoring fatality was, after all, a pedestrian casualty of an out-of-control car.

Cyclists, too, have always taken issue with motorists for much the same reason and for much the same period. Nor is the attitude of drivers to cyclists provably connected to enforcement, since the major criticism of cyclists (that they break some laws) is always accompanied by tap-dancing around the documented fact that most drivers consciously and deliberately break other laws. TRL report 549 of 2003 says this:

No reference to enforcement, it's all about social "out groups", largely a result of the enormous cost and consequent emotional investment of private motoring. And the major cost of motoring is finance and depreciation. Car companies know this and they invest millions in advertising to reinforce the message that buying a car has made you in some way better. Cyclists, by apparently opting out of this (and it's a false appearance, cyclists are actually more likely than average to own cars) are rejecting a message into which an immense amount of money and reinforcement has been poured. Seeing someone brazenly making their way around without a car induces massive.

Reduced interest in road safety and safe driving

 * Modern road safety is onerous even to responsible drivers. They are being pushed away from an interest in the subjects, rather than invited in. People who are disinterested are less likely to acquire skills, less likely to investigate the subject and less likely ultimately to perform well. While it will certainly be very difficult to determine the road safety consequences of such effects, it should be obvious that they are real and that they are negative.

Pardon me for being sceptical, but I'd be interested to see any evidence that this interest in road safety ever provably existed.

For the rest of it, the whole point of motor safety is that it is a complex multivariable system. There has probably never been a time when every collision had a single simple cause obvious before the event. As Minister of Transport under Churchill, opined that "[road accidents are] the result not of the taking of large risks, but of the taking of small risks very large numbers of times."

Smith, of course, by relentlessly promoting the idea of speeding as safe, or at least not bringing additional risk, is responsible for encouraging the taking of those small risks - possibly very large numbers of times.

Journeys take longer and cost more.

 * Most speed limits and speed limit enforcement has little effect on journey times, but there are important exceptions. One such exception is unnecessary enforcement of the HGV 40mph speed limit on single carriageway trunk routes, which can add as much as 35% to journey times for both HGVs and other road users. This loss of economic efficiency ultimately means less money in the economy to invest in schools and hospitals.

Oh really. A greater cost than, say, the additional fuel burned in accelerating to 50 and maintaining that speed when you have the aerodynamics of a house brick? And this accounts for any measurable proportion of the total cost of operating a goods vehicle distribution fleet? Numbers, please, audited by an independent body.

Reduced incentive to train drivers better.

 * The more we characterise drivers as “incompetents who must be regulated” the further we move away from the previous “individual responsibility” system of road safety that served us so well, providing excellent reductions in road casualties until about 1993. The present course of speed reduction tends to lead us to neglect the basic sound idea of obtaining improved safety standards by training to improve drivers’ attitudes and road safety culture.

First up, the new supposedly terrible system produces the same "excellent results" or better.

Second, there has never, to my recollection or knowledge, been any sustained effort to encourage continuing education for drivers, so the use of automated enforcement can't possibly have impacted that.

Third, regulation does not imply incompetence. Air travel is probably the most heavily regulated form of transport, and pilots are among the most highly trained and competent operators of any kind of machinery.

More lawless drivers - false number plates, improper registration, no insurance, car cloning etc.

 * It's obvious that we already have drivers who neglect or evade registration and other legal requirements in order to evade modern dumb speed enforcement. Once they have decided to behave outside the law we suggest that they may well behave in a more dangerous manner, and for example, might be much less likely to stop and render assistance after an accident. They are hit and run drivers in the making.

It's obvious is it? Ah, at last a citation: the International Journal of Because I Said So. Well, unlicensed, uninsured drivers without VED do exist and always have. So what drives this? Is it fear of being caught? Or is it the perception that without a car, regardless of your fitness to drive, you are nobody? Do you think car thieves are motivated by the chance of being caught on camera when road racing in their own cars? Most of them don't have their own cars.

The largest single source of uninsured driving according to the Motor Insurers' Bureau and the Association of British Insurers is middle-class parents mis-stating the main driver of a car driven by their children.

So this is just bogeyman stuff, random disconnected facts with an asserted but entirely unproven link. Sure, there will be a small number of banned drivers who can afford VED, insurance and the like, but use cloned plates anyway to evade bans. Is that the fault of cameras? Look at the number of people who are arrested for driving while disqualified due to alcohol, often being intoxicated on subsequent arrest. All we learn here is that middle class people, too, can become so vested in driving that they are unwilling to accept the consequences of repeated infractions. I don't think that's much of a revelation and I don't think that the few hundred fixed camera sites have any meaningful effect on the number of people driving illegally. There is some evidence that recession increases this.

Perhaps the solution is to reduce the stigma associated with not driving.

More safe drivers convicted with possible loss of job / home etc.

 * Whatever way you look at figures, it is clear that the vast majority of modern speeding offences are carried out by drivers who will never be involved in an excessive speed accident. One might reasonably infer that many cases of exceeding a speed limit take place in safe circumstances where no actual danger is caused. The Law has to be subservient to Justice. Applying the weight of law in these cases has serious consequences and sometimes results in loss of job. There probably already have been suicides triggered by the consequences of speeding convictions.

You don't lose your job for a single speeding conviction. Not even two. You need multiple convictions within a minimum space of time in order to lose your license, and even then job loss is far from certain - companies sometimes insure against this and provide drivers, in other cases a banned driver might hire someone independently to drive for them - young drivers offered the chance to be at the wheel of an expensive car insured by someone else.

As to suicides, Carl Baxter slashed his wrists after being convicted for deliberately running down a man and his child on a bike. I don't think that's a reason not to enforce the law in that case, the fact that people fail to think through the consequences of their actions, to their subsequent regret, has always been part of law enforcement.

But even that is to ignore the most glaring error in Smith's argument: it begins safe drivers. Are safe drivers really being banned? Is the sort of person who will not slow down even though he knows there is a risk of automated speed enforcement actually a safe driver? I have Paul Ripley's book. He says that safe drivers always observe the speed limit. I think I'd rather believe Paul Ripley than Paul Smith here.

Honesty and accuracy in official road safety messages suffers

 * ‘Speed’ has been deliberately demonised in support of speed camera policy. Huge claims are regularly made for crash reductions at speed camera sites, while the very important regression to the mean bias is never accounted for. Department for Transport insists on adding contributory factors for ‘inappropriate speed’ (a driver quality issue) to ‘speeding’ (a legal compliance issue) in order to exaggerate the importance of legal speed controls, and to justify their flawed speed camera programme.

Really? Speed enforcement dates back well over a century and has been resented, in much these terms, for pretty much all of that time. Speed is "demonised" because it is a necessary cause of serious injury and death - slow moving cars simply don't have the energy to cause much damage, whatever Smith may say to the contrary.

The campaigns that stick in my mind are those that speak of the consequences of hitting a child at various speeds. They are accurate and honest. Painfully so. I slowed down to well below 30 even where it's not legally required in some places because of those adverts. Unfortunately not everybody has sufficient imagination (or rather sufficient sense of reality and their own fallibility) to realise that yes, it can happen to you - even if you are entirely blameless for the collision, the severity of the outcome is almost always your fault. Road traffic collisions account for one in ten child injury admissions to hospital but half of those which result in fatality. Do you really want to be sitting there with a policeman patting you on the shoulder and saying never mind, it wasn't your fault?

Yes, speed does kill, genuinely. Not always and not inevitably but a motor car is probably the single most dangerous thing most of us will ever control.

Heavy load on Courts and CPS

 * Some solicitors have become notorious for clearing the name of certain celebrities who have been accused of speeding. This, along with websites such as Pepipoo encourages aggrieved motorists to become recalcitrant, raking over all details and having their day in court. The system is becoming overloaded.

Ah, so now the convicted criminal throwing his toys out of the pram is a reason for not enforcing the law? Pull the other one.

Camera overuse is leading to legal challenges to laws. Useful laws may be lost as a consequence.

 * Sophisticated members of society are mounting challenges to the law surrounding speed cameras. A particular current example is the ongoing challenge to the requirement for a vehicle owner to identify the driver at the time of an alleged offence. At the time of writing we’re awaiting a verdict from the European Court of Human Rights (ECHR) in the famous ‘right to silence’ case. If Section 172 of the Road Traffic Offenders Act 1988 is lost due to this case the Police’s ability to act in serious cases of hit and run crashes will be reduced.

A badly framed law has always been open to challenge. But see thatg word "overuse"? Yes, you're absolutely right, he's begging the question again. It happens that the first offence for which anyone has tried this on was speeding. If it wasn't speeding then surely it would have been one of the "useful laws" (yup, you;re right, begging the question again) instead.

Public danger, expense and resources consumed by speed camera vandalism

 * Speed camera vandalism is common. It’s possible that cases of vandalism will directly endanger the public. It’s certain that such vandalism will take funds from the wider economy with a small effect on funding available for public services including schools and hospitals.

A small effect? Try negligible, especially compared to, say, VED evasion, which has a direct and measurable effect on the exchequer in a way that removing cash for reinvestment by camera partnerships really does not.

And of course this is another example of Smith blaming the camera for the inappropriate and potentially dangerous reaction of criminals. I wonder why they do it? It's not as if there are websites fulminating against speed enforcement or anything, is it? Oh, wait...

A boom in number plate theft, car cloning and neglect of vehicle registration requirements

 * One of the fastest growing motoring crimes is number plate cloning. A number plate from a car of the same type and colour as the target car is made up and fixed to the target car. Any camera flashes it collects go to the registered keeper of the original car to deal with. Camera based enforcement positively promotes this sort of activity as people come to realise that they might be better off working outside of the system. This in turn consumes significant Police resources which are then unavailable for road safety work.

Just to knock out some more begging the question, speed enforcement is road safety work.

Now, Smith asserts that plate theft is "one of the fastest growing motoring crimes". The last time this was widely discussed appears to be 2004 when there were 33,000 recorded number plate thefts. The number of parents illegally claiming to be main drivers of cars used by their children is said to be in the millions. There are over 100,000 uninsured motor claims per year. So, if it is (or was) fast-growing, it's from a low base to a figure which is still small by comparison with something that rather a lot of middle-class parents seem to be doing.

In other words, it's another cite to the International Journal of Because I Said So.

Joy riders, drunks, reckless and lawless drivers unaffected.

 * ''If we get to a point where we have figures to extrapolate road safety improvements from speed reductions, it must be remembered that not all road users will be affected by enforcement schemes. Joy riders are in an innocent motorist’s car – camera paperwork goes to an entirely innocent person. Drunks are far less likely to be stopped as a result of the substantially reduced police presence. Reckless drivers who operate outside the law – cloned plates and throwaway cars – will not be touched by the cameras."

Actually SPECS cameras collect number plates so can pick up cloned plates, but did you see how this began? If we get to a point where we have figures to extrapolate road safety improvements from speed reductions. In other words, we say this doesn't happen but even if it does it's still wrong so there nyer.

Honestly.

Authorities are reluctant to risk ‘proving themselves wrong’

 * You would think that the side effects listed here would have been considered and evaluated by the authorities. But they have not been. We believe that this can only be because they consider that the results may prove that flagship road safety policies are ill founded and have failed.

This is a particularly good example of the fallacy of begging the question: Smith might just as well demand to know why the authorities have never set up any measures to combat the UFO menace, the answer is the same.

The most likely explanation especially given the number of times these and similar arguments have been asserted, is that the authorities have simply not accepted these claims, probably because they have not been published in a peer-reviewed journal where their merit (or otherwise) can be discussed by experts in the field rather than dilettante amateurs like Smith.

Failure to accept often far-fetched notions as gospel is something which, in Smith's world, always equated to wilful denial rather than valid difference of opinion, still less even the remotest possibility that Smith was wrong, because his faith in his own beliefs was unshakeable and he was the only authority that was not "fraudulent", at least according to him. I always thought it odd that the same organisation could produce so many "fraudulent" reports which purely by coincidence made the link between speed and risk and then one single shining gem that again purely by coincidence can be interpreted by those determined to do so as supporting the anti-enforcement lobby's case, although of course the organisation's dispute of that interpretation is right back in the "fraudulent" box. It was almost as if the judgement of what was good or bad was founded on how well it agreed with Smith's pre-existing agenda rather than any objective assessment of the work...

Professionals in any field tend to be somewhat dismissive of dilettante amateurs who call them frauds but can't bring themselves to publish these claims in journals where they can be dissected and analysed.

And even if the proof had been there, which of course it never was, why would it surprise anyone that politicians are unwilling to admit to their mistakes? Look at the Scott Report, where it became clear that the then government would rather withhold vital evidence of innocence in a criminal trial than admit that a minister lied to parliament. "Authorities are reluctant to risk ‘proving themselves wrong’" is simply a statement of the blindingly obvious and is not especially relevant to road traffic law enforcement any more or less than to any other area of public policy.