Scharf

Steven M Scharf is a relative latecomer to the helmet debates on Usenet. He describes himself as &quot;one of Earth's leading experts on bicycle helmets&quot; although it was not possible to probe this claim because as soon as he was challenged to provide evidence for his views, or show that he had read any of the research, he started killfiling anyone who challenged him.

Scharf is a master of logical fallacies, both using them and accusing others of using them. More of that in a moment. He also asserts that the techniques successfully used in the campaign against the Martlew bill cannot possibly work, because it denies what in his mind is a fundamental truth, that helmets work pretty much as advertised. He is unable to account for the experience in Australia, although there were some pretty impressive mental gymnastics used. This is an excellent example of the helmet zealot mindset and is informative as such.

The important thing to note here is that, although Scharf represents himself as &quot;the voice of reason&quot; (a siren call of bigots the world over, but that's by the by), his chosen examples for logical fallacies are always taken from the sceptical camp, even when the equivalent example from the pro-helmet camp is both clearer and more prevalent. In many cases he resorts to caricaturing sceptical arguments in order to make his point; this is mainly because, unlike pro-helmet arguments, so few sceptical arguments are couched in absolutist terms. Certainties are the preserve of the helmet zealot, not the sceptic. Yet Scharf requires the sceptic to disprove the zealot's case, rather than the normal scientific method which is to place the burden of proof on those proposing a theory.

There is only one unprovable theory commonly used by sceptics: risk compensation. This is described as a theory because it is impossible to prove as scientific fact. There is, however, a substantial body of evidence supporting the theory (including evidence specific to cycle helmets), and it is widely accepted in road safety circles as the most plausible explanation for a number of otherwise very problematic observations. The following is an analysis of his site bicyclehelmets.info as of mid April 2005.

Site Content Commentary

Bicycle Helmets Myths and Facts - A Fair and Balanced View of Bicycle Helmets This is a nod in the direction of Fox News (aka Faux News) and their slogan &quot;Fair and Balanced&quot; - which is suspected by many of being as ironic in intent as is Scharf's claim.

Purpose This site grew out of Usenet discussion of a proposed mandatory helmet law (MHL) for the province of Ontario, Canada. There is a great deal of debate, and a large number of personal attacks, regarding this issue. This site will present both sides of the issue in a clear, unbiased, manner, without all the acrimony. As you will see, it presents only one side. But that is an aside. The real purpose of the site was actually to give Scharf a forum for promulgating his views without being challenged, since each time he posted the arguments below he was soundly rebutted. He has done the same in respect of cycle lighting, a subject on which he also claims to be one of earth's leading experts. His settled view there is that no 6V dynamo lighting system can be adequate for safety at night. The more he was rebutted, the more he resorted to faux-&quot;balanced&quot; comments on his website along the lines of &quot;I am right and these others wrong&quot;. This is in line with his policy of killfiling those who challenge him effectively.

Why do I Bother? There are two reasons.

First, aggressive helmet opposition can convince some people that helmets are worthless. There is no controllable factor which is more closely linked to increased cyclist safety than helmet use. It would be wonderful if there were unlimited funds to widen roads, provide education to cyclists, and install bicycle friendly road features, but this money doesn't exist. Scharf appears to have bought into the idea that cycle facilities are the only way to make the road environment safe for cyclists. As John Franklin suggests, this is a false premise.

His assertion that &quot;there is no controllable factor more closely linked to increased cyclist safety&quot; is simply not supported by the evidence. Controllable factors include: reducing motor traffic speeds, training of drivers and cyclists, infrastructure changes, maintenance, conspicuity and lighting, and other interventions. Very few studies rate these for effectiveness, but all those which do (of which I am aware), place helmets last. Logical fallacy: proof by assertion.

Finally, &quot;Aggressive helmet opposition&quot; or &quot;anti-helmet activists&quot; are largely a figment of the zealots' imagination. I am one of those Scharf includes in this list; you will known by now that far from being &quot;anti-helmet&quot;, I own and often use one. I know only one or two people who are actually anti-helmet, and they are not active on Usenet.

Second, I am very much against compulsion. But the misinformation promulgated by the helmet opponents will almost certainly be used by advocates of mandatory helmet laws as evidence that the population is too stupid to act in their own best interests. Scharf's core argument here is that by arguing that the evidence is not uniformly in favour of helmet use, campaigners against compulsion will lead legislators to conclude we don't know what's good for us. In other words, we can only win by colluding in the conspiracy of silence regarding conflicting evidence. It appears that we may argue against helmet laws on the grounds that they are illiberal, but must ignore the documented fact that in Australia and New Zealand, they had no provable effect other than to deter cycling.

Essence of Arguments for and Against the Use of Helmets

Argument in Favor of Helmet Use Helmet advocates quote studies that measure the relative severity of injuries of helmet wearers versus non-helmet wearers. They have selected the tiny subset of cyclists that sustain injuries in accidents, and they conclude, correctly, that helmets reduce the severity of injuries, and reduce the fatality rate, when accidents occur. The helmet advocates employ the ignoratio elenchi fallacy*.

Argument against Argument in Favor of Helmet Use Experienced cyclists are rarely involved in accidents where a helmet would provide a real benefit in terms of safety. While most helmet proponents are not in favor of mandatory helmet laws for adults, their advocacy of helmet use increases the perception that cycling is dangerous and they helmets will mitigate most of the danger.

''* The ignoratio elenchi fallacy consists of claiming that an argument supports a specific conclusion when in fact it logically has little to do with the conclusion. This is one of the major fallacies behind mandatory helmet laws  ''So far, so good...

Argument Against Helmet Use Helmet opponents argue that the effectiveness of helmets should be based solely on the injury rates of the entire cycling population. The problem they have is that data that supports their position is not available. By employing the cum hoc egro propter hoc fallacy*, they are able to prove their point. First, they look the statistics in countries with mandatory helmet laws, both before and after the mandatory helmet laws were introduced. Second, they compare statistics between countries that have mandatory helmet laws and those that don't have mandatory helmet laws.

Argument Against Argument Against Helmet Use The problem with the helmet opponent's argument is that they try to equate correlation with causation, in ways that are so absurd that they are actually amusing. You cannot compare injury rates in New Zealand (which has a mandatory helmet law), and the Netherlands (which doesn't) and correlate anything! Even looking at New Zealand's statistics before and after helmet laws were enacted do not prove anything.

''* The cum hoc egro propter hoc fallacy asserts that because two events occur together that they must be causally related, leaving no room for other factors that may be the cause of the events. This fallacy is popular with helmet opponents.'' What helmet opponents? Who are these people? Name them. Who argues that measures of efficacy should be based solely on population figures? Give names and citations. This is a bit of straw man building, aimed at poisoning the well.

The point of population figures is that they give a check by which the observational studies can be measured. Observational studies predict massive savings in injuries, so we turn to population statistics to see to what extent that is borne out in practice. Answer: it is not. There is no correlation in real-world populations between helmet use and head injury rates. This is not the cum hoc fallacy but a simple reality check: proponents predict benefit, we measure whether there is independent evidence that such benefit exists. We find that independent evidence is lacking. The failure of observational studies has been documented, so this is an entirely valid criticism.

Neither are we comparing New Zealand with The Netherlands. The relevant comparison is between the entire NZ cycling population in 1993 and the entire NZ cycling population in 1994; the evidence comes not from country v country comparisons but from time series within single populations.

On what basis does Scharf assert that comparing New Zealand's head injury trends over time does not prove anything? This is argument by assertion again. Both NZ and Australia have helmet laws, they were strongly enforced at the time of passage leading to substantial rises in helmet use in a very short time, and comparing the trends for helmet use and head injury shows no correlation. If the predictions of the observational studies were anything like accurate, the effect would be glaringly obvious from the injury stats. It is not.

The Netherlands comment is a red herring, originating with the suggestion that if you are genuinely concerned with cycle safety it might be smarter to model your approach on countries with a good record of cycling safety and participation (e.g. Netherlands) rather than those with a much poorer record.

<td width="50%" valign="middle">Voice of Reason

Helmet use should be gently encouraged, but helmets should not be seen as an alternative to proper cycling technique. Helmet opponents need to accept the fact that helmets do reduce the severity of injuries when accidents occur. Helmet proponents need to accept the fact that helmets do nothing to reduce the number of accidents that occur, and that mandatory helmet laws can be counter-productive. <td width="50%" valign="middle">Scharf presents himself as honest broker, in an attempt to slipstream his bogus assumptions. The middle-ground fallacy is a technique whereby someone seeks support for an argument based solely on the fact that it lies between two extremes. In this case one of the extremes is largely illusory, and in any case the middle ground fallacy is fallacious because it provides no basis for judging the merit of either extreme. The truth could lie at either extreme or any point between them.

<td width="50%" valign="middle">The two biggest negative aspects of helmet promotion are the following: They cause uninformed people to believe that cycling is an inherently dangerous activity. They cause uninformed people to believe that helmets will mitigate all of the dangers that do exist. <td width="50%" valign="middle">There is at least one other: distraction from proven programmes and interventions which reduce injury at source, rather than seeking to mitigate some injuries after the fact.

<td width="50%" valign="middle">Helmet Use Studies I have a long list of links to helmet studies, and will be presenting many of the fallacies inherent in these studies. What is important to understand about these studies is the following: <td width="50%" valign="middle">After six months no such list is in evidence, but as ever the Bicycle Helmet Research Foundation got there first. See cyclehelmets.org for an analysis of the flaws in many of the leading studies.

I, too, have a long list of helmet studies. I have several hundred helmet papers, analyses and abstracts. Most of the members of BHRF have similar collections.

<td width="50%" valign="middle">There are no studies that do not show a reduction in injury severity coincident with increasing helmet use. <td width="50%" valign="middle">False. Example: Rodgers (1985)

<td width="50%" valign="middle">There is no evidence that helmet laws have any effect, one way or another, on the number of people that cycle <td width="50%" valign="middle">False. Example: Robinson (1994)

<td width="50%" valign="middle">There is no evidence that helmets contribute to increased torsional injuries (and no evidence that they do not, it hasn't even been studied). <td width="50%" valign="middle">False. The evidence is weak and untested but extant.

<td width="50%" valign="middle">Injury rates will never track helmet usage, in any study, one way or the other, because the studies don't account for time or distance. <td width="50%" valign="middle">Misdirection. If helmet usage rises from 43% to 95% in a single year, the effect in head injury trends must become visible then or very shortly afterwards. Even in complex multivariable systems a step change of 100% amplitude in one input will almost always be detectable in a supposedly linked output. The trends pre- and post-law for cyclists and pedestrians in NZ and Australia are indistinguishable; although it is probably possible to account for this with some mental gymnastics, Occam's Razor suggests a much simpler explanation.

<td width="50%" valign="middle">Myths and Facts about Helmet Statistics In this section I will examine the amazing myths that are promulgated by both sides in the helmet debate. I will be using a lot of material from Usenet threads on helmet use, dissecting the arguments and explaining the fallacies. <td width="50%" valign="middle">

<td width="50%" valign="middle">'''Pre-prepared &quot;helmets are worthless&quot; posts for use on Usenet. '''

To be fair, I should also have a list of &quot;helmets are the greatest thing since sliced bread&quot; posts, but no one ever claims this. The anti-helmet folks are much more into using fallacies to support their position. <td width="50%" valign="middle">Yes, you should. &quot;Helmets are the greatest thing since sliced bread&quot; posts are real and frequent, while &quot;helmerts are worthless&quot; posts are the result almost exclusively of zealots' inability to distinguish an agnostic from an atheist.

<td width="50%" valign="middle">Ad hoc &quot;I crashed and wasn't wearing a helmet and wasn't hurt, therefore helmets do not work&quot; <td width="50%" valign="middle">This is almost always posted as a response to the real ad-hoc fallacy (arguably hasty generalisation, actually), which is the classic &quot;helmet saved my life&quot; (HSML) anecdote. Importantly, I have never seen the stated argument used in any context other than as a reply to a HSML style post; to focus on the reply while ignoring the context is disingenuous.

<td width="50%" valign="middle">Affirmation of the consequent &quot;If helmets worked then no one who wears one would ever be hurt or killed, so it's clear that they don't work.&quot; <td width="50%" valign="middle">Again, this is almost always a response to a helmet zealot post. It is a response entirely designed to inspire critical thought in one who has clearly failed to apply it: they assert that helmets save massive proportions of lives, we point out that in Australia post-law the proportion of injured and killed cyclists wearing helmets is the same - 85% - as the proportion of all cyclists wearing helmets.

<td width="50%" valign="middle">Anecdotal evidence &quot;I crashed and wasn't wearing a helmet and wasn't hurt, therefore helmets are unnecessary.&quot; <td width="50%" valign="middle">Reversal again. This is almost always a response to HSML; I cannot recall it ever being advanced as an argument in principal.

<td width="50%" valign="middle">Argumentum ad antiquitatem &quot;For hundred of years cyclists did not wear helmets, so there must be no reason to wear helmets.&quot; <td width="50%" valign="middle">I do not recognise this statement. I do know of a variation on it: cycling has been popular for over 125 years, during most of which time nobody thought it especially dangerous.

<td width="50%" valign="middle">Argumentum ad hominem &quot;You claim that helmets work, but I happen to know that sometimes you don't wear one.&quot; <td width="50%" valign="middle">Reversal again. This is most commonly used against sceptics, who are falsely claimed to be anti-helmet; actually as much a straw man as an ad-hominem. It is very rarely used against helmet zealots, although there are valid exceptions</a>.

<td width="50%" valign="middle">Argumentum ad ignorantiam &quot;Of course helmets are worthless. No one can prove otherwise.&quot; <td width="50%" valign="middle">And another reversal. This is a subset of the burden of proof fallacy, actually, with a dash of straw man thrown in. What usually happens is that zealots say &quot;helmets are great, you can't prove otherwise&quot; (why should we have to? they are the ones proposing an intervention); the word &quot;worthless&quot; also amounts to a straw man since what we, the sceptics say is this: helmets do not prevent serious or fatal injuries; no-one has proved otherwise.

<td width="50%" valign="middle">Argumentum ad logicam &quot;Helmets don't reduce the severity of injuries in accidents.. Don't you know about the study that showed increasing helmet use didn't correlate with decreasing injury rates.&quot; <td width="50%" valign="middle">A non-sequitur. The two arguments are rarely linked, and nobody of whom I'm aware would make the emphatic statement &quot;helmets don't reduce the severity of injuries in accidents&quot;. Dogmatic statements tend to be the preserve of the pro-helmet lobby. One of the core messages put out by sceptics is that the evidence is equivocal, and that those who pretend otherwise should be distrusted.

<td width="50%" valign="middle">Argumentum ad numerum &quot;The vast majority of cyclists I know believe that helmets are worthless, so they must be worthless.&quot; <td width="50%" valign="middle">Reversal again. This is generally stated in the form &quot;the vast majority of doctors believe helmets work, therefore...&quot; The word worthless is also used falsely, since it is vanishingly rare for any sceptic to make such a dogmatic statement.

<td width="50%" valign="middle">Bifurcation* &quot;Either helmets prevent all deaths or injuries, or they're worthless. The former is untrue, so the latter must be true.&quot; <td width="50%" valign="middle">aka false dichotomy. I have never seen this used this way round, although I have seen it used by a zealot in the form &quot;either helmets prevent no injuries - which clearly is false - or they should be compulsory&quot;.

<td width="50%" valign="middle">Cum hoc ergo propter hoc* &quot;Cycling rates have declined in countries where helmet laws have been introduced. Clearly helmet laws cause a decline in cycling.&quot;

* Popular fallacy used in the Usenet helmet wars <td width="50%" valign="middle">According to Scharf., this is just a bizarre coincidence. Like the coincidence observed by TRL in areas where there are helmet promotion campaigns. And why not? The message that cycling is so dangerous that you must not do it without special protective equipment could hardly deter anybody from riding could it?

Scharf is accidentally right about the post-hoc fallacy, though. Almost all the pro-helmet studies embody precisely that, it is a core potential problem in any observational study.

<td width="50%" valign="middle">Denial of the antecedent* &quot;If a helmet ever saved me from injury or death then it would prove that helmets were useful, but this has never happened to me so helmets must be worthless.&quot; <td width="50%" valign="middle">Reversal again. Most statements of this form are HSML anecdotes from pro-helmet zealots.

<td width="50%" valign="middle">Ignoratio elenchi / Irrelevant conclusion Increased helmet usage didn't reduce the number of bicycle accidents or deaths, therefore helmets don't work. &quot;Widening roads, and increasing education, would reduce head injuries more than helmets, therefore helmets are unnecessary.&quot; <td width="50%" valign="middle">The former is a valid conclusion.

The latter is largely illusory, and refers to the fact that helmets are essentially a distraction form interventions which would reduce danger and injuries at source, rather than attempting to reduce the severity of injuries following crashes. Risk management 101: first reduce risk at source, then reduce exposure to risk, and only if the first two fail, apply personal protective equipment.

<td width="50%" valign="middle">Non causa pro causa &quot;Evidence shows that helmeted cyclists are more likely to hit their heads.&quot; &quot;Evidence shows that helmet promotion deters cycling&quot; &quot;Every cyclist who wears a helmet is in favor or mandatory helmet laws.&quot; <td width="50%" valign="middle">Evidence does show that helmeted cyclists are more likely to hit their heads, and two plausible explanations exist: helmeted heads are bigger and heavier, and risk compensation. To deny this evidence is futile.

Evidence does show that helmet prevention deters cycling. Given that it is pretty much universally accepted that cycling lengthens life, on average, that is an important point. The final statement relates not to the cyclists themselves but to (e.g.) the British Government's position on compulsion: that it is currently opposed primarily because wearing rates are too low, so enforcement would be problematic. Every cyclist wearing a helmet is &quot;voting for compulsion&quot;.

<td width="50%" valign="middle">Shifting the burden of proof* So if you think that helmets aren't worthless can you prove it? <td width="50%" valign="middle">This one is funny. Who is proposing an intervention? Who is suggesting a law? Is anyone suggesting making the wearing of helmets illegal? I think not. Where should the burden of proof lie, if not with those promoting helmets?

Or does Scharf think it should be up to sceptics to disprove the efficacy of aluminium foil hats in deflecting mind rays?

<td width="50%" valign="middle">&quot;Never argue with an idiot, they'll drag you down to their level and beat you with experience!&quot; I used to respond to the anti-helmet zealots on Usenet. Then I realized how hopeless it was to attempt to educate them. There is an old adage, &quot;Never argue with an idiot, they'll drag you down to their level and beat you with experience!&quot; By filtering out the posts of the core of idiots, I have cleaned up my newsreader, and eliminated the temptation to get involved in a flame war. <td width="50%" valign="middle">Zealots. Hmmm. I suppose the likes of Frank Krygowski and I are zealous in correcting ill-informed pro-helmet posts, so that might have some validity. For real zealotry, though, you need to speak to peopel who refuse to admit that a contradictory view even exists. People like BHIT, who assert that there is no possible reason why one should not wear a helmet and who simply denounce most of the evidence contradicting their favoured studies. As to educating, that would make a lot more sense if it weren't for the documented fact that many sceptics have arrived at that position having previously been enthusiatically in favour of helmet use or even compulsion. We have read the evidence and changed out minds. If the evidence ever becomes unambiguous in favour of helmet use we may well change our minds again. Zealots do not change their minds.

So there we have it: Scharf's commitment to debate amounts to putting his fingers in his ears and chanting &quot;Tra la la, I'm not listening&quot;. He starts arguments on Usenet but will not engage with those who disagree. Where in this &quot;myths and facts&quot; section is his debunking of HSML anecdotes, far and away the most common fallacious argument in helmet threads (probably the start of most of them)? Where is his analysis of the 1989 Seattle study, whose 85% and 88% figures are quoted by every leading helmet promoter? Where, in fact, is there any evidence of his debunking any of the myths put forward by helmet proponents?

<td width="50%" valign="middle">About the Web Master Steven M. Scharf is one of Earth's leading experts on bicycle helmets. He has bicycled all over the world, including Canada, Russia, Thailand, Japan, Taiwan, and China, sometimes with a helmet, sometimes without. <td width="50%" valign="middle">&quot;The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of doubts&quot; - Bertrand Russell. Here are some real experts; none of them make Scharf's inflated claim.