I think it’s also quite plausible that cellphone use hasn’t much increased the number of serious collisions leading to death, but has significantly increased the number of rear-enders and injury-causing crashes. It seems to me that it’s still rare to see a red light run by a driver on a cellphone, but it’s very common that the same driver would rear-end another car when traffic stops suddenly.

If no one drank 20 years ago, and today almost everyone who could drive a car drank, and data shows no substantial change in accident fatalities, even in the face of incontrovertible proof that driving drunk is a seriously bad idea for numerous reasons, and there is evidence some people are driving drunk (a few cars now and then spotted with cocktail glasses casually held out the windows, riotous parties in the back seats, etc) then one could (should!) indeed question the entire business of whether or not “drunk driving” is worth caring about to the point of a special law against it.

Of course, today we have utterly undeniable evidence of the fact that drunk driving is creating a massive pile of human bodies (I’ve selected one of thousands of hits here):

http://www.alcoholalert.com/drunk-driving-statistics.html

“In 2006, there were 13,470 fatalities in crashes involving an alcohol-impaired driver (BAC of .08 or higher) – 32 percent of total traffic fatalities for the year.”

Now, in the face of this kind of data, I don’t think anyone needs to conduct experiments as to why driving drunk is bad, model the chemical changes that create the deficiency in the brains of drivers, and stand before legislative committees begging for a law against drunk driving. The menace is clear, the decision is obvious.

Does cellphone use have similar support? Well, we have the instant report on the low-level cognitive stuff cited above, that shows the risk. But then we have real-world data. Here is a typical report:

http://www.ncsl.org/print/transportation/2006cellphone.pdf

Executive summary: even after years of research and monitoring, direct reports say about 1% of all crashes are cellphone related, but unspecified critics — perhaps bothered because the death toll isn’t as high as they would ‘like’? — argue about under-reporting.

Maybe. However, I would say that if using a cellphone while behind the wheel is as demonstrably dangerous as driving drunk, we would see the effect clearly, and it would closely track the deployment of cellphone usage in the broader population. Consider (say) 10% of all traffic deaths are because someone was on a cellphone. At rates near about 1995, we would expect, then, today, an excess of about 4200 people per year … or something like 46-47 thousand people killed on the roads in the USA.

Well, this appears not to be the case. More so, the fatality data is surprisingly flat — I haven’t run the tests, but to my eyes it isn’t totally out of the question that the real incidence is statistically indistinguishable from zero. (A statement, if true, doesn’t deny these kinds of accidents/deaths occur, but simply says that they are well inside the random noise of data gathered to date.)

Put it this way: it’s easy to spot someone who’s using a cellphone while driving. They struggle to keep in their lane, brake late & then too hard, and then drive too slowly for the traffic conditions as they compensate.

All the same sort of errors that are typically seen in DUI cases.

These don’t sound like a recipe for ‘good’ driving to me.

This is just a useless distraction. Here is a better question:

http://www.bridgeratings.com/Images/Cellphone/Numb%20Cellphone%20UsersTrend65.jpg

Over the last 20 years or so, we have a cellphone using population that probably went from exactly zero to virtually everyone who can drive a car.

http://geoff82.files.wordpress.com/2006/12/traffic-fatalities.gif

Over the same interval, the number of fatalities per year has remained essentially unchanged.

The Question:

Given cell phone use while driving is incredibly dangerous, where is the excess human wreckage?