It shows that modest, swift, carefully metered punishment is a lot more effective than the more traditional slow, but heavy punishment.

This is a big deal, because it shows a promising path to dramatically reduce the US prison population. Don't mind the drunk driving, think about all of the crimes out there, and all of the people who got a hammer dropped on them.

What little evidenced-based parenting advice is out there[1] suggests the same thing: taken individually, immediate minor punishments are as effective as major punishments. The big advantage comes with the fact that the parent can be more consistent as minor punishments are typically less work than large ones. Consistent annoyances are better at provoking change than occasional calamity.

1: If you thought the story for evidence-based medicine was bad, just skim through the parenting books at a book store.

> taken individually, immediate minor punishments are as effective as major punishments

The result here is that frequent minor punishments are 72% more effective than infrequent major punishments (see the article on Hawaii's 5 year study of this approach linked to by https://news.ycombinator.com/item?id=10064425).

My point was a single minor punishment is about as effective as a single major punishment. It's more complicated than that of course, (sometimes a single major punishment is more effective, sometimes a major punishment is actually counterproductive), but in any event, frequency seems to be much more important than severity.

We can take this data to a legislature and lobby for new treatment plans. A parent's hunch will not carry the same weight.

Evidence is currency, and so are demagoguery, posturing, campaign money, endorsements, promises, actual political platform, public sentiments, etc. Sadly, evidence does not trump all else. Luckily, even a modest amount of evidence currency can tip the scales of the outcome.

I would like to draw your attention to a media campaign waged in recent month by forces unknown to reduce harshness of the US criminal justice system. It's pretty much every week that I get to read an article about in the New York Times, and they all refer to some sort of research being released, or a small investigation into a particular injustice. Now we get this article from the Wall Street Journal. You know it's on when both the NYT and the WSJ converge on something. The tide is turning rapidly away from "Tough On Crime" to "Measured Response", and this evidence will be a contributing factor in this sea-change.

People become more steadfast in their opinions in the light of evidence.[1]

1. http://youarenotsosmart.com/2011/06/10/the-backfire-effect/

Yes. All the ridiculous advice my partner and I get about our children is unbelievable. And I'm someone who had to self-diagnose medical issues after years of being ignored by doctors and other "experts" who wouldn't look at evidence-based approaches. I thought it couldn't get worse. But it does. Oh, it does... so much worse.

Even as someone who's not a parent, when I have done babysitting for said parents, I'm able to use the Naughty Step without fuss.

I will say that I've found the Love & Logic series less terrible than most of the mainstream stuff, and you'll likely be able to find a local class, which (assuming you're not a single parent) can be huge on getting you on the same page with your significant-other. That made a big difference for me, as my wife and I were raised very differently.

A few snippets of advice:

* Operant Conditioning[2] works both ways: your child's behavior will affect you, and you will subconsciously avoid things that are distressing to you. Not taking this into account can cause a lot of problems.

* One thing at a time. e.g. don't potty train and teach table manners at the same time. Partly because it's harder on the kids, but mainly to prevent burnout for the parents.

* Focus primarily on having a positive relationship (versus a lot of training & discipline) with kids under about 2. It seems likely that Martha Sears[1] went overboard, but there is definitely some phenomena like "attachment." (Also, Attachment Parenting, Attachment therapy and Attachment-based therapy are all different things, and the terminology is not always agreed upon)

1: https://en.wikipedia.org/wiki/Attachment_parenting

2: https://en.wikipedia.org/wiki/Operant_conditioning

Most parents start out not really knowing much about the day to day details of being a parent. Your children learn really quickly how to achieve their aims.

As a parent the thing you do learn is to choose your battles. This is the cause of the eldest's lament that their parents were so much stricter with them than their younger siblings.

The real interesting part of this was the sobriety program. They found a way to identify bad habits at a relatively low cost. How to extend that model to other crimes is unclear. Parole basically exists for that purpose already, so the question is how it can be made more effective.

I just don't really see a clear path to general crime reduction here.

It probably does work better for something habitual and detectable like alcohol use, but I could see how it could be applied to gang membership and other forms of thuggery. Maybe spending our resources on light consistent touch programs would help more than the significant resources needed for full-time incarceration.

I'd be very nervous about extending this beyond drunk driving.

With DUIs, breathalyzers are essentially "proof" of the crime (to the extent that there is ever incontrovertible proof for any crime). And if you refuse a breathalyzer, you've violated the implied consent law, so there's no ambiguity there either. You may be able to get off on a technicality, but it's rather difficult to abuse this, short of a police officer falsifying breathalyzer results outright.

But for most other crimes, there's a lot of ambiguity, which is why we have a separation between LEOs and the court system. I'm already uncomfortable with the level of unchecked power that LEOs (and non-LEOs, like the TSA) have today. I really don't want them to have any more leeway with detention of civilians.

That is rather than send a person to jail upon conviction, a judge would put them on electronic probation, where each violation results into a small, immediate penalty. This will help in two way - prevent harsh sentences for simple missteps, and actual train people to follow the rules of the law by shortening feedback loop of the events.

It is unfair if you believe that the jail is supposed to rehabilitate people. It seems to me that people believe it in theory but don't believe it in practice.

For example, if they got convicted for second-degree murder because they had a hot-tempered friend in college who started a fight, they tried to held, and then the friend pulled a gun and shot someone. That's the kind of mistake some (many?) people are not likely to make again. But the record is permanent. And due to Fundamental Attribution Error, the world is inclined to believe that they were involved and did what they did with intent because they had and continue to have a flawed character.

I hate to be that guy, but the US prison population is large because certain powerful interests want it to be large. This is unlikely to change anytime soon. And it definitely won't change as a result of a study about drunk driving.

That said, it's still an interesting study. Like many people, I've lost a friend in an accident involving a drunk driver, so the subject is important to me.

1: https://en.wikipedia.org/wiki/California_Correctional_Peace_...

Taking a hint from dog training, if you punish a dog too late, they forget what they're being punished for. Also the severity also doesn't seem to matter. Bad is bad no matter how loud or how painful, so just keep it to a minimum and have it for convenience. Beyond that rewards work much better for training brains.

Consider:

> Hindsight will lead us to systematically undervalue the surprisingness of scientific findings, especially the discoveries we understand—the ones that seem real to us, the ones we can retrofit into our models of the world.

http://lesswrong.com/lw/im/hindsight_devalues_science/

Because the sentence against an evil deed is not executed quickly, therefore the hearts of the sons of men among them are given fully to do evil. - (Ecclesiastes 8:11)

Consider the case where a drunk driver gets home safely while obeying all traffic laws. Or even the case where a drunk driver is swerving- she can already be pulled over under a reckless driving statute.

A long time ago I read a statement from a judge sentencing two defendants for felony murder. They both shot and a man was dead, but there was a debate as to which bullet actually killed. "The only difference between the two was aim, and I am not going to let one off for being the poorer shot."

The more appropriate word for such laws are "regulatory crimes". They are meant to regulate behavior to prevent or reduce an evil that does not occur in all cases. This covers everything from practicing medicine without a license to drug possession, neither of which necessarily cause a physical harm in every case.

What you call luck, others call knowing their limits.

Basically just because someone hasn't been hurt, it doesn't mean that a behaviour is acceptable in society.

Besides that, you can't have your cake and eat it too. Why is manslaughter different than murder? The only difference is the intent--the end result is the same (someone's dead). When the state of mind is the same, why should luck in the outcome change the punishment?

> If you don't see the difference between an accident and doing something on purpose as requiring different punishment, well then you just fundamentally disagree with about all of mankind. If your car tire blows and you hit another car and kill someone through no fault of your own, apparently you think that's an equal crime to murder. Really... c'mon... really?

You miss the question here.

If I am driving my car and I take y attetion off the road for a moment to adjust my car radio, and then plough into another car...

1) ... And kill the driver of that other car

2) ... And injure, but don't kill the other driver

Why should y punishment for (1) be more severe than for (2)? My intent was the same, my poor behaviour was the same. Perhaps the difference between causing death and not causing death is the quality of the crashed driver's car.

Murder requires intent to kill. Man-slaughter also requires intent to cause harm. That's why there are seperate laws for vehicular death - causing death by dangerous driving or vehicular manslaughter. It's hard to meet the burden of proof required by murder or manslaughter.

It's a balance between the modern concept of punishments based on state of mind and actions, and the oldschool concept of an eye for an eye, punishment based on outcome.

Why do we punish people criminally? For doing bad things with malicious intentions. Intention is the most important thing in criminal law, because a guy who shoots at someone is equally a bad, culpable person worthy of punishment whether or not his aim is very good. On the flip side, someone who kills someone while driving in an accident that could happen to anyone is not worthy of punishment while someone who kills someone intentionally with their car is worthy of being punished as a murderer. Even though the end result is the same--the intention is what matters.

That's why drunk driving is illegal in and of itself. You make a conscious reckless decision to put other people at risk. Even if you don't hurt anyone, you're guilty of that act and that recklessness is a form of malicious intent justifying punishment.

If you are a novice driver and you choose to drive, you are consciously making a decision to put other people at risk.

There are different skill levels of driving. A highly skilled driver drinking can easily still be more skilled than someone who hardly ever drives or has poor reflexes in general.

So what are we punishing for, reducing your effective skills and then driving? If so, why don't we just charge everyone with a crime who fails to drive once a week to keep their skills sharp?

Agreed, however, the problem is what's considered "bad" isn't a matter of fact but of opinion. What is law is a matter of fact, but legal/illegal != bad/good.

Also, please stop calling it drunk driving, the law pertains to more than just alcohol which is why it matters and why it's unjust. Getting in a car and driving while sober can still get you a DUI because the law is stupid in defining what intoxicated/impaired means, this is especially important now that marijuana is legal for recreation in several states. Just because someone gets a DUI does not mean they were in any way reckless, beyond that it's just as reckless to drive tired as to drive intoxicated yet no one argues that should be DUI worthy, additionally "reckless" is a matter of opinion, not empirical fact.

The law is not logical, it is political, a popularity contest, not a logically evidence based means of helping society. There's certainly logic in keeping law self consistent for sure, which is where lawyers and judges have a role, but what becomes law has little to do with what is actually just. As you're a lawyer, you know this already.

Law is absolutely logical, you just have to be willing to look at the premises. And it is absolutely a structured expression of what people think is just.

The premise of criminal law is that it's an action taken with malicious intent that makes someone a criminal, not just causing a bad outcome. The law follows that logic to its conclusions. For example under the Model Penal Code, an attempt at murder is punished the same as a completed murder. Someone who, with intent to kill, points a gun and shoots at someone is no less criminal if he misses than if he hits. Criminal law is about punishing culpable conduct. It's illogical to punish two people differently who engage in the same culpable conduct because circumstances outside their control lead to one result versus another.

Now, you can disagree with the basic premise and apply a different logical framework. Just because the premise of the law is that intent is the most important thing does not mean you can't believe something different, such as actual harm being the most important thing. But you gotta wrestle with the implications of that. If harm is the most important thing, killing someone should always be murder. Shouldn't matter whether you hit a jogger wearing all black at night or whether you ran over your boss in broad daylight on purpose.

In my experience, the people who complain about the law being illogical and not the same as justice are the ones abandoning logic. They know what resuls they want, based on what "justice" means to them, (X should not be illegal) and get upset that's not the law. They don't take the time to look at the premises underlying the law to see if the rules logically follow from those premises. If you do that, you'd be surprised how often you conclude "well I think that premise is incorrect, but I can see how the rule follows logically from that."

Ps: also, reckless is a precisely definable concept: when conduct causes a measurable rise in the risk of some negative outcome. Where the line between acceptable and unacceptable rise in risk can't be precisely defined and must be established by social concnsus, but that doesn't make it an illogical concept as you imply by calling it an "opinion."

Great, but I am, and that's the disconnect. I agree with most of your comment, but the issue isn't that the premise is wrong, it's that the implementation is wrong. When I say that law isn't just, that's what I'm referring to.

The law may be logical in its premise in the abstract, but the actual implementation of DUI laws strays far outside those logical premises, the real world doesn't match the abstract. You're talking about what the law intends to be, I'm talking about what it actually is because imho that's what actually matters. I don't care about the good intentions behind the laws, I care about those being fucked by the poor implementation of said laws.

We don't have the highest incarceration rates in the first world because our laws are just, we have it because they aren't.

> Where the line between acceptable and unacceptable rise in risk can't be precisely defined and must be established by social consensus

That's just rephrasing what I said, something that relies on social consensus "is" just opinion and is not precisely definable, and by that I mean it isn't empirical, rather it's a popularity contest, i.e. political.

You are implicitly saying that it does harm. Indeed, it does harm the psych, which might lead to physical harm in the long run.

So do people who are tired, or old, or stressed, or any other number of reasons. People kill people all the time, singling out one group and basically charging them with pre-crimes is not just. The world is a dangerous place, jailing people so your "psyche" feels good is not just. The world is dangerous, how about you accept that and act accordingly rather than supporting jailing those who haven't hurt anyone.

That being said, generally speaking, a drunk driver who happens to not do anything wrong on the road is extremely unlikely to have any legal issues. Mostly they are going to have problems when they start swerving or exhibiting signs of being impaired. A tired driver who is swerving wildly is also likely to get pulled over.

> That being said, generally speaking, a drunk driver who happens to not do anything wrong on the road is extremely unlikely to have any legal issues.

That might be true if it weren't for things like DUI checkpoints, which make it untrue. You can be perfectly fine to drive and still get caught and punished because of what are essentially baseless mass searches.

If you meet the objective bar (no lawyer pun intended, though see California's attempt at doing this for the legal profession), then yes, you should be able to practice even without a multi-year degree.

It's just not possible to study at home and be educated enough to be a doctor. You need huge amounts of expensive resources (cadavers, equipment, live patients, etc.)

Only if one doesn't understand that medical schools are a cartel. There are far more people qualified and willing to be doctors than are allowed to be doctors, cartels have to keep their supply low to keep their wages high.

But this makes no sense for medical school, which teaches a lot that would be extremely difficult to test directly. Often the direct test would basically amount to going to medical school for some non-trivial amount of time anyways.

But the barriers inherent in medical training are really far and beyond pretty much any other field -- you need a bunch of dead bodies and a lot of very expensive equipment. By the time you pay for just the materials necessary for the series of tests and exams that would reasonably certify someone to be a doctor, you might as well make a degree program out of it.

So even if someone could get all of the book learning and also find a gaggle of doctors to study with and learn from, Medical School is one of the very few professional programs where having a degree program would still make sense, totally independent of learning and development, just from a logistical standpoint. In that sense Medicine is pretty unique (unless you count graduate degrees in experimental sciences, I guess).

Here's another comment from you in this thread:

> Only if one doesn't understand that medical schools are a cartel.

If all you have to fall back on is name calling, that makes me believe that your position is not very sound.

Heck, you could even argue that a particularly good driver and/or someone with a high alcohol tolerance may be safer even when impaired than some drivers are unimpaired; perhaps then the oldschool impairment tests - walk in a line, touch your nose, etc., are fairer than the theoretically more impartial breathalyzer. I'm not sure how I feel about that, but it seems like a reasonable argument at least.

It would also give police an opportunity to remove other dangerous drivers from the road, e.g. certain elderly or sleep-deprived drivers.

If you're in a VR driving sim with a cop standing next to you, you're going to pay attention and (virtually) drive, constantly paying attention to the speedometer and road signs. Rather than flipping through stations while turning around to tell a joke to the people in the back seat. Or just zoning out on a routine drive and not realizing you're 40 over.

It's why people often say "I'm a better driver when I'm drunk" -- not true for sloppy drunk, but probably true for slightly drunk because the fear of DUI causes them to intently focus when otherwise they would be on auto-pilot not devoting their full attention to the operation of a deadly machine.

No, we know that mild intoxication increases risk. A rule of thumb is 4 uk units increases risk by four times. A uk unit is a small amount of alcohol - 4 small glasses of weak wine (125 ml of 8% ABV * 4 is 4 uk units).

We know this because we put people in simulators and test them.

1. Regarding the last sentence of my prior post, I think you're also underestimating the extreme danger associated with inattentiveness while sober driving and/or how common distracted sober driving is. I 100% believe an very attentive person at .081 is less dangerous than an idiot texting and messing with the radio while driving. Which is to say, both are extremely dangerous.

See e.g. http://distracteddriveraccidents.com/texting-driving-dangero...

2. You / the grandparent are under-estimating the impact that attentiveness has in drunk driving accidents, and also underestimating the ability of drunk people to be attentive on demand. The VR test is not a good test of impairment because the drunk driver will be artificially attentive during the administration of the test.

Driving drunk is dangerous even before motor control is significantly impaired. And the reason for that is mostly psychological -- a bit reduced reaction time, okay, but the reduced attentiveness and judgement are really far more dangerous. The guy flying down the road 40 over "just wasn't paying attention". It's not that he would have noticed but didn't because "delayed reaction time". He never would have noticed because he was drunk and didn't have the good sense to pay attention while driving.

The problem with the VR test is that, when put on the spot, a drunk person can focus for a short or intermediate period of time. So that VR test isn't an accurate reflection of the actual typical behavior of an impaired driver.

Under-estimating the only-on-demand abilities of drunk people would be a deadly mistake when designing measurements of impairment. Measuring ability, not behavior, is not enough. Drinking changes behavior in ways that are dangerous while driving for for which a drunk person can often compensate (e.g. when confronted with a potential DUI, but often not when driving home on a routine route).

10,000 people killed annually is more than the number of soldiers killed in the entire war Iraq and Afghanistan over a decade.

If we only punish people for behavior that results in injury or damage then you're relying on each and every driver to correctly assess the risks of their behavior. People are notoriously bad at doing that, so any punishment will be ineffective as a deterrent.

It's a victimless crime, all right - but only until the next victim.

We need to obey the rules of the road so other drivers can reason about our behaviour. We could have no rules, but it would be much less efficient. This is the prisoner's dilemma and the law enforces cooperation.

Consider the case where a shooter sprays bullets in a crowded building wildly, but no one is actually hit.

Rightly or wrongly, drunk driving is considered endangerment of others.

endangerment comprises several types of crimes involving conduct that is wrongful and reckless or wanton, and likely to produce death or grievous bodily harm to another person.

Similar to crimes like attempted murder, we do not allow endangerment even though it may be the case that no one was hurt, because Person A took a high risk with Person B's life and we do not consider that acceptable.

Jail time is not really appropriate. Really, you should just lose access to public roads. But license revocation is adulthood revocation for all but the privileged few in SF, NYC, and maybe Chicago. In most of this country you can't get groceries or primary medical care without driving, and you certainly can't hold a job or provide a normal life for your children. A five-figure fine and even a month or two in jail is many times less severe than permanent loss of driving privileges; that's life-destroying (pretty much your only option is to drop everything and move to a high-cost-of-living area of a high-cost-of-living city to get adequate public transit coverage).

This is almost certainly true in America, but keep in mind that some other societies view driving somewhat differently. America represents less than one twentieth of the world's population. In major European cities, at least, car ownership seems to be the exception rather than the rule.

It requires some walking and biking, and some employment options do disappear. But most of those employment options are recoverable because the salaries (e.g. at suburban office parks) pay well enough to cover the cost of a cab from the nearest bus stop to the office every morning.

IME the lack of public will to walk a half mile, ride a bike in the rain, or wait ten minutes at a bus stop is by far the greatest barrier to use of public transit in America.

To reiterate:

IF you are not working in an office park off an interstate near an affluent suburb, most mid-sized American cities have enough transit to get by.

IF you are working in an office park off an interstate near an affluent suburb, then you're probably making enough that you can stitch together a combination of public transit and taxis to get to work from a low-rent area.

And yeah, long commutes suck when you're paying for child care. Which is why a lot of people choose a child care facility close to work.

Basically, my point was that I know a lot of Europeans whose commute involves > 1/2 mile of walking. And also a lot of Americans who live in cities that "don't have any public transit" because they have to walk 1/2 - 1 mile to get to a stop that takes them directly to work.

I'm not saying American public transit doesn't need to improve. Just that if Americans always demand door-to-door service, then public transit will never be good enough. And, more over, will always suck because that's an impossible demand.

Japan, for example, has reasonably high per-capita car ownership, but the yearly mileage is less than half of US usage.

I think it's reasonable to suspect that the standard deviation in the US is much higher too.

1 https://www.fhwa.dot.gov/ohim/onh00/bar4.htm

Travel via the common means is a right, not a privileged; such laws are unjust. That driving requires you make such promises is absurd.

Rights aren't unconditional.

> That driving requires you make such promises is absurd.

Here's why driving requires you to make these types of promises:

http://www.who.int/gho/road_safety/mortality/traffic_deaths_...

This is a common misunderstanding.

Rights are unconditional is the sense that they apply to everyone equally. But almost all are conditional in the sense that they don't give carte blanc permission without exception.

It might help to think of rights like axioms. (A & (A -> B)) -> B is true unconditionally. It is an axiom. But it is also conditional. It is universally true, and yet, B is not universally true.

Nearly all rights we have are unconditionally conditional in this way. They always apply to everyone (universally), but that doesn't mean there aren't conditions (that get applied universally).

For instance, being convicted of murder triggers conditions on most of the rights we have as citizens. Not all of them, but most.

Privileges are different from rights because they don't have to apply to each person equally in the way rights do. The two are separated by the fact that rights are extended universally and privileges aren't, not by the presence or absence of conditions.

Obviously, I apparently did not.

Therefore:

> I'm disagreeing with it as a matter of principle.

Please don't do that. Redefining words for yourself and then using your secret meanings in discussions is both irritating (because it wastes other commenter's time) and comes off as arrogant (because it demonstrates a clear disregard for other's time and a willful unwillingness to communicate effectively).

If you know something is a Right in the conventional sense even though it has conditions attached, state that you disagree with those conditions. Don't call it "not a right" to make a point. Doing so derails the conversation and wastes other's time.

> licenses are just another way for the state to make money and restrict my right to travel via the common means of the day.

That's obviously not true. You really think your $20 license fee pays for... anything? It's probably not even offsetting the cost of the licensing regime.

And restricting your travel obviously is not the goal of a driver's license. Where did you even get that idea?

"The State" doesn't require licensing because it wants to screw you out of money or keep you down. We, the people, require licensing because driving is dangerous and we don't want idiots to kill us.

> Driving should not require a license and an annual fee to the state

Of course, assuming you live in a democracy, you're free to try to change that.

I think you'll find this an impossible reform because given the danger of driving, licensing is an extremely important first step toward making roads safer.

> It didn't require a license to ride a horse or drive a buggy

Yeah, right, because horses and buggies didn't kill millions of people a year.

As an aside, in almost every state you can still ride horses on most public roads without a license.

We used to live in a democracy, we now live in a plutocracy, the government hasn't represented the will of the governed in quite a long time, that's why voter turnout is abysmal.

We're done here, we obviously won't come to any agreement.

Horses were taxed. Carriages were taxed.

http://www.genguide.co.uk/source/taxes-miscellaneous/177/

Etc.

Oh look, problem solved.

When intoxicated with alcohol your reaction times and concentration level are slowed. You may obey all road rules perfectly on your normal drive home when drunk, but one night something may be a bit different - a train crossing that isn't normally used, an inattentive driver who pulls out slightly two late, or a pedestrian who crosses a bit slower than most people - and you're now in an accident that would have been avoided had you been sober and had your normal reaction and reasoning skills.

It's not the stereotypical swerving red-light-running drunk driver that we target with drink-driving laws, as you've mentioned that's already covered by dangerous driving laws.

I'm not saying you're wrong, per se, but suggesting that you're right due to the absence of evidence is premature.

That said, while I don't know how official these numbers are, the NHTSA attributes 100,000 accidents to driver fatigue per year, and while the victim count is currently lower than your assertion for drunk driving, I think that driving while fatigued is underreported for the reasons I mentioned above.

Here's part of the problem. The actual statistic is that 10k deaths occur where one driver had a BAC of 0.08 or higher. It's purely an assumption that these deaths would not have occurred if the drivers were sober. Some, certainly; some, certainly not. We need better information.

The discussion lacks "scientific integrity", as Feynman would put it.

Driving while tired is definitely an issue though.

Yet. If we are ever able to reduce DUI to almost non existent the focus will move on other incident causes. I am sure there are already some people who would love to place camera inside every car to check if driver was paying attention.

I was in physical therapy next to someone who wasn't going to walk again because of a chronic drunk driver.

Distracted driving / texting while driving should be punished on the same level as drunk driving IMHO. It's really dangerous and until it's actively enforced people are going to get hurt often.

I ride my bike to work and it is so dangerous because of people texting. I've seen several accidents happen because one driver is texting and the one in front of them has to stop.

90% or so of the fatalities in london are caused by cyclists undertaking HGV's who turn into them.

That must be what people mean with the whole spraying bullets into a crowd thing. It's a victimless crime, except for the poor idiot that just had to lean into one bullet.

A cyclist undertakes the HGV - passes on the left. The cyclist is attempting to reach the cyclists junction box.

While making this manouver the traffic lights change from red to red-amber and then green. The HGV driver checks the mirrors but does not see the cyclist (who is in the HGV blindspot) and pulls off, turning left, crushing and killing the cyclist who is trapped between the HGV and the curb.

I've tried to word this neutrally. It's a tragic situation and there's learning for cyclists and HGV vehicles, and road designers here.

This isn't a tragic situation. It's a probable situation, and we know the root cause. It's the driver, of course, since the burden of lawfully operating his vehicle at all times falls on him, no matter the particularities. And it's the vehicle maker, because clearly the HGV isn't fit for traffic.

Cyclists need to understand that deliberately riding in unsafe ways to save a couple of minutes is dangerous.

Its more like going into a hells angels bar and picking a fight and being surprised you end up on hospital

That sounds to me exactly like distracted driving.

Another way to think of it is that you aren't punished for drunk driving, but for breaking the law, and the law has been decided by determining that drunken driving provides unacceptable risk overall. Whether individually a person would have caused harm is irrelevant, we create a enforce laws for the good of all, and without enforcement, or selective enforcement without due process, we have a slew of other problems.

Given the context of the surrounding discussion, I didn't think anyone was talking about situations of accidental distraction while driving.

I was a bit unclear in what I was doing though.

Firstly, I'm not sure what "victimless" means exactly, and I think you may be taking it too literally. As people in the many comments below have pointed out, there are plenty of crimes which are "victimless", e.g. taking a machine gun and "spraying" a crowd of people with bullets, but failing to hit anyone. Is this a "victimless" crime? At what point do we decide to set the bar?

Come to think of it, there are charges of "reckless endangerment", where someone it can be a crime to do something which endangers other people in a reckless manner. Drunk driving certainly falls under those, so I'd say that drunk driving laws are simply enhancements of "reckless endangerment" laws, providing more specific context.

Secondly, we as a society will often choose to make our laws based on certain factors, like what an average citizen knows For example, driving under the influence causes a variety of physical changes that are not obvious and not known by every single driver, e.g. slower response times. These are things that a reasonable driver might not even realise are impacting them, and might not feel when they are partially drunk. But we, as a society that has done scientific tests, know that these effects always happen. In this case, it makes sense to block actions that we know are harmful.

You've just terrorized a crowd of people, that's hardly victim-less; plenty of people there to complain about direct harm you've caused them, perhaps a few who will suffer years of PTSD. Plenty of lawsuits about to happen, victims by definition.

> At what point do we decide to set the bar?

If no one is aware enough to try and sue you, then you haven't hurt them, and thus your action is by definition victimless.

You seem to think DUI means alcohol only, it's a bigger issue than that and it's an absurd law that can get you in trouble literally days after you've been sober because it applies to all substances, not just alcohol. We have several states now with legal recreational marijuana, think about that. Many things people do while driving are far more influential on their response times than being "intoxicated", being tired is far more dangerous, being old is far worse on response times. DUI laws are unjust.

Plenty of drunk people have a nap, wake up a few hours later and think they're not drunk and drive, even though they still are drunk.

What is the alternative? Being frowned upon?

You say about how people automatically agree drunk driving should be a crime. But I always use to wonder how blindly we trust others on the road to behave responsibly. How often we leave our lives and our loved ones, to the fair judgement of a stranger.

Driving on roads will be impossible, without that trust. When you are drunk, and choose to drive when you judgement is poor, you are betraying the trust implicitly placed on you by every other person on the road. So isn't that something like a crime? Just thinking loudly.

I tell this as a moderate drinker myself.

The simple reason we need drunk driving laws is exactly because they are victimless. It rightfully establishes strict liability.

There is no right to access public streets while intoxicated with two tons of steel.

> Sobriety experienced not only a 12% drop in repeat drunken-driving arrests but also a 9% drop in domestic-violence arrests.

An interesting fact to consider when people talk about how substance abuse bans create "victimless crimes."

It's a little amusing to see people who have alcohol as a major part of their life respond negatively to other drugs with a much smaller social impact.

edit: or, from another perspective, let's say we find 8 hours of sleep a night, a better diet with fewer inflammatory foods, plenty of water, and an hour of exercise a day reduce violence by 40% in total. We criminalize not following this exercise program.

While it's damaging on a societal level if people don't abide by the Healthy Lifestyle, on the individual level most people would describe it as a victimless crime and say the government should not be able to regulate these decisions. (Or would they, if the effect was that strong?)

Whereas, on the opposite end of the spectrum, taking a pill with meth and PCP might increase the risk so much that very, very few people would call it "victimless" or wish to leave that decision up to the individual.

They can still access cheap easy alcohol (legally even?), like anyone can access illegal crack.

And pretty sure they don't have to do the program, they can just lose their license and take their appropriate punishment.

This is about offering a medical solution to people with a problem, not a ban.

It also catches up people without problems but as above they have chosen it as a better 'punishment' and given they don't have a problem, hopefully being caught once it will also give them a life lesson.

I don't see how this affects the victimless crime argument.

It's exactly as if you made a law forbidding people from living or hanging out together. Domestic-violence arrests would plummet, but the victimless argument would still be valid: living with someone else is not the problem, violence is.

Put another way: why are we seemingly okay with banning alcohol use for people who have had a DUI but we wouldn't be okay with banning masturbation or gay sex even for convicted felons? Because the latter are truly victimless activities. Substance use is, in contrast, an activity that creates risk to the public but is hard to regulate without major side effects. So targeted bans in substance abuse could be effective while overcoming th enforcement side effects of a gener ban, and would be legitimate because substance use isn't a "victimless crime."

A common issue with regulation is that the regulators cannot be trusted to act with wisdom or moderation or honesty. One of Washington State's own prohibition enforcement agents became its greatest bootlegger [1], for example. Roy served illegal booze to politicians, police, and citizens alike, and was treated with kid gloves when eventually arrested.

This dilemma -- of how to design a regulatory program to resist capture by the people entrusted to execute it -- makes it very difficult for me to understand how you can just hand-wave past the question of legitimacy.

Put another way: There is a train on the tracks that is going to run into one person. You could switch the tracks and save the person, but the train will then run into 5 people. What do you do?

[1] https://en.wikipedia.org/wiki/Roy_Olmstead

Vezzy-Fnord appears to be talking about the problem of bans when applied to everyone.

Also I'm quite amused how the original post referenced domestic violence and you immediately jump to it beating women.

In any event, I was referring to the use of substances and the right to determine one's own destiny.

Even the NCADV puts it at 85%, which is still under heavy dispute from researchers in the gender symmetry disciplines of IPV research. A 2010 CDC report still puts men much higher than this: http://www.cdc.gov/ViolencePrevention/pdf/NISVS_Report2010-a...

Not sure what tripe you've been reading.

> The ‘natural experiment’ of cross-cultural study finds levels of variance which rule out any direct causal effects of alcohol on behaviour. We have already noted that in some societies drunken aggression and belligerence are commonplace, while in others the same doses of ethanol result in quite opposite behaviour, characterised by calmness, passivity and good humour. [1]

[1] http://www.sirc.org/publik/drinking4.html

[2] http://www.ncbi.nlm.nih.gov/pubmed/9273914

Well, you only get hit by it if you drive drunk (and get caught) once. (Maybe that's what you meant by "targeted".) Once someone has driven drunk, you can lower the hammer on the driving part (loss of car/license), or on the drinking part. It seems less destructive to lower the hammer on the drinking.

Forcing violators to stop driving can seriously damage their lives (remove their ability to earn a living, see friends/loved ones, etc.), but hey, they're still allowed to get drunk!

It's seems in retrospect like needlessly cruel punishment, when "force violators to stop drinking" is the other option.

You could say the same of prison sentences for alcohol-related offenses. Which is more invasive?

Most violence related to illicit drugs is not caused by intoxication, but rather by fighting and stealing between drug dealers.

So there is good arguement that drug prohibition increases violence by creating a criminal black-market (where people turn to violence rather than police and laws to resolve disputes) and by encouraging alcohol use rather than use of safer, less violence-inducing drugs.

Previous violence is a better predictor. Both combined is stronger still.

Mental illness by itself is a weak predictor (vast mojority of people with metal illness are not violet) but when you combine mental illness with addiction, or previous violence, or both, you get a strong predictor of violent behaviour.

It's true that the illegality of drugs creates more violence. But that doesn't mean we should ignore the risks.

Targetted bans for individuals who have shown that intoxication makes them more likely to harm other people seem reasonable.

No kidding. It's in South Dakota, with a 10% (one of the highest) native american population. https://en.wikipedia.org/wiki/Alcohol_and_Native_Americans

" the problem still costs some 10,000 Americans their lives each year."

Imagine if terrorist bombings were killing the same number of people each year. Maybe a plane load every month or two. Would we ask the psychologists what might be a more effective way of deterring people from bombing things? Banning them from flying after they were caught carrying explosives through the airport doesn't seem to be working well enough.

> costs some 10,000 Americans their lives each year

with

> Maybe a plane load every month or two.

Would have to be 100 planes/year (at ~100 people/plan)! Imagine if you had 8 plane crashes/month due to drunk pilots!

A) Few but large scale killings - eg 100 planes of 100 people each seems worse than 10,000 cars of 1 person each.

B) A single identifiable group of outsiders that can be blamed. Most people drink so drinkers aren't outsiders.

But this last factor surprisingly doesn't seem to be a worry, as evidenced by people's willingness to drive, despite the fact that it sounds like it should induce terror:

C) Random sudden death without warning.

People actively drive the car, so they may think "I have a chance to affect the outcome", whereas being a plane passenger is basically vegetating for X hours in a very uncomfortable seat, and there is no way you can actively influence the outcome in a positive way.

Which ironically is what makes it a safer form of transport.

http://www.pbs.org/newshour/bb/law-july-dec13-hawaiihope_11-...

JUDGE STEVEN ALM: I thought to myself, well, what would work to change behavior? And I thought of the way I was raised, the way my wife and I would– were trying to raise our son. You tell him what the family rules are, and then, if there’s misbehavior, you do something immediately. Swift and certain is what’s gonna get people’s attention and help them tie together bad behavior with a consequence and learn from it.

MEGAN THOMPSON: These seemingly simple reforms in Hawaii soon produced remarkable results. An arm of the department of justice funded a study five years after the program launched. That study found that compared to people in regular probation, HOPE probationers were half as likely to be arrested for new crimes, or have their probation revoked. They ended up spending about half as much time in prison. And were 72% less likely to use drugs. The results from Hawaii caught the attention of criminal justice experts across the nation.

This is the same idea applied to people on probation. The short overview - typcally when people on probation commit an infraction (fail a drug test, fail to meet probation officer) there's no consequence the first time, or the second or the third. Far down the line the consequence happens - the person goes back to jail for a couple of monthes or years.

The alternative method this program uses is that the first problem results in going back to jail - but for a short period of time, the way I read about it was on the order of a weekend. As it turns out immediate inconvenience is way more effective then a long term possibility of justice coming down on you like a ton of bricks.

Like the first poster said this is about way more then just drunk driving. I think it's about more then just criminal justice.

In each case the drunk driver was FAR from their first offense and one of them even had a suspended drivers license but that didn't stop them.

Prison time to get them sober BEFORE they kill someone is the only answer I could come up with, but judges never seem to think it is a problem, even after they kill someone.

Neither of them got any significant sentence (just a few years) for literally murdering someone, it was like the law sees the other person on the road as taking their own chances or something just for driving.

scare tactics and huge fines (associated with pullovers and checkpoints which are super low probability events) just don't work, that's for sure. i know of people who literally drove drunk for YEARS until uber came along, then stopped entirely.

In my early twenties I would occasionally drive home “tipsy” — what felt like tipsy at the time; in retrospect, I was as drunk as a skunk. But there were no alternatives! I lived in a spread out western city, taxis were unheard of.

Threats and punishments don’t work on young people who have a psychological sense of invulnerability. The best way to curb drunk driving is through practical solutions that have immediate and obvious effectiveness.

All that said, I’m impressed with the psychology behind this program, and it’s nice to see policy that aims at curbing and solving destructive behavior instead of profiting off it in the form of fines.

Are you seriously suggesting you had no better alternative than to drive drunk?

There were at least three alternatives available: Don't drink, get a ride with someone sober, or walk/bike/etc.

To argue "but you knew when you were drinking that 3rd beer!!..." is all well and good, but the point of going out is to enjoy your friends company and have fun, not carefully monitor your alcohol consumption. The lack of taxis, buses, trains, etc in many states is a major contributor to drunk driving.

I think it’s unacceptable for me to drink anything at all when I plan on driving. That way I have never ever been in a situation where I had to drive while being drunk. It’s a quite simple heuristic …

I think it's acceptable to drink some amount before driving because I understand the physiological processes that turn alcohol into not-alcohol and therefore that at some point after I drink, I will be OK to drive. If that wasn't the case then having a drink would be a big decision, after which a person could never drive a car again.

The question is proximity. If we go for a few beers and then walk next door and have dinner (without beers) then I think it's safe to drive home. An hour drinking two beers, an hour eating dinner, and there should be little to no alcohol left in your system when you drive. Fine upstanding citizens do this all the time.

People are going to drink, they are going to do it in social environments beyond their homes, and they are going to return to their homes. A realistic solution involves recognizing these centuries-old truths and minimizing the inherent risk. Water flows downhill, and hitting it with a stick won’t change that.

So, yes, I would be for this program if the government agency administering it absolutely forbade all other law-enforcement from using their data to investigate people, and somehow I believed that they would really a) attempt to do protect the data, b) be capable of protecting the data, and c) allowed people freedom of movement.

http://www.sciencedaily.com/releases/2015/01/150127212158.ht...

If you're caught cell-driving, you either pay a large fine, say $750, or you submit to a program that bans talking while moving faster than 5mph for the next two months. The cell carrier will monitor the ban and issue citations by SMS.

So if you opt into this program, and then you're driving down the road, start talking on the cell phone, and then 30 seconds later BANG! you get a text message: "That call was $75 payable to the city hall before the end of the day". Yeah, you can't talk on the bus and stuff, but that's the "punishment" part for ya.

It's worth a try, I think.

The trick with a $75 fine for texting while moving is passengers. If you're willing to ban texting entirely while moving though, using technology to assess small fines could work!

The key is to training new habit and making it stick, according to the article, is to make punishment moderate, certain, and immediate. We can't do that across the board due to the limitations of technology, legitimate use, privacy concerns, presumption of innocence and so on. But we sure can do that to those who broke the law, especially when they opt into it.

There are three options that can be used alone or together.

1) Technology solution. Put digital eyes around the area of the steering wheel or otherwise pointing at the driver. Report drivers that frequently use a phone while driving. We can trivially build a phone identifier for a car today, given the drastic improvements in recognition technology. We could also easily judge drivers by their poor attentiveness to the road. This approach will freak people out in regards to being nanny state invasive.

2) Community shaming, education programs. This was half the battle with getting people to use seat belts regularly.

3) Traditional police enforcement. Cops start pulling people over for it more frequently, and hitting them with on the spot $100, $200, $500 fines. Perhaps it increases with each offense within a certain amount of time. The problem here is, the radical majority of people will simply slip through the enforcement cracks and not be caught. It would also be open to easy abuse, cops stuffing the budget with lots of fines. People see seat belts as a life saving device for the most part, so voluntary compliance mostly worked, but I'd wager most people do not see their phone habits as potentially life threatening. That requires greater disincentives to reach the same compliance, or a vastly larger shaming & education project.

> he penalties [...] are modest. Drinking results in mandatory arrest, with a night or two in jail

Which will make you loose your job. Not that modest to me.

Its a question of custom/culture. You need peer pressure that effectively says "only shtheads endanger others by climbing behind the wheel drunk" (forgive the language). That is what has an impact on the target audience...not fines (big/small/immediate/delay...)