Okay, let's apply a gross reality check. You're an alien. You're
observing Earth with an infinitely powerful telescope from Alpha
Centauri. You have a simple question. Since 1950, has human
civilization - or American civilization, which amounts to pretty much
the same these days - advanced or declined?
Apparently the easiest way for Sam Altman to answer the question is to
trade it for a different one. He is not alone in this. He asks:
since 1950, has human technology advanced or declined? Clearly, the
alien, you, I, and Sam Altman all have the same answer to this
question.
Any question with an obvious answer is a stupid question. "Is an iPad
more advanced than a Smith-Corona?" is a stupid question. Who asks
stupid questions? Obviously, blithering idiots.
But we can compose an interesting question by factoring out the stupid
question. Which world would Sam Altman rather live in? 2013, with
iPads and teh Internet? Or 1950 - with iPads and teh Internet?
In a sense, this 1950 is just as real as the "real" 1950. Neither
exists. Sam Altman cannot pack his bags and move to either the real
1950 or my imaginary super-1950. Both exist only as thought
experiments. It is not hard to construct or define the super-1950,
though - one run of a time machine, with a printout of Wikipedia,
would be pretty much all the real 1950 needed. Send the technology
back to 1945, and you'll have iPads by '55 at the latest. Those guys
got things done.
The interesting (and scary) question this thought-experiment asks is
whether, aside from technical progress, human civilization has
advanced or declined since 1950. In actual reality, this too is a
stupid question. The answer is no less obvious - I assert.
There are at least two massive fallacies embedded in this quote (and afaict in the larger essay, though I did not read the whole thing):
- Due to better modern reporting (and the 24h news cycle), despite generally lower rates of crimes and various other bad stuff, we feel like their rates are higher now, so we get all nostalgic for the 1950s.
- More importantly, 1950 was a good time for certain kinds of people and not so much for others. Within the US, internet or no, 1950 is not a great time to travel to if you are LGBT, disabled, female, non-Christian, or ill with any of a variety of chronic conditions, and it's especially shitty if you are black (or, to a lesser extent, of another minority). Outside the US, Europe and Japan were still basically starving from post-war deprivation, the Soviet bloc was hitting the worst parts of Stalinism, China was approaching something like a thousand-year low in terms of civilisation, and most of the rest of the world were moribund colonies of the dying European empires.
If you're a middle-class able-bodied white straight Christian (preferably Protestant) man in the US, though, sure, 1950 sounds super.
So what you're saying is that most people who valorize the 1950's are assuming they'll be Don Draper and not some nameless black daughter with no economic or educational prospects whose dad was just lynched?
I think the people who valorize the 60's should actually listen to Don Draper. From some 5'th season episode, roughly paraphrased (working from memory):
Roger Sterling: "Buying a Jaguar is how you know you've arrived."
Don Draper: "I grew up on a farm. Indoor plumbing is how I know I've arrived."
despite generally lower rates of crimes and various other bad stuff
You have to be careful with statistics. One of the thought experiments posed later in that same Moldbug post is this: how much of the Earth's inhabited area is safe for a person to wander about at night, alone? And how much was in 1950? His conclusion is that much more of the Earth's inhabited area (in particular, large parts of many cities--Detroit is his example) is unsafe now vs. in 1950. Crime rate statistics don't usually get broken down to that level of detail, so you might not see the pattern there. (Although in other posts Moldbug also quotes the statistic that the crime rate in Britain today is something like 50 times higher than it was in the Victorian era.)
Personally, I like to think of real facts, not thought experiments. In terms of global poverty, disease, hunger, active wars, inequality, the year 2012 is the best year recorded in human history [1] [2] [3] [4] [5] [6]
What happens with the US (and the UK) is a global redistribution of wealth. The US economy is basically stagnating, but developing countries are booming.
Although in other posts Moldbug also quotes the statistic that the crime rate in Britain today is something like 50 times higher than it was in the Victorian era
And yet just before the Victorian era, crime was so bad that people were sent to literally the other side of the earth for crimes as minor as stealing a pair of scissors.
It would also suggest that the murder rate would have been ridiculously low - in modern times, it's 1.2/100k. In 1831, the start of the era, that would mean that the UK had a mere 3 murders of its 13 million people for the entire year. Difficult to believe. In 1900, the end of the era, it would still be less than 10 for the 30-odd million. According to this article [1], homicide rates weren't too far from what they are today, neatly bracketing the modern rate (which has the advantage of better reporting; historical homicides were likely higher)
A 50-fold reduction in crime is astounding, enough so that it sounds like baloney. Particularly for a country famous for shipping its criminals around the world during the 'low crime' period.
It would also suggest that the murder rate would have been ridiculously low - in modern times, it's 1.2/100k. In 1831, the start of the era, that would mean that the UK had a mere 3 murders of its 13 million people for the entire year. Difficult to believe.
Difficult indeed. Wouldn't 1.2 per 100k translate to 156 when scaled to 13 million?
If you're just calculating the overall crime rate, yes. If you want to calculate the murder rate, the penny candy stealing rate, and other rates of individual crimes separately, of course you can do that too. (Though you might find it difficult to find numbers for penny candy stealing.)
crime was so bad that people were sent to literally the other side of the earth for crimes as minor as stealing a pair of scissors.
You're confusing the crime rate with how crimes are punished. The fact that punishments were severe for what we today consider minor crimes was a major part of the reason why crime rates were so low.
It would also suggest that the murder rate would have been ridiculously low
Murder is only one crime; the factor of 50 is aggregated over all crimes.
Firstly, there isn't "a crime rate". You can't aggregate "a crime rate" across all crimes. Different crimes have different rates - you can't fold a murder into a rape into an assault into a burglary into a failure to pay a debt. If you think that there is "a crime rate aggregated over all crimes", then there's some fundamental things you don't know about criminology.
The thing about murder is that it's fairly unambiguous and usually recorded. Just about every other crime has interpretations and wild variations in recording - it's hard to compare across similar culture countries, let alone across time, without voicing an armload of caveats.
One example is comparing the assault rates in Australia versus that in the US. Australia has a much higher assault rate... until you look at the details. An assault in Australia carries several charges, the most severe of which sticks, but each of those charges makes it into the total stats. In the US, the stats are only published on aggravated assaults - those with a weapon or serious bodily harm. Two guys having a fistfight outside a bar will add to the Australian stats, but not the US stats. It's the same story with the same country across time - not to mention wildly different reporting rates, especially when you're considering a period where modern policing was still in its infancy. Murder is one of the most consistently defined and reportable crimes, hence why I used it.
was a major part of the reason why crime rates were so low.
That is very much putting the cart before the horse. Victorian England was rife with theft, unsurprising for a place with plenty of poverty and no welfare. I'd buy a 50-times increase in drug crime, because that's a crime now more than it was then. I wouldn't buy a 50-times increase in debtors crimes, because debtors prisons are no longer a thing and bankruptcy is no longer a crime. It's just hard for anyone with any familiarity with Victorian England to see it as 50 times safer than modern day England - this was a tumultuous time, with the industrial revolution obsoleting jobs left, right, and center. There was huge amounts of poverty, and there was legal reform at the time because the rise in petty crime against harsh penalties meant that there was a lot of dissatisfaction from the disproportionate response at the bench. There were huge problems with alcoholism (gin being very cheap) amongst the urban poor causing fights.
So, in context, is it safer to walk around Victorian England than modern-day? Well, the assault rate would have to be astonishingly low, because theft certainly wasn't, and we've already seen that murder was equal to or greater than today's rate.
Do you have a link to Moldbug's comment where he discusses this 50-fold decrease? I tried to find it as I wanted to see how it was sourced, but there's a lot of waffle - the front page is only two posts (and one announcement) that total 20k words. I have no idea where in the archives it might be.
Yes, there is. There are purposes for which it may not be the most important metric, but it certainly exists.
> You can't aggregate "a crime rate" across all crimes.
Sure you can. And if you can measure the rates for individual crimes, its trivial to then aggregate them across crimes. Whether that's useful depends on the purpose you are applying it to, but it certain can be done.
> Different crimes have different rates - you can't fold a murder into a rape into an assault into a burglary into a failure to pay a debt.
you certainly can fold the rates of instances of a specific subtypes of the event "crime" into a general rate of occurrence of the whole class, though, to be fair, failure to pay a debt is generally not a crime, its a tort, and so it would be invalid to include it a measure of crime.
> The thing about murder is that it's fairly unambiguous and usually recorded.
Its fairly unambiguous -- assuming that there is not a missing body situation -- that someone has died. It is less unambiguous that they died through homicide, and even less unambiguous still that the homicide was criminal homicide, and even less unambiguous yet that the specific subtype of criminal homicide was murder.
> In the US, the stats are only published on aggravated assaults - those with a weapon or serious bodily harm.
This is true of the FBI's Uniform Crime Reports (and publications based on those reports), but not BJS's National Crime Victimization Survey (and publications based on that survey.)
Yes, there is. There are purposes for which it may not be the most important metric, but it certainly exists.
Can you please furnish a link where someone describes 'the crime rate', actually giving a number? Preferably from an LEO website and not a sensationalist newspaper. I've not seen a 'the crime rate' number before, and I'd like to see how its presented.
[ambiguity of murder]
Which is why I said 'fairly' unambiguous - and it's one of the least ambiguous crimes, especially when comparing across time, countries, or culture.
'Indictable offense' is a particular kind of crime, and one that changes over time as laws change and reporting improves. It even says that improved reporting is responsible for part of the effect in the first dot-point after the graph. The graph itself is explicitly titled "known to police" because of exactly this issue. To take that graph and claim a 50-fold difference in actual crime is extremely disingenuous.
Moldbug is saying that the Victorians nearly abolished crime and that the modern UK government can't - so is the solution the same as in Victorian times: just don't record so much of it?
'Indictable offense' is a particular kind of crime, and one that changes over time as laws change and reporting improves.
Yes, but the change can be in both directions. Improved reporting means more offenses get recorded; but changes in how society perceives crime means things that used to be indictable offenses (such as stealing a pair of scissors) no longer are--either the laws are taken off the books entirely, or they are no longer enforced because people don't think it's "fair" to convict someone of a crime if all they did was steal a pair of scissors. These two effects work in opposite directions.
It even says that improved reporting is responsible for part of the effect in the first dot-point after the graph.
But it doesn't say how much, because it can't; there's no way to know. And it does not say what difference changes in the laws and in society's perception of crime made, or whether that difference was, as I suggested above, in the opposite direction.
is the solution the same as in Victorian times: just don't record so much of it?
I think the solution Moldbug is implicitly suggesting is to make it clear to everyone that if you commit a crime, you will be caught and you will be punished; in other words, he is saying the problem today is with society's attitude towards crime.
This is just a more clever way to project the same nostalgia as adventured's comments. What evidence do we have that 1950's America was a font of innovation and civilization that 2013 cannot match? "Civilization" is an odd term for a place with millions living under legalized segregation, and an entire half of the population largely consigned to domestic labor.
Not to mention, much of the prosperity of the '50's was a "fringe benefit" of most of the rest of the industrialized world blowing itself up for the second time in 30 years.
And of course there's survivorship bias: No one remembers all the terrible inventions of the '50's. (Had they come up with Smell-O-Vision yet?) All we remember is the transistor radio and the hydrogen bomb. Look back on the 2010's in 60 years and you'll see all sorts of amazing stuff that made its mark on the future, I'm sure.
It's: what do you get for your labor, what can it buy and for how much, what kind of savings do you have, how hard do both parents have to work to make ends meet, what does an education cost and what income can you get for it, how many jobs are available?
Simple questions that can be easily answered with real data that was available in 1970 and today.
I started to make a list of all the things you can get in 2013 that you couldn't get at any price in 1970 (notable ones included Gleevac and a Blu-ray set of the Godfather trilogy) but I am just going to give up. If you think you'd be happier living as the median American in 1970 than as the median American in 2013, especially in the Socratic satisfaction sense of knowing the experiences of both, I just can't imagine how to convince you otherwise.
With respect, I think you're coming at this from the wrong angle. In my opinion, availability of blurays, or whatever, has no correlation whatsoever with the general health and happiness of a society. In fact, the essay I linked specifically seeks to factor that out, on the presumption - that I share - that it is basically irrelevant. Are you a miserable, miserable man today because you can't get 2025's GreenRay UHD or super-ultra-holo-3d or the iPad 20 or whatever? Would any of those things make you a long-term happier person? Of course not. Technological knick-knacks do not bring long-term life satisfaction. Convenience does not equal happiness. You will not die with a smile on your lips because you own a retina display.
I'd advise you to read the essay I linked. It might change your thinking on a few things.
If I get a certain variety of leukemia, and I'm in a society which hasn't discovered Gleevac yet, I will probably die. That's a pretty awful consequence. Not getting to see the Godfather movies in glorious 1080p is perhaps a lesser tragedy, but all these little things add up. Sure, there are all sorts of "holistic" measurements of society's overall happiness, but I'll come back to the Socratic question: Would you really be happier in 1970, knowing what you're missing from 2013? And again, we're assuming straight up that you're not in some woefully underprivileged demographic that would be seriously screwed by having to live under 1970's mores.
Can we please stop using "happiness" as some kind of value. It is such a loaded and slippery term. Happiness is a fluid and multi-dinensional state of being. It doesn't quantify well.
Not a downvoter, but your observation is (a) false, and (b) irrelevant, if it were true.
To fill in the gap in what I wrote: Even if the physics and the roadmap were known, the infrastructure and background engineering expertise needed to go from vacuum tubes to fabs making cheap ICs at nanoscale in volume would take longer than a decade to build.
Moore's law has been in operation since about 1958. That's 36 density doublings, at one per 18 months. You don't get 36 doublings in 10 years, because that would be one doubling every 3 months.
The only reason I bothered to comment is that I tire of the outlandish claims and "everything you know is wrong" posturing of UR. It was really only worth the original two-word rebuttal.
Do you really think that people have become that much stupider? Moore's Law has been going for decades. Multiple Nobel prizes have been awarded for discoveries that made its continued existence possible. The fact that you don't understand the accomplishment does not make it less significant.
I am typing on a small device whose internal electronics exceed in complexity all electronic devices that existed in 1940. Combined. Yet this is a commodity item.
Stop and think about that. And then you might recognize the absurdity of claiming that a few tips and a can do attitude would have created the results of decades of modern technological progress in 10 years.
Just to add to your point, think about how these electronics are actually designed and developed. You need lots of CAD and CAM to get working chips out the door, and you don't get those without already having working chips. Essentially you need earlier iterations of electronics to even build the next iteration. A single iPad isn't going to help much, unless it's loaded with schematics and data for entire generations of electronics leading up to it.
I agree. It would certainly take decades for them to master the technology. It might even take them 50 years. A lot of the knowledge required is implicit in the industry taken as a whole. Put me in the stone age with a manual for making bronze. I would be proud of myself If I managed to forge something useful inside of a lifetimes work. Where does the ore come from? How do I extract it? Precisely how do I get the fire hot enough? Similarly, if I knew how to create a warp drive, there is absolutely no guarantee that I could actually build one with current day tech. It might take a decade just to produce one of the required components.
I concur with this point of view. I very much doubt most smartphones are used to access the wealth of knowledge on the Internetz. They are used to play games. Source: my time on the NYC subway system.
