Popularity of duets, 1959 to 2011

On Norah Jones as an example of the rise in fearful-avoidant women, Ziel left a comment about how painfully uncomfortable she is in this duet for "Love Hurts" with the admittedly oddball Keith Richards. She should've just gotten over it for a few minutes in order to give a good performance for the crowd, but she totally freezes.

Compare that to the 1983 duet, "We've Got Tonight", by Kenny Rogers and Sheena Easton. Rogers was 44 and gray-haired, while Easton was 23 and a vulnerable little thing. She doesn't even look like she's just putting up with him for the sake of the crowd -- she's getting really into the song, letting him hold her hand and hug her. Today her counterpart's reaction would be omg creepy guy!!! where are you when i need you harry potter?!?!?! But back then people didn't find normal affection weird.

They can sound cheesy, but duets are not made with pop music perfection in mind. They're more of a demonstration to the audience that normal barriers can be let down -- for a little while, anyway -- so that two growing-distant groups can overcome their differences and be a little more team-minded in the future. If the duet partners can do it, why can't you?

It's rarely two groups who are bitter enemies for good reason, but more like those who are drifting out of touch and need to count on each other more. Men and women, younger and older, one sub-culture and another, or blacks and whites (NB: never Mexicans or East Asians).

Banding together, putting aside petty differences, the carnivalesque temporary weakening of barriers -- all themes that keep showing up in relation to the trend in the violence rate, so let's see if it's here too.

I looked at the songs of the Billboard Year-End Hot 100 (1959 was the earliest year with all songs), and found which ones had more than one artist listed -- this excludes songs performed by a group that makes duets normally, like Simon & Garfunkel. I'm looking just at those that represent two normally separate musicians coming together. (I also left out two songs from Disney movies, sticking just to pop music for teenagers and older.)

Then I checked if the Wikipedia page described the song as a duet (or if unsure, the Pandora page). This excludes songs where two stars come together just for hype value, but where there is no sense of the two interweaving their voices, interacting with each other, and if anything playing down their own egos. We all know what "duet" means.

I've weighted each song based on how high in the rankings it was [1], and added up these weighted values for each year. Here is how this index of popularity has changed over time:


It starts low but shows a jump in 1967, and continues to rise through the '70s, plateauing at a high throughout the '80s. Even 1990 is still pretty high. But by 1991, there are zero duets for the first time in over two decades, the downward trend after that is clear. There are still fluctuations up and down, but both the peaks and troughs get steadily lower in the post-'91 period.

So the link to the crime rate checks out. Still, it looks like there are only fits and starts during the '60s and early '70s; most of the action is in the mid-'70s through 1990. It seems like people get more into duet mode during the apocalyptic second half of a rising-crime period, while in the first half they're still hoping that the experts can work things out, so that banding together at the grassroots level won't be necessary.

Aside from serving as a model to the audience that they shouldn't keep their guard up and be so isolated all the time, some of these duets sound great on their own:

"Easy Lover" by Philip Bailey (of Earth, Wind & Fire) and Phil Collins (of Genesis). Collins has a few moments in the video where he overdoes the "goofy white guy" act, but overall there's no self-consciousness about a black guy and a white guy making a kickass rock song about heartbreaking women.

"What Have I Done to Deserve This?" by Pet Shop Boys and Dusty Springfield. In this different-age duet, it's the woman who's older, by 15 years. But this was back in the pre-cougar days when women still aged gracefully. Both made their careers during rising-crime times, so it was easy to fit together.

[1] For 100 songs, a good formula seems to be exp(-0.02*R), where R is the rank number.

What ever happened to making home movies?

Over Christmas vacation my family watched a bunch of old home movies from when my brothers and I were little, around 1982 to '85. We made movies after that as well, but they must be under a pile somewhere. The last I remember was probably Christmas of 1990 or '91 being taped, and anything after that would have been very sporadic.

My mother has a DVD compilation of all of her old 8mm home movies from the '60s and early '70s. The same kind of events are shown there as on ours -- every birthday, every Christmas, and random special occasions. (My mother even taped my brothers "performing" in the daycare center's 2 year-old Olympics during the summer of '84. In hindsight, it's something we all felt like fast-forwarding through after a few minutes, but moms were more motherly back then and recorded more rather than less.)

Again that seemed to fade out during the '90s, and not just because we were too old to tape. My mother's home movies show her in high school, when her older brother had already become a father, etc.

By now it seems just about dead. My nephew is nearly 4 years old, and there isn't a single home movie of him, whether for a birthday, Christmas, or anything else. Sure, there are a handful of clips, but none is longer than about a minute, they are usually divorced from any context, and so there's no sense of recreating an experience. Also, we were with my cousin and his four kids (ages 6 to 16, I think) for Christmas, and there were no home movies being made. People still take pictures, but no movies.

The movies of me and my brothers go for about 10 minutes or more at a stretch, adding up to several hours per tape, and you get a good feel for what was going on, and how everyone was interacting with one another.

I tried to google around to see if there are any data to pin this feeling down, but I didn't find anything. Still, it was just so common to make home movies, and I haven't seen anything at all of my nephew. Some of my brother's friends have little kids too, and I think he would've mentioned it if they were very different in that respect. Like, "Yeah, they make lots of home movies of their kids, but as for us..."

Obviously it has nothing to do with economics, since camcorders only get cheaper over time, and we either stay the same or get richer. Not to mention easier to operate -- my dad had to hold that Betamovie camcorder with two hands and rest it on his shoulder, while wearing the giganto VCR itself (which held the blank tape) strapped over his shoulders, like a proton pack or something.

Technological change isn't it either, since people choose what to adopt, how often to use it, and in what ways. It's true that a smartphone is dumb as a camcorder, but nothing keeps people from still buying camcorders and using them the way they used to. "My iPhone wasn't built for that" is just a lame excuse.

Once again the changes seem to reflect the trend in the violence rate. Looking through YouTube for home movies from the '50s (i.e. before '59 or '58), they aren't as frequent as the ones from the '60s through the '80s. It's not just that the cameras weren't as common in the '50s vs. the '60s. They don't show very private or intimate events, whereas in the ones from the '60s the people are more open and not so concerned about being caught on film. Here's an example of Christmas from the mid-'60s. And then people in home movies only became less self-conscious through the '70s and '80s.

Nobody else was ever going to watch your home movies but you, so this isn't a difference between one time period being more private and another more exhibitionistic. It's more about how close you wanted to be to your family members. If you compare the '50s family sit-coms to the ones from the '80s, the family members are much more distant and neutral toward each other in the former, and more affectionate and looking out for one another in the latter. During the '90s and 2000s, they've returned to "it's nice that you're here, but I don't really need you."

Making home movies, then, was a way to preserve something that looked like it was being increasingly threatened. Not consciously, of course. You just care more about preserving something that feels transient, and take it for granted when it feels permanent.

That has lasting consequences, too. You can gather the family around and watch those old home movies as a rite of renewal for your togetherness. The Silent Generation never seemed to feel that way about their childhood as they aged, probably because they never got very close to each other in the first place. "Too mushy." I don't think most Millennials will either. Just as the Silent Generation mostly gets nostalgic about radio programs, the Millennials will get nostalgic mostly for TV and video games. Not so much the relationships they had.

This is further evidence not to believe the hype about the "family values revolution" since the mid-'90s. They're definitely hovering over their children all the time, but if parents so strongly value their family, then why aren't they making home movies like they used to?

Is file-sharing the sole cause of the music industry's decline?

Here's the abstract from a newer paper by the always enlightening Stan Liebowitz (free PDF):

The file-sharing literature has focused mainly on whether file-sharing has decreased record sales, with less attention paid to the size of any decline. Although there is still some contention, most studies have concluded that file-sharing has decreased record sales. What has not been noted is that most estimates indicate that the file-sharing has caused the entire enormous decline in record sales that has occurred over the last decade. This heretofore hidden result is due to the lack of a consistent metric that would allow easy comparability across studies. The task of this paper is to provide such a metric, translate the results reported in the literature into that metric, and then summarizes the results from this exercise.

The studies that suggest that the whole decline in sales is due to file-sharing are earlier, back in the wild west days of the early 2000s. The more recent studies from 2008 and 2009 still put file-sharing's role as accounting for 65-75% of the declining sales.

Most of what the music industry puts out is junk, but sooner or later the zeitgeist will change and we'll get the next big thing. First it was jazz, then rock, and next who knows. But for that to happen, all of the infrastructure has to still be around. Good music, when people can create it again, won't record and distribute itself.

That's the worst part of this whole mess -- having to stick up for bloodsucking record companies. But what choice is there when the other side is file-sharing dorks who don't care if the basic infrastructure melts away, all so they can save a few bucks on their faggot album by Bruno Mars or Avenged Sevenfold?

Percent of teenagers with driver's license still plummeting

I think I've already got two other graphs like this on the blog somewhere, but the 2010 data are in.


The first drop-off takes place in 1990, actually a bit before the peak in the crime rate. This is another example of cocooning behavior slightly preceding a drop in violence rates. During the mid-2000s, it looked like it might have been slowing down, but it's only fallen at a faster rate within the past several years. The overall decline gets sharper the younger the age group you look at -- least steep for 19 year-olds, up to the most steep for 16 year-olds.

A driver's license used to be sought after as a rite of passage and a ticket to freedom, especially in high school when you don't have a college campus that's at least somewhat accessible by foot or school shuttles. By now, among high schoolers old enough to get one (16-17 year-olds), only 38% have followed through. Talk about delaying growing up.

Don't teenagers find it embarrassing anymore to have to get driven around by their parents, or burden their one friend who drives with ride requests? Truly a lazy and shameless generation.

It would be neat to find data on rates of attempting the test for the first time, passing rates, and re-try rates for those who failed. Like, are fewer of them trying to get a license at all? Even among those who try, if they fail, are they more likely to put off the second attempt until much later?

Due to a careless mistake, I flunked my first test -- and right on my 16th birthday! -- but I showed up the next weekend, or maybe the one after that, and passed it with no problems. Today I don't think a 16 year-old would have that minimal amount of perseverance. They'd need a year of self-esteem therapy before they'd feel comfortable giving it another go.

Similar music from similar environments

Stumbled upon this hit from the Jazz Age, where a clingy girl wonders why her crush doesn't make a move on her, even after sending him such forward signals:

"He's So Unusual" by Helen Kane, 1929

In tone it's not so different from this hit from the New Wave Age:

"Johnny Are You Queer" by Josie Cotton, 1982

One of the biggest changes that people undergo when violence rates start rising is to desire more social interaction and attachment. So even people with low self-esteem will still want to reach out and touch someone; they'll just come off as needy.

But I miss clingy-needy girls, now that I've seen the alternatives. When people no longer desire social contact, the high self-esteem ones become dismissive and avoidant, like Fergie and other attention whores, while the low self-esteem ones become fearful and avoidant, or mousy, kind of like Norah Jones. They both don't trust others, and wouldn't feel comfortable letting their guard down to get close to someone. That really comes across in their singing, which has a very limited range of pitch and never gets very high.

That songbird type of inflection that says "hey, notice my voice and come over to talk to me" only comes from those who desire contact, whether they have lower self-esteem like the needy ones or higher self-esteem, who show a "secure" attachment style, like how Belinda Carlisle sounded.

Did cigarette warning labels curb smoking rates?

That wasn't the question I had in mind when looking at these data, but one I realized I could take a crack at after picking apart something that I had felt like looking at for no reason at all. *

The General Social Survey asked two questions about regular smoking -- do you regularly smoke now (SMOKE), or if you don't now, did you ever regularly smoke in the past (EVSMOKE). Adding the "Yes" answers for both together, and taking them as a percent of everyone asked about smoking, we can find out who has ever been a regular smoker in their life -- past or present.

Here is the percent of ever-regular-smokers within 5-year birth cohorts, shown by the first year of their group. So "90" means 1890-94, "15" means 1915-1919, etc. I've split them into males and females because their history of adoption and abandoning of smoking is different (something I hadn't thought of at first).




The first two male cohorts don't have huge sample sizes, so I wouldn't make much of the jump in the 1895-99 group. Males born around the turn of the 20th century already started with a high chance of ever taking up smoking during their lives, around 40%. That rose to a plateau of just under 50% for men born between 1925-39. Already by the cohort of the earlier '40s there is a decline, and that only plummets in each cohort after.

Women, by contrast, started this period out very unlikely to ever get into smoking. That changed with the Jazz Age, which began around the mid-1910s, about the time that girls born around the turn of the century would be teenagers and be exposed to a whole new set of role models. Although it was born of that time, the glamor of female smoking outlived the Roaring Twenties. The cohort most likely to regularly smoke at some point was the 1940-44 group. After a half-century of steady growth, the first decline comes with the 1945-49 group, which only falls further with each one after. There is a jump in the late '50s cohort, though -- maybe when they were young teenagers during the Counter-Culture, smoking became briefly fashionable again for girls (not for boys, according to their graph).

Since the cohorts are fairly large, breaking them up into single-year groups still leaves sample sizes over 100. To pinpoint when the declines began, for males it looks like the 1941 cohort, and for females the 1947 cohort.

The first cigarette warning labels came in 1966 and have remained since. By that time, the 1941 cohort of males had already started to give up trying smoking in their teens and early 20s, when most people who ever become regular smokers begin. I also doubt it was the earlier famous 1954 British Doctors Study linking smoking and lung cancer -- hardly any 13 year-old American boys would've heard about it.

For females, it's a little more plausible that the warning labels had an effect, since the 1947 cohort would've been 19 when they came out, still having some of their vulnerable period left. We can definitely rule out the academic studies from the '50s, since they would've been 7 when the famous one came out.

Still, we should prefer one cause to two. The labels can only possibly explain the female decline, requiring some separate cause for the male decline. The disappearance of cigarettes seems so close in both groups that we should just treat it as one change, just like when men and women both got scared of dietary fat and began eating more carbohydrates.

What that cause is, I don't know. It's not related to my usual stories about rising or falling crime rates. But it is worth noting that two "obvious" answers -- a huge study linking smoking and lung cancer, and dire warnings right there on the package -- are wrong. That doesn't surprise me, since those are both top-down influences, and those cannot change people's attitudes so dramatically and for so long. (And it's clear that this is an attitudinal change, not some lame economic change.)

In fact, since the warning labels came after the decline was already in progress, we see yet another case of the top-down solution being a mere delayed and symbolic reaction by the technocracy. The experts, whether in the government, academia, private sector, or whatever, just jump on the already-rolling bandwagon and shamelessly try to hog the credit for having set it in motion.

It could be as simple as a fashion cycle, where no external factors are needed to explain the rise and fall pattern. It would be just like an epidemic. That would be a local, bottom-up explanation, and those tend to pan out.

I don't like "fashion cycle" explanations when a bunch of cycles all fit on top of each other, like the ones I've been looking at for awhile with crime rates, etc. But if there isn't some other cycle to link it to, I'm OK with treating cigarette smoking as a fashion that came and went. There may well be some other collection of cycles that fit with the smoking cycle, and I'll keep my eyes peeled. For now, though, I have no clue. Someone else should look into it, though. But something that changed at the grassroots level, not some ineffectual policy or campaign by outsiders.

* As an aside, that's how real science gets done -- you're just playing with all kinds of different stuff, and some of it goes places you didn't intend. Blind variation, then selective retention. Starting with a question and collecting data that bear on it -- so-called "hypothesis testing" -- usually goes nowhere.

Some cool pattern could turn up in your data, but because you're in tunnel-vision mode when testing hypotheses, you'll miss what's right under your nose. You'll only think of it months or years later, when you're no longer thinking so narrowly about that project, just letting your mind wander. But why not start off that way, instead of wasting so much time with blinkers on?

Burger King home delivery and Starbucks pub

In a move that recalls the anti-social drive-in culture of the mid-century, Burger King has begun testing a home delivery service.

There's virtually nothing to be saved in terms of time or money. You have to live within 10 minutes of the restaurant, and they only aim to get your food to you within 30 minutes. So on your own, less than 10 minutes there and back, plus the time to get your food inside or via the drive-thru, and you get it faster. There's a $2 charge for the delivery, and you won't be spending that much on gas to do it yourself. What they're selling is the "luxury" to stay locked indoors all day.

You can't really fault the company, since they're only trying to keep up with consumer tastes, and people these days don't want to be out in the open. They've had to reverse their self-conception from a place that offers an experience special enough to leave the house for, to a stripped-down catering / outsource service.

I usually drop in a couple times a week, eating there, and it's not as enjoyable as it used to be because all of the windows are blotted out with ads. It's on a busy corner, and would make a greater spot for people-watching, but it's too inefficient to have the windows open up to the outside world for the handful of dining room customers, when they could be used to push the latest deals to the penny-pinching drive-thru majority.

One good thing, though, is the music. You can't expect good music in public, but Burger King's is usually listenable and unobtrusive. If they're not going to go for an exciting atmosphere inside, they should just play '80s adult contemporary hits. Something soothing and uplifting. I did hear "Mad About You" by Belinda Carlisle once -- I don't know who could hear that song in public and not break a smile.

Unfortunately you can't say the same about Starbucks music. It's always so cerebral (bebop) and self-conscious (folk) that it wakes you right up from the dream that you'd like to settle into when you're lounging around out of the house. It's not toe-curling, but it would be such an easy thing to fix to make the experience more satisfying.

Still it is my go-to hang-out since it's the closest thing to a neighborhood meeting place that you can find nowadays. They may begin to focus even more on that communal aspect by offering more food and booze. They're starting with beer and wine; hopefully they'll sell Irish coffee. In any case, it'll help mellow everyone out. I do like the direction the store's experience has taken, where it used to be a lek for attention-whoring professionals, and has slowly become a more chill, all-ages hang-out.

The one big misstep they've made is offering free WiFi. That just invites the parasites and cocooners. Reading newspapers, books, writing with your hand, etc., is all fine. But once you're hunched over a laptop, you might as well bring portable cubicle walls as well. It's off-putting and depressing to walk into a coffeehouse and behold a computer lab.

Once the society shifts away from cocooning, I'll bet it will start at a place like Starbucks, similar to the use of diners as young people came out of hibernation in the late '50s. They weren't as carnivalesque as the food court at a mall would later become, but they were at least keeping the embers of sociability burning in a withdrawn era, compared to the drive-ins and strip malls.

The last period of falling crime and cocooning lasted 25 years, and we're already about 20 years into this one. So hopefully around the end of the decade we'll begin to see signs of life again.

Do pastoralists make better taxi and truck drivers?

East Asians and Central Americans flock to all sorts of low-paying jobs, but you hardly ever see them driving taxis or trucks (long-distance, anyway).

The stereotypical cab driver is from north and eastern Africa, the Middle East, and the more northern and western parts of South Asia. Occasionally from Western Africa. The stereotypical truck driver is a restless Scotch-Irish hillbilly.

Being adapted to a more nomadic way of life certainly helps if you're going to be driving around all day. That weeds out the East Asians and some Mexicans, who are designed for sedentary life in large-scale intensive agricultural societies.

That still wouldn't weed out most sub-Saharan Africans and other Central Americans, whose horticulturalist ways also involve a good deal of moving about. They certainly enjoy going out cruising, but they wouldn't want to transport others for pay. They're too wary of strangers.

A cab ride is only a little different from hitch-hiking, so trust is crucial. Low trust will make the cabbie think that the customers are going to kill him, rob him, or skip out without paying. Low trust will also keep customers from getting in, thinking that they're just going to get ripped off.

It's a kind of guest-host relationship, where the taxi driver is the merciful host willing to help out the stranded guest, expecting a little something in return. Cultures of hospitality are the same as cultures of honor, since they're just two forms of an obsession with reciprocity. Hospitality is kindness repaying kindness, starting out generous; and honor is harm repaying harm, starting out threatening. And those both are found almost entirely in pastoralist societies.

The most helpful taxi driver I've known was from some herding region in East Africa. He was wiry, had a thinner and more pronounced nose than other black-skinned Africans, and spoke with pharyngeal consonants, unique to the Afroasiatic family that includes Semitic, and is mostly spoken by pastoralist groups. (It wasn't Arabic. Maybe Amharic.)

I wanted to make sure he got a tip, but the meter was running a little close to what I had on me, so I asked to be let out a couple blocks before the spot that I'd first said. He immediately sensed why I'd changed my mind, and said he'd turn off the meter and take me home, that I could pay with whatever I had -- "It's too cold for you to walk tonight." After I apologized that the tip wouldn't be that much for a late night trip, he brushed it off saying, "Oh no, that's too much anyway."

He may have been an extreme case, but you can't hold down a job like that if you don't have at least a milder level of that generosity. There will be too many complaints about how rude and inhospitable you are, and you'll get fired. Most times you may not even know that the person is generous because they won't be put to any test. You'll only find out when you're a little short. They're really one of the few groups of workers who are willing to cut you a break when you're in a bind.

Now, driving trucks long-distance doesn't tap into the guest-host psychology, but it does require drivers who can deal with and even enjoy long stretches of solitary, adventuresome activity. That's perfect for someone built to follow a herd of livestock, which isn't a big-group affair.

It's not hermetic either, though: bouts of close socializing punctuate the on-your-own flow of time. For herders, it's reuniting with kin, boisterous communal festivals, and the all-important guest-host relationships like stopping at a caravanserai to mingle and rest. For truck drivers, it may have been a honky tonk bar or a rest stop with a greasy spoon diner, places where you can interact face-to-face and even get worked up into a crowd-vibe with people trusting one another enough to cut loose and have some rowdy fun.

I wish we'd grown up living closer to my father's father, who was a truck driver. We got a good deal of exposure to footloose, unpretentious living through my mom's side (Appalachian hillbillies), but it would've been better if we'd gotten more from my dad's side too.

Civilization crumbles during Wikipedia blackout

"Imagine a world without free knowledge" -- if that means a world without Wikipedia, well, it would be like 2003. Even if it meant a world without the web or the internet itself, it would be like 1994.

We haven't gotten any richer, happier, or more productive since either date, so really who cares? You'd lose the minor buzz you feel from visiting your favorite sites, but they're more than replaceable in the real world.

The main reason people freak out so much when pondering the disappearance of the internet is that we live in an age of cocooning, and they're too dismissive or fearful of those real-world substitutes for farting around on the Facebook etc. They really would have nothing to do.

As for Wikipedia, it is incapable of increasing our wisdom, for technological reasons alone (forget who controls and edits it). It belongs with websites that only allow you to tunnel narrowly around an initial search, rather than browse broadly -- Amazon, Netflix, iTunes, Google, eBay, Pandora, and any other site that has so massive a scale of items on offer that you cannot hope to browse through them.

You must instead tell the search bar where you want to go, and trying to click away from that target will still keep you confined to a narrow range around where you started. Similar or related items, customers who like this also like that, also by this artist, and so on.

The best they can do to expose you to things you didn't even know about is to have a "featured item" or "new items". That's like the movie being played at a video rental store -- better than nothing, but not as good as browsing their selection.

A real encyclopedia allows browsing. In fact that's what you end up doing most of the time after the initial stage of "Oh I wonder what it says about this, Oh I wonder what it says about that!" That's true whether it's a general one or a subject-specific one. So browsing is fractal -- an encyclopedia only about animals still allows you to explore parts of the animal world you didn't even know about, and so could not have purposefully searched for.

In high school I had a Benet's Reader's Encyclopedia that I used purposefully as a reference for a little bit, then quickly switched to just flipping through pages at random and following through on the entries that sounded cool. Blind variation and selective retention -- the basic ingredients for evolution by natural selection. And you just can't get that first part with massive-scale sites based on a search bar.

It's not as though a person couldn't tunnel before if they wanted to; all of the specialized knowledge in Wikipedia is out there in books or databases. It's just faster and sometimes cheaper to access through Wikipedia. I use it for the #1 songs on the Billboard charts, to familiarize myself with the zeitgeist over time. But I could find that out some other way, though it might take a couple days instead of minutes, and maybe cost me something, though nothing prohibitive.

So in exchange for a tiny boost in convenience -- remember how well-run and productive society was before the web -- we've sacrificed the ability for our sight to wander into places we didn't even know were there. The search-and-tunnel websites make our view so hyper-specialized that we often cannot see what is right under our nose.

We try to make the best out of the internet, now that it's here, but it would probably have been better if we hadn't adopted it in the first place.