Anyway, this will consequently be a slightly shorter attempt today, but there are variations of a line floating around. You have likely encountered it: it's the guns, stupid.
Hi. I'm stupid.
In causal terms, of course, we distinguish between "necessary" conditions, and "sufficient" conditions, and a gun is a necessary condition for a shooting, or at least close enough to it. If you try hard, you can fire a bullet without a gun, but like Kirk fighting the Gorn, your rate of fire isn't going to be anything close to an engineering feat by modern standards, let alone 23rd Century standards. You might be better off with the Star Trek double-handed backfist.
By that logic, though, you can look at deaths by heroin overdose and say, "it's the heroin, stupid," but that facile argument doesn't seem to have taken hold among the ITGS crowd, nor does the policy conclusion of a crackdown on heroin. Why not? Because we know a lot about over-prescription of opioids, and so forth, and when we aren't talking about guns, lefties remember that black markets are a thing. Then again, the ITGS crowd doesn't have the same visceral, personal hatred of drugs that they do of guns. If they did, they'd respond differently to the opioid crisis.
Anyway, though, as I said, I'm stupid. And you know what stupid people do? Two things. We poke bears, and we look at numbers. Look up "stupid," in the Oxford English Dictionary, and it says, "a mathematician poking a bear." You get a picture of Neil deGrasse Tyson running from said bear, and then another mathematician wondering what was so interesting that a physicist had to poke at it.
Where's my stick?
Ah. It's that box right up in the blog editor that says, "link." Got it.
So, here's me and my simple, statistical mind. If there are changes in shooting deaths over time, is that accompanied by changes in the number of guns, over time? I'm actually a trained statistician, and this is what I do. There's a whole field of time series econometrics, and addressing this kind of thing is harder than you think.
But I'm going to make it really easy. Kind of...
First, there's a bit of a data problem. How many guns are there, per capita in the US, in any given year? I did some searches, and it's harder to find than I thought. Remember, we don't have a universal registry. That's... kind of a hot-button issue. So it's all estimates, done sporadically. The best I could find with a quick search (I have real work to do, and this ain't it) was basic reporting on percentages of households with firearms, year-to-year. In a simple, comprehensible form, here's a nice graph from a German consulting firm called "Statista." I'd never heard of them, and I can't vouch for them, but the numbers look basically like what I've seen. Take this with some grains of salt, but the numbers look like what I consistently see elsewhere. Just don't put too much in any given point estimate. This stuff is usually compiled through surveys, and... Yeah. Mmmmm… salty.
What do we see? Basically, fluctuation over time in the neighborhood of 40%. That's consistent with what you'll see in any other given year, regardless of your source. Yeah, around 40% of US households have at least one gun in them. It goes up and down, but that's been around consistent since 1972. Has the number of guns themselves, per capita, changed dramatically in any given period of time over that time frame? Maybe, but in order for that to be compatible with the household-level data, it would have to be either households trimming down collections or hording. There are times when that happens. Elect a Democrat, and the NRA/militia crowd freaks out and buys a bunch of guns. Also, there are collectors, hunters, etc., who account for large overall proportions of the guns. Someone whose primary hobby is hunting isn't just going to have one gun, and if those people guy more guns in any given period of time, the household-level data won't change, but guns-per-capita will.
So are these the droids I was looking for? Not quite, but it does make a partial point, and you'll see where I'm going soon anyway. The distinction won't matter.
Next up, remember a point I have made several times: "crime" versus "gun violence." This is one of those political rhetoric tricks. Gun control advocates have convinced the political world to use the phrase "gun violence" instead of "crime" because the phrase primes people to think that the solution is gun control rather than more police, stricter sentencing, etc. Hence, you won't hear Democrats say the word, "crime," and you won't hear Republicans say the phrase, "gun violence." However, if you refer to FBI Uniform Crime Reporting, you can find out what actually happens over time with any given broad category of offense. Remember the 2016 Republican Convention, when Trump portrayed the country as having descended into a Mad Max-style, post-apocalyptic hellscape because of... "crime?" And he did it again with his inauguration speech? Remember that phrase, "American carnage?" The Democratic response at the time was to cry foul because the actual statistics, which you can get from those lovely people at the FBI showed that we have seen a long-term secular decline in violent crime, including murder, including murders committed with... guns. Yet, once you shift to the phrase, "gun violence" rather than crime, the left will absolutely think we live in some sort of hellscape of, well, we can't use the word, "crime," because, "crime," is a verboten word, but "gun violence," even though that category of action is subject to the same long-term secular decline according to the FBI's Uniform Crime Reporting.
There is a category of offense that is probably on the rise, although it is difficult to assess because it wasn't always necessarily reported as such, though. "Mass shootings." What constitutes a "mass shooting?" Um... Here's a reasonable definition: more than one person shot by random act, unconnected to any other crime. There's your more-than-one, and it gets you away from gang violence, domestic violence, robberies, and other stuff with personal elements. In the current political environment, any one of these, anywhere in the country gets mentioned in the national press. Would that have happened in 1950? Suppose someone just snapped, took a gun and went into a bar, a restaurant, whatever, shot a few people... Would that necessarily have been a national news story without a pre-existing political atmosphere that deems "gun violence" to be a central, political issue? Not necessarily. That's not a "mass shooting," then, it's a "multiple homicide" with no apparent motive. In which case, we don't actually have the best long-term data.
But, over the last, say, 20 years, has there been an increase? Probably.
Mathematically, keep in mind what that means. (Hey! Bear! Are you asleep? Poke!) Overall, homicides went up as the crack cocaine epidemic hit in the 1980s, and then began a long-term secular decline, for reasons that are more difficult to explain. (I'll leave it to you if you believe the Freakanomics argument on abortion.) Within that, the rare event of "mass shootings" became more common over the last few years, but the number of deaths attributable to that in a country of 330,000,000 is small enough that when you focus on the long-term secular trend, you still see the decrease in homicides.
And while all of this was happening, households with firearms stayed the same.
The last couple of years have seen an increase in the very rare event of a "mass shooting," which you think of as common because of what I call "the paradox of news."* However, that increase has not happened alongside an increase in households with firearms. We've been fluctuating at around 40% for half a century. Similarly, the increase in crime generally, and homicides specifically in the 1980s, followed by declines in both-- none of that was accompanied by appreciable changes in the proportion of households with firearms. Fluctuation around 40%.
Look, it might seem intuitively obvious that when deaths by homicide increase, that should be attributable to a proportionate increase in guns in the country, and when homicides decrease, that must be because guns are proportionately rarer. Empirically, this is a simple cut at the data, but there just isn't support for it, and any attempt to reconcile this stuff with the data runs into the logical problem that talking about homicides generally, and then "mass shootings" means speaking at cross-purposes because we have seen a long-term secular decline in homicides, which the left remembers any time Trump talks about "crime," but an increase in the rare event of a "mass shooting," making it mathematically impossible to find a variable to explain both. They run in opposite directions.
Math, people.
*Again, this is my phrase. "The paradox of news" is the paradox that an event is newsworthy if it is sufficiently different from normal events to merit news coverage, but that coverage convinces you that the event is normal, even though lack of normality is exactly why it was news.