The issue here is semantics and definitions of words.
"Crime" is far too broad a word for there to be an overwhelming consensus as to whether it's going up or down. That's the main issue.
If Scott A. had said "actual murders, property crime (defined as ____ ), ... and NOT perceptions of these" then there would be a more fruitful conversation.
All this stuff about greek is a red herring. "Crime" is a collection of discrete events that occur or don't occur. There are more or fewer of them per time period. Whether or not those events are recorded correctly or that people are more or less aware of them can be debated, but the actual numbers are the numbers.
No, the issue, as outlined in the post, is real problematic behavior of real people on the internet who are inclined to tell anyone who is skeptical regarding the data (whatever it may be) that they should more or less discount their personal observations, reasoning, and experience when it goes counter to the data.
The post is about the author, not crime. The critique of Scott. A's posts is an example of the kind of online content that led the author to become "apostate to the Church of Graphs".
> they should more or less discount their personal observations, reasoning, and experience when it goes counter to the data.
OK, I look at two objects [1] and posit that object B is larger than object A. I see it with my very eyes, I directly experience this feeling of largeness and smallness. How dare any data, any calipers or rulers (must be oppressive rulers!) tell me that my perception is wrong, and the sizes are equal?
The whole thing is based on the idea that seeing with one's own eyes is somehow not interpretation, but unadulterated truth. This is, unfortunately, not exactly so. No matter who you ask, Buddhist practitioners or cognitive scientists, anyone who paid attention to the problem know that "direct experience" is not very direct.
Tools to rectify biases in perception exist, and statistics (when properly implemented) are one such tool. But accepting one's own bias is psychologically hard; it's much easier to think that all these other people have a bias, or several. (It's an important part of growing up though.)
I'm now in LA. There are illegal food stalls over over the place.
Some people like them but irrelevant, a crime is being committed, nothing is being done.
So every day I see these crimes. They weren’t here 10 years ago.
Hence, my experience is crime is up since I visibly see it every time I go out.
This is the first example provided. It is not new, and it is legal.
Don't mean to be curt, just, puzzled me to read to say the least. Googled it myself 2 months ago. [1]
In general, the problem is that the strong arguments in the essay are epistemically local - they say specific things about specific measurement gaps - but they're translated into a general license to privilege vibes over data. And that move is where the essay falls apart for me.
I think this is an important point lurking behind a lot of disagreements about these kinds of issues: basically, there are a fair number of things that are legal that people don't want to be legal, and there are a fair number of things that are not legal that people do want to be legal. The first category likely includes, for instance, all manner of tax trickery practiced by the wealthy; the latter category includes things like going 75 mph on the freeway.
There are also cases where it's not entirely clear what most people want, but where (I would say) the legality should be based on what most people want, but it is instead based on a complex apparatus of legal jousting and machinations by small groups of people. I would put the food stalls in this category. If more people want the food stalls in LA than do not, then they should be legal; if more people do not want them, then they should be illegal. But their legality should not depend on which advocacy group was able to muster a bigger war chest to fund their legal fees and win a court judgment one way or the other.
I believe this is a symptom of fundamental failures in our system of law and government that have caused it to be quite unresponsive to the actual desires of the citizenry. This causes us to waste a lot of time and energy fighting over things like "crime" without making much progress because we are working against the grain of the social/legal apparatus that some people put in place over a long period of time.
Don't read following as a caricature/driveby, really appreciated the thought and framing and it wins out over what I'm about to say, I'm just putting my thoughts after 2 minutes musing as concisely as possible:
There is something to be said for that's how stuff works today.
i.e. "a complex apparatus of legal jousting and machinations by small groups of people" can be reframed from (my rewording) "confusing thing I'm excluded from" as "people who give a shit doing the work to change things" - if it was popular to get rid of food stalls in LA*, should be pretty easy, people are pretty plugged in these days
There's the impossible extreme of "we live poll everything all the time", and you've made me really curious about a shift in that direction looks like.
* it wouldn't be, they're not, like, disheveled people slaving over a stove with unclearly sourced hot dogs. Generally, juice and fruit outside park entrance, ethnic food under tent next to sidewalk, miniature hot dog stand at sporting event. If someone said something like they did in real life, you'd ignore it because it's fringe, or, tell them to move to Newport Beach (ritzy suburb). Even just ~15 years ago, in Buffalo, it was perfectly polite to say "sounds like you should move to the suburbs."
There's also tons of nuance that doesn't get caught in "feelings" - if the homeless outside your gated community are repeatedly murdering each other, you have a "high crime rate" that you may not care about at all.
But if suddenly they stop murdering each other and only kill you (or someone like you) during the year, the crime rate has gone way, way down, but your perception of it has skyrocketed.
People keep telling me I'd be safer if I chose a self-driving car, citing overall car crash stats. Those include people who DUI, text while drive, are way too old, or drive a Tesla or Altima, where I'm none of the above.
Store owners in certain "low crime" areas are taking up the serious cost of locking merchandise because they think the problem is really that bad. Maybe they're wrong, but I trust the money more than the public stats.
Then you might be surprised to know that a lot of those numbers were exaggerated. If you're trusting people to tell you accurate things based on numbers that they won't share.... you're gonna have a bad time.
Even in those cases, they aren't wildly off the mark, at least the ones that survive. Maybe they spend a little extra on some overpriced enterprise solution, but it at least works. Locking up items is a big decision, cause it not only costs money but also discourages customers.
This also distorts the shoplifting statistics. If two stores have equal rates of shoplifting but one has boxes and the other doesn't, they are not at equal risk. The one with boxes is at more risk. Flat shoplifting rates while boxes proliferate mean things got worse. (You can argue that boxes do nothing, but you would need strong evidence, cf parachutes.)
The number of shoplifting incidents is also a weak metric. Most shoplifting is not of serious economic concern. The problem is with repeat offenders and those doing it for profit. The value of merchandise lost is a much better metric but stores may be reluctant to share this due to concerns about insurance rates and public perception.
Crime statistics are very hard. And state capacity is declining, sadly. We can't expect bloggers to pick up the slack.
Graphs can be abused and statistics can be misleading, and some things are hard to quantify and measure. But the author never makes any convincing case why the statistics would be wrong or misleading in this case: "I’m not here to argue with Scott’s statistics. I think they’re about as accurate as we could hope to make them. I’m here to argue that you don’t require them to make sense of the world".
His main argument is that many people feel crime is increasing, and that in itself is a good argument to disregard any falling numbers as obviously incorrect without any further justification being necessary.
The obvious problem is that people almost always say that crime is increasing, and they have consistently been shown to misjudge the actual trend for decades on end: "In 23 of 27 Gallup surveys conducted since 1993, at least 60% of U.S. adults have said there is more crime nationally than there was the year before, despite the downward trend in crime rates during most of that period." If we bought into the author's argument we would never be able to reach any other conclusion than that that crime has always been increasing and will always continue to increase.
During the satanic panic the the 1980's the populace at large were convinced that large swaths of satanists were routinely sacrificing and abusing children. The police was convinced it was a real problem and had special "satanic experts" to combat the issue, a huge amount of parents were genuinely afraid of their childrens' safety, and there were thousands and thousands of cases of reported ritual abuse. In reality and in hindsight there were zero evidence of satanic cults abusing children. The author's argument could, completely unmodified, be used to argue that we should listen to the people's lived experience instead of the evidence and conclude that the satanic cults must actually have been a real societal danger back then. Or is he only against disregarding someone's lived experience in favor of evidence when it is his lived experience?
It doesn't even matter if he is right in this case. Maybe the all the statistics is flawed and his feeling of rising crime rates is justified. The problem is that he offers no actionable heuristic that allows us to separate his intuition from other people's intuition that has been obviously wrong in hindsight, like the satanic panic.
Sides aside (heh), arguing that personally experiencing something means that it is reasonable to claim a wider-than-you trend is utter insanity. And that's exactly what this post spends 5,680 words arguing for.
Misunderstanding of data is a big problem. That's not a problem with relying on data. That's a problem with widespread innumeracy. Pretending like you're some kind of universal subject to whom all things happen and all thoughts occur is really not the solution. Your eyes may not be lying, but you are still a very small fish in a very big pond, and your own personal experience is less than epsilon against the world at large.
There is a positionally valid form of knowing from experience of a thing happening: "I have seen a thing happen therefore the thing happens sometimes."
And there is an extremely invalid form, which is the form that the post defends and holds dear: 'When you generalize about “how people are likely to treat a stranger in need” or “how should one live to be happy” based on examples from your own life.'
The problem words there are "generalize", "likely", and "should".
It's a known cognitive bias, not something to lean into.
But if you want something shorter than a three word phrase, there's also a single word for it: egocentrism. It's bad. Let's please not uphold it as our guiding standard.
My main problem with Scott Alexander is this: To draw correct conclusions from data, a necessary (though insufficient) condition is to be an expert in the field from which the data is drawn and/or to which the data applies. Otherwise, you might not know how accurate the sources of the data are and, more importantly, whether you're considering enough context (i.e. whether you have all the right data to draw your conclusion). At the very best, you can consider the objections you've heard, but are these (all) the right objections? For example, when I read Paul Krugman on international trade or central banks, at least I know that he's an expert in that subject matter so he knows what context may be more or less relevant. When he's not an expert in some subfield of economics, at least he knows who the experts are and refers to them.
Scott Alexander is not an expert in almost anything he writes about. As far as I know, he's not done any scholarly work outside his area of practice, psychiatry. In relation to this post's subject, Alexander is not an expert in criminology, law enforcement, political perception, or sociology. Then again, neither is the author of this post (at least they don't say what their relevant credentials are). It seems neither of them even know who the experts are. I can understand why they find the question interesting, but they're ill-equipped to provide answers. Both personal perception and data can obviously be misleading, which is precisely why people who truly want to understand something spend years becoming experts.
It seems to me that both Alexander and the author of this post are, actually, members of the same church whose members are those who believe that people can draw correct conclusions from a smattering of data without the necessary scholarship and expertise, and that you can understand something complicated without putting in all the effort required to understand it: the Church of Dunning–Kruger Dilettantism.
Of course, anyone is free to write their thoughts on anything, and readers are free to form opinions on what they read. What this reader sees here is two people arguing over something that both know far too little about to offer the relevant insight. What is interesting to me is that someone who's not particularly knowledgeable on the subject of crime took the time to write a long rebuttal to another post about crime written by someone else who knows just as little. I can guess that's because that church is large.
> It seems to me that both Alexander and the author of this post are, actually, members of the same church - the church of those who believe that people can draw correct conclusions from a smattering of data without the necessary scholarship and expertise, and that you can understand something complicated without putting in all the effort required to understand it. It's the Church of Dunning–Kruger Dilettantism.
We are all like that, we have no other options, haven't we? I mean, either we try to understand the world around us, or we are not. We can't be experts in everything, so in most cases we are go by Danning-Kruger Dilettantism.
Scott made the dilettantism into a profession, he has its methods and he sharpens them. He debates things with other dilettantes, and it helps them to improve themselves. To me, personally, it is one of the main attractions of the blog. I'm dilettante in a lot of topics, but still I don't want to simply ignore them, because I'm not an expert.
> What is interesting to me is that someone who's not particularly knowledgeable on the subject of crime took the time to write a long rebuttal to another post about crime written by someone else who knows just as little.
It is not about crime really. The author we discussing talks about methodology, they are on a meta level of a discussion, the crime discussion is just one data point for a meta-discussion.
Your post is the part of the same meta-discussion about methodology, though your attack comes from the other direction.
I have a different problem that I would also describe as a Church of Graphs.
I keep reading essays in which the author makes some claim and supports it by displaying a graph. The graph is not explained other than as proof that the claim it supports is correct. The axes are unlabeled, or labeled with meaningless abbreviations.
Apparently enough people find this persuasive that the practice has become widespread. But why?
This is an interesting article. I feel like the point the author thinks he's making isn't maybe the one he's actually making, or at least not the one he ought to be making.
The problem is he sets this up as a contrast between, on the one side, quantification, evidence, "graphs", and the like; and, on the other side, "your eyes", "lived experience", and so on.
But these are not necessarily in opposition. There is nothing unquantifiable about "lived experience" or people opinions about crime, nor is there any reason to dismiss such data as irrelevant to policy decisions.
Even if the "church of graphs" showed crime on a clear upswing, it would be absurd to say, "Crime has gone up, therefore we must build a new prison." To justify that action requires more than just that bare fact; it requires some kind of causal analysis that explains why that action would play a causal role in producing some desirable effect (like reducing crime).
On the flip side, it is not absurd to say "Surveys show that the perceived level of crime has gone up, so we should explore policies to address that." This is especially true if you swap "perceived level of crime has gone up" for "perceived quality of life has gone down", because perception is in some measure an irrefutable judgment on quality of life. (That is, if you think your quality of life has gone done, then to at least some degree it factually has, because part of what it means to have a good life is to know that your life is good and to be happy about that.) Such a swap is likely warranted, because many of the author's examples of "crime" in the article make more sense as examples of quality of life. Seeing things locked up in stores is not experiencing crime or even perceiving an increase in crime; it is experiencing a decline in quality of life which may plausibly be an effect of an increase in crime, but that's not the same thing.
So just having data doesn't tell you what to do, and just having feelings and perceptions doesn't mean you shouldn't do anything. What's missing in both cases is the causal explanation of how the data and/or the perceptions arose.
Whenever I see people talking about "lived experience" I get a bit leery, because often that seems to be a lead-in to an argument of the form "I personally experienced X, therefore large-scale change Y should be implemented." The fallacy there is not starting from perception or from gut feelings; it's starting from just your own perceptions and gut feelings. If you can get data that shows a lot of people share your perceptions and gut feelings, then we can have something to work with. What we do with that information can vary: sometimes there is a causal theory to be developed and action to be taken that can trickle down into a change in those perceptions; sometimes the answer is better education or messaging that makes clear to people that their perceptions were inaccurate. But the problem is not a "church of graphs".
With regard to the issue of crime as discussed in this article, it seems likely to me that the data adduced in support of the "there is no crime problem" position is missing something important that has a genuine impact on people's quality of life. This doesn't mean the data we have is wrong or irrelevant; it just means it's not the whole story. If you have a bunch of data on temperatures in different places around the world and you use that to pick the best place to live, you may be disappointed if you get there and find it's raining all the time. That doesn't mean your data was bad (temperature surely is a major determinant of what makes us like a certain climate) but that it's incomplete (you need more than just temperature).
The solution to this is not to give up on data, it's to bring more data into the fold. Data on people's perceptions is immensely useful as a starting point for policy. It's not an endpoint, but then neither is any other data.
An interesting article. I wonder if there's a valid point in here buried somewhere underneath the endless obnoxious comparisons of his opponents to cultists.
I’m not familiar with this writer or the writer this is about so can someone help me out with something? This article is fawning over this Scott person, talks about how the audience is more intelligent than average and about how ‘impeccably sourced and credentialed’ this Scott person’s arguments are.
Am I missing an in joke somewhere or do people actually write like this?
> Members of The Church of Graphs live by one primary commandment: thou shalt not believe your lying eyes.
Ah, how familiar this is from some colleagues in tech.
Demands of evidence are asymmetric: make a bold claim that's aligned with the group, and it slips by; make a hint of a misaligned claim, and you get chided for not being a researcher in the field and spreading misinformation.
> I increasingly find myself in disagreement with Scott’s essays on social issues and public policy, despite broadly sharing his small-L liberal outlook.
Well, there's your problem. Scott isn't a "small-L liberal." He does a decent job at masquerading as one, but ask a fan to recount his "greatest hits" and they're all boring old orthodox conservatism: race realism [1], IQ [2], anti-identity politics [3], etc.
(No, I'm in the mood to debate his positions on any of this, it's all been done to death and further debate isn't going to change anyone's mind, let alone his. The citations are there to establish that he is aligned with these views, whether or not it's warranted.)
One fringe benefit of belonging to "The Church of Graphs" that I don't think the author really touches on is that believers can do motivated reasoning _very_ easily. Scott is an expert at laundering his motivated reasoning through well-researched citations and data that supports his points, but he's not so great at giving the other side a fair hearing.
Author loses me when he starts pearl clutching about the harms of seeing boarded-up windows, and his wife having to walk past unhoused drug users. (Who, let us be clear, are the ones who are actually experiencing harm.)
This guy isn't a liberal, he's a guy looking to justify his discomfort by dressing it up with a bit of rehydrated bible-school epistemology
The issue here is semantics and definitions of words.
"Crime" is far too broad a word for there to be an overwhelming consensus as to whether it's going up or down. That's the main issue.
If Scott A. had said "actual murders, property crime (defined as ____ ), ... and NOT perceptions of these" then there would be a more fruitful conversation.
All this stuff about greek is a red herring. "Crime" is a collection of discrete events that occur or don't occur. There are more or fewer of them per time period. Whether or not those events are recorded correctly or that people are more or less aware of them can be debated, but the actual numbers are the numbers.
No, the issue, as outlined in the post, is real problematic behavior of real people on the internet who are inclined to tell anyone who is skeptical regarding the data (whatever it may be) that they should more or less discount their personal observations, reasoning, and experience when it goes counter to the data.
The post is about the author, not crime. The critique of Scott. A's posts is an example of the kind of online content that led the author to become "apostate to the Church of Graphs".
> they should more or less discount their personal observations, reasoning, and experience when it goes counter to the data.
OK, I look at two objects [1] and posit that object B is larger than object A. I see it with my very eyes, I directly experience this feeling of largeness and smallness. How dare any data, any calipers or rulers (must be oppressive rulers!) tell me that my perception is wrong, and the sizes are equal?
The whole thing is based on the idea that seeing with one's own eyes is somehow not interpretation, but unadulterated truth. This is, unfortunately, not exactly so. No matter who you ask, Buddhist practitioners or cognitive scientists, anyone who paid attention to the problem know that "direct experience" is not very direct.
Tools to rectify biases in perception exist, and statistics (when properly implemented) are one such tool. But accepting one's own bias is psychologically hard; it's much easier to think that all these other people have a bias, or several. (It's an important part of growing up though.)
[1]: https://en.wikipedia.org/wiki/Ebbinghaus_illusion
Don't mean to be curt, just, puzzled me to read to say the least. Googled it myself 2 months ago. [1]
In general, the problem is that the strong arguments in the essay are epistemically local - they say specific things about specific measurement gaps - but they're translated into a general license to privilege vibes over data. And that move is where the essay falls apart for me.
[1] https://la.streetsblog.org/2024/07/22/l-a-street-vendors-cel... (note: this just removed the last barriers, temporary events (i.e. sports), farmers markets, schools)
> It is not new, and it is legal.
I think this is an important point lurking behind a lot of disagreements about these kinds of issues: basically, there are a fair number of things that are legal that people don't want to be legal, and there are a fair number of things that are not legal that people do want to be legal. The first category likely includes, for instance, all manner of tax trickery practiced by the wealthy; the latter category includes things like going 75 mph on the freeway.
There are also cases where it's not entirely clear what most people want, but where (I would say) the legality should be based on what most people want, but it is instead based on a complex apparatus of legal jousting and machinations by small groups of people. I would put the food stalls in this category. If more people want the food stalls in LA than do not, then they should be legal; if more people do not want them, then they should be illegal. But their legality should not depend on which advocacy group was able to muster a bigger war chest to fund their legal fees and win a court judgment one way or the other.
I believe this is a symptom of fundamental failures in our system of law and government that have caused it to be quite unresponsive to the actual desires of the citizenry. This causes us to waste a lot of time and energy fighting over things like "crime" without making much progress because we are working against the grain of the social/legal apparatus that some people put in place over a long period of time.
This is very fair and I generally agree.
Don't read following as a caricature/driveby, really appreciated the thought and framing and it wins out over what I'm about to say, I'm just putting my thoughts after 2 minutes musing as concisely as possible:
There is something to be said for that's how stuff works today.
i.e. "a complex apparatus of legal jousting and machinations by small groups of people" can be reframed from (my rewording) "confusing thing I'm excluded from" as "people who give a shit doing the work to change things" - if it was popular to get rid of food stalls in LA*, should be pretty easy, people are pretty plugged in these days
There's the impossible extreme of "we live poll everything all the time", and you've made me really curious about a shift in that direction looks like.
* it wouldn't be, they're not, like, disheveled people slaving over a stove with unclearly sourced hot dogs. Generally, juice and fruit outside park entrance, ethnic food under tent next to sidewalk, miniature hot dog stand at sporting event. If someone said something like they did in real life, you'd ignore it because it's fringe, or, tell them to move to Newport Beach (ritzy suburb). Even just ~15 years ago, in Buffalo, it was perfectly polite to say "sounds like you should move to the suburbs."
There's also tons of nuance that doesn't get caught in "feelings" - if the homeless outside your gated community are repeatedly murdering each other, you have a "high crime rate" that you may not care about at all.
But if suddenly they stop murdering each other and only kill you (or someone like you) during the year, the crime rate has gone way, way down, but your perception of it has skyrocketed.
There's a serious problem where people look at statistics and assume the constituent components are 100% evenly distributed.
This is rarely the case.
People keep telling me I'd be safer if I chose a self-driving car, citing overall car crash stats. Those include people who DUI, text while drive, are way too old, or drive a Tesla or Altima, where I'm none of the above.
Store owners in certain "low crime" areas are taking up the serious cost of locking merchandise because they think the problem is really that bad. Maybe they're wrong, but I trust the money more than the public stats.
Then you might be surprised to know that a lot of those numbers were exaggerated. If you're trusting people to tell you accurate things based on numbers that they won't share.... you're gonna have a bad time.
https://www.nytimes.com/2023/12/08/business/organized-shopli...
Businesses do make irrational decisions with their money, that’s a big part of why enterprise sales works.
Even in those cases, they aren't wildly off the mark, at least the ones that survive. Maybe they spend a little extra on some overpriced enterprise solution, but it at least works. Locking up items is a big decision, cause it not only costs money but also discourages customers.
This also distorts the shoplifting statistics. If two stores have equal rates of shoplifting but one has boxes and the other doesn't, they are not at equal risk. The one with boxes is at more risk. Flat shoplifting rates while boxes proliferate mean things got worse. (You can argue that boxes do nothing, but you would need strong evidence, cf parachutes.)
The number of shoplifting incidents is also a weak metric. Most shoplifting is not of serious economic concern. The problem is with repeat offenders and those doing it for profit. The value of merchandise lost is a much better metric but stores may be reluctant to share this due to concerns about insurance rates and public perception.
Crime statistics are very hard. And state capacity is declining, sadly. We can't expect bloggers to pick up the slack.
Graphs can be abused and statistics can be misleading, and some things are hard to quantify and measure. But the author never makes any convincing case why the statistics would be wrong or misleading in this case: "I’m not here to argue with Scott’s statistics. I think they’re about as accurate as we could hope to make them. I’m here to argue that you don’t require them to make sense of the world".
His main argument is that many people feel crime is increasing, and that in itself is a good argument to disregard any falling numbers as obviously incorrect without any further justification being necessary.
The obvious problem is that people almost always say that crime is increasing, and they have consistently been shown to misjudge the actual trend for decades on end: "In 23 of 27 Gallup surveys conducted since 1993, at least 60% of U.S. adults have said there is more crime nationally than there was the year before, despite the downward trend in crime rates during most of that period." If we bought into the author's argument we would never be able to reach any other conclusion than that that crime has always been increasing and will always continue to increase.
During the satanic panic the the 1980's the populace at large were convinced that large swaths of satanists were routinely sacrificing and abusing children. The police was convinced it was a real problem and had special "satanic experts" to combat the issue, a huge amount of parents were genuinely afraid of their childrens' safety, and there were thousands and thousands of cases of reported ritual abuse. In reality and in hindsight there were zero evidence of satanic cults abusing children. The author's argument could, completely unmodified, be used to argue that we should listen to the people's lived experience instead of the evidence and conclude that the satanic cults must actually have been a real societal danger back then. Or is he only against disregarding someone's lived experience in favor of evidence when it is his lived experience?
It doesn't even matter if he is right in this case. Maybe the all the statistics is flawed and his feeling of rising crime rates is justified. The problem is that he offers no actionable heuristic that allows us to separate his intuition from other people's intuition that has been obviously wrong in hindsight, like the satanic panic.
Sides aside (heh), arguing that personally experiencing something means that it is reasonable to claim a wider-than-you trend is utter insanity. And that's exactly what this post spends 5,680 words arguing for.
Misunderstanding of data is a big problem. That's not a problem with relying on data. That's a problem with widespread innumeracy. Pretending like you're some kind of universal subject to whom all things happen and all thoughts occur is really not the solution. Your eyes may not be lying, but you are still a very small fish in a very big pond, and your own personal experience is less than epsilon against the world at large.
There is a positionally valid form of knowing from experience of a thing happening: "I have seen a thing happen therefore the thing happens sometimes."
And there is an extremely invalid form, which is the form that the post defends and holds dear: 'When you generalize about “how people are likely to treat a stranger in need” or “how should one live to be happy” based on examples from your own life.'
The problem words there are "generalize", "likely", and "should".
There's a phrase for this ilk of anti-logic: the False Consensus Effect. https://en.wikipedia.org/wiki/False_consensus_effect
It's a known cognitive bias, not something to lean into.
But if you want something shorter than a three word phrase, there's also a single word for it: egocentrism. It's bad. Let's please not uphold it as our guiding standard.
My main problem with Scott Alexander is this: To draw correct conclusions from data, a necessary (though insufficient) condition is to be an expert in the field from which the data is drawn and/or to which the data applies. Otherwise, you might not know how accurate the sources of the data are and, more importantly, whether you're considering enough context (i.e. whether you have all the right data to draw your conclusion). At the very best, you can consider the objections you've heard, but are these (all) the right objections? For example, when I read Paul Krugman on international trade or central banks, at least I know that he's an expert in that subject matter so he knows what context may be more or less relevant. When he's not an expert in some subfield of economics, at least he knows who the experts are and refers to them.
Scott Alexander is not an expert in almost anything he writes about. As far as I know, he's not done any scholarly work outside his area of practice, psychiatry. In relation to this post's subject, Alexander is not an expert in criminology, law enforcement, political perception, or sociology. Then again, neither is the author of this post (at least they don't say what their relevant credentials are). It seems neither of them even know who the experts are. I can understand why they find the question interesting, but they're ill-equipped to provide answers. Both personal perception and data can obviously be misleading, which is precisely why people who truly want to understand something spend years becoming experts.
It seems to me that both Alexander and the author of this post are, actually, members of the same church whose members are those who believe that people can draw correct conclusions from a smattering of data without the necessary scholarship and expertise, and that you can understand something complicated without putting in all the effort required to understand it: the Church of Dunning–Kruger Dilettantism.
Of course, anyone is free to write their thoughts on anything, and readers are free to form opinions on what they read. What this reader sees here is two people arguing over something that both know far too little about to offer the relevant insight. What is interesting to me is that someone who's not particularly knowledgeable on the subject of crime took the time to write a long rebuttal to another post about crime written by someone else who knows just as little. I can guess that's because that church is large.
> It seems to me that both Alexander and the author of this post are, actually, members of the same church - the church of those who believe that people can draw correct conclusions from a smattering of data without the necessary scholarship and expertise, and that you can understand something complicated without putting in all the effort required to understand it. It's the Church of Dunning–Kruger Dilettantism.
We are all like that, we have no other options, haven't we? I mean, either we try to understand the world around us, or we are not. We can't be experts in everything, so in most cases we are go by Danning-Kruger Dilettantism.
Scott made the dilettantism into a profession, he has its methods and he sharpens them. He debates things with other dilettantes, and it helps them to improve themselves. To me, personally, it is one of the main attractions of the blog. I'm dilettante in a lot of topics, but still I don't want to simply ignore them, because I'm not an expert.
> What is interesting to me is that someone who's not particularly knowledgeable on the subject of crime took the time to write a long rebuttal to another post about crime written by someone else who knows just as little.
It is not about crime really. The author we discussing talks about methodology, they are on a meta level of a discussion, the crime discussion is just one data point for a meta-discussion.
Your post is the part of the same meta-discussion about methodology, though your attack comes from the other direction.
Ah. You refer to Rationalism.
Thank you for this, I was struggling to put my finger on it.
I have a different problem that I would also describe as a Church of Graphs.
I keep reading essays in which the author makes some claim and supports it by displaying a graph. The graph is not explained other than as proof that the claim it supports is correct. The axes are unlabeled, or labeled with meaningless abbreviations.
Apparently enough people find this persuasive that the practice has become widespread. But why?
Here I was, baited by the idea that it was a rejection of knowledge graphs and the semantic web...
This is an interesting article. I feel like the point the author thinks he's making isn't maybe the one he's actually making, or at least not the one he ought to be making.
The problem is he sets this up as a contrast between, on the one side, quantification, evidence, "graphs", and the like; and, on the other side, "your eyes", "lived experience", and so on.
But these are not necessarily in opposition. There is nothing unquantifiable about "lived experience" or people opinions about crime, nor is there any reason to dismiss such data as irrelevant to policy decisions.
Even if the "church of graphs" showed crime on a clear upswing, it would be absurd to say, "Crime has gone up, therefore we must build a new prison." To justify that action requires more than just that bare fact; it requires some kind of causal analysis that explains why that action would play a causal role in producing some desirable effect (like reducing crime).
On the flip side, it is not absurd to say "Surveys show that the perceived level of crime has gone up, so we should explore policies to address that." This is especially true if you swap "perceived level of crime has gone up" for "perceived quality of life has gone down", because perception is in some measure an irrefutable judgment on quality of life. (That is, if you think your quality of life has gone done, then to at least some degree it factually has, because part of what it means to have a good life is to know that your life is good and to be happy about that.) Such a swap is likely warranted, because many of the author's examples of "crime" in the article make more sense as examples of quality of life. Seeing things locked up in stores is not experiencing crime or even perceiving an increase in crime; it is experiencing a decline in quality of life which may plausibly be an effect of an increase in crime, but that's not the same thing.
So just having data doesn't tell you what to do, and just having feelings and perceptions doesn't mean you shouldn't do anything. What's missing in both cases is the causal explanation of how the data and/or the perceptions arose.
Whenever I see people talking about "lived experience" I get a bit leery, because often that seems to be a lead-in to an argument of the form "I personally experienced X, therefore large-scale change Y should be implemented." The fallacy there is not starting from perception or from gut feelings; it's starting from just your own perceptions and gut feelings. If you can get data that shows a lot of people share your perceptions and gut feelings, then we can have something to work with. What we do with that information can vary: sometimes there is a causal theory to be developed and action to be taken that can trickle down into a change in those perceptions; sometimes the answer is better education or messaging that makes clear to people that their perceptions were inaccurate. But the problem is not a "church of graphs".
With regard to the issue of crime as discussed in this article, it seems likely to me that the data adduced in support of the "there is no crime problem" position is missing something important that has a genuine impact on people's quality of life. This doesn't mean the data we have is wrong or irrelevant; it just means it's not the whole story. If you have a bunch of data on temperatures in different places around the world and you use that to pick the best place to live, you may be disappointed if you get there and find it's raining all the time. That doesn't mean your data was bad (temperature surely is a major determinant of what makes us like a certain climate) but that it's incomplete (you need more than just temperature).
The solution to this is not to give up on data, it's to bring more data into the fold. Data on people's perceptions is immensely useful as a starting point for policy. It's not an endpoint, but then neither is any other data.
An interesting article. I wonder if there's a valid point in here buried somewhere underneath the endless obnoxious comparisons of his opponents to cultists.
I’m not familiar with this writer or the writer this is about so can someone help me out with something? This article is fawning over this Scott person, talks about how the audience is more intelligent than average and about how ‘impeccably sourced and credentialed’ this Scott person’s arguments are.
Am I missing an in joke somewhere or do people actually write like this?
Scott Alexander, https://www.astralcodexten.com/ is one of the most influential bloggers in tech spaces.
I was a keen reader, but don't follow so much anymore.
That said, I don't think his blogging influence is that large, but whose is?
He's likely fawning over Scott because he wants the post to read by readers of Scott.
"A spoonful of sugar helps the medicine go down"
Idk, the whole article is super weird, even though I agree with the point
> Members of The Church of Graphs live by one primary commandment: thou shalt not believe your lying eyes.
Ah, how familiar this is from some colleagues in tech.
Demands of evidence are asymmetric: make a bold claim that's aligned with the group, and it slips by; make a hint of a misaligned claim, and you get chided for not being a researcher in the field and spreading misinformation.
Ironically, it is a Scott Alexander post that articulates this phenomenon best: https://slatestarcodex.com/2014/08/14/beware-isolated-demand...
> I increasingly find myself in disagreement with Scott’s essays on social issues and public policy, despite broadly sharing his small-L liberal outlook.
Well, there's your problem. Scott isn't a "small-L liberal." He does a decent job at masquerading as one, but ask a fan to recount his "greatest hits" and they're all boring old orthodox conservatism: race realism [1], IQ [2], anti-identity politics [3], etc.
(No, I'm in the mood to debate his positions on any of this, it's all been done to death and further debate isn't going to change anyone's mind, let alone his. The citations are there to establish that he is aligned with these views, whether or not it's warranted.)
One fringe benefit of belonging to "The Church of Graphs" that I don't think the author really touches on is that believers can do motivated reasoning _very_ easily. Scott is an expert at laundering his motivated reasoning through well-researched citations and data that supports his points, but he's not so great at giving the other side a fair hearing.
[1] https://www.astralcodexten.com/p/how-should-we-think-about-r...
[2] https://www.astralcodexten.com/p/how-to-stop-worrying-and-le...
[3] https://www.astralcodexten.com/p/against-against-boomers
Author loses me when he starts pearl clutching about the harms of seeing boarded-up windows, and his wife having to walk past unhoused drug users. (Who, let us be clear, are the ones who are actually experiencing harm.)
This guy isn't a liberal, he's a guy looking to justify his discomfort by dressing it up with a bit of rehydrated bible-school epistemology
His wife and the unhoused drug users can both can be worthy of sympathy and consideration.
Yes, seeing someone unhoused is the REAL homelessness