As someone who works in data science (bioinformatics), I couldn’t agree more. Even in technical fields, analysis is never just mechanical number-crunching—it’s shaped by human discernment: which variables to include, how to frame the question, what assumptions to privilege. At its core, data science is as much an art of judgment as it is a science of measurement. Unlike the physical sciences, where controlled experiments and natural laws provide a stable foundation, data science deals with messy, contingent, and socially constructed data. In practice, this means models are always provisional, interpretive, and deeply dependent on human choices. In theory, too, the field was never purely positivist: it emerged as a hybrid discipline precisely because it had to combine rigorous statistical methods with the softer skills of inference, domain knowledge, and interpretive judgment. That’s what makes it powerful—but also what demands humility, a quality often missing in political data analysis.
That's exactly right, I'm very much an amateur regarding social science and data analysis and I have a very STEM oriented way of looking at things. So I guess "positivism" with it's quantitative way of looking at things is a natural fit for me and other "quant types" like Nate Silver, who I still follow. I think it's more of a style of problem solving, but I think it's easier to be overconfident if you have that quantitative approach. And I think Eric Levitz has done some very good work, here is a nice in-depth takedown of neoliberalism: https://nymag.com/intelligencer/2023/02/industrial-policy-designer-economy.html
The obsession among people like this with polling and data stems from the same self defeating impulse that leads them to believe that policy can’t be good unless it’s complicated and abstruse and shows off just how darn *smart* the boys wonder who devised it are. Nothing simple can be good; if people who aren’t in the club can comprehend it, it must be naive and dumb and childish.
Liberalism is dead in the water until people like this kill the teacher in their heads that they’re trying to impress for an A+ on their report card.
The implication of this, which I think is the absolute truth, is that a more durable, successful and broadly appealing left-of-center politics needs to be, to put a fine point on it, genuinely just objectively dumber and worse.
We, politics-knowers of the world, will reject it when it comes, MUST reject it when it comes, the disgust of the politics-knowers (Ganz and Levitz alike) will act as its fuel.
I know that's a trollish thing to say, but I really do believe it's true.
There is an underpants gnomes aspect to the whole Popularism thing, where it is tautological that political candidates seen as more "moderate", or closer to the views of the responding voter (a more valuable way of asking the question IMO) will be more successful, but Popularism lacks a method of HOW to accomplish that beyond an endless procession of Heidi's Heitkamp being fed into a wood chipper every cycle, forever. It's a Moneyball of losing 100 games every year by being ever more dedicated to the crusty old scouts.
The signal polling result of our age was Donald Trump being rated as significantly more "moderate" than Hillary Clinton by 2016 voters. That breaks the model that Popularism purports to offer, and I think identifying rhetoric and an ability to form a public opinion rather than merely forensically chase its shadow gets us closer to understanding why.
The question going forward for the Democratic Party isn't what should it say and do, it's how can it even be heard in the first place?
The discussion of “the vulgar positivism of the center right” captures what grates me about Matt Yglesias, his substack Slow Boring, and the discussion there. Thank you parsing it.
Somehow the techy numbers side of the purported center left got tons of political capital. And they are boring, unappealing and unpopular a lot of the time. It makes sense this could be a solid chunk of why the left is losing the culture war and national elections. Trump made Fauci into a reality tv show character of this stereotype after all …as a foil to contrast against himself. Being technically correct, *actually*, in politics, means so little without defining the context, laying out vision.
And that is what, imo, the left is missing, a unifying , widely appealing vision (preferably coming from an inspirational leader)
The focus on polling changes not only how news events are covered, but what events get covered in the first place. A lot of this is reflected in criticism of how horserace vs stakes-based priorities shape political news media coverage and not just in what the parties do, though they are related.
Issue polling has always struck me as one of those cases where the very act of measurement disturbs and distorts the object (public opinion) being measured. The subjective experience of being interrogated on a series of explicitly political ‘issues’ is so completely alien to how normal people engage with politics that I can’t help but think it’s unhelpful. Using them to say that a majority of people think this or that is insane - no, a majority of people have not been asked the question, will never be asked the question and probably don’t even conceive of it as being a choice they can or should make. In that sense horse-race polling is more justifiable, since at the very least the public will really be asked to put themselves in one box or the other at some point.
It definitely seems as if poll respondents are getting more sophisticated at understanding that basically everything they are being asked is being done for the purpose of sorting them into the tribal political binary and will be analyzed that way, so they answer accordingly.
So much was made of Republicans changing their opinions on the state of the economy based on which party controls the White House, but since wonks consider economic polling to be so rigidly determinative of electoral outcomes, who is fooling who there exactly?
I'm split on this: on the one hand, it's true that polling is super limited in what it can tell us about the world, and a too-narrow technocratic vision of politics just comes across as soulless, A/B-tested, and dead-eyed.
On the other hand, it's true that lots of people use the above to ignore inconvenient facts about how well-disposed the public is to their views in a way that is strategically damaging to overall left goals. I don't think you do this, but I think it's fair for the data folks to read what you say in that context.
I like your rhetorical style: I think the fact that you do analysis that is rigorous, methodical, deeply informed, but almost entirely humanist is one of the best features of your writing; and I agree that there's something deadening and corporate in abandoning that mode for vulgar popularism.
But... On the other hand, we live in a world where the MCU, the epitome of soulless corporate product, is pretty popular!
Lincoln could refer to Shakespeare and the KJV secure in the knowledge that those were cultural touchstones--that other people would understand the language and the references. He also lived before TV and mass media.
For better or for worse, I think we've both lost some of the cultural literacy that made that kind of rhetoric resonate with people; and also the rise of statistics and technocracy and big science means that we've incorporated a lot of positivist vocabulary into our sense of what makes rhetoric effective.
I don't want to caricature your view: I know you're not saying to just add a bunch of "wherefores" and "howsoevers" and biblical quotes to your arguments--but I think even if you ask, what does a modernized version of a Lincoln or FDR speech look like, it probably is gonna sound a lot more technocratic and positivist than the old version!
I think what I'm trying to say is, I don't like the idea of polarizing the conversation around the dichotomy of "positivism vs humanism"--i think both have their place, in terms of strategy and persuasion.
As a scholar of media and communications, it’s notable how the entire history of modern mediated politics is saturated with exactly this “both-and” and not the old (C.P. Snow-pilled) “either/or.” Efficient mediated communication is almost always predicated on _both_ pure artful rhetoric appeal _and_ pseudoscientific audience measurements.
A striking feature of John’s original argument is how similar it is to that of Jean Baudrillard, who also argued that the very notion of the audience itself – the public itself – is mediated. It’s an enactment, a construction, that _all_ industries of publicity (marketing, propaganda, mediated entertainment, mass politics) require to even exist in the first place. And if we go back even further, we of course find Walter Lippmann lurking in the background.
My only peeve with John’s writing here is that the “measurement error” he notes – can indeed itself be mitigated statistically. Given a somewhat unchanging target population, somewhat unchanging external conditions, strict adherence to method (e.g. truly random sampling), and faithfulness to replicability, the same set of indeed flawed questions _repeated over time_ can reveal insights about the population. If the same questions are asked year in, year out, the fluctuations of the answers, in aggregate, can indeed be revealing. In short, longitudinal studies do have some validity, at least in this respect. But the U.S. might be too diverse and divisive a continent, and too broken by the enactments noted in the comment above, to make such approaches work. The approach I’ve outlined here might perhaps only be workable in smaller and more linguistically homogeneous countries. Maybe Pew has a study to prove it!
While the right continually agitate to drag the Overton window* to the right, the centrists respond with "triangulation" - i.e. let's hippie punch down to our left and trail along rightwards following the window and never fighting it's rightward drift. These motherfuckers would triangulate us all onto the cattle trucks, if we let them. Screw their political cowardice. And above all screw their "it's not ideology, we're just following the data" bullshit. The ideology that poses as being above ideology is the most dishonest of all (as you rightly said). They're not following the data, they're following the people (and media) dragging the window to the right.
This is another reason why one of my pet peeves is the evisceration of the word "performativity" to just another synonym for play-acting. J L Austin's original distinction between language that was merely constative - registering or describing the "is" - and that which is about bringing about the change it wants to see in the world - the "ought" is vital for understanding the basics of political militancy in the field of rhetoric. Was Marc Anthony really come to bury Caesar? Was he hell. That's the point of rhetoric. To MOVE people. To change their opinions. To sway them to see the same facts from a different perspective, a different set of values. Lincoln didn't write the Gettysburg Address on the back of the goddam polls. And certainly at the start of the Civil War, had their been polls, they would definitely have indicated that a majority of white voters in the North were not in favour of liberating the slaves. Does that mean Jefferson Davies was "following the data"? Seriously, fuck these guys. And the hobby horse they rode in on.
----
* Named after an American RW PR guy who dedicated his life to doing just that (dragging the window of centrist "common sense" to the right). Although, as a political strategy, the AF cognoscenti will know that the first FR figure to articulate this strategy for power publicly, was the Austrian neo-nazi Jörg Haider
"𝑉𝑢𝑙𝑔𝑎𝑟" positivism? Is there any *other* kind of positivism than "𝑉𝑢𝑙𝑔𝑎𝑟"? I think not. Your response hits all the right notes, most especially (to speak pro-speak) on the topic of "𝐭𝐡𝐞 𝐭𝐡𝐞𝐨𝐫𝐲-𝐥𝐚𝐝𝐞𝐧𝐧𝐞𝐬𝐬 𝐨𝐟 𝐨𝐛𝐬𝐞𝐫𝐯𝐚𝐭𝐢𝐨𝐧" discussed most thoroughly by philosopher of science and fighter pilot N. R. Hanson. To wit, there are no 𝒑𝒖𝒓𝒆 empirical observations, all perception is mitigated by theoretical frameworks.
Or to put it clearly and simply, we need only remember the 𝗧𝗵𝗿𝗲𝗲 𝗨𝗺𝗽𝗶𝗿𝗲𝘀: Ump 1: "I call 'em the way they ARE." Ump 2: "I call 'em the way I see 'em." Ump 3: "They are what I call 'em."
Eric, Ump 1, basically accuses you of being Ump 3, but of course, any reasonable (!) person can see that you're Ump 2 in actual fact.
Once upon a time I was paid by a rocket company to do stat for the USAF. Then I taught stat, logic, and physics at an R2 for 43 years. I *love* this kind of stuff.
The bit about mutilating reality for ideological interests made me think of the second book of Spinoza's Ethics (P28) "The ideas of the modifications of the human body as they have reference only to the human mind, are not clear and distinct, but confused." I believe the Penguin edition even substitutes 'mutilated' for 'confused.'
Not sure if the wording was explicitly referencing this, but I think it corresponds well to your point about not getting a semblance of clear meaning from polls, which are several layers of ideas of modifications away from direct reference to the mind.
I’ve always found it interesting that the one thing the polling/data guys appear to have no curiosity about whatever is looking at or trying to measure the corrosive impact they themselves have had on political processes and outcomes.
It seems that progressives like polling just fine when it jibes with their preferred policies. As someone who is a polling skeptic, I would have more respect for polling skepticism from the left if they applied it to the polling results they believe supports their policy agenda. The same epistemological problems you identify in this piece, like what does the pollster mean by the terms used in their questions and how do we know the respondents understand them in the same way, exist when the poll is showing support for universal health care or any other social democratic policies.
Exit polling seems to me one of the best gauges we have as to why people voted how they did, because you are talking to people immediately after they voted and asking them for their reasoning. It seems like an inconvenient fact for progressives that swing voters who opted for Trump thought of Harris as too far to the left. This seems like a potent argument in favor of more moderation, not less, regardless of the finer points of what people think Real Leftism™ is.
It seems to me that the obsession with polling is a symptom of a wider problem, an impulse that becomes more and more overwhelming as the elections are higher, as the audience gets wider, as one has more to lose and as the stakes become immense, I'm talking of the urge of "playing it safe". It wouldn't be sound for politicians to be kamikazes, but there's a point where not taking risks becomes a liability and stifles all of the life out of a campaign. You can't blame her so much when you see what were the stakes (and how she was thrown in), but Kamala Harris' candidacy obviously died from an overdose of playing it safe : "you're ahead in polls, play it safe", "he's speaking nonsense, he's digging his grave, play it safe", "democracy depends on you, play it safe", "don't give geniune interviews, script what you say, play it safe", "follow what the consultants are saying, play it safe", "don't tell too precisely what you will do, people might disagree, play it safe"
We're seeing that same urge in the apathy of the democrat leadership : "he's digging his grave, we'll reap the benefits in the midterms, let's say nothing, play it safe", something Sen. Chris Murphy framed correctly (in his usual nice boy way) when he said the Democrats had to be ready to take more tactical risks (they, of course, also have that near insurrection of young Dems who want to primary some of their useless coastal seniors). I was not so surprised when reading the list of the senators who voted for Bernie Sanders' proposition to block arm sales to Israel to see most of the names of the senators who had me be going "oh cool, what that guy is doing is interesting" in the recent monthes : they're the ones who realized they would die from not changing the statu quo.
This has probably always been a staple of politics, but it seems to me it's taken a new level now that there is so much more money involved in political campaigns : "so much money has been invested in that campaign, you can't take the risk of being yourself, play it safe", "don't be too populist, donors might want to invest against you, play it safe". Someone else in the comments told of the MCU, I think it's an acute comparison. Massive money has the same effect in politics as in blockbusters : it creates products with a broad appeal, but no soul. There is too much money involved for a person to be able to put much of a mark on them. The public is not attached to them for their own merits, it is because of a legacy and because of a longing for more geniune avatars of a more distant past. They meet reasonable success for a time, but eventually people get bored. The new blood and the next geniune surprise hit will come from a smaller budget product from an independent studio (even if there are less and less independent studios that can produce medium budget films... a big but different problem)
The problem with playing it too safe is that it goes against the very fabric of politics and of democracy : that of having a thriving public debate, where people identify problems and share openly views that are conflicting, advancing to a solution via this exchange of views. If a politician will not debate, will not engage, will not defend views with fierceness and openness, then what's the point ?
Of course, it's easier for underdogs who have nothing to lose to take risks - that's why Trump who came in politics as an outsider started out doing so, and could continue on his brand ; that's why Zohran Mamdani was so good during the primary, and we can see he understandably plays more safe now that he's the frontrunner. Which is why, I suppose, it's worth wondering where are the places where political talent can cut its teeth, and how it can be supported to move up.
I'm always reminded of Lasch's Agony of the Left when I run into these crossfires and grateful if I even know and recognize the target and think it deserving of the broadside, but also cringe a little with you if I don't recognize the target for fear the wrathful critique might smear or somehow insult me. The vanity thing cuts both ways with writers and readers. Thankfully, in this instance I know the Vox policy wonk numbers people and without resorting to a bunch of name-calling this articulates exquisitely my nagging sense that their liberal commitments are not so much to liberal human rights, labor rights, workers, protecting the social safety net or protecting the environment but their libertarian liberal commitments are to free capital and the pursuit of monopoly rents. Their polling popularism is a cudgel to reinforce austerity and efficiency tradeoffs defined by billionaires.The Kochs et al. And their never-talk-about-taxing-the-rich "moderation" further fuels the underclass reaction and revolt against liberal elites. Although, for that matter, so does almost any historically literate writing about politics. It's all CRT to the hoi polloi. But some liberal elite needs to speak up against this plutocratic BS. Glad you have.
As someone who works in data science (bioinformatics), I couldn’t agree more. Even in technical fields, analysis is never just mechanical number-crunching—it’s shaped by human discernment: which variables to include, how to frame the question, what assumptions to privilege. At its core, data science is as much an art of judgment as it is a science of measurement. Unlike the physical sciences, where controlled experiments and natural laws provide a stable foundation, data science deals with messy, contingent, and socially constructed data. In practice, this means models are always provisional, interpretive, and deeply dependent on human choices. In theory, too, the field was never purely positivist: it emerged as a hybrid discipline precisely because it had to combine rigorous statistical methods with the softer skills of inference, domain knowledge, and interpretive judgment. That’s what makes it powerful—but also what demands humility, a quality often missing in political data analysis.
That's exactly right, I'm very much an amateur regarding social science and data analysis and I have a very STEM oriented way of looking at things. So I guess "positivism" with it's quantitative way of looking at things is a natural fit for me and other "quant types" like Nate Silver, who I still follow. I think it's more of a style of problem solving, but I think it's easier to be overconfident if you have that quantitative approach. And I think Eric Levitz has done some very good work, here is a nice in-depth takedown of neoliberalism: https://nymag.com/intelligencer/2023/02/industrial-policy-designer-economy.html
The obsession among people like this with polling and data stems from the same self defeating impulse that leads them to believe that policy can’t be good unless it’s complicated and abstruse and shows off just how darn *smart* the boys wonder who devised it are. Nothing simple can be good; if people who aren’t in the club can comprehend it, it must be naive and dumb and childish.
Liberalism is dead in the water until people like this kill the teacher in their heads that they’re trying to impress for an A+ on their report card.
The implication of this, which I think is the absolute truth, is that a more durable, successful and broadly appealing left-of-center politics needs to be, to put a fine point on it, genuinely just objectively dumber and worse.
We, politics-knowers of the world, will reject it when it comes, MUST reject it when it comes, the disgust of the politics-knowers (Ganz and Levitz alike) will act as its fuel.
I know that's a trollish thing to say, but I really do believe it's true.
There is an underpants gnomes aspect to the whole Popularism thing, where it is tautological that political candidates seen as more "moderate", or closer to the views of the responding voter (a more valuable way of asking the question IMO) will be more successful, but Popularism lacks a method of HOW to accomplish that beyond an endless procession of Heidi's Heitkamp being fed into a wood chipper every cycle, forever. It's a Moneyball of losing 100 games every year by being ever more dedicated to the crusty old scouts.
The signal polling result of our age was Donald Trump being rated as significantly more "moderate" than Hillary Clinton by 2016 voters. That breaks the model that Popularism purports to offer, and I think identifying rhetoric and an ability to form a public opinion rather than merely forensically chase its shadow gets us closer to understanding why.
The question going forward for the Democratic Party isn't what should it say and do, it's how can it even be heard in the first place?
The discussion of “the vulgar positivism of the center right” captures what grates me about Matt Yglesias, his substack Slow Boring, and the discussion there. Thank you parsing it.
Somehow the techy numbers side of the purported center left got tons of political capital. And they are boring, unappealing and unpopular a lot of the time. It makes sense this could be a solid chunk of why the left is losing the culture war and national elections. Trump made Fauci into a reality tv show character of this stereotype after all …as a foil to contrast against himself. Being technically correct, *actually*, in politics, means so little without defining the context, laying out vision.
And that is what, imo, the left is missing, a unifying , widely appealing vision (preferably coming from an inspirational leader)
The focus on polling changes not only how news events are covered, but what events get covered in the first place. A lot of this is reflected in criticism of how horserace vs stakes-based priorities shape political news media coverage and not just in what the parties do, though they are related.
Issue polling has always struck me as one of those cases where the very act of measurement disturbs and distorts the object (public opinion) being measured. The subjective experience of being interrogated on a series of explicitly political ‘issues’ is so completely alien to how normal people engage with politics that I can’t help but think it’s unhelpful. Using them to say that a majority of people think this or that is insane - no, a majority of people have not been asked the question, will never be asked the question and probably don’t even conceive of it as being a choice they can or should make. In that sense horse-race polling is more justifiable, since at the very least the public will really be asked to put themselves in one box or the other at some point.
It definitely seems as if poll respondents are getting more sophisticated at understanding that basically everything they are being asked is being done for the purpose of sorting them into the tribal political binary and will be analyzed that way, so they answer accordingly.
So much was made of Republicans changing their opinions on the state of the economy based on which party controls the White House, but since wonks consider economic polling to be so rigidly determinative of electoral outcomes, who is fooling who there exactly?
I'm split on this: on the one hand, it's true that polling is super limited in what it can tell us about the world, and a too-narrow technocratic vision of politics just comes across as soulless, A/B-tested, and dead-eyed.
On the other hand, it's true that lots of people use the above to ignore inconvenient facts about how well-disposed the public is to their views in a way that is strategically damaging to overall left goals. I don't think you do this, but I think it's fair for the data folks to read what you say in that context.
I like your rhetorical style: I think the fact that you do analysis that is rigorous, methodical, deeply informed, but almost entirely humanist is one of the best features of your writing; and I agree that there's something deadening and corporate in abandoning that mode for vulgar popularism.
But... On the other hand, we live in a world where the MCU, the epitome of soulless corporate product, is pretty popular!
Lincoln could refer to Shakespeare and the KJV secure in the knowledge that those were cultural touchstones--that other people would understand the language and the references. He also lived before TV and mass media.
For better or for worse, I think we've both lost some of the cultural literacy that made that kind of rhetoric resonate with people; and also the rise of statistics and technocracy and big science means that we've incorporated a lot of positivist vocabulary into our sense of what makes rhetoric effective.
I don't want to caricature your view: I know you're not saying to just add a bunch of "wherefores" and "howsoevers" and biblical quotes to your arguments--but I think even if you ask, what does a modernized version of a Lincoln or FDR speech look like, it probably is gonna sound a lot more technocratic and positivist than the old version!
I think what I'm trying to say is, I don't like the idea of polarizing the conversation around the dichotomy of "positivism vs humanism"--i think both have their place, in terms of strategy and persuasion.
As a scholar of media and communications, it’s notable how the entire history of modern mediated politics is saturated with exactly this “both-and” and not the old (C.P. Snow-pilled) “either/or.” Efficient mediated communication is almost always predicated on _both_ pure artful rhetoric appeal _and_ pseudoscientific audience measurements.
A striking feature of John’s original argument is how similar it is to that of Jean Baudrillard, who also argued that the very notion of the audience itself – the public itself – is mediated. It’s an enactment, a construction, that _all_ industries of publicity (marketing, propaganda, mediated entertainment, mass politics) require to even exist in the first place. And if we go back even further, we of course find Walter Lippmann lurking in the background.
My only peeve with John’s writing here is that the “measurement error” he notes – can indeed itself be mitigated statistically. Given a somewhat unchanging target population, somewhat unchanging external conditions, strict adherence to method (e.g. truly random sampling), and faithfulness to replicability, the same set of indeed flawed questions _repeated over time_ can reveal insights about the population. If the same questions are asked year in, year out, the fluctuations of the answers, in aggregate, can indeed be revealing. In short, longitudinal studies do have some validity, at least in this respect. But the U.S. might be too diverse and divisive a continent, and too broken by the enactments noted in the comment above, to make such approaches work. The approach I’ve outlined here might perhaps only be workable in smaller and more linguistically homogeneous countries. Maybe Pew has a study to prove it!
Just to say that I liked your MCU comparison and referenced it in my comment (with a slightly different take), if you want to take a look :)
https://open.substack.com/pub/johnganz/p/against-vulgar-positivism?r=3hnvrl&utm_campaign=comment-list-share-cta&utm_medium=web&comments=true&commentId=147479239
Thank you for pointing out the insidious use of the term "prior" to mean belief. This has been a real pet peeve of mine.
While the right continually agitate to drag the Overton window* to the right, the centrists respond with "triangulation" - i.e. let's hippie punch down to our left and trail along rightwards following the window and never fighting it's rightward drift. These motherfuckers would triangulate us all onto the cattle trucks, if we let them. Screw their political cowardice. And above all screw their "it's not ideology, we're just following the data" bullshit. The ideology that poses as being above ideology is the most dishonest of all (as you rightly said). They're not following the data, they're following the people (and media) dragging the window to the right.
This is another reason why one of my pet peeves is the evisceration of the word "performativity" to just another synonym for play-acting. J L Austin's original distinction between language that was merely constative - registering or describing the "is" - and that which is about bringing about the change it wants to see in the world - the "ought" is vital for understanding the basics of political militancy in the field of rhetoric. Was Marc Anthony really come to bury Caesar? Was he hell. That's the point of rhetoric. To MOVE people. To change their opinions. To sway them to see the same facts from a different perspective, a different set of values. Lincoln didn't write the Gettysburg Address on the back of the goddam polls. And certainly at the start of the Civil War, had their been polls, they would definitely have indicated that a majority of white voters in the North were not in favour of liberating the slaves. Does that mean Jefferson Davies was "following the data"? Seriously, fuck these guys. And the hobby horse they rode in on.
----
* Named after an American RW PR guy who dedicated his life to doing just that (dragging the window of centrist "common sense" to the right). Although, as a political strategy, the AF cognoscenti will know that the first FR figure to articulate this strategy for power publicly, was the Austrian neo-nazi Jörg Haider
"𝑉𝑢𝑙𝑔𝑎𝑟" positivism? Is there any *other* kind of positivism than "𝑉𝑢𝑙𝑔𝑎𝑟"? I think not. Your response hits all the right notes, most especially (to speak pro-speak) on the topic of "𝐭𝐡𝐞 𝐭𝐡𝐞𝐨𝐫𝐲-𝐥𝐚𝐝𝐞𝐧𝐧𝐞𝐬𝐬 𝐨𝐟 𝐨𝐛𝐬𝐞𝐫𝐯𝐚𝐭𝐢𝐨𝐧" discussed most thoroughly by philosopher of science and fighter pilot N. R. Hanson. To wit, there are no 𝒑𝒖𝒓𝒆 empirical observations, all perception is mitigated by theoretical frameworks.
Or to put it clearly and simply, we need only remember the 𝗧𝗵𝗿𝗲𝗲 𝗨𝗺𝗽𝗶𝗿𝗲𝘀: Ump 1: "I call 'em the way they ARE." Ump 2: "I call 'em the way I see 'em." Ump 3: "They are what I call 'em."
Eric, Ump 1, basically accuses you of being Ump 3, but of course, any reasonable (!) person can see that you're Ump 2 in actual fact.
Once upon a time I was paid by a rocket company to do stat for the USAF. Then I taught stat, logic, and physics at an R2 for 43 years. I *love* this kind of stuff.
Onward!
An indulgence in polling has delivered us a politics from which the ideas of originality or counter-intuition are absent.
The bit about mutilating reality for ideological interests made me think of the second book of Spinoza's Ethics (P28) "The ideas of the modifications of the human body as they have reference only to the human mind, are not clear and distinct, but confused." I believe the Penguin edition even substitutes 'mutilated' for 'confused.'
Not sure if the wording was explicitly referencing this, but I think it corresponds well to your point about not getting a semblance of clear meaning from polls, which are several layers of ideas of modifications away from direct reference to the mind.
I’ve always found it interesting that the one thing the polling/data guys appear to have no curiosity about whatever is looking at or trying to measure the corrosive impact they themselves have had on political processes and outcomes.
It seems that progressives like polling just fine when it jibes with their preferred policies. As someone who is a polling skeptic, I would have more respect for polling skepticism from the left if they applied it to the polling results they believe supports their policy agenda. The same epistemological problems you identify in this piece, like what does the pollster mean by the terms used in their questions and how do we know the respondents understand them in the same way, exist when the poll is showing support for universal health care or any other social democratic policies.
Exit polling seems to me one of the best gauges we have as to why people voted how they did, because you are talking to people immediately after they voted and asking them for their reasoning. It seems like an inconvenient fact for progressives that swing voters who opted for Trump thought of Harris as too far to the left. This seems like a potent argument in favor of more moderation, not less, regardless of the finer points of what people think Real Leftism™ is.
It seems to me that the obsession with polling is a symptom of a wider problem, an impulse that becomes more and more overwhelming as the elections are higher, as the audience gets wider, as one has more to lose and as the stakes become immense, I'm talking of the urge of "playing it safe". It wouldn't be sound for politicians to be kamikazes, but there's a point where not taking risks becomes a liability and stifles all of the life out of a campaign. You can't blame her so much when you see what were the stakes (and how she was thrown in), but Kamala Harris' candidacy obviously died from an overdose of playing it safe : "you're ahead in polls, play it safe", "he's speaking nonsense, he's digging his grave, play it safe", "democracy depends on you, play it safe", "don't give geniune interviews, script what you say, play it safe", "follow what the consultants are saying, play it safe", "don't tell too precisely what you will do, people might disagree, play it safe"
We're seeing that same urge in the apathy of the democrat leadership : "he's digging his grave, we'll reap the benefits in the midterms, let's say nothing, play it safe", something Sen. Chris Murphy framed correctly (in his usual nice boy way) when he said the Democrats had to be ready to take more tactical risks (they, of course, also have that near insurrection of young Dems who want to primary some of their useless coastal seniors). I was not so surprised when reading the list of the senators who voted for Bernie Sanders' proposition to block arm sales to Israel to see most of the names of the senators who had me be going "oh cool, what that guy is doing is interesting" in the recent monthes : they're the ones who realized they would die from not changing the statu quo.
This has probably always been a staple of politics, but it seems to me it's taken a new level now that there is so much more money involved in political campaigns : "so much money has been invested in that campaign, you can't take the risk of being yourself, play it safe", "don't be too populist, donors might want to invest against you, play it safe". Someone else in the comments told of the MCU, I think it's an acute comparison. Massive money has the same effect in politics as in blockbusters : it creates products with a broad appeal, but no soul. There is too much money involved for a person to be able to put much of a mark on them. The public is not attached to them for their own merits, it is because of a legacy and because of a longing for more geniune avatars of a more distant past. They meet reasonable success for a time, but eventually people get bored. The new blood and the next geniune surprise hit will come from a smaller budget product from an independent studio (even if there are less and less independent studios that can produce medium budget films... a big but different problem)
The problem with playing it too safe is that it goes against the very fabric of politics and of democracy : that of having a thriving public debate, where people identify problems and share openly views that are conflicting, advancing to a solution via this exchange of views. If a politician will not debate, will not engage, will not defend views with fierceness and openness, then what's the point ?
Of course, it's easier for underdogs who have nothing to lose to take risks - that's why Trump who came in politics as an outsider started out doing so, and could continue on his brand ; that's why Zohran Mamdani was so good during the primary, and we can see he understandably plays more safe now that he's the frontrunner. Which is why, I suppose, it's worth wondering where are the places where political talent can cut its teeth, and how it can be supported to move up.
I'm always reminded of Lasch's Agony of the Left when I run into these crossfires and grateful if I even know and recognize the target and think it deserving of the broadside, but also cringe a little with you if I don't recognize the target for fear the wrathful critique might smear or somehow insult me. The vanity thing cuts both ways with writers and readers. Thankfully, in this instance I know the Vox policy wonk numbers people and without resorting to a bunch of name-calling this articulates exquisitely my nagging sense that their liberal commitments are not so much to liberal human rights, labor rights, workers, protecting the social safety net or protecting the environment but their libertarian liberal commitments are to free capital and the pursuit of monopoly rents. Their polling popularism is a cudgel to reinforce austerity and efficiency tradeoffs defined by billionaires.The Kochs et al. And their never-talk-about-taxing-the-rich "moderation" further fuels the underclass reaction and revolt against liberal elites. Although, for that matter, so does almost any historically literate writing about politics. It's all CRT to the hoi polloi. But some liberal elite needs to speak up against this plutocratic BS. Glad you have.