Andrew Eggers and Justin Grimmer write:
Lott (2022) introduces ‘simple tests’ of voter fraud and applies them to measure the extent of fraud in the 2020 election. Using the tests, Lott (2022) claims to have discovered 10,000 extra votes for Biden in Pennsylvania and Georgia based on mishandling of absentee votes, 6,700 extra votes for Biden in Pennsylvania from inappropriate allowances for provisional votes, and 255,000 excess votes due to artificially large turnout across several counties in six key states. All three claims are demonstrably false: the first claim relies on a specification error that Lott (2022) inadvertently reintroduced into the analysis after we identified and corrected the same error in the first public draft of this paper; the second claim is based on analysis that, by its own logic, would show larger fraud in favor of Trump; the third depends on selective reporting, as Lott (2022) shows insignificant results when differential trends in turnout across states are acknowledged. As we document in the Appendix, Lott’s (2022) analysis is riddled with errors and fails to accurately report the research that was conducted. At several points the paper misstates the regression specification being used (as we confirm through replicating Lott’s (2022) results), misreports regression coefficients (always in a direction favorable to his argument), and incorrectly reports variables used to produce a result. After correcting these errors, we show Lott’s (2022) analysis fails to provide any evidence of voter fraud in the 2020 election.
It’s not a surprise that Lott’s 2022 article has such huge problems, given his track record, both with this particular claim and his work more generally (for example here). And that’s not even getting into his record of lying on social media.
What’s disturbing is that this paper is scheduled to appear in Public Choice, which is a legit academic journal. I guess Lott’s a good writer, but that shouldn’t be enough. Real journals should avoid publishing papers that are riddled with errors. This isn’t the Journal of Economic Perspectives we’re talking about here, right?
The editors of Public Choice should be ashamed of themselves to publish something like this. I get that it’s a conservative journal, the editor’s a conservative, etc. But they have no more business publishing such flawed work than the liberal Lancet editor has publishing fatally-flawed politicized public health studies.
Don’t get me wrong, I believe in free speech. Public Choice has every right to publish this article. They can publish articles claiming the Loch Ness monster is real, pi = 3, Jesus was married, the moon landing was fake, JFK was killed by some dude on the grassy knoll, beautiful people are 36% more likely to have a girl baby, whatever they want. It’s a free country! But to publish this just shows very poor judgment. You’re allowed to show very poor judgment, but I don’t recommend it.
And, yeah, being published in Public Choice lends this paper some credibility. Just remember, the loan of credibility goes both ways! Lancet used to be considered a serious journal; nowadays, it’s understood to be a mix of serious science and not-so-serious politics. The Proceedings of the National Academy of Sciences gets a big loan of credibility from . . . the National Academy of Sciences. But every time PNAS publishes a junk article (hello, himmicanes!), that degrades a little bit the credibility of the NAS: that organization chose to give the keys to their journal to some academic bigshots who consistently show poor judgment, and that reflects poorly on the organization. Action and reaction. Just as Lancet is known to occasionally publish really bad papers because they push its liberal agenda, similarly we can now see that Public Choice is willing to publish a really bad paper that pushes its conservative agenda.
And, to me, that’s too bad. We already have zillions of political news outlets. I hate to see scholarly journals go this way.
Look, if you want to express your political views, go for it. Let’s say you’re the editor of Public Choice, a respected academic journal, and you have strong conservative views. You might think that Donald Trump would be a better president than Joe Biden, or you might vaguely feel that Biden’s election was illegitimate because, oh, I dunno, people voted based on the coronavirus epidemic and that wasn’t fair because the vaccine was already developed before the election. If you feel that, then, fine. Publish an editorial in the journal. You can do that, you’re the editor! You could share your views on political economy, arguing, for example, that high marginal tax rates and business regulation will stifle innovation, that the ability of liberal politicians to tax the rich and use the proceeds to woo voters is dangerous, and that for this and other reasons we should prefer corrupt Republicans to corrupt Democrats (given that these are the two choices on offer), etc etc. Say what you really feel, that’s fine! Or maybe you don’t support Trump for president, you just like the idea of publishing a claim of fraud as a poke in they eye to the liberal establishment. Fine, poke in the eye all you want! But do you really have to publish junk science? Really?? This is why you decided to edit a journal??? I feel the same way about Lancet: if you want to run an editorial saying that guns are a menace, go for it. Just present the evidence that you really have, not bad papers that make claims that are not supported by the data, so you can claim the science is on your side even when it isn’t.
Grimmer writes:
Our paper documents a series of the problems with the paper that indicate it is extremely sloppy and Lott merely running models until he gets a preferred result . Beyond the massive issue (Lott’s estimating nonsensical coefficients), he claims to have used different variables in regressions than he does and even appears to have flipped the sign on some reported coefficients so that they are consistent with what he wants to show—that there was anti-Trump fraud.
It’s impossible to know from here whether Lott is fooling himself or whether he is aware that he’s just fitting model after model until he get to whatever result he wants. Even if he knows what he’s doing, he could think it’s morally justified on the grounds that, in his view, the other side does it too.
There’s also selection bias. Most social scientists, liberal or conservative, would avoid writing such a bad paper on such a consequential topic. But there’s some market for academic papers saying that Trump should’ve won the election. Not a huge market, but some market. So you’ll see some papers on the topic.
I appreciate that Eggers and Grimmer did the exhausting work of shooting this down. Unfortunate that it’s even necessary, but I guess someone has to do it, just as in the 1960s people went to the trouble of debunking JFK conspiracy theorists, in the 1970s people went to the trouble of debunking spoon benders, etc. In this latest story, the journal Public Choice plays the role of the mainstream publishers that released books on the Bermuda triangle, ancient astronauts, etc. In that case I guess the motivation was money and fame, pure and simple. Here I’m guessing the motivation is political, with maybe a bit of self-righteous contrarianism (of the Yeah-we-publish-the-stuff-that’s-too-hot-for-the-mainstream-journals.-Got-a-problem-with-that? variety) thrown in.
P.S. More here.
My shorter abstract to this paper: Is Rosh doing any better this Shanah than last Shanah? The authors present answers to Four Questions.
Jonathan:
Say what you want about the quality of John Lott’s research, he was the the best professor Mary Rosh ever had. We know this because Rosh herself said so. And teaching should count in this world too, right? Academic shouldn’t all be about research.
Come to think of it, might this be how Public Choice ended up publishing a paper that is “riddled with errors and fails to accurately report the research that was conducted”? Maybe they received 3 glowing referee reports from Mary Rosh?
Usually you wouldn’t ask one person for three reports, but Mary Rosh is such an exceptional personality that I could believe she could be trusted to write three different reports, each using different superlatives.
I’m not in academia, so I had to google Mary Rosh. I dunno, maybe that whole thing gives Lott some street cred when it comes to the topic of artificially large turnout.
Jeff:
In all seriousness, given his demonstrated willingness to misrepresent himself and lie, it’s possible that Lott doesn’t fully understand the many people out there who don’t do that. He might think of us as suckers or rubes, or maybe he thinks we’re all liars too who just don’t admit it.
Just downstairs you had someone recommending Lott’s analysis, and objecting to the idea that past problems might be reasonably included in assessment of the credibility of Lott’s analysis.
It may be that Lott simply doesn’t care.
It shouldn’t be lost that his analysis is in effect support for Trump, who for decades cultivated and operationalized the explicitly elaborated Roger Stone/Roy Cohn policy of weaponizing lying and the strategy of doubling down when called on it.
Lying is a phenomenally effective political tactic. Some people (presumably like the commenter downstairs) don’t care about the lying as long as their identity-based biases are confirmed. An open question for me is whether being more open and explicit about the lying is some kind of growing.
Joshua:
Just speaking in general (without any reference to Lott, one way or another), I agree with your points that (a) lying can be effective (or, at least, people often seem to lie for instrumental reasons), and (b) in some areas such as politics and negotiations, lying is so commonplace that the liars themselves may well not feel that they’re doing anything wrong by lying. They’ll still deny that they were lying, but they could well just feel that lying about lying is just part of the game.
I then wonder what they think of people like us who don’t routinely lie. Do they pity us as suckers or rubes, do they simply not believe us (i.e, they think we lie just like they do), or maybe do they envy us for existing in a world in which we can simply say what we think without feeling the need to lie in order to get what we want?
Andrew –
>… they think we lie just like they do
Of course all this will vary by the individual, so I wouldn’t know how to apply a general formula.
I think there’s a pretty bright line between deliberately lying as an explicit strategy and just convincing yourself that you’re telling the truth. Inventing a sock puppet, I don’t think, can be explained by motivated reasoning. It’s likely “motivated lying,” and not akin to convincing yourself to treat uncertainty selectively – as seen with confirmation bias. So Lott is in a kind of special category.
If I had to guess (based on what I’ve encountered) I’d say it’s a matter of “winning” being the goal on a zero sum playing field – and it can be justified (as you suggest) by the view that everyone lies anyway so you’re actually being more honest if you don’t pretend you’re not lying. That seems to be a rationalization I’ve seen applied to Trump a lot.
So the follow on is a sense of impugnity that there’s certainly no harm that comes from lying and (1) it helps “win” and (2) it actually curries favor as a winning strategy and paradoxically, is seen as being honest.
So what if Lott created a sock puppet to praise himself online? It’s not like that should reflect on his credibility in any way!
In fact, I always look for people who create false identities to promote themselves as a first criterion for assessing credibility. My financial advisor created 3 false identities to write reviews online and the last lawyer I used created 4. I’m excited because I recently just found a plumber who has created no less than 7 sock puppets to write reviews.
Creating John Barron helped to establish the credibility of Donald Trump!
Next you’re going to say that claiming to have proof that Obama’s birth certificate was fake somehow called Trump’s credibility into question.
Is it possible this says more about the influence, if you will, of influencers on scientific consensus? Are Lott and Trump perhaps doing a service by exposing a flaw in a recommendation-based system of trust? Instead of just trusting that the paper in question is actually ridden with errors, why not make this blog about the raw numbers that prove the errors, so we don’t just talk about whom we trust more?
“It’s impossible to know from here whether Lott is fooling himself or whether he is aware that he’s just fitting model after model until he get to whatever result he wants.”
Given Lott’s history of inventing Mary Rosh, faking an entire survey on defensive gun use, magically changing data tables after the fact, claiming one of his papers was published in a peer-reviewed journal when it wasn’t, and having a basic data error at the heart of his claims on mass shooters targeting gun-free zones (he treated every death in a mass shooting as an individual mass shooting), it’s almost certainly option B.
Take his recent statistical analysis of Permitless Carry laws (which thankfully has not been published, yet… but has still been cited in state legislative hearings on these bills): https://crimeresearch.org/2022/01/changes-in-crime-and-killings-of-police-after-constitutional-carry-adopted/
In my organization’s rebuttal, we contacted researchers at Stanford and Johns Hopkins to analyze the STATA files. Just like in the voter fraud paper, it was filled with systematic errors that rendered the entire analysis useless: https://www.gvpedia.org/gun-myths/permitless-carry-myth/
I mean Lott can’t even spell Pittsburgh’s county correctly in the abstract: https://www.alleghenycounty.us/
Doug:
But it’s good enough for Public Choice!
I have two comments here:
– Surely the real story here is that America can’t even run an election without all this vote-count chaos. What century are we living in?
– What this guy did is the standard practice at least in psycholinguistics, and probably other areas too. Once one has decided on a position, every single model or data-set that one produces is magically consistent with the desired story. It is extremely rare for a scientist to publish a paper that goes against his/her pet beliefs. Basically, if you believe in theory X, then you will find evidence favoring theory X. It seems that is how science works in practice. For example, in psycholinguistics there is a debate about whether syntactic processing precedes semantic processing or both come into play simultaneously. Those who believe in syntax-first *always* find evidence for that claim, and those who believe in synsem acting simultaneously *always* find evidence for *that* claim, with the exact same experiment design (for psycholinguists: “the evidence examined by the lawyer” sentences)! It’s comical to watch everyone just magically finding evidence for their pet belief. The only conclusion I can come to is that each side is gaming the data; there is no other explanation.
Shravan Vasishth says,
“Surely the real story here is that America can’t even run an election without all this vote-count chaos”
What is the evidence for that? The vote count was fine. Trump just said it was repeatedly without any evidence.
Shravan Vasishth says,
that what Lott did is “standard practice at least in psycholinguistics. . . The only conclusion I can come to is that each side is gaming the data; there is no other explanation.”
If it is, then psycholinguistics isn’t a science. There is another explanation. Not all questions are well-formed questions that can be studying and subject to measurement. So, if no progress can be made on psycholinguistic questions, then that suggests that maybe the hypotheses have not be formulated (or cannot be formulated) in a way to be testable. That is not a reflection on all science.
To be fair to psycholinguistics, if you aren’t replicating each others results and testing your hypotheses (rather than a default “null hypothesis”) that could explain these reproducible results, then it is not really science.
Unfortunately, that is exactly what describes 99%+ of what gets referred to as science these days. I call it bizarro science, others have called it “cargo cult science”: https://sites.cs.ucsb.edu/~ravenben/cargocult.html
There was, I swear, a post at some point about an article observing papers on publicly charged topics to be substantially more likely to have abstracts and conclusions not supported by their own data. Can anyone point me to the paper, or the post?
Econ Journal Watch, which Andrew steered me to, has assured me that they are willing to publish high-quality critiques that do not fit comfortably with their political views. This would be a wonderful opportunity for Eggers and Grimmer to test that claim. It’s a very low-hassle journal to publish in, so it wouldn’t be a big waste of their time.
Maybe to you and your regular blog-readers, but I think the reputational damage you imagine in its future hasn’t entirely come to pass.
Wonks:
Yes, this reminds me of the famous line that the future has arrived, it’s just not evenly distributed. To the readers of this blog, Lancet is not always to be trusted, PNAS is uneven, Gladwell is a punchline, NPR is notorious for promoting iffy claims, etc. But in much of the outside world, Lancet and PNAS are gospel, Gladwell’s a savant, NPR always knows what they’re doing, etc.
This article relies on a self published paper by the article editor. He couldn’t even be bothered to get it so you can download it. Sort of like Joe Biden’s actual fraud on Hunter’s laptop this piece fails to point to any statistical problems (I don’t accept hidden papers on drop box as legitimate and don’t know anyone who does) You could have published on Google and provided a public link and done better. Being a professor at Columbia doesn’t make it any more likely you’re fraud claims are true, in fact it suggests that you claim it because of your political leanings. Maybe you need to go view the videos of the mules carrying ballots that obviously aren’t secret to a drop box. The secret ballot is critical to preventing election fraud and the breakage of it by ballot harvesting was not only a felony but a grave injustice to the election (BTW we know the current democratic candidate for GA governor (she is unopposed) was a leader in doing it. The ballots went from her organizations offices straight to the drop box (phone cell geo data showed it).
Rinkevichjm:
I don’t know what you’re talking about here. The Eggers and Grimmer paper is right there at the link at the top of the page. You can download it with literally two keystrokes—it’s not hidden at all.
Anyway, blog commenters can say whatever they want but I remain disappointed in Public Choice, which has a history of being a legit academic journal, publishing garbage science. If they wanted to public an op-ed saying that they were unhappy that Biden won the election and that they’ve hear unsubstantiated rumors of foul play etc., then go for it. They could put it in the same section of the journal where they put JFK conspiracy theories, ghost sightings, the daily horoscope, etc. Presenting it as serious social science, though, that’s annoying. And you being annoyed some political leader doesn’t make that Lott article any good.
One can view credibility in a number of ways – the journals, the authors and their experience, the research methodology, bias etc. While Eggers and Grimmer are academics with research experience and have presented their methodology for the analysis, they’re analysis is based on refuting claims presented, rather than an independent analysis of empirical evidence or for that matter statistical probability. John Lott was the senior advisor for research and statistics for the US DOJ Office of Legal Policy and has decades of experience with voter fraud. While both journals (PNAS and Public Choice) have their history of unverified bias and inconsistent claims based on “research”, the real world credibility of John Lott are far superior and staking his career and reputation on the publication did not matter to him, as the evidence led him to his conclusions. We all can debate that ad nauseam….but let’s think about voter fraud in the context of probability, opportunity, pressure and rationalization. Where are the real fraud experts? What are their opinions?
We have $billions in financial fraud and corruption in every industry, every year, despite extensive governance, federal laws, financial audits, risk controls, monitoring of processes and transactions, investigations, etc. And yet every year we have financial fraud escalating…because there is opportunity.
Now take elections, where votes are the currency, where there are thousands of disparate processes across every jurisdiction, with very little oversight, control, best practices, election boards that are political in nature, and quite honestly very little oversight and investigations into fraud. Not complacent, let’s wait to refute the unverified claims, but the proactive sort that roots out corruption at its root. It doesn’t happen.
Does anyone believe for a minute that the most dishonest, corrupt profession in the world (politics), where power and $ are at stake in every election, guided by special interests that need to influence a specific outcome to ensure power and control, that our elections are free from fraud, that the votes can’t be tampered with? And when it’s a loosely regulated?
We want to believe that our democracy is sandro-sanct and free from stains, but it’s not. Our idealistic perceptions of our democratic processes are preventing rational analysis and open honest discussions. Elections is a dirty business, with bad people and they want to win at all costs…and they don’t want anyone questioning or trying to root out the mess. There’s too much at stake, at all costs. That’s why John Lott is a target – whether you agree with his analysis or not.
It’s time for everyone to get their head out of the sand. Elections have fraud, and the probability is that it’s great, if someone is willing to look for it and call it out.
Amory:
1. I’m surprised to hear you say that Lott has “real world credibility,” given that he’s published low-quality work in the past and that he is well known for claiming for have conducted a survey, while offering no evidence that the survey ever existed.
2. I agree that there’s lots of fraud in the world. If Lott had just written an op-ed saying there’s lots of fraud in the world, that would be fine. But he made specific claims about election fraud, and Eggers and Grimmer carefully read through and found serious problems with those claims: “riddled with errors,” as they put it.
Even if his published work is considered subpar – that shouldn’t exclude his real world experience with analyzing and identifying voter fraud – that has more credibility than the academic research. He was a practitioner, not just a researcher. Likewise, I wouldn’t want to discount the research efforts of Eggers or Grimmer because they’re not anti-fraud professionals in practice, but that doesn’t mean their credibility has greater weight. In all of these discussions, the weight and opinion of anti-fraud and legal experts should always have greater credibility. But where are these opinions?
Amory:
Setting aside Lott’s history of bad work, his new paper got a careful reading, and Eggers and Grimmer documented lots of problems with it. It’s not about opinions or credentials or general “credibility”; it’s that they found specific problems with the paper.
Did Eggers and Grimmer accurately describe my paper? Did they seriously attempt to replicate my regressions? No and no.
Round 2: Responses to Eggers & Grimmer’s…
http://ssrn.com/abstract=4193314
Responses to Eggers & Grimmer’s…
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4095558
Have u read my paper or just accepted their claims?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3756988
As to my “history of bad work,” I would put my publications in peer-reviewed journals up against almost anyone’s, but I think that you are confusing research that people disagree with the conclusions of versus the quality of the research.
Amory:
1. There is no such thing as an “anti-fraud professional.”
2. As Eggers and Grimmer discuss, there have been many different claims of massive fraud in the 2020 election. None of these claims have held up to scrutiny, but that does not stop people from continuing to come up with new stories that don’t hold up. It’s like the Loch Ness monster, or ESP, or the Bible code, or all sorts of other things that some people are highly motivated to find, so they look really hard and think they’ve found it.
t.
I would suggest that we read his rebuttal or acceptance to the items called about Eggers and Grimmer. He acknowledged and accepted where there was a specific issue with his research (what was included/excluded) but offered that would not have changed or altered the overall result, and therefore the conclusions would not changed either. Is this accurate? Was this major or minor? Is there another rebuttal on the way?
This could be a matter of competing methodologies that are at “logger heads” arriving at very different conclusions and is open to interpretation, or is each party is coming to the table with a perceived bias as a premise for their research? Then none of us will get anywhere. IMO, that’s the crux of it – no one is “trusting” either the premise or the research of the other party for perceived bias…which taints research projects.
The adjacent polling site research and data analysis is a major premise for which Lott has drawn his conclusions – that raw data itself is abnormal and would represent a major outlier comparatively, therefore it warrants further investigation and shouldn’t be discarded. However, I wouldn’t draw a conclusion that fraud wasn’t a factor either unless the variables could be explained.
In the end, fraud can only be proven legally with “intent” regardless of the statistical research and methodologies used by either party…even if the data points you in a particular direction, it’s only a guide…you still need to investigate and interview to prove intent. No one has done that part to corroborate the data. Nor will they.
Amory, thanks for your comments. Here are the responses that I have had to Eggers and Grimmer.
Round 2: Responses to Eggers & Grimmer’s…
http://ssrn.com/abstract=4193314
Responses to Eggers & Grimmer’s…
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4095558
I don’t have the patience to wade through Lott’s paper, the critique, or Lott’s response so I don’t claim to have any informed opinion about either. But Amory has stated three positions that bother me greatly:
1. That Lott deserves to be believed due to his credentials (past experience, etc.). His background means that he should be taken seriously, if for no other reason than people will pay attention. But we just have too many examples of people with good degrees, extensive publication records, and ample experience doing shoddy work for me to let that one pass. Lott has a checkered record at best, and none of Amory’s cited accolades convinces me that he should be granted “credibility.”
2. Amory alludes to fraud being common in elections. I agree – I’m sure that every election has had dead people voting, people voting twice, and even organized efforts to turn out voters to support particular candidates that go beyond what would be called community organizing. But that does not equate these examples with “fraud” in the sense used in this past election. Organized fraud on the scale needed to effect this last presidential election goes way beyond the types of fraud that I believe happen in every election (more isolated incidents, more local in nature). My prior is that an organized attempt to use election fraud to sway the presidential election requires far too much operational excellence to be believed. Far more reasonable is to believe that influencing media (including social media) is a common attempt to influence election results. Whether this counts as fraud or not, I’m not sure, but I think it is now commonplace to disseminate falsehoods and spread rumors in order to influence results. The fact that every election likely has individual incidents of fraud does not translate into the type of large scale organized fraud being claimed here.
3. Amory says that it is necessary to prove intent in order to prove fraud. Why? I’m no legal expert so I’m not thinking in legal terms here. In the commonsense wording, of course for fraud to occur, there must have been some intent involved. But why do I have to prove intent to claim fraud? Fraud can be proven simply by showing that someone falsified their data, for example. If they falsified their data, then fraud has occurred. Intent may be of interest, but why does that need to be proven? Wouldn’t data falsification be bad enough on its own? Amory’s position amounts to it not being sufficient to show that someone made up their data, changed the data, invented images, etc. – it would be necessary to show that they intended to do so. How does one make up data by accident? Intent can help identify the difference between accidents (inadvertently double counting observations, for example) and fraud (fabricating a bottomless soup bowl, for example), but the burden of proof should not lie with those that discover the mistakes. If fabrication or deception has occurred, the burden should be on the party committing it to prove that it was an accident and not intentional. Amory’s burden of proof is the reverse.
I have to add one thing. I did look briefly at the Lott paper and the critique and responses. There is one thing about the Lott paper that I find disturbing regardless of the technical merits of the critique and his responses to that. Comparing places where fraud was been alleged with places where it has not strikes me as poor research. What is the official source of “alleged fraud?” Is there an official fraud register somewhere that is objectively maintained? Lott’s approach seems circular to me. Some people didn’t like the way the vote turned out in some areas, so they allege fraud. Then Lott investigates those areas and find that the votes seem to follow a different pattern than in other “similar” places where fraud was not alleged. Should we expect anything different? Wouldn’t someone objecting to an election results allege fraud in precisely those places where the vote looked strange to them?
Lott’s approach seems to embody the approach used by many Trump supporters: alleging fraud makes it real. The fact that so many people (to this day) do not accept the results of the election is taken as “evidence” that fraud occurred. I only see it as evidence that these people are unwilling to accept the results. When that unwillingness becomes evidence, then we have unfortunately crossed into the area we now find ourselves – reality is whatever you want it to be, there are no facts, and analysis is merely a way to legitimize beliefs.
Out of the 17 people that have responded to this post, only 3 (including Lott himself) have offered any support for Lott’s work. Given that Lott has a track record of inventing supporters, I allege that Amory is really Lott. If the probability of a Lott supporter commenting in this blog is 5%, then the probability that at least 2 of the 17 commentators support Lott is around 21% – enough for me to cast doubt on Amory’s authenticity. I allege fraud. Further analysis will probably find that a number of Amory’s statements are similar to those made by Lott (probably better research could be done by a linguist). But there is a problem: I need to prove intent and I’m not sure how I can do that.
Dale:
I was thinking that “Amory” might actually just be Lott—as is well known, Lott has a history of commenting on his own work using a fake name (“Mary Rosh”)—but my guess is no, just because there was such a long delay between Amory’s comment on 3 May and Lott’s response on 22 Aug. On the other hand, it could be that Lott posted the comment as “Amory” on 3 May, hoping to get some further replies, and then he could’ve posted under his own name on 22 Aug out of frustration that the thread wasn’t being continued. That sequence somehow seems less plausible—at this point, why not just invent another sock puppet and keep the conversation going that way. So I’m guessing that “Amory” is a real person who’s not Lott (also possibly not named Amory). But I guess we’ll never know.
I nominate Lamb Chop as this blog’s resident Sock Puppet, but I’m willing to put her up against Fraac, Mary and all comers in the next speaker’s bracket if you aren’t convinced by her congressional testimony.
https://www.congress.gov/congressional-record/volume-144/issue-108/extensions-of-remarks-section/article/E1544-2
Miss Lamb Chop: Mr. Chairman, I have been entertaining children for 35 years, which is a long time in the life of a 6 year old. I would like to say that we really need your help and your care and concern, and we need the best that you grown-ups have to offer. And if you give it to us, we will give the good stuff back. Not only to you, but to our own children as well.
Ms. Lewis: Lamb Chop, I couldn’t have said that better myself.
Miss Lamb Chop: I know.
Ms. Lewis: Say good-bye, Lamb Chop.
Miss Lamb Chop: Good-bye, Lamb Chop.