Summary: COVID-19 teaches us about the propaganda and arrogance that hobble us. The cost in lives and money might be worthwhile if we learn from it and make a better America.
A nation lit only by propaganda
This “it’s China’s fault” campaign (details here and here) is eerily similar to the hysteria whipped up – manufactured – by the US government against Iraq before the 2003 invasion. The press were uncritical cheerleaders. Much of the US public became a mob (or more accurately, a flock). Calmer voices pointed out that the evidence was weak and that we were being manipulated. Manipulated again, as we were with the Tonkin Gulf incident and the invasion of Afghanistan (9/11 was plotted in Germany, with training in Florida).
The same tactics are being used against us again. Again with equally obvious motives – to arouse people against a new Cold War foe (to justify our massive military spending) and distract people’s attention from the massive failures of the Trump administration during the COVID-19 epidemic.
The saddest aspect of these campaigns is how easily our elites stage them. Can a people so easily manipulated and so unwilling to learn from experience govern themselves?
Climate Science to the rescue
Many climate scientists have been vocal during the pandemic, with unsolicited offers of their expertise. For instance, in this article by James Annan (h/t to Climate Etc). Here is the money paragraph.
“We have been doing a very straightforward MCMC calibration of a simple SEIR model (equivalent of energy balance box model in climate science, pretty much). The basic concept is to use the model to invert the time series of reported deaths back through the time series of underlying infections in order to discover the model parameters such as the famous reproductive rate R. It’s actually rather simple and I am still bemused by the fact that none of the experts (in the UK at least) are doing this. I mean what on earth are mathematical epidemiologists actually for, if not this sort of thing? They should have been all over this like a rash.”
Well, something here is like a “rash.” When a layperson is “bemused” that experts are not doing what seems obvious to him, it probably means that he should ask an expert and dial down the self-confidence. In the climate wars, such statements often precede “analysis” showing that there is no greenhouse effect or that the world is totally doomed really soon.
The staff at CDC and WHO frequently discussed this issue that bemused Annan, explaining that R0 is of little use for laypeople. It is a variable, not a model parameter in the sense Annan appears to use it. Rather than being simple to calculate, R0 is complex. It varies by circumstances and can change over time. The CDC often cites a paper saying that it is “easily misrepresented, misinterpreted, and misapplied.”
More broadly, there are many fields using quantitative models (e.g., macroeconomics). Personally, climate scientists are among the last I would consult. The most obvious similarity of climate models to epidemiological models of COVID-19 is the Imperial College model made famous by Prof. Neil Ferguson. Created 13 years ago, undocumented code, unvalidated, lacking peer-review (per Ferguson) – but it made headlines and had a massive influence on UK public policy. Although it now appears grossly flawed (as seen in the “actual results don’t matter” justifications), it made him famous and resulted in a shower of grant money.
More broadly, the COVID-19 crisis has followed the same pattern as the public climate wars. Experts are center stage. Then they are displaced by a wider pool of authorities. Then wild claims by laypeople dominate, displacing actual experts (such as those at the IPCC and WHO). The climate wars reached that point after 3 decades. COVID-19 reached that stage in only 14 weeks, as claims that “It Is China’s Fault” increasingly dominate the news. This suggests that our dysfunctional response to major crises results from deeper problems in our society. If not fixed, we might have difficulty coping with more serious threats than COVID-19.
For More Information
Ideas! For some holiday shopping ideas, see my recommended books and films at Amazon. Also, see a powerful and disturbing story about “Birth of a Man of Steel …for the Soviet Union.”
If you liked this post, like us on Facebook and follow us on Twitter. See all posts about Reforming America: steps to a new politics, about the importance of clear vision, and especially these…
- Important advice: Learning skepticism, an essential skill for citizenship in 21st century America. About “extraordinary claims require extraordinary proof”.
- Important: Our leaders so often lie, but we still believe them.
- Swear allegiance to the truth as a step to reforming America.
- We live in an age of ignorance, but can decide to fix this – today.
- Ways to deal with those guilty of causing the fake news epidemic.
- The secret source of fake news. Its discovery will change America.
- We have been blind, but can learn to see.
- A new phase of the epidemic begins with propaganda.
Propaganda rules America! Read all about it!
Propaganda
Media Control: The Spectacular Achievements of Propaganda
Think you are spot on with your idea of deeper problems within our society.
That reality is reflected in our crazy economic beliefs, (hedge funds deserve a bailout, share buybacks enhance corporate prosperity, etc) and in our increasing intolerance of differing opinions.
Whether anyone can implement an effective national policy subject to these kinds of constraints is arguable.
Lincoln’s ‘House divided’ speech is a relevant warning from our history, but that is apparently not taught much any more, to our great disadvantage.
Etudiant,
A future post, perhaps tomorrow, might discuss this. Lots of discussion about the possibility of a revolution in America.
We are like kittens that look in the mirror and see lions. People too lazy to work the political machinery bequeathed us by the Founders are unlikely to risk their lives and fortunes – and honor, if any – in a revolution.
Well do peasants protests. Perhaps cosplay street parties like Occupy and the Tea Party.
We might riot – another way to blow steam, common in our inner cities since the 1960s. I need not tell you how well prepared the police and Guard are for that.
But a serious revolution? That’s talk to boost our self-esteem – about the GREAT DAY IN THE VAGUE FUTURE WHEN STAND UP AGAINST THOSE BAD FOLKS – as a counter to our awareness that we are screwups.
Pingback: COVID-19 is a harsh teacher. Let’s learn from it. – Investment Watch | Covid-19 Business Relief Resources
“[R0] is a variable, not a model parameter in the sense Annan appears to use it. Rather than being simple to calculate, R0 is complex. It varies by circumstances and can change over time.”
While I think your mentioning of Annan (and climate scientists in general) is a red herring, it’s obvious that he knows what he is talking about. He has ready made tools for modelling and he’s been using them for an interesting new problem. At the moment even his results match well. And while I don’t like his tone either, nevetheless what you’ve read into a fcukin blog post about climate science is way way beyond the post itself. I know Annan is just an example but IMHO the above, I’m sure, applies to the other climate scientists with “unsolicited offers”, mutatis mutandis.
This said, the other part of your post is simply beyond justification. Neil Ferguson is an epidemiologist, an actual expert, a respected one. Compared to him, you are the often mentioned layperson in this equation. His recent work about COVID-19 may be “Undocumented code, unvalidated, lacking peer-review” but this is exactly what we expect here (ie. lack of time and the urgency of the situation). He and the whole lot of other scientists have built upon their existing, validated, documented, and peer-reviewed research. It’s not the output of a single ad hoc model he had pulled out from his ass. This is application of literally decades of research by a broad international cooperation of scientists, who with their own models have produced very similar results. FYI peer review in science-engineering takes months, in my narrower field it’s rarely below a year.
Furthermore, his results don’t “now appear[…] grossly flawed”. You, an outsider, are smearing an expert here. And no one has claimed for non-existent flaws the “actual results don’t matter” justifications. And just to quote a “headline grabbing”, “alarmist” result that doesn’t matter from him (that he was saying for grand money): “And the British response, Ferguson said on 25 March, makes him “reasonably confident” that total deaths in the United Kingdom will be held below 20,000.” At the moment this looks quite okay.
I have some bad news for you… academics tend to be some of the worst programmers out there. I made a comment on the GISS source code here: https://fabiusmaximus.com/2015/09/24/scientists-restart-climate-change-debate-89635/#comment-336356
Five years later, the code is still a mess. No public access to the commit history, no integration tests around the model, and it’s still written in FORTRAN.
The linked Nature article states
I’m not sure what to make of that. If Microsoft is involved that tells me they probably used a modern language. OTOH if they need to ‘tidy up the code’ it’s probably more than needing to format everything nicely, especially considering that article was written over 2 weeks ago.
Iaxo,
I’m impressed. Almost every claim you make is wrong.
“While I think your mentioning of Annan (and climate scientists in general) is a red herring”
It was an example to my central point in that section, and hence not “misleading or distracting.”
“it’s obvious that he knows what he is talking about.”
Since I gave a clear example otherwise, that’s an odd statement.
“His recent work”
False. The model was not “recent work.” He wrote the code 13 years ago.
“about COVID-19 may be “Undocumented code, unvalidated, lacking peer-review” but this is exactly what we expect here (ie. lack of time and the urgency of the situation)”
That’s false. There were more current models by other epidemiologists. Do you believe that the only model available was
{9?}13 years old, undocumented, not peer-reviewed? At a key moment, he got headlines by making dire predictions far beyond what other epidemiologists said – and so had a powerful influence on Western governments. He then refused to release his code (has he yet done so?), preventing other experts from commenting. This is exactly what we should not “expect.”Also, he has done this twice before – making confident dire predictions on minimal evidence, which prove to be grossly wrong: BSE in the late 1990s, the 2001 epidemic foot and mouth disease (FMD), and (to a lesser extent) 2009 Swine Flu.
“FYI peer review in science-engineering takes months, in my narrower field it’s rarely below a year.”
The model was
many13 years old(9, from memory). More than adequate time for peer-review.“who with their own models have produced very similar results.”
Predictions have varied widely (Ferguson’s is at the high end), as the WHO and CDC have reported. But other experts have been far more cautious in making predictions.
“his results don’t “now appear[…] grossly flawed”. ”
False. Compare predictions and results isn’t even addition. I believe every nation (except perhaps some tiny ones) has cases of COVID-19. The richest and the poorest. Some who have taken drastic action and some who have done little or nothing. No nation has experienced anything remotely like the extent of severe (clinically obvious) infections and fatalities that his model predicted. And the rate of spread and increase in fatalities are both slowing. So unless a second wave erupts that is worse than the first (implausible), then the model’s predictions were wrong by several orders of magnitude.
“And no one has claimed for non-existent flaws the “actual results don’t matter” justifications”
False. His walk-back on his prediction was contrary to the evidence. While the measures taken in Britain can be used to explain the lower numbers, that does not explain the low numbers from nations which have taken few or no drastic measures.
“At the moment this looks quite okay.”
How nice that near the end of the epidemic he can predict what is pretty much obvious to everyone. Congrats!
Tweet from Neil Ferguson on March 22:
https://twitter.com/neil_ferguson/status/1241835454707699713
“Almost every claim you make is wrong.”
Unfortunately every claim you make about scientific methods in general and computer modelling
in particular shows that your understanding of these is vague.
My narrower field is in computer science (not modelling), so I can help you understand how
publications are reviewed, what is published etc.
Source Code
Source code is extremely rarely published (apart from trivial programs, code snipets, and
pseudo code) and it’s almost never the subject of peer review. Sometimes yes, but that’s very
rare. And this is for reason. It’s very unlikely that a reviewer can understand a lengthy, non
trivial code in reasonable time, and in engineering practice while code review is important
(and tedious) no one expects the code to be bug free.
Until formal verification becomes a routine, code is mostly explained in quite broad terms in
publications (specifications, description of modules, interaction of modules, pseudo code,
qualitative description of program objects etc., there’s no established method). Even the
specifications are rarely formal, except for the most general descriptions.
So the fact that Ferguson or whoever else hasn’t published the code doesn’t mean anything, and
doesn’t mean lack of peer reviewed publication. This is a key misunderstanding on your part.
Some researchers provide link or github repo for the source code and this is becoming more and
more frequent in the last 5-10 years but this is not at all expected from them by the
scientific community. Maybe in a few years, together with formal methods, it will be a
requirement.
‘Cos Ferguson and all the other scientists publish the underlying mathematical model, perhaps
with some implementation considerations, with all the parameters, the required initial data
etc., and all the statistical considerations together with the results. He explicitly
mentioned that it took longer to get the contact patterns in Thailand than to program and run
the simulation, and this data is an extremely important part of the publication. So this is
what’s published and peer reviewed.
The mathematical model is usually enough for experts to assess the model, and this part is the
least exciting and most alien for laypersons like you. The actual source code is usually
assumed to be more or less working and reasonably bug free by the reviewers and experts, and
at least big errors can be easily discovered with simple “sanity checking” runs (more or less
those that can be done in Excel by any reader of the publication).
This is running quite long now, so I detail the rest in a different post.
Iaxo,
The difference between us is that I document what I say, while you appear to be making stuff up. I’m uninterested in another point by point debunking of your long comments – since facts appear to just roll off you, while you reply by making stuff up. But a one point deserves attention, and illustrates your method.
“Source code is extremely rarely published (apart from trivial programs, code snipets, and pseudo code) and it’s almost never the subject of peer review.”
That’s false. Code for climate models has been published. Ferguson is publishing his. Many peer-review journals require publication of key software used in papers. As in the prestigious Nature family of journals.
“This is running quite long now, so I detail the rest in a different post.”
It won’t be posted unless you document what you say. You have proven to be unreliable, and I’m not going to bother fact-checking you any more. Life is too short.
Thanks for the link to Nature, very interesting. Incidentally, it confirms the UK Government has not been relying simply on Ferguson, though he has been very influential:
And to avoid too much reliance on one model, Ferguson says, the UK government took advice from a number of modelling groups, including teams at Imperial and the LSHTM (see, for example, ref. 7). “We all reached similar conclusions,” he says.
I am certainly an out-and-out layman when it comes to epidemics. But as a citizen of a democracy, one has to do ones best to think critically about policy and the theories and models that policy is based on.
There seems to be genuine scientific uncertainty. In the UK, this has not led to a debate in which propaganda and blaming China play any significant role, but it has led to debate about policy.
Basically if you think infections in the population are high, the corollary is that to fit with the reported deaths, the mortality rate must be low. In scenarios like this, the risks of relaxing the lockdown are relatively small, and the strategy of protecting those at risk is justifiable.
In these scenarios there is a large pool of immunity, which lowers the impact of continued social contact. My impression is that such views are very much in a minority at the moment.
If you think infections are low, then mortality rates must be high, and the risks in relaxing the lockdown are large, huge numbers of deaths, a wave of infections because there still is no immunity, and particularly this would imply running out of critical care beds.
The progress of the UK infection is given in the daily press conference slides, though the information is imperfect owing to lags and gaps in collection:
https://www.gov.uk/government/collections/slides-and-datasets-to-accompany-coronavirus-press-conferences#history
The interesting UK confirmed observation is that most deaths, like 95%, are of those of people over 70 with pre-existing conditions. But the key question is what the infection rate and thus immunity pool in the population is, and this is simply not known yet.
Faced with all this, and having to make some decision, the UK government decided yesterday to remain in total lockdown for three more weeks. Not an easy decision. Its getting very expensive indeed. Demands for an exit strategy are getting louder. It looks like it may well end up being a reversal to the original one, protect the at-risk and relax for everyone else. We shall see in three weeks.
Pingback: COVID-19 is a harsh teacher. Let’s learn from it. – Investment Watch – Female Investors Daily
The money statement in the Nature article:’ “Forecasts made during an outbreak are rarely investigated during or after the event for their accuracy, and only recently have forecasters begun to make results, code, models and data available for retrospective analysis,” Edmunds and his team noted last year in a paper6 that assessed the performance of forecasts made in a 2014–15 Ebola outbreak in Sierra Leone. They found that it was possible to reliably predict the epidemic’s course one or two weeks ahead of time, but no longer, because of the inherent uncertainty and lack of knowledge about the outbreak. ‘
That uncertainty is larger with new diseases. Ebola has been studied for years. Also, with CoVid19, the termination of vectors by fatalities and asymptomatic disease vectors is much different than Ebola. I do not doubt that these and many more variables were discussed, included, excluded, and rationalized. The problem with “inherent uncertainty and lack of knowledge” did not decrease with the discussions, inclusions, exclusions and rationalizing. The rule is still “Garbage in; Garbage out.”
These models, like Climate models, should be used for what they are good for which is generating responses to scenarios to give ideas for potential futures. In fact, reading between the lines, that is what the disease models are being used for. Which means Annan’s statements do show that he misunderstands the purpose of the model, and that Ro is variable due to human factors which include such things, as population density, familial relationships, eating habits, social habits, etc.
YMMV
“and distract people’s attention from the massive failures of the Trump administration during the COVID-19 epidemic”
and the massive failures of the EU administration.
The only people without any governmental screw-ups of any kind here are the South Koreans, as far as I know. China did well once they accepted the reality of the issue; we are… doing whatever we’re doing.
SF,
I think that’s harsh. China was the first hit, so fumbles were inevitable. Every disaster is marked by-know-it-alls who pour through the files and show how it was obvious.
Oddly, these genei never seem capable of real time demonstrations of their superior brilliance. Only in hindsight are the superior to those on the front lines.
But you point to an important point. TheEast Asian nations learned from SARS. America seems incapable of learning from experience. Y2K, 9/11, the anthrax episode, the botched occupations of Iraq and Afghanistan, the mortgage bubble, the crash, and now COVIS-19. It is an amazing series of failures, which we seem incapable of even recognizing- let alone learning from.
This is tangible evidence of the senescence I have written about for so long.
Perhaps eventually Americans will see it. Then perhaps Americans will do something about it.
Richard,
This website is about America. But your point is significant!
But our failure is bigger, because we were so much better prepared in a January. Preparation on which we spent vast fortunes. Poor execution “trumped” lavish preparations.
Pingback: COVID-19 is a harsh trainer. Let’s be informed from it. – All My Daily News
Hello,
Firstly: there’s a nice example from a ~10M people country, the Czech Rep., with 150k+ tests and 6k5 cases / 173 dead and published R0 ~ 0.7 now (czech for yourselves: https://koronavirusvcr.cz/en/)
Then, I can’t believe anyone of sound mind would fall for those grossly overblown modelling scenarios (as per UK and US BS) to initiate lock-downs — see: A very recent Study has proven that the First World Lock-Down makes no sense at all.
The Chinese could be excused from the over-reaction as they were against an unknown; however, the EU and US have no excuse at all. E.g. 2015 influenza in EU (EU_MoMo) killed over 300k people and nobody even blinked, while this COVID-19 “scared” everybody, as it seems. (FYI, just now the EU+ has just over 90k dead)
So F.M., you were surely onto something:
There must be a ‘BETTER’ EXPLANATION than madness!