In this blog, I will encourage you to take the warnings in that earlier blog -- henceforth SDP -- and greatly broaden them. I will also expand upon the perspectives of the "deniers" based on my interpretation of their presentations and discussions.
I will NOT engage in an assessment the research results -- I don't have the necessary knowledge. And since I can't moderate such in the comments, such are declared to be off-topic.
Readers: I make heavier use of footnotes here than usual. Many of the explanations and illustrative examples are too long to include in the main body -- individually or cumulatively -- and only part of the audience will be interested enough in the points to read the detail.
The title on each footnote is intended to remind you of the context. If you want to revisit the context, the footnote numbers in the main body have a prefix of "foot#" to facilitate search.
Aside: I decided to delay this blog until the comments on the SDP blog had played out, both to have the opportunity to respond to the whole thread and to avoid the potential confusion of comments that assumed that people were also reading the comments on the other blog. My writing was then interrupted by "events".
----First, some general background.----
"Science advances funeral by funeral."(foot#1) While science is self-correcting, those corrections may take many years, even decades. Why? One cause is that the dominant scientists in the field are unwilling to accept the change because they are psychological committed to a failing paradigm and/or because the particular change might negatively affect their status and legacy. Awareness of the importance of social and political factors in the conduct of scientific research became widespread with Thomas S. Kuhn's 1962 book The Structure of Scientific Revolutions and is an ongoing subject matter. For example, the 2009 earthquake in L'Aquila Italy killed 308 people and injured 1500 shortly after the government used a panel of seismologists to claim that there was no threat of a major quake and it was safe to return to vulnerable homes. The scientist hadn't bothered to make an assessment, and let the government's false claims stand. When those seismologists were judged to have contributed greatly to the death toll by their grossly negligent and irresponsible behavior, seismologists around the world rallied to their defense, including the head of that section of the prestigious AAAS (American Association for the Advancement of Science).(foot#2)
Contrary to the carefully cultivated notion of scientists being truth-seekers divorced from various categories of politics and other social pressures, scientists have to cope with those influences in their choices of what to pursue, of how to structure their research and of how to report it. The accommodations can be so small as to seem to be meaningless, or as necessary tradeoffs, with the danger becoming apparent only through the cumulative effect.
Moral: Don't let faith in science-the-process become faith in scientists-as-people.
If a group of research professionals and graduate students are having a discussion of this phenomenon, you will typically hear many stories not only from the participants' fields, but stories they have heard from other fields (for example, from a scientist spouse or multi-disciplinary collaborators). These are stories not just of delayed corrections, but of careers badly damaged and even destroyed. I've heard some real horror stories from the physical and biological sciences. My career provides a minor example.(foot#3)
The core argument of The Structure of Scientific Revolutions is that major scientific advances don't come from gradual evolution, but rather from large punctuated changes -- hence the "revolution" in its title. Once a "paradigm" has been established for a scientific field, research to expand and refine the paradigm discovers problems that cannot legitimately be treated as part of the paradigm, but typically get shoehorned in. At some point, the weight of the inconsistencies becomes so great that the paradigm collapses, to be replaced by a new one. From this perspective, it is wrong to speak of a "scientific consensus" because what you are seeing is a phase where problems with the paradigm are being warped to fit in, or are being suppressed by those committed to the paradigm. Similarly, there is no "settled science":
-- A more detailed examination of the data -- for example with improved instruments -- may reveal complexity that forces a reinterpretation of that data. For example, some cancers have been determined to actually be a group of similar cancers but with different characteristics, such as fast-growing vs. slow-growing. Recognizing this provided an explanation for why different groups receiving the same treatment had very different results: Those groups' research subjects had different mixes of those cancers.
-- Unknown factors/variables are discovered. A classic example is that Fermi's early work on nuclear fission was producing very different results between his lab in Italy and the one in England. The cause? The Italian lab's workbenches were marble whereas the ones in England were wood. Hence the discovery of the importance of moderators such as wood for achieving a sustained chain reaction.
-- More data may turn up exceptions -- example: until black swans were discovered in Australia, it was held that all swans were necessarily white.
-- A better interpretation of the data may be discovered.
When I hear people who are scientists using either of those two terms, I assume that they are not speaking as scientists but as partisan advocates, either within their field or to society in general.
Aside: In the leadership/management literature, I have encountered advice that a consensus should be a warning of ^groupthink^ or of other perspectives having been suppressed, excluded, or at least not represented.
"Publish or perish" is often cited as the Prime Directive for academics and many categories of research scientists. Problem: Although peer-reviewed publications are a very large factor in the advancement of a researcher's career, the peer review process has long been known to be capricious, sometimes to the extent of verging on being random.(foot#4) In my field and some others, there are lists of seminal papers that were initially rejected, with some listings including the reviewer's comments. These make for funny, painful, informative reading. Peer review is also poor at detecting fraud, with preeminent publications, such as Science and Nature, withdrawing published papers. Reviewers typically don't have the resources to do much more than a basic quality check. The efficacy of the peer review process is dependent on the author being honest and forthcoming and the reviewers being competent and diligent. Additionally, peer review is supposed to be only a first filter -- validation is supposed to come from others reproducing the results. But it is now rare to attempt to reproduce results. It can be very expensive. It contributes nothing noteworthy to a funder's list of achievements. It takes researchers away from activities that would advance their careers.
One way to reduce the chance of negative reactions from reviewers is to pursue research that is a small, safe increment to the field. This also allows researchers to heavily cite the papers of prominent people in the field, which benefits them and thereby making them more favorable to your work. Funders lament the absence of high risk-high payoff proposals, with some acknowledging they play a large role in perpetuating a system that discourages such proposals.(foot#5) Saying: "If we had had the current research environment during the Polio epidemic, we would have had a variety of better iron lungs, not a vaccine."
----Scientists and the popular press----
Much of the science news in the popular media comes from an unrepresentative group of scientists in that field, or has passed through several layers of "interpreters" where each can innocently add errors and other misrepresentations. For example, a public relations person at a university interviews the scientist to create a press release that is both a summary and "attention-grabbing". Reporters then edit the press release enough to qualify as an article under their byline. Editors or others may further edit the articles for various purposes. The ^game of Telephone (US; British Chinese Whispers)^ illustrates how such sequences can produce major distortions.(foot#6) Quotes from researchers create the illusion that the reporter interviewed them, but routinely they have been strategically provided in the press releases, as evidenced by the exact same quote appearing in the various derived articles.(foot#7)
In the mid-1980s, Computer Science research groups started electronically sharing papers and talk announcements, which was a professional boon for keeping up with each other. But some also distributed their universities' press releases, and I got to see how much the articles in local newspapers were derived from those.
I know a range of scientists and technologists who have had professional interactions with the media. For the overwhelming majority, it was a bad experience, with them feeling abused and betrayed by the reporters. Many are worried that readers, especially their scientific colleagues, would believe that what appeared in the article represented what they had actually said. Part of an earlier blog summarizes one of my negative experiences.(foot#8)
The news media prioritizes entertainment over information, seeing it as more profitable. When prominent scientists discuss their experience with live interviews and panels, and common refrain is that during a break, the show's producer tells them that they need to be more controversial, often accompanied by the warning that if they don't, their on-air time will be cut and/or they won't be invited back. This is hardly unique to scientists or current times.(foot#9) The media cultivates extreme and outrageous positions and controversies between the extremes and other outliers.
Moral: Don't assume that shills, extremists ... are responsible for the bad coverage of a controversy.
From the mass of scientists who have difficulty producing overviews responsive to what the media wants -- or are unwilling to do so -- there emerge those who are promoters who excel at marketing and selling the field to the broader public, sometimes deliberately and sometimes as a side-effect of promoting themselves.
Recognize that funders have their own funders: They have to "sell" the research that they are funding to get new and continuing funding. For example, a US government agency funding a research project at a university has a budget from its parent organization which has ... which gets its money from Congress which may be influenced by special interests ... and occasionally the electorate. Good publicity may not just facilitate each stage of this process, but may be essential. Consequently, bad behavior of the effective promoters may well be ignored or excused by other scientists.
Example from my field: One of the most prominent researchers was widely suspected of "over-claiming" results. This was confirmed during a demonstration during a conference when his funder insisted upon a trivial deviation from his script, revealing that it wasn't a working system but one simply following that rigid script. The consequences? None that I heard of. He had established himself as a top go-to expert for the New York Times -- and hence many other media outlets -- and publicly discrediting such a person could undermine the financial health of the field.
From current news: The director of the high-profile Media Lab at MIT had to resign because of his involvement with Jeffrey Epstein (billionaire pedophile, died in jail). Of what has been made public, the part relevant here is that he extensively socialized with Epstein, providing the Lab Director with even more access to the rich, powerful and famous. How much of this could be written off to the proper activity of a funder raiser -- for an already well-funded Lab -- and how much was ego? Especially troubling was that by being there as a guest of Epstein, the Lab Director's celebrity was enhancing Epstein's reputation.
----Finally, addressing points in the other blog----
Reminder: I am using SDP to refer to "^Skeptics, Dupes and Paranoids^" by another PAOnline blogger. That blog selected and summarized advice on how to better assess discussion on climate change, with those selections being common arguments in the meta-discussion of the climate science debate.
SDP section "Cultivating friendly scientists" presents some of the ways that junk science can be solicited and produced by manipulating well-intentioned scientists in addition to the outright shills. However, my discussion above should alert you to the many other hard-to-detect ways that science can be distorted/perverted. Explanations of how this is happening can come off as conspiracy theories, which they may well be. But recognize "Even paranoids have enemies."
Problem: Because the size and complex interactions of the climate change research, it can be hard to allay these fears, and an appeal to authority (for example, "the consensus") will likely be seen by many as evidence or confirmation that something is rotten.
--Media biasing science--
From SDP section "Playing the media": "When science deniers take their debates direct to the public via the media, they will insist that the media 'tell both sides' or 'give us our fair share' . The media have fallen for this as they work to avoid the appearance of bias."
Question: How do you know if the perspective being shut out is because it lacks merit or because of the media being biased or influenced against it?
Claiming that one side is distorting media coverage seems to imply that the media should be deciding which scientific claims warrant publication. This in turn implies that the media is qualified to make such determinations. Which in turn requires that the media know, and turn to, experts who can give unbiased assessments. When I consider the vast amount of junk science that receives blaring headlines, my conclusion is that the media is structurally unable to fulfill this role.
Another interpretation of this argument is that the media will not present the various sides in a way that the reader can detect which positions are nonsense. In the days of print media, its space limitations and its inability to effectively link to expanded discussions would have made me sympathetic to this argument. Now, this is an argument that the media lacks the competence or integrity to do this.
Rhetorical question: If the media wants/demands sensationalism, can those who know how to play the media's game be accused of "playing the media"?
--SDP Section I Attacking other scientists==--==
In a proper debate, an ad hominem attack implies that the attacker doesn't have a valid response to the attackee's points. However, this rule doesn't always apply to public policy debates such as climate change because the media controls what most people hear of the debate. Ad hominem attacks, unsupported accusations ... are easier to report than the scientific debate, and are much more attractive to the media as a whole. Without substantive points to respond to, ad hominem counter-attacks may be the only responsive responses. For example, when a scientist's public pronouncement on scientific policy isn't about the actual science but reek of "bias, of seeking publicity, of issuing 'political statements' ".
--Outsiders aren't necessarily wrong--
From SDP section "Emphasizing lack of proof": "...The general public may fail to understand that science typically works on the preponderance of evidence, it rewards theories that explain many observations, it moves forward based on peer review and continuous research. ..."
This is what the typical history of science would lead you to believe. Pruning out the many times it went off the rails is appropriate for teaching how-we-got-here, but that sanitizes the actual dynamics. It hides the difficulties of those outside the scientific establishment: apostates, those with unpleasant personalities, amateurs, women, engineers, technicians, ... Unfortunately, the simplified history biases the public to assume that science is a pure meritocracy, and if you are an outsider, your ideas and critiques are likely wrong.
When the establishment of a field declares that there is a consensus, this gives them psychological grounds for rejecting papers and other participation by those outside that consensus. How much of this is happening? Unknowable. If a scientist believes that a research paper is likely to be rejected for political reasons within the field, writing and submitting it is not only a waste of time, but a step to professional suicide.
In the comments on SDP blog, there was extensive back-and-forth about who was responsible for the alarmist claims in the media. How such claims made it into the public sphere is typically invisible to the public and is irrelevant to the effects of such alarmism on subsequent public discourse. Thus, it is utterly irrelevant here. The focus here is on the public better understanding the public discourse on climate change.
From section "Slowing thing down": "... They will denigrate those who wish to move faster as 'alarmists' or 'hysterics' and portray themselves as the thoughtful, rational, unemotional participants. Unfortunately, they may be aided in this by overly simplistic and alarmist statements made on the other side."
Unfortunately, the public face of climate science long ago cemented its reputation for alarmism. On the web, there are lists of prominent failed predictions. Examples: The Arctic Ocean was to have become ice-free years ago, sea level rise was to have already flooded large areas. Climate change advocates need to accept that this history is a legitimate reason for skepticism and find legitimate ways to allay it.
Snark: Who are we to believe: the "settled science" -- that such events have already occurred -- or our own lying eyes? (apologies to Groucho Marx).
Even worse, in supposedly scientific presentations to the public, substantial exaggerations are excused as being needed to impress upon the public with how urgent and dire the situation is. For example, at one presentation, a 5-foot rise in sea level was illustrated as being well over the head of a 6-foot tall man.
Climate change advocates have, and continue, to make claims that induce skepticism. For example, claims such as "Today is the hottest day in New York City in the past 100 years" often pop up, only to be challenged by the skeptics (real or pseudo), for example, posting front-page headlines from the New York Times archive reporting even hotter days during those years.
Problem for the public: Is this someone seeking to keep climate change in the headlines by misrepresenting the record, or a case of the record having been adjusted to keep it comparable across the years, or of temperatures from different measuring points being used?(foot#10)
There have been substantial advances in climate science over the past decades: discoveries and major refinements in the understanding of known phenomena. Yet the public presentation of these advances seems to be pitched as "proving" that the climate change situation is even worse. What are the odds of them all/predominantly being in the same direction, unless they are being made to fit a narrative?
In the face of a history of alarmist predictions used to justify demands for massive, fundamental changes to the economy and society, why isn't advocacy for caution regarded as rational and responsible, rather than obstructionism?
To understand how skepticism of "the consensus" can be legitimate, you might start with the pattern of ways that climate scientists' own public presentations have damaged their credibility. A very common one is cherry-picking the year to serve as their baseline for warming. For multi-century graphs, that baseline is too often in the depths of the Little Ice Age. For multi-decade graphs, 50 years ago (1970-ish) is popular because that was near the end of the cooling trend in the 1950s and 1960s, one which produced (alarmist) headlines about an approaching ice age.
Then there is the sleight-of-hand with data. A too common example is that a scientist introduces the presentation as being about human-caused (anthropogenic) climate change (or global warming) and establishes the term "climate change" as a shorthand for this. Then to enhance his argument, he uses that term to refer to all components of climate change, such as the rebound from the Little Ice Age. Technically, this is not a false statement, but it is deceptive, disingenuous ...
For me, such behavior marks a scientist as disreputable. If someone outside the field -- such as me -- can detect such deceptions, there are likely more in the presentation that weren't detected. And when a field seems to tolerate this behavior, that warrants high skepticism of its claims.
Problem: Along with most of the public, I see such a minuscule portion of the presentations that it cannot be taken to be representative (statistically significant). The evidence of bad behavior is too much too ignore, but far short of a conclusion. Arrrrgh!
--All models are wrong, but some are useful. (statistician George Box)--
Climate change predictions start with scientists from a wide range of disciplines providing their results to researchers who specialize in modeling. A well-known danger of this approach is that the model-builders don't have the knowledge of the individual disciplines to perform sanity checks on the model. For example, a large part of the field of Economics has become based on model-building, with failures of various models being attributed to those model-builders not knowing the economic history that would have alerted them to their model's flaws.
Additionally, the climate change model-builders have been using problematic data, which they should have been informed of and whose resulting uncertainties should be reflected in the public discourse.(foot#11)(foot#12)
--Skepticism resulting from the inferred agendas of advocates--
I assume that it is difficult/impossible to determine how resilient the collective process is to individual pieces of bad science -- intentional or not. Conceivably, bad science at a few key points could negate a vast body of good research. The model-building process is one such obvious key point -- it is where so much is brought together.
The public focus of climate change research can easily be interpreted as agenda-driven: political, philosophical or other world-view. For example, the focus on CO2 from burning fossil fuels, and within that, a focus on vehicles and then industry. The efficiency of cars has improved greatly over the years, but I occasionally still hear cars referred to as "gas guzzlers", which sets off alarm bells that I am hearing hyperbole, not science. Yet questions about diminishing returns and shifting some resources to other components of climate change are largely ignored or dismissed or labeled as coming from "deniers" -- at least in the public sphere. With CO2 being seen as the largest component of climate change, there is good reason to guess that the best opportunities for limiting climate change would come from reducing it. However, that should be little more than an intuition or first guess.
An analogy: In ecology there is the concept of ^keystone species^ -- ones that have outsized impact on the environment. It is often surprising what those are, and why.
There are climate change advocates who are researching cost-benefit tradeoffs, and these results could be used to prioritize the various potential actions and programs. Where do I hear discussion and advocacy for such approaches? Mostly from people labeled as "skeptics" and "deniers" by those within "the consensus".
Question for yourself: Have you heard of these?
The 2019 ^Green New Deal^ of Representative Alexandria Ocasio-Cortez (D-NY) and Senator Ed Markey (D-MA) is an example of what can easily be viewed by many as a ^stalking horse^ for an agenda promoting major social programs. The proposal contained a host of social measures that had only tenuous connections to addressing climate change, such as free college and guaranteed jobs. And the climate change portion was so slapdash that it launched rafts of jokes and derision, including ones from climate change supporters.(foot#13)
Question: If the sponsors of such legislation were truly concerned about fighting climate change, why attach unrelated, controversial measures that greatly increase opposition to the package?
Some of the skepticism of climate change programs results from them appearing to be more about increasing the size and reach of a centralized government, especially the federal government. For example, proposed national policies that fail to take into account -- and accommodate -- regional and local differences, resulting in them being counterproductive. One such example is the push for conversion to electric vehicles in regions where the electricity comes predominantly from coal-fired generators, resulting in those electric vehicles calculated as having increased the carbon footprint over gasoline-powered vehicles. However, these critics tend to overlook the various local and individual actions, probably because those are ones that those critics approve of.
Aside: Bureaucracies may originate from a desire to regularize administration -- to eliminate favoritism and other inconsistencies -- and wind up eliminating needed flexibility.
Some of the skepticism results from the appearance of "solutions" having been chosen and then the justifications tailored to fit. For example, when the cost comparisons are cherry-picked, such are pitching lower operating costs while ignoring increased capital cost that more than offset the savings.
These are legitimate and reasonable concerns for enough people that advocates should accommodate them in the discourse. Advocates should not fall into the trap of lumping them in with those who seem to be simply obstructionists, or those with such strange world views that there is inadequate common ground for discussion. For example, the equivalent of religious zealots.
--Statistics, uncertainty, risk assessment--
"Never expect a reporter to be able to convey anything but the simplest of statistics" seems to be common advice in corporate media training. "Convey" includes first understanding and then transforming that understanding into something their audience can understand.
Add to this that statistical illiteracy is also widespread in many scientific disciplines, and it sometimes seems that the understanding extends little beyond how to manipulate data so that the results are "statistically significant", and thus warranting publication.(foot#14)
It is well-established (settled science?) that we humans have really, really bad intuitions about probabilities, uncertainties, and risk assessment. Bear in mind that by the time research results reach the public sphere, there have been multiple opportunities for such to have been misinterpreted or misreported.
--Characterization of the skeptics--
When you are trying to convince someone, you need them to listen to you and trust that there is some value to them in what you are saying. So, it seems unwise to announce your contempt for them and denounce them as vile, unethical and/or stupid. Yet climate change activists too often start off that way, portraying themselves and their beliefs as the smart, virtuous ones.
Given the established history of alarmism, climate change advocates should understand that "Forget all that. This time the predictions are right!" is unlikely to be accepted, and more likely to trigger:
-- "The definition of insanity is doing the same thing over and over again and expecting different results" or possibly
-- "In individuals, insanity is rare; but in groups, parties, nations and epochs, it is the rule." (Friedrich Nietzsche).
What I have seen (in the public sphere) from the advocates gives me no sense that the scientists understand what went wrong with earlier predictions nor with the excessive certainty that they ascribed to those predictions. I would think that this is important to many who are following the debate.
Questions and objections about "the consensus" that seem reasonable to me are too often written off as coming from obstructionists harassing the climate scientists. If so, some of these are specific enough to be quickly refuted. Yet when I do web search, what I find are predominantly ad hominem attacks against those raising questions/objections. Often volumes of attacks, representing far more effort than a useful response. Should I take such attacks as evidence that the advocates don't have a good response? Or is it a matter of ego overriding legitimately responding? And if it is ego, does that imply anything about the underlying research or conclusions?
It is inevitable that I will be accused of being a "denier". Well, substantial local weather/climate changes have been evident in my yard for years. Frosts previously killed my Bougainvillea plant back to the most sheltered places; now, after a brief pause for what the calendar claims to be winter, it takes off where it stopped in late fall. I have to deal with dandelions much earlier. Plants that need a winter chill to bring on dormancy are deprived of that and no longer come back. And before this summer, we had a long enough stretch of cool summers that it pushed me to switch my mix of tomato plants to ones that were short-season and tolerant of cooler weather.
Why am I unimpressed by the "consensus of climate scientists" claim? Because scientists are highly specialized. Can you expect a specialist in reconstructing California climate history from tree rings to be able to assess the correctness of the model of how currents in the North Atlantic Ocean change in response to changes in meltwater contributions from both the Arctic and Antarctic? Can you expect the latter to be able to assess the correctness of models of how much meltwater will be produced and when? The advice echoed in SDP legitimately warns against scientists operating outside their specialties ... if they are challenging "the consensus". Why should it be different for those who endorse "the consensus"?
In case you haven't already guessed, I regard simply citing "the consensus" as an instance of the ^Appeal to Authority^ fallacy. It is also uncomfortably close to an equivalent to God's revealed truth to his Earthly representatives.
Final reminder: Appropriate discussion is on how the public can better interpret and assess the public discourse on the topic of climate change. This includes not just indicators of the credibility of claims of various perspectives, but also when none of the claims deserve belief.
I have neither the inclination nor the resources to moderate a discussion of the science involved. Similarly for past debates unless they provide useful examples for going forward.
1. "Science advances funeral by funeral" - Paul A. Samuelson, Nobel Prize winning economist (1970).
2. L'Aquila Earthquake:
In 2009, the Italian city of ^L'Aquila^was being rocked by an ^extended series of earthquakes^. Knowing the history of devastating earthquakes in that city and the surrounding region and the vulnerability of many of the buildings in that town, many residents were afraid to sleep in their houses, choosing to sleep in cars and elsewhere outside. The Italian government dispatched prominent seismologists to assess the situation. Possibly irritated that an amateur was warning of an imminent big quake (based on radon releases), they made no effort to assess the situation(one of the attendees was subject of a criminal investigation and the surveillance recording of him also captured the meeting). The seismologists returned to Rome and the accompanying PR official announced that the seismologist had determined that no big quake was pending and that residents could return to their homes. Despite being aware of this announcement, the seismologists failed to issue a correction or even request that the government issue a correction. A few days later, a 6.3 magnitude quake struck in the middle of the night, damaged 3-10K buildings, with some collapsing. Because of the assurances, many who had been sleeping outside had returned to sleeping in vulnerable buildings, and were among the 308 killed and 1500 injured.
My assessment is that those seismologists saw their role as political -- serving the government that appointed them to prestigious positions -- rather than providing legitimate scientific advice to the populace.
In 2011, the seismologists and a government official were charged with involuntary manslaughter. Seismologists from around the world sprung to their defense arguing that since earthquakes were currently unpredictable, then a failure to predict an earthquake was no different from falsely predicting that an earthquake would not occur (the philosophy of morality is decisively to the contrary). In newspaper editorials and science articles, there were arguments derived from the template "We shouldn't punish people for lying because doing so would discourage people from speaking the truth because they will be afraid of being falsely accused of lying."The variants in this particular case seemed to boil down to that punishing these seismologists for their conduct would discourage scientists in the future from providing expertise because they would seek to avoid accountability and responsibility for their actions.
The seismologists were convicted, but that was overturned on appeal.
3. Career damage from being outside the dominant paradigm: a personal example:
When I was working on my Ph.D. in Natural Language Understanding/Processing (NLU / NLP) area of Artificial Intelligence (AI), the field was divided between the "Scruffies" and the "Neats". The Neats were advocating for more use of formalisms, while the Scruffies had the dominant paradigm at that time and tended to be dismissive of the Neats because earlier, simple formalisms had failed to scale. As a junior professor, my first proposal for research funding failed: One reviewer registered tepid support and the other reviewer was a Scruffyand strongly opposed funding it, writing that its general area wasn't viable. My second research proposal failed because the reviewer stated that I was "unfamiliar with the literature". How could that be? The statement the reviewer was referring was mine, but had been removed from its context -- thereby reversing the claim -- by the author of widely circulated draft of a paper describing his new approach. Additionally, his citation of the claim was to a journal article by two of my collaborators on a topic separate from my research. Aside:That the false claim quickly achieved widespread acceptance was a reminder to me of how rarely citations are checked.
Details:My thesis observed that a straight-forward simplistic computation implementation of the formal theory was impossible because the smallest non-trivial database would have required many more items than there were atoms in the universe. However, only a tiny number of these items were actually used, and having a database that had only the needed items was both eminently feasible and had interesting features that extended the basic theory. Unfortunately, a prominent professor had come up with a competing theory that had formalisms, but was also appealing to many Scruffies. In a footnote on the first page of his initial paper, he quoted a phrase of mine to claim that the theory I was working within was computationally infeasible. I received an early draft of that paper and contacted the professor multiple times -- including face-to-face at a workshop -- and asked to correct both the claim and the attribution. He didn't. My guess is that the professor was sloppy and negligent, not malicious. However, the reviewer of my research proposal was not just negligent, but irresponsible: If he had read the first paragraph of the abstract, he would have been aware of the professor's error.
Reviewers who recommend rejecting a paper without fully reading even the abstract was a common-enough situation that it didn't evoke surprise, much less outrage. It happened to me at least once more, but by then I was more established and the consequences were minor, but still annoying -- an interesting research result didn't make it into the literature.
Consequences of the funding rejections:As a graduate student, I had seen this happen to junior faculty who, despite being regarded as rising stars in the field, were denied tenure (a digression that is too long, even for me).I knew that these proposal failures had put me so far behind that I had a negligible chance of gaining tenure. I loved teaching and was a good teacher -- measured both by evaluations and by the growth in enrollment in my primary class from 75 to 400 in two years. Despite this, I knew I had to leave academia, abandon my research area and start a new career in a related subfield. I spend 4 years in Advanced Development before being able to return to Research (non-academic).
Remember/recognize:There was nothing special or unusual about my situation: It was but one instance of an existing widespread problem. After being characterized as a "crisis" for many years, the National Science Foundation (NSF eventually implemented multiple programs attempting to reduce it.
4. Well-known failings of the peer review process:
-- "^Peer-review practices of psychological journals: The fate of published articles, submitted again^", Douglas Peters and Stephen J. Ceci, Journal of Behavioral and Brain Sciences, 1982.
For 12 highly regarded journals, an article that had been published there was resubmitted 18-32 months later. Three were detected as resubmissions. Of the remaining 9, 8 were rejected -- an 89% disagreement between the first and second sets of reviewers.
-- "^The NIPS Experiment^" (2014)
Additional discussion: "^The NIPS Experiment^" - Communications of the ACM (Association for Computing Machinery) (2015)
10% of the papers submitted to a conference were giving to a second set of reviewers, resulting in a 26% disagreement rate, which translates into 50-67% of the accepted papers being rejected if they had gone to different reviewers. - From my recollections: A common practice for reviewing submissions for a conference was to send each paper to two reviewers who were to respond on a three-point scale, roughly "Good/Accept", "Bad/Reject", and "Adequate/Marginal Accept". If the two reviewers returned one "Good" and one "Bad", the paper would then be sent to a third reviewer who was intended to be a tie-breaker. This process was supervised by the conference's Program Committee, and I was hearing of committee members worrying about the increasing the number of papers for which a third reviewer was necessary. A few reported the statistics. One that I remember was that the substantial majority -- 80%? -- required a third reviewer.
5. Funder incentivizing narrow, incremental, safe research: personal example:
Because of the ups-and-downs of funding in my primary research area (too long a story), I had drifted into Computer Security. My new DARPA-funded project had surpassed in several ways several long-running projects within the same DARPA funding program. The program manager was not only happy with the approach and results, but my project was explicitly part of the briefing his superiors gave to their superiors, with at least one more iteration up the bureaucracy. The program manager was new and thought the collection of projects he had inherited was too narrow. His analogy was ^picket fence^where some of the vertical boards were 6-foot high, but surrounded by ones that were very short: 1-2 feet high. He emphasized that while the tall sections represented many intellectually challenging issues, the ease of bypassing them rendered further improvements as having poor cost/benefit ratios. He urged applicants for the next round of funding to focus on the underdeveloped areas. Since my project already targeted one such area, I was surprised when it didn't get new funding. I misread the politics: Apparently, the program manager's guidance to us was not that of his managers.
Aside: I left research for a startup, but the 2001 recession hit before I could complete turning the approach into a commercial product.
6. Distortion in press releases and derived articles
In April 2019, MIT distributed a press release on the use of a "virtual telescope" to image a black hole. Over 200 scientists were involved, and the press release was (obviously) intended to highlight contributions of those from MIT. One was a woman and MIT -- like many universities -- has a practice of highlighting women in science.Unfortunately, "highlighting" became a "simplification"that was a serious misrepresentation by making her significant contribution much broader. Because such press releases routinely include pre-packaged quotes from the scientist and others, many apparently assumed that she had talked directly to the reporter and was taking credit for contributions of others, resulting in her being pilloried on the Internet.
7. The Scientist to Press Release to Newspaper Article pipeline:
Back in the late 1970s or early 1980s, a research group had looked at business news articles found that roughly 80%were derived almost entirely from corporate press releases, removing the boilerplate and doing some mild editing and rewriting for length or whatever. In this period, many university Computer Science departments were gaining access to what would become today's Internet -- except that there was no Web, and for many email was slower (and less reliable) than USPS. CS research groups started distributing items of professional interest to each other: papers, talk announcements, ... Some were also distributing their university's press releases about their research, allowing me to compare the press release to what appeared in various newspapers (local, the NY Times and occasionally others). It was the same situation: slightly reworked press releases were presented as journalism.
If you want to see this for yourself, go to ^Google News^and find a science story that has a substantial collection of articles under "View full coverage" and select several that seem to be from science-focused publications. You will likely find massive overlap, including the exact same quotes from the scientist.
Aside: I find the full coverage feature very useful when I think that the first article I read had misinterpreted something or elided important details.
8. Scientists versus media: Personal example:
Subsections 5-6 (paragraphs 7-10) of section "Why am I so skeptical and critical of the news media? Experience" of my blog "^This time we're not lying. HONEST! No, really!^" (2018-01-19).Non-clickbait headline: "Highly partisan media, the Gell-Mann Amnesia Effect & examples from my history".
9. Panelists and interviewees needing to be controversial is long-standing:
In the early 1980s, ^ABC News Nightline^ with Ted Koppel had an interview with a Conservative Evangelical Christian couple heading a group pushing to limit what books and content were available to Texas public school students. For "balance", Nightline had several Texas librarians. After introducing the librarians, Koppel prompted them with "And now for your opening blast." However, the librarians refused to cooperate and made an on-one-hand and on-the-other presentations to that and the questions. It wasn't long before Koppel stopped directing questions of the librarians. In response to criticism, the show's producers claimed that this was the result of "technical difficulties", although none were apparent during the show -- quite the contrary.
10. Legitimate adjustments to the temperature record:
Just to give you some ideas about why adjustments may be needed. In the village where I grew up in the 1960s, there were two commonly cited temperature sources: the lighted display in front of the bank at the main intersection, and the grass airstrip outside town (for light single-engine planes). Being more visible to residents, the former is what typically got printed in the regional newspaper. The latter usually being a bit cooler. It possibly provided the "official" for the US Weather Service. That airstrip is now paved and considerably larger, and thus warmer. The bank and the area around it substantially changed when it was rebuilt after being inundated by a major flood. What if that grass airstrip had become a housing development,re-establishing itself further out of town? Another adjustment needed.
11. Problematic data: Historic ocean temperatures:
This is used in creating models both the behaviors of the oceans themselves and how the oceans affect weather and climate.
In a social setting years ago, an oceanographer remarked that climate models were using ocean temperature data that was known to be unreliable: It was OK for the narrow purpose for which it had been collected, but it suffered both from being inaccurate and was being treated as being more precise than it was. I was reminded of this by a recent article about a researcher who had worked out a way to reduce some of those problems: "^Tracking Global Warming Means Finding The Flaw In Old Data^" - NPR, 2019-08-19.Initial/alternative title: "How much hotter are the oceans? The answer begins with a bucket").
12. Problematic data: Tree ring correlation to climate:
Where there were good records of the weather, tree rings were found to have strong correlations -- strong enough to be considered as valid proxies for times and places where there aren't direct records. However, in recent decades, there has been an unexplained divergence, with tree rings indicating lower temperatures than what was being recorded. In the 2009 "Climategate" -- ^hacked emails from the prominent Climate Research Unit at the University of East Anglia (England)^-- one of the controversies centered on a researcher's reluctant decision to modify the tree-ring data chart to use recorded temperatures instead of the values from the (diverging) tree-ring data. The critics jumped on this silent substitution, which I believe missed the important issue: How were scientists to know whether and when such divergences occurred in the past? Since the weather inferred from tree rings is being used to formulate the climate models, those models could simply be houses built upon sand.
13. Green New Deal: slapdash climate proposal example:
On replacing air travel with high-speed rail:
- Senator Hirono (D-Hawaii) amusingly remarked that a high-speed rail from the mainland would be impractical.
- In rural US, connecting the sparse population would have trains with more crew than passengers.
- In Alaska, the many remote villages are connected by small planes -- even gravel roads don't make sense.
Deficient due diligence or disregarding details?
14. Statistically significance: data manipulation
For more information, start your web search with "p-hacking"; secondarily, "reproducibility crisis" (in Psychology).In the popular media, you will see these effects in contradicting announcements of research results in fields such as medicine, nutrition, and other health fields.
"Noooo! You mean that unlimited chocolate isn't good for my health?"
An ^abbreviated index by topic and chronologically^ is available.
----Boilerplate on Commenting----
The ^Guidelines^ for comments on this blog are different from those on Town Square Forums. I am attempting to foster more civility and substantive comments by deleting violations of the guidelines.
I am particularly strict about misrepresenting what others have said (me or other commenters). If I judge your comment as likely to provoke a response of "That is not what was said", do not be surprised to have it deleted. My primary goal is to avoid unnecessary and undesirable back-and-forth, but such misrepresentations also indicate that the author is unwilling/unable to participate in a meaningful, respectful conversation on the topic.
A slur is not an argument. Neither are other forms of vilification of other participants.
If you behave like a ^Troll^, do not waste your time protesting when you get treated like one.