Bad Science vs. Good Science: A Guide for the Layperson (Part 2)

This week's column by Ctein

Continuing from last time...into the realm of genuine crackpots. The problem with crackpots is that they can truly blind you with science, or at least with lengthy discourse and arguments about minutiae that you can't possibly fully dissect. Nor should you even feel obligated to try.

Crackpots exhibit many of the symptoms I described in Part 1. They're misunderstood geniuses; they see the truths that the whole scientific community has managed to miss or is suppressing because the rest of the community can't recognize their brilliance. Largely, this is the province of fiction. It's not impossible, but it's bloody damned unlikely.

Often they suggest darker, more malign motives keeping them out of the limelight. Again, it makes for gripping novels, but in the real world, scientific conspiracies are few and far between.

Sometimes they'll claim that other scientists have been bought off by the lure of grants and prestigious positions. The problem with this thinking is that all research involves money. Everyone gets paid for what they do. You can't use money to demonstrate that one scientific position is correct and another isn't. It's an ad hominem game that anyone can play, declaring the other side's information to be corrupted by money. And both sides do. It doesn't really help you, as a layperson, distinguish Good Science from Bad. It just reinforces people's prejudgements.

Some folks are just born crackpots, but the more serious cases start off as good scientists. They come up with a novel hypothesis that is entirely legitimate. But, when the science doesn't pan out, they can't let go of it. They just know they're right, so they keep looking and looking. Once in a blue moon their quest proves justified. More often, it destroys their careers and their reputation. It's extremely sad, because they frequently are brilliant and (formerly) productive scientists who just couldn't accept that they were wrong about something.

Most of these folks are harmless and just an annoyance in the scientific community. Occasionally, however, they're genuinely dangerous. Peter Duesberg was a well-respected virologist. In the 1980s, early in the AIDS crisis, he observed many peculiarities in the way AIDS worked that didn't fit standard models for infectious virus disease. He became convinced that AIDS was caused by environmental and lifestyle factors, not HIV.

In the early days of the epidemic, that was not implausible. Many researchers allowed for the possibility. But as work progressed, one after another those peculiarities got explained (the god of the gaps problem) and more and more infectious disease experts concluded, correctly, that Duesberg was wrong. Now, the overwhelming consensus is that HIV causes AIDS.

...Except Dr. Duesberg won't let go. Thirty years later, almost no one in the biomedical field believes him. This would not be a problem if he'd only destroyed his own reputation. Unfortunately, he has millions of believers worldwide. Some whole nations have adopted his crackpottery as national policy. Tens of thousands of people have died because of his Bad Science.

Arguably the worst kind of Bad Science is the outright, blatant fraud. This is relatively rare, but when it happens it can be extremely pernicious because it's very hard to detect. Most of the meta-warnings aren't present. These are people who know how to walk the walk and talk the talk, so even experts can't tell that they're in the presence of fraud. It only gets discovered when further work in the field contradicts their "discoveries."

The worst of these, in modern times, is the autism-vaccination connection promulgated by Dr. Andrew Wakefield. Like the anti-HIV rhetoric, this has killed tens of thousands of people outright and will kill more in the future. Dr. Wakefield published an utterly fraudulent paper in 1998 that reported that vaccines caused autism. Numerous further studies failed to replicate the connection, or did so erratically or at a much lower level than what Wakefield reported (that trend-line-in-results problem I described in Part 1). His results had been pretty well discredited when it was discovered that they were fabricated to begin with.

Folks like Duesberg and Wakefield go beyond Bad Science; so far as I'm concerned, this is genuinely Evil Science. I neither know nor care what their personal motivations are; I just know that they do massive harm in the world.

Most of that harm comes because laypeople honestly can't tell when they're being presented with Bad Science. Maybe these two columns will help to reduce that problem by some minuscule amount.

©2013 by Ctein, all rights reserved

Featured Comments from:

Geoff Wittig: "As a physician, I can categorically state that the utterly bogus vaccine-autism 'link' has caused immense mischief, suffering, and unnecessary death. Evil science is right. Vaccines have prevented more premature death than any medical tool we have at our disposal, including antibiotics. So it makes us want to beat our heads against the table when people tell us they 'don't believe in vaccines.' It's rather like not believing in gravity.

"Pertussis (whooping cough) is a bacterial infection that can be lethal in infants. As of 30 years ago it had become extremely uncommon in the U.S. due to high vaccination rates with a very potent vaccine. Many of my contemporaries had never seen a case of it. But last year we experienced a nasty little epidemic of pertussis in my area. I personally saw about a dozen cases. Several factors were in play, one of them the poorly understood tendency of pertussis to peak about every 12 years or so. Roughly 15 years ago we changed to a safer pertussis vaccine that is much less likely to cause fever in infants receiving it...but which unfortunately is also a bit less immunogenic, with immunity that is more likely to wane over time. In small part the spike in cases was an artifact of better testing which can now detect pertussis much more reliably. But mostly it was due to a an expanding pool of susceptible individuals who had never been immunized at all, or who had 'aged out' of prior immunity.

"In our area, ground zero for the epidemic was an affluent community with a large number of vaccine skeptics who had 'protected' their kids from vaccination. Such a pool of non-immune individuals can then propagate the infection into the larger population around them.

"Bad science isn't just a pie-fight in a technical journal. It can have dreadful real-world consequences."

Alan Hill: "Bad science happens all the time. Mostly it gets filtered out by the system (repeated experiments, referees etc.). But no system is perfect: in the case of Andrew Wakefield's paper in The Lancet, a most reputable journal, the system fell down.

"However the real problem comes when bad science is exploited by vested interests or people with an axe to grind. Wakefield's paper would have had very little consequence if certain sections of the British media, particularly The Daily Mail, had not hyped it up wildly, with global consequences. I suspect that a bad scientist is usually a naïve fool rather than an evil genius, but I would never trust a non-specialist who promotes an unorthodox scientific idea until I had examined his motivation very carefully."

Comments

[AM Townsend]

Don't know where you draw the bounds on conspiracy theories, but the Stalinist suppression of Mendelian genetics in favor of Lysenko's "hybridization" theories was certainly a full throttled suppression effort. Set the soviets behind by a decade in their agricultural production, and more importantly, has demonstrated to many serious scientists with the dangerous assumption that official approval means the something is scientifically valid.

[Animesh Ray]

A great article. Bravo Ctein. One should point out, however, that Duesberg's science is not really "bad science" in that sense. Some will argue, myself included, that Duesberg actually continues to critique certain "bad science" related to AIDS virus research, and his critique of the orthodoxy, though has been harmful for medical policy, is really a purist's skepticism (some would claim, fanatical skepticism). Unfortunately, in medical science a purist's approach isn't always the most effective one in a timely fashion. Unlike mathematics, science is not "logical" or axiomatic, but is subject to 'best judgement' according to the weight of empirical evidence, where all variables might not be excluded before conclusions are drawn. In contrast to his somewhat destructive position on AID's virus, Duesberg the man continues to be an active scientist in at least one other area--theory of cancer heterogeneity, and his ideas therein are credible (e.g., http://www.ncbi.nlm.nih.gov/pubmed/23388461).

[John Camp]

Then there are the totally inadvertent errors by well-meaning scientists that turn out to have very expensive consequences. I know (personally) of a cancer researcher who found an intriguing and unexpected anomaly in some animal-based research she was doing, that pointed toward a possible advance in a certain kind of cell treatment. Not only that, but a separate team, at the same facility, confirmed the anomaly. It turned out that the animals they were using (mice) had been contaminated by a physically harmless virus that ran through the entire mouse population and had gone undetected...but had the effect of producing the cellular anomaly. Although the virus was physically harmless, it certainly wasn't economically harmless, as several researchers then spent several researcher-years trying to extend the results, before the truth was discovered. The inadvertently bad science, I am told, did produce a couple of PhDs.

[John Camp]

re: my post above.
It occurred to me about a nano-second after I pushed the button that this kind of science isn't the kind of bad science you're talking about -- it's simply an error, though an expensive one, that was eventually sorted out and agreed upon. So maybe that's good science.

[Joseph Kashi]

These are excellent articles by Ctein about Bad Science and why it will so often be believed for emotional and ideological reasons, even when disproven. Part of the problem, I believe, is that science and mathematics are too-often reflexively rejected by those who do not understand these areas.

Prior to the autism-vaccine hypothesis, many mental health therapists in the 1970s through 1990s, especially those in the Freudian and "eclectic" traditions based upon received dogma and therapists's intuition rather than real studies, diagnosed autism as resulting from distant "refrigerator" parents.

That alleged autism link, too, was disproven but only after a huge amount of damage was done to both children and their parents. I am personally familiar with a few such instances.

It's worth noting that scientific fraud, as in intentionally concocted false data, is a federal crime whenever the research includes federal grant funds.

As a result of a series of US Supreme Court decisions since in 1993, both state and federal judges are now required to act as gatekeepers to exclude from evidence Bad Science of the autism-vaccine sort. In order to be admissible as evidence now, the alleged science proof must be generally accepted by practitioners in the relevant scientific area, replicated, and falsifiable - basically, the courts use Karl Popper's criteria that are generally accepted as the basis of the modern scientific method.

Unfortunately, intentionally cooked data happens from time to time, which is why we insist upon consistent replication of experimental results, especially when the results seem novel or unexpected. The autism study from the UK, to which Ctein refers, was apparently one such example, assuming that press reports are accurate. The doctor in question, if I recall correctly, was in consequence later stripped of his medical license by British authorities. But, far too much damage was done, and continues to be done, especially by credulous parents who preferred to find some deep third-party reason rather than the genetic determinants that now seem the underlying cause.

Other situations also do major social damage. As an example, the alleged link between silicon breast implants and various auto-immune diseases like lupus was prominently asserted by the pathologist who claimed to prove the auto-immune reaction and was a frequent, paid "expert" in damage claims against the manufacturers.

That alleged link was repeatedly disproven by a series of large CDC and other studies that showed no significant systemic effect in actual people. That pathologist who made that assertion was not accused of cooking data, if I recall correctly. It seems that he used improper and inaccurate experimental techniques to do experimental techniques for which he was not appropriately trained and hence had systemic false data results that neutral scientists could not replicate.

However, the result of that testimony with credulous juries around the US was a series of huge damage awards not ultimately grounded in fact and science, causing the bankruptcy of several companies and the loss of many jobs before being slowly corrected.

For whatever cultural and political reasons, perhaps an unfamiliarity with statistical proof, many still refuse to accept the CDC and other studies and continue to adhere to the discredited urban legends of causation, seeing a government and business conspiracy instead.

In the long run, though, both merely careless technique and intentionally cooked data do major cultural damage in undermining confidence in science generally, which is the only real proof-based knowledge system that we have.

[MarkB]

As the (layperson) spouse of a respected scientist, I thank you for writing in plain prose what I've come to understand over time, and wished I could communicate to my mate. Sometimes scientists are not the easiest people to talk to about... science.

[Eric Brody]

Bravo Ctein for mentioning Wakefield. I agree, he seems to be a genuinely evil man. His license to practice medicine has been revoked. His original work was financed by the plaintiffs suing the vaccine companies. While I have no love for big pharma, especially their direct to consumer advertising, vaccines simply save lives. I used to tell parents who resisted immunizations that they should talk to their grandparents about polio in the 1950's or go to a third world country where preventable diseases kill children every day.

[Chris Gibbs]

Ignaz Philipp Semmelweis?

[EricF]

One perfectly fitting example of good scientist gone crackpot is Linus Pauling who became obsessed with vitamins after having earned two Nobel prizes. He was convinced that massive amounts of vitamins could cure cancer and increase life expectancy among other miracles.

His previous and deserved reputation as a chemist and biochemist unfortunately gave a lot of credit to his late theories. I recently read a good article about it:
http://www.theatlantic.com/health/archive/2013/07/the-vitamin-myth-why-we-think-we-need-supplements/277947/?1

[Bear.]

Just YOUR opinion... :)

[Andreas]

Ctein, are you going to address the issue of media reporting of science? My loathing of 'Dr' Wakefield (now struck off the medical register, I believe) is upstaged only by my feelings towards the mainstream media who seemed to be delighted to pour fuel on that particular fire with no apparent thought to the dreadful consequences for the people who read and believed their published drivel which was supportive of Wakefield's bogus claims.

Worse still, IMO, is the fact that the media were eventually in a position to clear up the public's perception of vaccines. Sadly, they did not; the debunking of Wakefield's theories were reported with a tiny fraction of the intensity of the original vaccine 'scare' story.

BTW, for those interested in bad science I recommend Ben Goldacre's 'Bad Science' or the same author's 'Bad Pharma' (I have yet to read the latter, but I hear good things about it).

[mastaliu]

The Good-Bad binary works for single examples of rogue "evil" scientists (and the examples by Ctein are certainly bad science), but doesn't help us to understand the history of science and how norms develop in scientific thought and how these norms we take for granted can be harmful even when they are the standard consensus.

Homosexuality was labeled a mental disorder by the American Psychiatric and Psychological associations until the 1970s. This was a norm, fitting human sexuality into rigid classifications that were ignorant of biological and social patterns. Norms went against scientific understanding and were influenced by prevalent ideologies. What about that harm to thousands of men and women, a social and psychological harm that continues to reverberate in our society today? Is this the evil science of an individual? A more contemporary example of scientific norms and standards that cause serious harm might be Intersex Genital Mutilation.

I don't need to go into the ugly historical genealogy of standard Western scientific dogma and myth, from eugenics to biological racism. The point is that science, whether in the natural sciences or humanities, and its norms need to be understood with their political dimensions, that science (not fringe science, but mainstream consensus too) gets dirty and tangled up with all sorts of things we're probably uncomfortable with. Think about why those affluent parents are starting to reject vaccines and follow bad science? Why is it that they believe this rejection of medicine and instead free ride on everyone else's immunization?

The sooner we bring science out of it's false objective standing and artificial distance as an exclusive specialized knowledge (better science education might help) and accept and understand its political and social dimensions, the sooner we'll (us laypeople) be able to critically think about and judge bad and good science.

[Henry Rogers]

Ctein, Two very interesting articles but undermined by a failure adequately to distinguish between deliberate scientific fraud and obsessive wrong-headedness. Both can have horrible consequences but allegations of the former which subsequently don't stand up in court can have financial consequences too. Btw one can't help recalling that in days long ago one justification for burning heretics at the stake was that they were endangering the immortal souls of others. Being a perishing nuisance to the civil powers that were was another of course.

[Alan Hill]

Bad science happens all the time. Mostly it gets filtered out by the system (repeated experiments, referees etc). But no system is perfect: in the case of Andrew Wakefield's paper in The Lancet, a most reputable journal, the system fell down.
However the real problem comes when bad science is exploited by vested interests or people with an axe to grind. Wakefield's paper would have had very little consequence if certain sections of the British media, particularly the Daily Mail, had not hyped it up wildly, with global consequences.
I suspect that a bad scientist is usually a naive fool rather than an evil genius, but I would never trust a non-specialist who promotes an unorthodox scientific idea until I had examined his motivation very carefully.

[William]

Painting oneself into a corner is an unpleasant experience so many people refuse to believe they have unmovable walls at their back and are surrounded by wet paint.

When these people harm others, it is evil.

[Andrew Molitor]

Crackpots are interesting. Sometimes, they are bellwethers.

Some of them are just nutters, sure. Some of them have an idee fixe which is supported by nothing but their own madness, but others have an idee fixe which is supported by the evident and obvious fact that something is fishy here. To coin a phrase, there's a lot of Crummy Science out there, which is being driven by money, by politics, or even by ethical concerns. When the real science is dubious, the crackpots are encouraged to hold on to their radical positions.

Crummy Science is science that looks like good science, the people doing it are reasonable, the results are reasonable, and it appears that the appropriate processes are being followed. It's Crummy because the results are wrong because the people doing it, while respectable and experts in the field, are a little bit biased, a little bit sloppy, and under a great deal of pressure to write a paper or ten.

When Crummy Science is in play, reality probably lies not-very-close to the crackpot position, but may or may not be all that close to the mainstream.

To pick a not-very-controversial example, it turns out that most epidemiological studies are wrong. They measure, essentially, experimenter bias and problems in design. These studies bear all the earmarks of Good Science, and people wailing against them frequently look like crackpots. You're a crackpot if you think eggs aren't going to kill you, or you're a shill for the egg producer's association, etc. Oh wait, that study was actually crap which measured nothing but its own experimental error. Oops.

The magic of Science is that it does sort itself out.

[Nicholas Condon]

"Again, it makes for gripping novels, but in the real world, scientific conspiracies are few and far between."

Was there ever a 20th-century (or even a 19th-century) case of a true scientific conspiracy, where the leading lights of a field intentionally and with malice aforethought banded together to suppress a result that challenged the orthodoxy? I know it used to happen for religious motivations back in the day, but I can't think of a single example of such conspiracies in the modern era.

[Nic]

Science... became scientism, which means it didn't just pursue its own truths, it aggressively denied that there were any other truths at all... Ken Wilbur

Lies, Damned Lies, and Medical Science
Much of what medical researchers conclude in their studies is misleading, exaggerated, or flat-out wrong. So why are doctors—to a striking extent—still drawing upon misinformation in their everyday practice? Dr. John Ioannidis has spent his career challenging his peers by exposing their bad science. http://www.theatlantic.com/magazine/archive/2010/11/lies-damned-lies-and-medical-science/308269/?single_page=true

[ctein]

Dear Chris & Henry,

I don't know if Semmelweis was a crackpot or not. I don't know enough about his life.

Having your work non-accepted by the scientific establishment isn't what makes someone a crackpot (although it's a requirement). What makes them a crackpot is when they steadfastly and excessively hang onto a particular belief when the scientific data and analysis consistently says otherwise. I believe (not having done any particular research) that Semmelweis was just ignored (and castigated) not disproven by anyone else's work.

Henry, I would've thought the distinction would be terribly clear to everyone. It's the common colloquial one. A fraud is someone who lies to you to convince you that they are to be believed. Just that simple and no more complicated. Crackpots, on the other hand, sincerely and deeply believe in something that is wrong. The dedicated crackpot wouldn't lie to you; they have no reason to. They truly believe that what they know (or think they know) is correct.

That doesn't mean they won't give you bad information, but they don't think it's bad. Frauds know it's bad.

Duesberg became a crackpot. Wakefield became a fraud. As did Newton. Who also became a crackpot in his later years. (One doesn’t preclude the other, they're just different.)

Which takes us back to what I said in the very beginning, Sometimes frauds are correct. Sometimes bad science is correct. It's just not where you should place your money. Someone who claims they are the next Newton or Semmelweis is almost invariably wrong. Proof of existence is not proof of likelihood. Somebody holds a winning lottery ticket. What are the odds that it's your next-door neighbor?

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

[ctein]

Dear Nicholas,

I knew of a few scientific conspiracies that have successfully suppressed results.

It is exceptionally uncommon, though, and they were not done with malice and not because of a challenge to the orthodoxy.

I will not give any details. I'm on the conspirators' side.

pax / Ctein

[Henry Rogers]

Ctein, I'd have thought you were probably wrong in your definition of fraud. As commonly used, and that's the sense I used it, deceit is committed for financial gain, not just for the sake of being 'one up'. Accusations of fraud need to be made with great care because they imply criminality.

I'm sure you'll agree, however, that there are plenty of people who are factually correct in their assertions who display all the irritating behaviour patterns of cranks and obsessives.

[ctein]

Dear Henry,

No, we're not going down that semantically nitpicky path. Your preferred definition is not the only one, and it's not my preferred one. Words have multiple definitions so that those may be used. Doesn't matter what any one person thinks is "best" or "most common."

(Not a remark you made, but one that one sees very frequently on line from folks who erroneously assert that someone is misusing a word.)

So, kind readers, read the word the way I meant it. It's entirely clear from context how I used it.

We don't split hairs, here on TOP, just pixels.

pax / Ctein

[Norm Snyder]

I'm late to this discussion, but it is also important to consider how high the official, i.e. governmental, support of bad science, in this case Eugenics during the 20th century, can be. The case of Carrie Buck, a young woman with cognitive disability who was mandatorily sterilized by the State of Virginia, eventually went to the US Supreme Court, after she had been subjected to surgery (as were thousands of people accross the country). The court, in Buck vs. Bell, voted 8 to 1 in support of the law, and the court's majority opinion, written by Oliver Wendell Holmes, Jr., reads:

We have seen more than once that the public welfare may call upon the best citizens for their lives. It would be strange if it could not call upon those who already sap the strength of the State for these lesser sacrifices, often not felt to be such by those concerned, to prevent our being swamped with incompetence. It is better for all the world, if instead of waiting to execute degenerate offspring for crime, or to let them starve for their imbecility, society can prevent those who are manifestly unfit from continuing their kind. The principle that sustains compulsory vaccination is broad enough to cover cutting the Fallopian tubes.

This court decision has never been explicitly overturned and it was not until the 1970s that the last of these State laws was taken off the books. It still astounds me that so many can be taken in by bad science like eugenics, or "research" on vaccinations that should be dismissed, and then persist in their views, as in the latter case, even after the "science" they have treated like gospel is proven to be rubbish.

[Nic]

One of the first duties of the physician is to educate the masses not to take medicine. William Osler

http://www.brainyquote.com/quotes/quotes/w/williamosl159321.html

At any given moment, there is a sort of all pervading orthodoxy, a general tacit
agreement not to discuss large and uncomfortable facts… Anyone who
challenges the prevailing orthodoxy finds himself silenced with surprising
effectiveness. - George Orwell