The other day I was shocked to read that several prominent geologists in Italy had been convicted of manslaughter for failing to adequately predict an earthquake that ultimately claimed the lives of 300 people. Or maybe the prediction was made but the press did not properly communicate it to the public. To the extent I can figure out who is culpable, it doesn’t seem to be the scientists. After all, saying or implying that a particular event is “not probable” does not mean it won’t happen, but many citizens in the area of the quake apparently either didn’t understand that or chose to put an overly-optimistic spin on what they heard. Further, it appears that the resulting loss of life was way out of proportion to the severity of the tremor—an earthquake of magnitude 6.2 like this Italian one probably wouldn’t make headlines in California or Japan. The people who are really at fault are probably the administrators who are responsible for establishing Italian building codes.
This line of thinking led me down various tracks, such as what is science, pseudoscience, rules of evidence, correlation vs. causation, probability, and finally, great examples of how science has been wrong in the past. Actually, some of the examples that occurred to me did not come about because science got it wrong—science either did not exist at the time or was so primitive it essentially did not exist—but rather people’s ideas, theories, folklore, or myths were wrong. But it wouldn’t be very entertaining to poke fun at a primitive conception of the earth being flat or the world resting on the back of a turtle. Or of the sun revolving around the earth or passing through an underworld as it “moved” from sunset to sunrise. Then there was the belief that malaria was caused by “bad air” [mal (bad) + aria (air), get it?). Or the idea that a cute little human called a homunculus supposedly resided in each sperm cell (preformationism). And so on.
But we don’t have to look to the distant past to find incorrect beliefs about some really basic things—in our own lifetime there have been many ideas that were accepted by a lot of people, but have been since been rejected by the scientific community. Like ulcers being caused by “stress.” Or homosexuality being caused by “how we were raised.”
And so I thought it would be fun to look at some “theories” that have been widely accepted at some point during the past 75 years or so and that many of us remember. The three I have chosen collectively illustrate how science works, how press reports can influence beliefs, and how theories may become politicized. With apologies to Clint Eastwood, I have called them the ugly, the good, and the bad.
First the ugly. One of the most egregious and widespread misconceptions was the Lysenko doctrine, which lasted right up until the mid 1960s. Now, many of you may not of heard of Lysenko, but millions upon millions of Soviet citizens from the 1930s to the mid 1960s did. And most Russian scientists know his name well. This is because he perpetrated perhaps the greatest scientific hoax in the history of the modern world, and in consequence probably contributed to the many famines that swept the Soviet Union in the middle years of the twentieth century.
And I use the term “scientific hoax” lightly, since there was very little science associated with it. Basically, Lysenko convinced the political system in the Soviet Union that what was at the time considered genetics was “bourgeois science” and should be rejected. What he advocated instead was the concept of “the inheritance of acquired characteristics.” He believed that if he exposed seeds of wheat, for example, to cold conditions, then not only would they germinate better, but their offspring would germinate better. Further, the seeds exposed to cold would produce hardier plants, and this hardiness would pass down from generation to generation. Or if he fertilized crops and they had enhanced growth, then this growth pattern would be transmitted to subsequent generations. Or if he trimmed a tree so that it had reduced branching, then its offspring would have reduced branching too. His claims came so fast that scientists who wondered whether his ideas actually worked didn’t have either the time or the resources to refute him. That is, they didn’t have time or money to do what is central to good science—the replication of his work.
Now, the idea of the inheritance of acquired characteristics (known as Lamarckism) had actually been popular since the 1800s. In fact Darwin believed in this mode of inheritance, which he advocated in the last chapter of On the Origin of Species (1859). But as gene theory was developed in the early 1900s, this mode of inheritance was gradually rejected except in the Soviet Union, where Lysenko’s influence was so great that scientists who refused to accept his ideas were often imprisoned or sent to labor camps. Consequently, an entire generation of geneticists either left the Soviet Union for fear of their lives and reputations, or were “lost.” (The United States benefited, however, because many of them came to North America.)
As a consequence, the Soviet Union was plunged into massive ignorance about genetics and its practical applications to plant breeding. Students were not taught modern genetics. When DNA discoveries were made during this time period, biologists in the Soviet Union did not learn about them—or if they did, they were officially rejected. It is a striking counterpoint to the huge Soviet successes in space and their accomplishments in other fields such as physics, math, and engineering. One wonders why the same methodologies they used in physics were not applied to biology. I think the best answer is cultural—the inheritance of acquired characteristics comported with Soviet belief in the infinite malleability of all things, including human thought.
Anyway, Lysenkoism was finally rejected after repeated famines and agricultural failures. By the late 1960s Lysenko himself had been demoted, and the Soviet Union started to catch up with the rest of the world, from way, way behind.
Interestingly, there is increasing evidence that Lamarckian-type inheritance has validity in some very limited circumstances. There is a new field called epigenetics that is starting to show how this mode of inheritance can work. This in itself is a wonderful example of how science progresses by fits and starts—even though a theory is proposed and initially rejected, continued research may in fact show there is something to it. Or it may not. But this does not happen without relentless cycles of idea generation, testing, rejection, refinement, replication by other scientists, and new idea generation, with the new ideas being substituted for the previous, more primitive ones. And so the knowledge edifice is gradually built.
And now for the good—continental drift.
I’m sure that all of you know that the continents “drift” over time. Every school child has looked at the west coast of Africa and the east coast of South America and noted that they “match up.” This has been observed since 1596. Also, by the early 1900s it was recognized that fossils of similar reptiles and plants can be found in Antarctica, South America, Africa, India, and Australia. For example, there are two extinct species of reptiles that lived in both South America and Africa, but nowhere else. The extinct fern Glossopteris is found on five continents. (In fact Robert Scott collected the first fossils of this fern in his ill-fated race to the South Pole; remarkably, even though the explorers abandoned vital gear in their failed race to safety, they died with 35 pounds of fossils next to their side. Imagine that, placing such importance on science in the face of crushing fatigue and starvation! Scott has been widely criticized for this—risking lives for “rocks.”)
So clearly ONE explanation as to why fossils of the same species have been found on five different continents was that the continents were once connected. But then, another explanation was that in the distant past there were no oceans separating the continents, so that these plants and animals were just widely distributed over a range that was once waterless.
And here we have an excellent example of how two different theories can explain the same phenomenon. And this is the root of how science progresses—in order to resolve the dilemma, one of the theories has to be rejected, or “falsified.”
The term “continental drift” was first proposed by Alfred Wegener in 1912. From 1912 to 1968 the question of whether or not the continents actually “drifted” was hotly debated because there was no widely-accepted theory for the mechanism driving continent movement. One theory was that the continents moved due to the spin of the earth. Another attributed continental drift to tidal forces created by the moon; another was that continents were floating on top of liquid magma; and still another was that cooling of the earth caused the continents to move around.
Fortunately, during this same time period the concept of “tectonic plates” was also being developed, the idea being that the earth’s crust is divided into “plates”, like a jigsaw puzzle. There are seven major plates, which are then further divided into dozens and dozens of smaller plates. Another concept developed in the early 1960s was that of “sea floor spreading,” which resulted from the observation that a continual upwelling of magma in the mid-Atlantic over millions of years had created a ridge that in turn has caused the sea floor to “spread” out on either side.
So by the late 1960s there was the theory that continents “drifted”; it was known the earth’s crust is divided into “plates,”; and it had been established that the sea floor is spreading. But there was no ONE theory tying all of these observations together and providing a mechanism driving the movement of continents.
And then in 1968 the problem was solved. By analyzing the travel patterns of seismic waves created by 400-mile-deep earthquakes, it was demonstrated the seafloor was being pushed under the continental crust. And so these researchers proposed a theory—that sea floor spreading caused tectonic plates to move, which in turn caused continents to move. Further, there was a “recycling” of oceanic crust as it spread out and then dipped down under the continents deep back into the earth. At times some of this oceanic crust got “scrapped” off onto the continental crust and added to it.
So today, plate tectonics is probably the single most unifying concept in geology, explaining everything from the stratification of rocks on earth, the existence of volcanoes, and the building of mountains, to the distribution of fossils as well as living plants and animals.
And finally, let’s look at vaccines—the bad. From their use to vaccinate against smallpox during the Revolutionary War, to the eradication of polio in the 1950s and 1960s, and the near elimination of measles, mumps, smallpox, diphtheria, and rubella in the western world, you’d think the use of vaccines would merit unqualified respect in modern health care. However, in recent years there has been an upwelling of anti-vaccine fervor, and it is estimated that 20-25% of the people in the U.S. believe that vaccines cause autism.
The story really begins with the publication in 1998 of a paper in a very respected medical journal, The Lancet. In this article, Andrew Wakefield and 12 other authors reported that of 12 children studied, 8 showed a link between the use of an MMR (measles, mumps, rubella) vaccine and a “new condition” called autistic enterocolitis. Prior to the publication of the paper, a press conference was held to announce the results. A public furor against the vaccine was ignited, and its fires are still burning.
It is critical to point out that (1) after publication, Wakefield’s paper was subsequently denounced by most of his co-authors, and was eventually retracted by The Lancet in 2010; (2) Wakefield’s results have never been duplicated by any subsequent study; and (3) an analysis of the underlying research uncovered fraudulent handling and fabrication of data. Many many studies, too numerous to list here, have tried—and failed—to find either a statistical correlation between vaccines and autism or an underlying causal mechanism.
What is amazing is that a significant proportion of the public still believes that vaccines cause autism, in spite of the large body of evidence to the contrary and in the complete absence of any scientific evidence supporting this notion. The problem is that it is quite easy to find ANECDOTAL evidence—that is, people say “my child got a vaccine and then they got autism.” It is hard to believe this could be convincing to anyone other than the family involved, but it appears to be. It seems that neither the press nor a large segment of the population knows the difference between correlation and causation, that it usually takes hundreds, if not thousands, of test subjects under careful experimental control and analysis (double blind experiments) to come to any valid conclusions; that false positives occur due to chance alone, and etc., etc. And on top of everything else, it seems the press is more interested in sensationalism rather than reporting the facts.
In the pre-internet days, it is understandable how the public could have been swayed by “yellow journalism.” After all, no one can be an expert in everything and locating relevant information used to require quite some effort. But today almost every published study and paper is available on the web—for everyone to look at and judge for themselves. Of course not everything on the web is legitimate, and sorting through the anecdotal and the fraudulent can be daunting, but there are a couple of excellent rules of thumb for evaluating the reliability of information found on the internet: if it comes from a private individual’s website, it is to be discounted; if it comes from the Mayo Clinic (or another esteemed institution) or a respected peer reviewed journal—especially if many legitimate studies have come to the same conclusion—then one’s confidence is increased greatly.
So in summary, what we have in these three examples is the “ugliness” of Lamarckism, which represents bad science being promoted by a corrupt establishment; the “goodness” of continental drift, which is an example of excellent science and the struggles involved in developing and testing new theories; and the “badness” of the vaccine-autism debate, which is a lesson in bad science being promoted by a free press and an otherwise educated public’s irrational reaction.
This got me to thinking about “rules of evidence.” And how people often do not think about why they believe something to be true.
And there ARE rules of evidence that are generally accepted by all scientists, or at least all reputable scientists. This is a good thing because there are certain values or qualities that scientists around the world SHOULD adhere to because they undergird all good science. As posted by the global “InterAcademy Panel: The Global Network of Science Academies” on Oct. 17, 2012, these are:
“honesty, fairness, objectivity, reliability, skepticism, accountability, and openness.”
Sound familiar? These are, actually, values that ALL people should adhere to. Not just scientists. But science has another mechanism that is a “check” on their work that the public may not be aware of and that most people are not subjected to. And this mechanism is that fellow scientists will always eventually get around to checking each other’s work and attempt to duplicate it. And publish the results.
Repeatability is central to the entire scientific enterprise. And in each of the three stories told here, scientific theories have either failed or survived based on this one overriding factor.
This line of thinking led me down various tracks, such as what is science, pseudoscience, rules of evidence, correlation vs. causation, probability, and finally, great examples of how science has been wrong in the past. Actually, some of the examples that occurred to me did not come about because science got it wrong—science either did not exist at the time or was so primitive it essentially did not exist—but rather people’s ideas, theories, folklore, or myths were wrong. But it wouldn’t be very entertaining to poke fun at a primitive conception of the earth being flat or the world resting on the back of a turtle. Or of the sun revolving around the earth or passing through an underworld as it “moved” from sunset to sunrise. Then there was the belief that malaria was caused by “bad air” [mal (bad) + aria (air), get it?). Or the idea that a cute little human called a homunculus supposedly resided in each sperm cell (preformationism). And so on.
But we don’t have to look to the distant past to find incorrect beliefs about some really basic things—in our own lifetime there have been many ideas that were accepted by a lot of people, but have been since been rejected by the scientific community. Like ulcers being caused by “stress.” Or homosexuality being caused by “how we were raised.”
And so I thought it would be fun to look at some “theories” that have been widely accepted at some point during the past 75 years or so and that many of us remember. The three I have chosen collectively illustrate how science works, how press reports can influence beliefs, and how theories may become politicized. With apologies to Clint Eastwood, I have called them the ugly, the good, and the bad.
First the ugly. One of the most egregious and widespread misconceptions was the Lysenko doctrine, which lasted right up until the mid 1960s. Now, many of you may not of heard of Lysenko, but millions upon millions of Soviet citizens from the 1930s to the mid 1960s did. And most Russian scientists know his name well. This is because he perpetrated perhaps the greatest scientific hoax in the history of the modern world, and in consequence probably contributed to the many famines that swept the Soviet Union in the middle years of the twentieth century.
And I use the term “scientific hoax” lightly, since there was very little science associated with it. Basically, Lysenko convinced the political system in the Soviet Union that what was at the time considered genetics was “bourgeois science” and should be rejected. What he advocated instead was the concept of “the inheritance of acquired characteristics.” He believed that if he exposed seeds of wheat, for example, to cold conditions, then not only would they germinate better, but their offspring would germinate better. Further, the seeds exposed to cold would produce hardier plants, and this hardiness would pass down from generation to generation. Or if he fertilized crops and they had enhanced growth, then this growth pattern would be transmitted to subsequent generations. Or if he trimmed a tree so that it had reduced branching, then its offspring would have reduced branching too. His claims came so fast that scientists who wondered whether his ideas actually worked didn’t have either the time or the resources to refute him. That is, they didn’t have time or money to do what is central to good science—the replication of his work.
Now, the idea of the inheritance of acquired characteristics (known as Lamarckism) had actually been popular since the 1800s. In fact Darwin believed in this mode of inheritance, which he advocated in the last chapter of On the Origin of Species (1859). But as gene theory was developed in the early 1900s, this mode of inheritance was gradually rejected except in the Soviet Union, where Lysenko’s influence was so great that scientists who refused to accept his ideas were often imprisoned or sent to labor camps. Consequently, an entire generation of geneticists either left the Soviet Union for fear of their lives and reputations, or were “lost.” (The United States benefited, however, because many of them came to North America.)
As a consequence, the Soviet Union was plunged into massive ignorance about genetics and its practical applications to plant breeding. Students were not taught modern genetics. When DNA discoveries were made during this time period, biologists in the Soviet Union did not learn about them—or if they did, they were officially rejected. It is a striking counterpoint to the huge Soviet successes in space and their accomplishments in other fields such as physics, math, and engineering. One wonders why the same methodologies they used in physics were not applied to biology. I think the best answer is cultural—the inheritance of acquired characteristics comported with Soviet belief in the infinite malleability of all things, including human thought.
Anyway, Lysenkoism was finally rejected after repeated famines and agricultural failures. By the late 1960s Lysenko himself had been demoted, and the Soviet Union started to catch up with the rest of the world, from way, way behind.
Interestingly, there is increasing evidence that Lamarckian-type inheritance has validity in some very limited circumstances. There is a new field called epigenetics that is starting to show how this mode of inheritance can work. This in itself is a wonderful example of how science progresses by fits and starts—even though a theory is proposed and initially rejected, continued research may in fact show there is something to it. Or it may not. But this does not happen without relentless cycles of idea generation, testing, rejection, refinement, replication by other scientists, and new idea generation, with the new ideas being substituted for the previous, more primitive ones. And so the knowledge edifice is gradually built.
And now for the good—continental drift.
I’m sure that all of you know that the continents “drift” over time. Every school child has looked at the west coast of Africa and the east coast of South America and noted that they “match up.” This has been observed since 1596. Also, by the early 1900s it was recognized that fossils of similar reptiles and plants can be found in Antarctica, South America, Africa, India, and Australia. For example, there are two extinct species of reptiles that lived in both South America and Africa, but nowhere else. The extinct fern Glossopteris is found on five continents. (In fact Robert Scott collected the first fossils of this fern in his ill-fated race to the South Pole; remarkably, even though the explorers abandoned vital gear in their failed race to safety, they died with 35 pounds of fossils next to their side. Imagine that, placing such importance on science in the face of crushing fatigue and starvation! Scott has been widely criticized for this—risking lives for “rocks.”)
So clearly ONE explanation as to why fossils of the same species have been found on five different continents was that the continents were once connected. But then, another explanation was that in the distant past there were no oceans separating the continents, so that these plants and animals were just widely distributed over a range that was once waterless.
And here we have an excellent example of how two different theories can explain the same phenomenon. And this is the root of how science progresses—in order to resolve the dilemma, one of the theories has to be rejected, or “falsified.”
The term “continental drift” was first proposed by Alfred Wegener in 1912. From 1912 to 1968 the question of whether or not the continents actually “drifted” was hotly debated because there was no widely-accepted theory for the mechanism driving continent movement. One theory was that the continents moved due to the spin of the earth. Another attributed continental drift to tidal forces created by the moon; another was that continents were floating on top of liquid magma; and still another was that cooling of the earth caused the continents to move around.
Fortunately, during this same time period the concept of “tectonic plates” was also being developed, the idea being that the earth’s crust is divided into “plates”, like a jigsaw puzzle. There are seven major plates, which are then further divided into dozens and dozens of smaller plates. Another concept developed in the early 1960s was that of “sea floor spreading,” which resulted from the observation that a continual upwelling of magma in the mid-Atlantic over millions of years had created a ridge that in turn has caused the sea floor to “spread” out on either side.
So by the late 1960s there was the theory that continents “drifted”; it was known the earth’s crust is divided into “plates,”; and it had been established that the sea floor is spreading. But there was no ONE theory tying all of these observations together and providing a mechanism driving the movement of continents.
And then in 1968 the problem was solved. By analyzing the travel patterns of seismic waves created by 400-mile-deep earthquakes, it was demonstrated the seafloor was being pushed under the continental crust. And so these researchers proposed a theory—that sea floor spreading caused tectonic plates to move, which in turn caused continents to move. Further, there was a “recycling” of oceanic crust as it spread out and then dipped down under the continents deep back into the earth. At times some of this oceanic crust got “scrapped” off onto the continental crust and added to it.
So today, plate tectonics is probably the single most unifying concept in geology, explaining everything from the stratification of rocks on earth, the existence of volcanoes, and the building of mountains, to the distribution of fossils as well as living plants and animals.
And finally, let’s look at vaccines—the bad. From their use to vaccinate against smallpox during the Revolutionary War, to the eradication of polio in the 1950s and 1960s, and the near elimination of measles, mumps, smallpox, diphtheria, and rubella in the western world, you’d think the use of vaccines would merit unqualified respect in modern health care. However, in recent years there has been an upwelling of anti-vaccine fervor, and it is estimated that 20-25% of the people in the U.S. believe that vaccines cause autism.
The story really begins with the publication in 1998 of a paper in a very respected medical journal, The Lancet. In this article, Andrew Wakefield and 12 other authors reported that of 12 children studied, 8 showed a link between the use of an MMR (measles, mumps, rubella) vaccine and a “new condition” called autistic enterocolitis. Prior to the publication of the paper, a press conference was held to announce the results. A public furor against the vaccine was ignited, and its fires are still burning.
It is critical to point out that (1) after publication, Wakefield’s paper was subsequently denounced by most of his co-authors, and was eventually retracted by The Lancet in 2010; (2) Wakefield’s results have never been duplicated by any subsequent study; and (3) an analysis of the underlying research uncovered fraudulent handling and fabrication of data. Many many studies, too numerous to list here, have tried—and failed—to find either a statistical correlation between vaccines and autism or an underlying causal mechanism.
What is amazing is that a significant proportion of the public still believes that vaccines cause autism, in spite of the large body of evidence to the contrary and in the complete absence of any scientific evidence supporting this notion. The problem is that it is quite easy to find ANECDOTAL evidence—that is, people say “my child got a vaccine and then they got autism.” It is hard to believe this could be convincing to anyone other than the family involved, but it appears to be. It seems that neither the press nor a large segment of the population knows the difference between correlation and causation, that it usually takes hundreds, if not thousands, of test subjects under careful experimental control and analysis (double blind experiments) to come to any valid conclusions; that false positives occur due to chance alone, and etc., etc. And on top of everything else, it seems the press is more interested in sensationalism rather than reporting the facts.
In the pre-internet days, it is understandable how the public could have been swayed by “yellow journalism.” After all, no one can be an expert in everything and locating relevant information used to require quite some effort. But today almost every published study and paper is available on the web—for everyone to look at and judge for themselves. Of course not everything on the web is legitimate, and sorting through the anecdotal and the fraudulent can be daunting, but there are a couple of excellent rules of thumb for evaluating the reliability of information found on the internet: if it comes from a private individual’s website, it is to be discounted; if it comes from the Mayo Clinic (or another esteemed institution) or a respected peer reviewed journal—especially if many legitimate studies have come to the same conclusion—then one’s confidence is increased greatly.
So in summary, what we have in these three examples is the “ugliness” of Lamarckism, which represents bad science being promoted by a corrupt establishment; the “goodness” of continental drift, which is an example of excellent science and the struggles involved in developing and testing new theories; and the “badness” of the vaccine-autism debate, which is a lesson in bad science being promoted by a free press and an otherwise educated public’s irrational reaction.
This got me to thinking about “rules of evidence.” And how people often do not think about why they believe something to be true.
And there ARE rules of evidence that are generally accepted by all scientists, or at least all reputable scientists. This is a good thing because there are certain values or qualities that scientists around the world SHOULD adhere to because they undergird all good science. As posted by the global “InterAcademy Panel: The Global Network of Science Academies” on Oct. 17, 2012, these are:
“honesty, fairness, objectivity, reliability, skepticism, accountability, and openness.”
Sound familiar? These are, actually, values that ALL people should adhere to. Not just scientists. But science has another mechanism that is a “check” on their work that the public may not be aware of and that most people are not subjected to. And this mechanism is that fellow scientists will always eventually get around to checking each other’s work and attempt to duplicate it. And publish the results.
Repeatability is central to the entire scientific enterprise. And in each of the three stories told here, scientific theories have either failed or survived based on this one overriding factor.
No comments:
Post a Comment