Checking The Facts
My Concise Oxford Dictionary defines "facts" as "truth" and "reality", as a "thing that has known to have occurred, to exist, or to be true". Most people would have no problem with this definition, particularly when it comes to scientific facts. After all, gravity is gravity. Oxygen is oxygen. And that is all there is. But this simplistic perception is dangerously obsolete.
Scientific facts do not exist in "reality" as absolutes simply waiting to be "discovered". Facts become facts not just through the procedures of science but also through a social process; and every fact has a social history associated with it. The really interesting fact is that scientific facts come in a variety of forms -- as false facts, changing facts, sliding facts and contested facts.
A common false fact is oxygen. Every student learns that Lavoisier, the French chemist executed in the Revolution, discovered oxygen. But the name actually refers to a nonexistent element whose meaning comes from a false theory of chemical composition. For "oxy" means "sharp-maker"; its best translation is in German, Sauerstoff -- sour stuff. Lavoisier believed that this gas, when supporting the combustion of elements, made them into acids. This worked for sulphur and phosphorus. But Humphrey Davy showed that it did not work for "muriatic acid" -- what we now call hydrochloric acid. Its active constituent, chlorine, had no oxygen in it. So Lavoisier's theory was refuted, and oxygen deprived of its meaning, just two decades after its discovery. But try to find this in a chemistry textbook.
Textbooks themselves are a great place to find out how much the "facts" change. In some subjects, particularly chemistry, what is given to students as facts at one level is rejected as simplistic at the next. In other subjects, a given fact changes from one theoretical structure to another. For example, the "fact" that light behaves as a wave is conveniently forgotten in photovoltaic effects where it is treated as consisting of photons; but the "fact" that it behaves as a particle is ignored in explaining the phenomenon of Newton's rings. So the nature of calculation determines which "fact" is regarded as "fact". Even the facts that are by definition "constant" can change. Hubble's constant, which is the yardstick for measuring the size and age of the universe, has been bungee-jumping for decades.
Sometimes well-known facts become false in retrospect, thanks to the advances of science. A wonderful recent example is the total reorganisation of the classification of plants, thanks to the use of DNA techniques. All sorts of plants, which had been considered closely related because of superficial similarities, are now discovered to have very different DNA. The botanical map will have to be withdrawn; many of its previously accepted facts are now seen to be false.
Sliding facts appear where science meets public policy. In the 1970s and 1980s, newborn babies were given vitamin K to prevent bleeding. But in the 1990s, research linked vitamin K to risks of cancer in children. Similarly, when first used in the 1960s, breast implants were "safe". But in the 1990s they began to be linked to connective tissue diseases. In both cases, different studies prove different things -- and none of them have "proved" breast implants or vitamin K to be either "safe" or linked to statistically significant "higher risks". The "facts" slide from one position to another.
Given the risks involved, it is important for sliding facts to be pinned down. Indeed, when it comes to risks, all facts are contested facts. For example, since the late 1970s we have been hearing that children living near electric power lines have a higher risk of developing leukaemia. But what are the facts? Well, a number of epidemiological studies appear to confirm the fears -- but none of them have transformed the fear into a fact. The latest study, published in July 1997 in the New England Journal of Medicine, found a 24 per cent increase in risk. But this result is statistically non-significant and researchers have used the study to argue that there is "little evidence" for a risk of leukaemia from electric cables. However, a similar study by the World Health Organisation, which also produced a statistically non-significant result, actually provided "substantial" evidence for a link between passive smoking and lung cancer. So why is the latter generally accepted as a "fact" and the former not given half a chance?
A great deal of science depends on statistical inferences. But no statistical survey or test can ever "prove" a causal relationship between two factors. So results are always stated in terms of levels of "confidence". Different problems are conventionally investigated to different "confidence limits". Now, whether an inquiry is accurate to 95 or 99.5 per cent depends on the values defining the investigations, the costs and weight placed on social, environmental or cultural consequences. In other words, it is social and political power, backed by hard cash, that defines the level of confidence to which a risk problem is investigated.
For instance, when a dangerous chemical plant is placed in an area with an aware and politically active citizenry, the risks are worked out to a high level of confidence. However, when the same plants are located in an area where the citizens themselves are ignorant of the dangers and do not command political power, the confidence levels are much more relaxed. The people of Bhopal and Chernobyl know this to their cost. In the case of passive smoking, no one but the wicked tobacco companies will cast doubt on any "proof' of carcinogenicity. However, there are powerful interests with a lot at stake financially if power lines turn out to be carcinogenic.
So we can't take facts for granted. Neither can we uncritically equate them with "truth", "reality" and "existence". Indeed, we should assume that facts, because they are largely (although not wholly) socially constructed, will change and could be false. I think it is always necessary to ask how probable is a given "fact", how well supported is it by evidence, how well does it withstand competent criticism, how transparent are its promoters in explaining how their results were derived, and how sincere are they about engaging in dialogue with critics. Even then, I will keep my mind open for possible unexpected changes.
By Ziauddin Sardar
Scientific facts do not exist in "reality" as absolutes simply waiting to be "discovered". Facts become facts not just through the procedures of science but also through a social process; and every fact has a social history associated with it. The really interesting fact is that scientific facts come in a variety of forms -- as false facts, changing facts, sliding facts and contested facts.
A common false fact is oxygen. Every student learns that Lavoisier, the French chemist executed in the Revolution, discovered oxygen. But the name actually refers to a nonexistent element whose meaning comes from a false theory of chemical composition. For "oxy" means "sharp-maker"; its best translation is in German, Sauerstoff -- sour stuff. Lavoisier believed that this gas, when supporting the combustion of elements, made them into acids. This worked for sulphur and phosphorus. But Humphrey Davy showed that it did not work for "muriatic acid" -- what we now call hydrochloric acid. Its active constituent, chlorine, had no oxygen in it. So Lavoisier's theory was refuted, and oxygen deprived of its meaning, just two decades after its discovery. But try to find this in a chemistry textbook.
Textbooks themselves are a great place to find out how much the "facts" change. In some subjects, particularly chemistry, what is given to students as facts at one level is rejected as simplistic at the next. In other subjects, a given fact changes from one theoretical structure to another. For example, the "fact" that light behaves as a wave is conveniently forgotten in photovoltaic effects where it is treated as consisting of photons; but the "fact" that it behaves as a particle is ignored in explaining the phenomenon of Newton's rings. So the nature of calculation determines which "fact" is regarded as "fact". Even the facts that are by definition "constant" can change. Hubble's constant, which is the yardstick for measuring the size and age of the universe, has been bungee-jumping for decades.
Sometimes well-known facts become false in retrospect, thanks to the advances of science. A wonderful recent example is the total reorganisation of the classification of plants, thanks to the use of DNA techniques. All sorts of plants, which had been considered closely related because of superficial similarities, are now discovered to have very different DNA. The botanical map will have to be withdrawn; many of its previously accepted facts are now seen to be false.
Sliding facts appear where science meets public policy. In the 1970s and 1980s, newborn babies were given vitamin K to prevent bleeding. But in the 1990s, research linked vitamin K to risks of cancer in children. Similarly, when first used in the 1960s, breast implants were "safe". But in the 1990s they began to be linked to connective tissue diseases. In both cases, different studies prove different things -- and none of them have "proved" breast implants or vitamin K to be either "safe" or linked to statistically significant "higher risks". The "facts" slide from one position to another.
Given the risks involved, it is important for sliding facts to be pinned down. Indeed, when it comes to risks, all facts are contested facts. For example, since the late 1970s we have been hearing that children living near electric power lines have a higher risk of developing leukaemia. But what are the facts? Well, a number of epidemiological studies appear to confirm the fears -- but none of them have transformed the fear into a fact. The latest study, published in July 1997 in the New England Journal of Medicine, found a 24 per cent increase in risk. But this result is statistically non-significant and researchers have used the study to argue that there is "little evidence" for a risk of leukaemia from electric cables. However, a similar study by the World Health Organisation, which also produced a statistically non-significant result, actually provided "substantial" evidence for a link between passive smoking and lung cancer. So why is the latter generally accepted as a "fact" and the former not given half a chance?
A great deal of science depends on statistical inferences. But no statistical survey or test can ever "prove" a causal relationship between two factors. So results are always stated in terms of levels of "confidence". Different problems are conventionally investigated to different "confidence limits". Now, whether an inquiry is accurate to 95 or 99.5 per cent depends on the values defining the investigations, the costs and weight placed on social, environmental or cultural consequences. In other words, it is social and political power, backed by hard cash, that defines the level of confidence to which a risk problem is investigated.
For instance, when a dangerous chemical plant is placed in an area with an aware and politically active citizenry, the risks are worked out to a high level of confidence. However, when the same plants are located in an area where the citizens themselves are ignorant of the dangers and do not command political power, the confidence levels are much more relaxed. The people of Bhopal and Chernobyl know this to their cost. In the case of passive smoking, no one but the wicked tobacco companies will cast doubt on any "proof' of carcinogenicity. However, there are powerful interests with a lot at stake financially if power lines turn out to be carcinogenic.
So we can't take facts for granted. Neither can we uncritically equate them with "truth", "reality" and "existence". Indeed, we should assume that facts, because they are largely (although not wholly) socially constructed, will change and could be false. I think it is always necessary to ask how probable is a given "fact", how well supported is it by evidence, how well does it withstand competent criticism, how transparent are its promoters in explaining how their results were derived, and how sincere are they about engaging in dialogue with critics. Even then, I will keep my mind open for possible unexpected changes.
By Ziauddin Sardar
Labels: science
posted by Mary Mezack at 4:16 PM
0 Comments:
Post a Comment
<< Home