Entry 899, on 2008-11-27 at 20:20:12 (Rating 3, Science)
Earlier this year (2008-03-16) in a blog entry titled "Quirky and Brilliant" I discussed the famous theoretical physicist, Richard Feynman. Currently I'm listening to a second audio book about this demented genius who made major contributions to quantum physics (including discoveries relating to quantum electrodynamics which he received a Nobel Prize for in 1965). This time I'm going to concentrate more on his philosophical ideas (even though he had a poor opinion of philosophy) and some of his controversial opinions about what he calls "cargo cult science".
So what is a cargo cult? The term refers to the beliefs and behaviours of some of the natives on Pacific Islands (especially) after the Americans visited during World War II. When the Americans arrived in their planes they brought a time of plenty for the islanders which ended once the war was over. Since then, the islanders have tried to encourage the "gods" to return by imitating what they saw when the Americans were there. They build runways and false planes out of bamboo. They had a villager sitting in a box built like a control tower with headphones made from coconuts on his head which they reasoned should encourage the gods to return.
Its both funny and sad that these people should waste their time this way (actually, it might not be a total waste of time because I'm sure some social anthropologists have returned to study the interesting phenomenon of this cult). The point Feynman makes is that he sees a lot of areas in modern society as not being that much different from a cargo cult. Groups of people involved in these areas create the superficial appearance of doing real research or seeking real truth but they are really just fooling themselves, just like the cargo cult does.
So, OK, I haven't mentioned what these areas of knowledge are so let's cut to the chase and list them. Feynman seemed to have a particular dislike for educational experts, philosophers, politicians, some psychologists, and paranormal investigators. He also implied that some people who claim to have a scientific religious view indulge in the same false logic. I would add to this most business people, many economists, and a certain percentage of people from every other area as well, including some scientists!
Feynman died in 1988 so the ideas I'm discussing here are several decades old, but it seems to me that they are more true now rather than less. He readily admitted that the real scientific method is difficult - perhaps even impossible - to apply to social sciences and other more amorphous areas, but I think the criticism still stands.
So what are some examples of where people make errors in their studies? I will take paranormal investigation as an example because that area should obviously be treated as a science because its clearly open to logical and empirical investigation.
What are some of the attributes of experiments in the paranormal? Well a lot of paranormal information comes from anecdotes rather than real controlled studies for a start. Secondly, the experiments that are done tend to be very variable in their results: the same experiment repeated by another researcher will get a different result from the original experiment. And lastly - and most importantly - as the quality and thoroughness of the experiment increases the observed anomalous effects tend to get less to the point of being indistinguishable from background "noise".
In these situations a researcher in almost any other field would have to conclude that the effect under study most likely doesn't exist. There would be the possibility of altering the experiments to make them more sensitive or more accurate but the initial conclusion would be that ESP, psi, spirits, or whatever other phenomenon is being tested isn't real.
But this doesn't tend to be what actually happens. Feynman quotes the example of a leading paranormal experimenter who encourages a teaching institution to only concentrate on the students who can get positive results. If we took this approach to other areas of knowledge we could prove any crazy belief. In real science negative results are just as important (maybe more important) than positive. At the very least both positive and negative results need to be considered or nothing is really gained.
But its worse than that. According to Feynman, the same person suggested that it was unreasonable to require an experiment in the paranormal to be repeatable. In other words, if one person or group got a positive result we shouldn't have someone else repeat the experiment to test for errors in methodology, bias, weird statistical phenomena, etc. Why not? Surely every researcher should be ready to admit that errors are possible. There are certainly experiments in the past which have produced incorrect results because of incredibly subtle errors in methodology. In an area where the effects are so slight and subconscious human effects are so prominent, careful repetition is essential.
When he talked about science Feynman made several observations. First he said its important not to fool yourself because you are the easiest person to fool. A researcher who really wants to see a particular result can easily make that happen without really realising he is even doing it. Its also important to have scientific integrity and not try to fool other researchers, so the experiment should be carefully described and potential sources of errors should be admitted. Also its important not to fool the layman when discussing the outcomes of the research with the public.
There are few people who would say that fooling yourself, your colleagues, or your fellow citizens is a good idea, so why are so many people hesitant to follow protocols which will ensure this doesn't happen? There are many reasons. Some researchers work in an area they have a special emotional attachment to - they might really want spirits to exist for example - so the danger of bias in their research is high. Others might rely on positive results for funding or the recognition of their peers. Others might feel the need to maintain the status quo and feel insecure about rejecting the existing zeitgeist.
Clearly social sciences and the paranormal are far more prone to these sorts of errors than the physical sciences. First, there is more likely to be an emotional aspect to the researcher's interest in the social area; and second, subtle social and subject-experimenter effects tend to happen a lot more when studying another human than when studying (for example) a proton!
So the paranormal and social sciences often say that the scientific method is inappropriate to their area but the reality is that the truth seems to be more the opposite: they need it even more. Feynman knew this. Maybe more people should listen to his lectures!
There are no comments for this entry.
You can leave comments about this using this form.
Enter name, email (optional), enter number, comment, click Add. You can leave the name blank if you want to remain anonymous. Enter your email address to receive notifications of replies. Comment should appear immediately (authorisation is inactive).
BLOG PAGE