This article is from the Scientific Skepticism FAQ, by Paul Johnson Paul@treetop.demon.co.uk with numerous contributions by others.
Homeopathy is sometimes confused with herbalism. A herbalist
prescribes herbs with known medicinal effects. Two well known
examples are foxglove flowers (which contain digitalin) and willow
bark (which contains aspirin). Folk remedies are now being studied
extensively in order to winnow the wheat from the chaff.
Homeopathists believe that if a drug produces symptoms similar to
certain disease then a highly diluted form of the same drug will cure
the disease. The greater the dilution, the stronger this curative
effect will be (this is known as the law of Arndt-Schulz). Great
importance is also attached to the way in which the diluted solution
is shaken during the dilution.
People are skeptical about homeopathy because:
1: There is no known mechanism by which it can work. Many homeopathic
treatments are so diluted that not one molecule of the original
substance is contained in the final dose.
2: The indicator symptoms are highly subjective. Some substances have
hundreds of trivial indicators.
3: Almost no clinical tests have been done.
4: It is not clear why trace impurities in the dilutants are not also
fortified by the dilution mechanism.
Although homeopathy involves little more than doing nothing, it was
invented in the days when doing nothing was usually better for the
patient than conventional treatment. It therefore represented a
significant advance in medical practice. Since then conventional
medicine has improved beyond recognition, while homeopathy is still
equivalent to doing nothing.
Reports of one scientific trial that seemed to provide evidence for
homeopathy until a double-blind trial was set up can be found in
Nature vol 333, p.816 and further, and the few issues of Nature
following that, about until November of that year (1988).
SI ran a good article on the origins and claims of homeopathy:
Stephen Barrett, M.D., "Homeopathy: Is It Medicine?", SI,
vol. 12, no. 1, Fall 1987, pp. 56-62.
 
Continue to: