It is regularly argued that homeopathy works by the placebo effect, the claim being that the remedies have no active ingredient and so the patient gets better by self-delusion or some other inexplicable reason. There are several important points that are overlooked in this assertion:
It cannot explain provings, where specific reactions are recorded which have no relation to the subjects of the trial feeling better – indeed they usually feel worse.
It cannot explain why different remedies produce different reactions in the same person, since the reaction should be the same for all remedies if it is solely a placebo effect. For example, it does not explain why a person immediately gets better from a burn after the correct remedy, but does not get better at all more quickly after the wrong remedy. 
It cannot explain why a person may have different reactions at different times after taking the same remedy.
It cannot explain the differences in reactions to different potencies, which can range from no reaction through beneficial reaction to unpleasant reactions. 
It cannot explain why the two different types of potency (centesimal and decimal on the one hand, and LM on the other) need to be given in significantly different ways, because they generally produce significantly different patterns of reaction. 
It cannot explain the fact that the patient’s symptoms may change rather than simply get better. In particular it cannot explain why patients sometimes get worse as the immediate reaction before they get better.
It cannot explain why homeopathy works on babies, toddlers and animals, since they can have no preconception of the treatment being beneficial, and so are not disposed to a placebo effect. In this case the claim is made that the benefit is only perceived by the practitioner, the parent or the owner of the animal, without there being a real improvement. Such a claim ignores the fact that the placebo effect is not a self-deception but a real modification of symptoms.
It cannot explain why homeopathy works on people who are unconscious.
It cannot explain the action of potentised remedies on tissue samples, plants or seedlings. [3,4,5]
Orthodox medicine cannot distinguish between a patient getting better after a ‘placebo’ and one getting better after an ‘active’ treatment. In other words, both treatments are considered equally successful, and therefore both should be considered equally ‘active’. The fact that the method of a so-called placebo’s action is unknown should be a reason for investigating that action further, not for denying its validity.
Critics of homeopathy maintain that the degree of dilution involved in making remedies means that they cannot be ‘active’ treatments, but must work by the placebo effect. However, this is a narrow view of treatment based on an approach entirely oriented on chemistry, and it ignores the application of physics to questions of effects on living organisms. It cannot be a coincidence that this narrow view is consistent with the pursuit of a phamacological approach based on chemical drugs.
Critics of homeopathy also maintain that potentised remedies cannot work because homeopaths cannot explain how potentisation works. However, they are quite happy to allege that any success is entirely due to the placebo effect without themselves being able to explain how a placebo works.
1. James Tyler Kent, New Remedies, Clinical Cases, Lesser Writings, Aphorisms and Precepts (New Delhi: B. Jain Publishers, repr. edn 2001), pp. 357-59.
2. Robert Barker, LM Potencies: A selection of articles giving a practical guide to their use (Sheringham: The Homoeopathic Supply Company, 1997), pp. 8 and 13-14.
3. See, for example, Belon, P, J Cumps, M Ennis, P F Mannaioni, M Roberfroid, J Sainte-Laudy, F A C Wiegant, ‘Histamine dilutions modulate basophil activation’, Inflamm. Res., 53 (2004), 181-188.
4. L Betti, L Lazzarato, G Trebbi, M Brizzi, G L Calzoni, F Borghini and D Nani, 'Effects of homeopathic arsenic on tobacco plant resistance to tobacco mosaic virus. Theoretical suggestions about system variability, based on a large experimental data set', Homeopathy, 92 (2003), 195-202.
5. M Brizzi, L Lazzarato, D Nani, F Borghini, M Peruzzi and L Bettib, 'A Biostatistical Insight into the As2O3 High Dilution Effects on the Rate and Variability of Wheat Seedling Growth', Forsch Komplementärmed Klass Naturheilkd, 12 (2005) 277-283.
Why it works