Homeopathy: Avogadro, Placebo, Proponent Arguments and Herbal Medicine

What is homeopathy?

Homeopathy is an alternative medical system developed by Samuel Hahnemann at the end of the eighteenth century. Its name derives from the Greek words hómoios (“similar”) and páthos (“suffering”). Homeopathy is not a general term for gentle medicine and should not be confused with herbal medicine or naturopathy.

The system rests chiefly on two connected claims:

The principle of similars: a substance that produces certain symptoms in a healthy person is supposed to treat an illness with similar symptoms in homeopathic form—“like cures like.”
Potentisation: the selected source substance is repeatedly diluted and shaken or triturated. Homeopathic doctrine claims that this process increases rather than decreases its effect, even when no molecule of the source substance is expected to remain.

Remedy selection is based on so-called drug provings in healthy people. Observations and subjective symptoms reported after taking a substance are assembled into a “remedy picture.” Practitioners compare that picture with a detailed case history and select the closest match. Two people with the same medical diagnosis may therefore receive different homeopathic products.

Hahnemann presented the similarity principle as a postulate in 1796 and later developed the system in the Organon of Medicine. His observations predated modern chemistry, molecular biology, pharmacology, randomised trials and statistical analysis. They were not blinded, placebo-controlled experiments and did not demonstrate that similarity causes healing or that dilution and shaking generate hidden medicinal powers.

Scientific assessment: the two foundational claims conflict with central findings of chemistry and pharmacology. Increasing dilution does not increase a substance’s effect without a demonstrable mechanism. Homeopathic preparations have not shown a reliable specific therapeutic benefit beyond placebo, contextual and natural-course effects. Homeopathy is therefore classified scientifically as a pseudoscientific treatment system.

The following sections examine these claims step by step: how globules are made, what Avogadro’s limit means, whether shaking can transmit information, how proponents argue, and how placebo effects, risks and herbal medicine differ from homeopathy.

How globules are made

Homeopathic production begins with a mother tincture or another starting preparation. In a decimal D dilution, one part is mixed with nine parts solvent. In a centesimal C dilution, one part is mixed with 99 parts solvent. Each stage is followed by a prescribed shaking procedure known as succussion.

Globules are small pellets made mainly from sucrose, lactose or similar excipients. They are moistened or coated with the liquid preparation and then dried. The finished pellet therefore consists mainly of sugar or another carrier.

What does the Avogadro constant mean?

A mole is a counting unit for atoms, molecules, ions or other entities—rather like a dozen, but enormously larger. One mole contains exactly 6.02214076 × 10²³ entities. This number is the Avogadro constant.

The constant is not a mysterious chemical wall. It converts an amount of substance into a number of particles. In an idealised example starting with one mole, a dilution of 10⁻²⁴ leaves an expected average of about 0.6 source molecules in the same reference amount.

An average of 0.6 does not mean that a vial contains a fraction of a molecule. Molecules are distributed randomly. At an expected value of 0.6, about 55% of samples contain no source molecule, around 33% contain one, and the rest contain two or more. One further D step reduces the expectation to 0.06, so about 94% contain none. At C13, the idealised probability of no source molecule is about 99.4%.

D potencies: 1 : 10 per stage

Potency	Total dilution	Expected molecules*	Probability of 0 molecules*	Interpretation
D1	10⁻¹	about 6 × 10²²	effectively 0%	very many molecules
D12	10⁻¹²	about 6 × 10¹¹	effectively 0%	very many remain
D23	10⁻²³	about 6 on average	about 0.25%	only a few molecules
D24	10⁻²⁴	about 0.6 on average	about 55%	Avogadro region reached
D25	10⁻²⁵	about 0.06 on average	about 94%	nearly all samples contain none
D30	10⁻³⁰	about 6 × 10⁻⁷	more than 99.9999%	effectively certain to contain none

C potencies: 1 : 100 per stage

Potency	Total dilution	Expected molecules*	Probability of 0 molecules*	Interpretation
C1	10⁻²	about 6 × 10²¹	effectively 0%	very many molecules
C6	10⁻¹²	about 6 × 10¹¹	effectively 0%	very many remain
C11	10⁻²²	about 60 on average	effectively 0%	source material may remain measurable
C12	10⁻²⁴	about 0.6 on average	about 55%	Avogadro region reached
C13	10⁻²⁶	about 0.006 on average	about 99.4%	almost all samples contain none
C30	10⁻⁶⁰	about 6 × 10⁻³⁷	effectively 100%	astronomically below one molecule

*Idealised calculation starting with one mole in the same reference amount and random molecular distribution. The exact point in a real product depends on starting concentration, volume and manufacturing method.

Beyond Avogadro’s limit does not mean that every vial becomes empty at one sharp step. It means that the expected count falls below one and, with each further dilution, the probability of finding no source molecule rapidly approaches 100%.

Can shaking transfer “information”?

Shaking can mix a liquid, form bubbles, dissolve gases and introduce traces from the vessel or environment. What has not been demonstrated is a mechanism by which it stores the identity of a vanished substance as stable, specific and biologically readable information.

Such a claim requires a physical storage medium, a code distinguishing different source substances, transfer of that code to a dry sugar pellet, long-term stability, and a biological reader. None of these steps has been established experimentally. Liquid water has a rapidly fluctuating hydrogen-bond network; spectroscopic work shows that persistent structural correlations are lost within roughly 50 femtoseconds. This is not a plausible medium for a substance-specific message that must survive repeated dilution, shaking, drying, storage and digestion.

Could unknown “water information” exist in principle?

Science can discover surprising phenomena. But a proposed effect must be measurable, reproducible and predictive. Blinded samples should be distinguishable; different starting substances should generate different physical signatures; the signature should survive vessel changes, drying and storage; and its biological effect should remain when expectation and treatment context are controlled.

No such reproducible chain has been demonstrated. If preparations cannot be distinguished chemically or physically but allegedly produce different effects, the measurable causal link is missing. If dilution increases the effect while the material carrier disappears, a testable mechanism of encoding and amplification is required.

What did Hahnemann actually establish?

Samuel Hahnemann developed homeopathy before modern molecular chemistry, pharmacology, randomised trials and statistical clinical evaluation were established. He reported personal experiences and drug “provings”, but these were not randomised, blinded, placebo-controlled or statistically analysed.

In the Organon of Medicine, Hahnemann asserted that rubbing and shaking developed hidden medicinal powers and transformed material substance into a “spirit-like” essence. This was a speculative interpretation within his vital-force model, not a physical demonstration of information storage.

The selectivity problem

But suppose for a moment that the physically implausible “impossible” were possible: water really could store substance-specific information through shaking, and further dilution made it stronger.

Real water is never exclusively H₂O. It contains traces of dissolved gases, minerals, ions and organic material, together with traces from glass, closures, air, cleaning and previous processing. If tiny traces are potentised, all unintended traces should be potentised as well. The result would be an uncontrolled mixture of information.

If only the intended substance is stored, a selector is required that distinguishes the manufacturer’s intention from accidental contamination. Matter does not know human intention, and no such selector has been demonstrated.

Why do people report improvement?

Improvement after taking globules certainly does not prove that the globules caused it. “After this” does not mean “because of this”. Symptoms may change through natural recovery, regression to the mean, concurrent effective treatment, expectation, attention, ritual, behavioural change and selective memory.

Placebo is therefore not the only explanation for observed improvement; it belongs to a broader group of contextual and natural-course effects. The key conclusion is that a specific medicinal effect of highly diluted homeopathic preparations beyond placebo, contextual and natural-course effects has not been reliably demonstrated.

How do proponents argue?

It would be inaccurate to say that all proponents reject randomised or double-blind trials. Homeopathy research organisations collect such trials and cite selected positive studies and meta-analyses. Common arguments include:

Individualisation: the remedy is selected from the whole symptom pattern, so a fixed remedy for everyone allegedly does not represent classical homeopathy.
Whole treatment package: the consultation, relationship and personalised prescription are said to be inseparable parts of the treatment.
Positive studies and “plausibility bias”: proponents argue that critics discount favourable results because the mechanism appears physically implausible.
Children, animals and laboratory experiments: improvement in subjects who supposedly lack conscious expectations is presented as evidence against placebo explanations.

Why double-blinding remains possible

Individualisation and double-blinding are compatible. A homeopath can take the case and select the individual remedy. An independent pharmacy can then randomise the patient to that remedy or an indistinguishable placebo, while patient, practitioner and assessor remain blinded. Such trials have been performed.

Pragmatic trials can test whether the entire package of consultation, ritual and prescription is useful in real life. But only a placebo-controlled comparison can isolate a specific effect of the globule. If every negative result is explained after the event by a wrong remedy, insufficient individualisation, short treatment or unsuitable outcome, the claim becomes practically unfalsifiable.

Blinding can be difficult for surgery or physical therapy. For visually identical globules, it is comparatively straightforward and particularly appropriate.

“But it works in animals and children”

This argument uses an overly narrow idea of placebo. A placebo response includes more than conscious belief by the patient. Natural course, regression to the mean, additional care, changed caregiver behaviour, selective observation and subjective assessment can all produce apparent improvement.

In dogs with osteoarthritis, owners and veterinarians frequently reported improvement during placebo treatment while objective force-platform measurements did not show a corresponding change. Children can also develop expectation- and conditioning-based placebo responses, and parental expectations can influence reports and behaviour.

An animal or infant not understanding homeopathic theory is therefore not proof of efficacy. In patients unable to report reliably, randomised blinded studies with objective outcomes are even more important.

What do controlled studies show?

Controlled trials aim to distribute expectation, attention, natural improvement and other influences equally. Major governmental and scientific assessments have not found reliable evidence that homeopathy treats a specific health condition beyond placebo. Personal experiences need not be invented, but they cannot establish a specific effect of the globule.

Herbal medicine is not homeopathy

Homeopathy, herbal medicine and naturopathy are often confused. Phytotherapy uses plants or extracts containing measurable active substances. These can be absorbed, measured, studied in dose-response relationships and overdosed. High-potency homeopathy instead relies on similarity and potentisation, often after the source material has disappeared.

Natural substances are neither automatically effective nor automatically safe. Each preparation and indication must be evaluated:

Example	What actually acts?	Evidence and limitations
Atropine from deadly nightshade	A defined alkaloid blocks muscarinic acetylcholine receptors.	Atropine is an effective medicine, including for symptomatic bradycardia and selected emergency uses. The plant itself is poisonous; benefit and harm depend on dose.
Peppermint oil	Measurable essential-oil constituents affect smooth muscle and sensory nerves.	Standardised preparations have recognised use for minor abdominal spasms, flatulence and pain, especially in irritable bowel syndrome. Adverse effects and contraindications occur.
Yarrow	Plant constituents with possible antispasmodic and anti-inflammatory actions.	Traditional use includes mild menstrual cramps and digestive complaints. EMA classifies this as traditional use because sufficient clinical trials are lacking. Traditional use is not the same as proof.
Ispaghula husk	Water-binding fibre increases stool bulk and softness.	It can treat ordinary constipation and help soften stool. Adequate fluid and spacing from other medicines are important.
St John’s wort	Pharmacologically active plant compounds affect enzymes and transport proteins.	Some evidence suggests benefit in mild or moderate depression, but results are not fully consistent. It can dangerously weaken or interact with many medicines.

The crucial difference: in herbal medicine, constituents, dose, absorption, effects, adverse effects and interactions can be measured. That is ordinary pharmacology, regardless of whether a molecule comes from a plant, fungus, microorganism or laboratory.

Is homeopathy dangerous?

The pellet itself

A typical high-potency globule consists mainly of sugar or lactose and usually contains no pharmacologically relevant amount of the named source material. A spoonful would therefore mostly be a dose of sugar, not a meaningful scientific test. People with diabetes, intolerances or relevant metabolic disorders must consider the excipients.

Not every product labelled homeopathic is highly diluted. Low potencies, mother tinctures and liquid products can contain measurable active ingredients or alcohol and can therefore cause direct adverse effects, poisoning or interactions.

The larger danger: effective treatment is replaced

The main risk is often lost time. Harm occurs when necessary diagnosis, surgery, vaccination or effective medication is delayed, stopped or replaced. This can be serious in severe infections, cancer, acute cardiovascular disease, acute abdominal conditions and other progressive or life-threatening illnesses.

Published case reports document both direct harm from some products and indirect harm when effective treatment was replaced by homeopathy. The risk is especially serious when worsening symptoms are reinterpreted as a desired reaction or when patients are told that evidence-based medicine cannot help them.

Why the German term “Schulmedizin” is problematic

The German word Schulmedizin is now often used neutrally, but it has a polemical history. It can suggest a rigid doctrine—merely one school among several. Modern medicine is not a closed belief system: its recommendations are expected to change when better evidence appears. More accurate German terms are wissenschaftsbasierte Medizin and evidenzbasierte Medizin. Their defining feature is not infallibility but systematic testing and correction.

Conclusion

The Avogadro constant is a counting relationship: one mole contains exactly 6.02214076 × 10²³ entities. In the idealised model, the expected source-molecule count falls below one at D24 or C12. One step later, nearly all samples contain none.

No demonstrated storage medium, code, selective transfer or biological reader supports the proposed “information”. People, children and animals may improve or appear to improve after homeopathic treatment, but this does not establish causation. A specific therapeutic effect beyond placebo, context, observation and natural course has not been reliably demonstrated.

Homeopathy