Skepticism About Homeopathy

Homeopathy is one of the most misunderstood things filed under “natural medicine.” It looks and sounds medical — little bottles, Latin names, dosing schedules, practitioners who take a careful history — and it sits on pharmacy shelves right beside remedies that do work. But it rests on two ideas that collide head-on with chemistry, physics and biology, and after more than 200 years the evidence has landed in one place. This article walks through what homeopathy actually is, why its core mechanism can’t work as advertised, what the major scientific reviews found, and — the part that matters most — where the genuine harm lies.

First, a word on what “skepticism” means here, because it’s often heard as an insult. Skepticism is a process, not a position. It’s applying critical thinking to work out whether a claim holds up — following the evidence to a conclusion, rather than hunting for evidence to justify a conclusion you already hold. A skeptic can end up believing almost anything, provided the reasoning and evidence support it. So this isn’t an attack on the people who use or practise homeopathy. It’s an honest look at whether the modality does what it claims.

What Is Homeopathy?

Homeopathy is a system of medicine that treats a condition using extremely diluted preparations of various substances. Two principles hold it together.

The first is “like cures like.” The idea is that if a substance causes a certain reaction in a healthy person, then the same substance — hyper-diluted — will produce the opposite effect in a sick one. Coffee, a plant famous for causing wakefulness, is used in homeopathic form as a sleep remedy.

This flatly contradicts the foundations of herbal medicine (traditional and modern) as well as conventional pharmacology. Those systems use substances that actually exert an effect to counter a disease process — a sedative herb to calm, a stimulant to rouse. “Like cures like” is a philosophical analogy, not a demonstrated pharmacological law.

The second principle is the “rule of infinitesimals”: the more diluted a preparation is, the more potent it’s said to become. Many homeopathic remedies are diluted so far that not a single molecule — not one atom — of the original substance remains. Between dilutions the preparation is “potentized” by vigorous shaking (succussion), sometimes by tapping it against a leather pad, sometimes by machine. Manufacturers advertise the extreme potency their potentizing machines can reach.

Because there’s nothing left of the starting material, homeopathy leans on the idea that water carries a “memory” of the substance it once touched — an essence that survives dilution and can be transferred onto a sugar pill as the water evaporates off. That claim is the crux of the whole system, and it’s where the science turns unforgiving.

The Dilution Problem

Dilutions are written with an X or a C. X is a 1-in-10 dilution per step; C is 1-in-100. A 6X remedy has been diluted 1:10, six times over — a final ratio of 1 part substance to 1,000,000 parts water.

To swallow a single drop’s worth of the original substance from a 6X remedy, you’d need to drink around a million drops — roughly 33 litres. A human stomach holds about one litre, so you physically cannot consume even one drop of the source substance at any one time. And 6X is a mild dilution.

The C dilutions are where it stops making sense. 30C — an everyday homeopathic potency — is a dilution of 1 part in 10⁶⁰. For scale, there are an estimated 10⁸⁰ atoms in the entire observable universe. It gets more extreme still: 200C preparations exist, and the well-known flu remedy Oscillococcinum is one of them.

There’s a hard chemical wall here called Avogadro’s number (~6.022 × 10²³). Past a dilution of roughly 1 in 10²⁴ — which happens at about 12C — you would, on average, expect fewer than one molecule of the original substance to remain. Everything beyond that point is, statistically, pure diluent. A 30C remedy is astronomically past that wall.

Samuel Hahnemann devised homeopathy in the 1790s, before modern chemistry, atomic theory and Avogadro’s work existed. He could not have known that his most-diluted preparations contained nothing of the original substance at all. That’s forgivable for 1796. What’s harder to defend is holding to the same idea two centuries after we learned better.

Does Homeopathy Actually Work?

This is the only question that ultimately matters, and it has been studied a great deal. Set mechanism aside for a moment — if the remedies reliably helped people, the mechanism would be a fascinating puzzle worth solving. So what does the evidence show?

The single most-cited study is a 2005 meta-analysis in The Lancet. Shang and colleagues matched 110 placebo-controlled homeopathy trials against 110 conventional-medicine trials. When they restricted the analysis to the larger, higher-quality trials — the ones least prone to bias — homeopathy’s apparent benefit vanished, while conventional medicine’s benefit held. Their conclusion was that the findings are “compatible with the notion that the clinical effects of homoeopathy are placebo effects” [1]. The same issue carried an unusually blunt editorial titled “The end of homoeopathy,” urging doctors to be honest with patients about the lack of benefit [2].

That was not the end of the matter — it was the beginning of a run of national and supranational reviews, and they all point the same way:

• Australia’s NHMRC (2015) ran an overview of systematic reviews spanning 225 studies across 61 conditions. Its verdict: there are no health conditions for which there is reliable evidence that homeopathy is effective, and it should not be used to treat conditions that are chronic, serious, or could become serious [3].
• The European Academies’ Science Advisory Council (2017) concluded there is no robust, reproducible evidence that homeopathy works for any condition, called the proposed mechanisms scientifically implausible, and flagged that harm can come from delayed treatment and from unproven claims on labels [4].
• The UK House of Commons Science and Technology Committee (2010) found homeopathy performs no better than placebo and recommended the NHS stop funding it [5]. NHS England duly moved to end routine prescribing of homeopathy in 2017, a decision later upheld by the High Court [6].

What about the specific, popular products? Oscillococcinum — a 200C duck-liver preparation sold worldwide for flu — got its own Cochrane review in 2015. The conclusion: “there is insufficient good evidence to enable robust conclusions to be made” about it for preventing or treating flu; the handful of supporting studies were low quality [7].

To be fair to the strongest counter-argument: a 2014 meta-analysis by Mathie and colleagues — several of whom worked in homeopathy research — pooled 22 trials of individualised homeopathy and found a small, statistically significant effect (odds ratio 1.53). It’s the paper advocates most often cite. But the authors themselves put the crucial caveat right in the abstract: only 3 of the 22 trials were rated as reliable evidence, the overall quality was “low or unclear,” and they called for new high-quality trials before drawing conclusions [8]. A small signal built almost entirely from low-quality trials is exactly the pattern Shang 2005 showed evaporates once you filter for rigour.

So the honest summary is this: the best-case evidence for homeopathy is a weak, low-quality signal its own authors won’t stand behind, and the large high-quality reviews consistently find nothing beyond placebo.

A note on why the literature is so full of “positive” studies: many are small, poorly blinded, or designed in ways that favour the hoped-for result. That’s not a homeopathy-specific accusation — conventional medicine has produced biased, badly designed studies too (the 1960s sugar-industry-funded research that deflected blame for heart disease from sugar onto fat is a notorious example [9]). The difference is what happens next. Peer review and replication are meant to catch those flaws, and when a whole field’s positive results only survive in the low-quality tier, that is the finding.

Homeopaths often counter that their patients feel better, and many genuinely do. But feeling better after a treatment isn’t evidence the treatment caused it. The illness may have run its course; the unhurried, attentive consultation may itself help; and the placebo effect is real and powerful. Untangling those is exactly what a placebo-controlled trial is for — and when you run that trial properly, the remedy adds nothing.

How Is It Supposed to Work?

Homeopathy’s own explanation runs like this: a hyper-diluted substance triggers a tiny reaction in the body, and the body’s response overshoots the trigger, producing the opposite effect. In the coffee example, the remedy causes a minuscule “wakefulness” nudge, and the body’s rebound tips it into sleepiness.

There are two deep problems with this.

The closest real phenomenon this resembles is hormesis — a documented, biphasic dose-response where a low dose of something can stimulate a response that a high dose suppresses. Homeopathy advocates sometimes invoke it to borrow scientific respectability. But hormesis requires the substance to be physically present at a low-but-real dose, with an actual dose-response relationship. Homeopathic high potencies contain no molecules of the original substance, so there is nothing there to produce a hormetic effect. Treating hormesis as validation of ultra-dilution is a category error.

Homeopathy is also frequently compared to vaccination — a small exposure priming a bigger response. That comparison fails for the same reason: vaccines contain real, measurable antigens and produce a well-understood, reproducible immune response. Homeopathic remedies contain nothing measurable and produce no such response. Homeopathy cannot logically — or safely — be equated with vaccination.

The Quantum Mechanics Detour

When pressed on mechanism, some practitioners reach for quantum physics — the claim that “quantum” effects let shaken water retain information about a substance it no longer contains. Terms like quantum entanglement, coherence domains and water memory get deployed as explanations.

To check whether there was anything to this, the original version of this article consulted a physicist at the University of Alberta. The short answer: quantum mechanics operates at scales far too small to govern the behaviour of a glass of water at body temperature, where classical thermodynamics rules. Quantum entanglement does mean particles can be correlated by past interaction — but nothing about that is special or medically relevant, and it doesn’t grant water a durable, therapeutic “memory” of a dissolved compound.

The EASAC review put the same point formally: the proposed mechanisms are inconsistent with established science on dose-response and drug-receptor interaction [4]. As the physicist Sam Harris has observed of this move, the fact that quantum mechanics is strange and hard to understand, and a claim is also strange and hard to understand, does not make the two mutually supporting. Borrowing the vocabulary of physics is not the same as having physics on your side.

Why Is Homeopathy So Popular?

If it doesn’t work, why has it lasted? The history helps. Hahnemann invented homeopathy when mainstream medicine still meant bloodletting, mercury and purging — treatments that routinely did more harm than good. Against that, remedies that were merely inert looked safe, and often kinder. For a century, homeopathic schools spread across North America.

As chemistry, physics and biology matured, homeopathy stopped making sense, and the practice declined — schools closed or absorbed other modalities. It survives today as a niche, and much of its modern appeal is a mirror image of dissatisfaction with conventional medicine.

Is It Harmless?

This is where skepticism stops being an academic exercise. The remedies themselves are usually just water and sugar, so it’s tempting to shrug — where’s the harm? But the harm is real, and it’s mostly indirect.

Partly in response to cases like the teething tablets, the FDA changed course in 2019, withdrawing the decades-old policy that had let homeopathic products be marketed with little oversight and adopting a risk-based approach that prioritises action against higher-risk products — those aimed at children, at serious conditions, or containing ingredients of safety concern [11].

Concluding Remarks

Homeopathy asks us to accept two things that the last two centuries of science firmly contradict: that “like cures like,” and that a substance grows more potent the more it’s diluted — even past the point where none of it remains. When those claims are put to the proper test, the remedies perform no better than placebo, and the largest, most careful reviews agree.

None of this is a strike at the people who use or practise homeopathy. Many are drawn to it for understandable reasons — a bad experience with conventional care, a wish for gentler medicine, a practitioner who actually listens. Those needs are real and worth taking seriously. But the honest conclusion, after all the evidence is in, is that homeopathy’s benefits are the benefits of placebo and attention, not of the remedy — and that its most important risk is standing between someone and a treatment that would have helped.