Chapter 6 — Garlic as Medicine

Chapter 6 — Garlic as Medicine — Claim and Evidence

"Extraordinary claims require extraordinary evidence." — Carl Sagan, Cosmos (1980)

G arlic is the most mythologised food on earth, and the myth is old. It was pressed on the pyramid labourers of Egypt, prescribed by the physicians of Greece, hung against plague and against the evil eye, and marketed in our own century as a natural antibiotic, a heart tonic, a cold cure, a cancer shield and a general elixir. No other vegetable carries so heavy a freight of promise. The trouble with a claim repeated for four thousand years is that repetition is not evidence, and antiquity is not proof; a belief can be very old and still be wrong. This chapter is the counterpart, in a book of praise, to the honest accounting a naturalist owes any specimen. It asks of each medical claim not "is it traditional?" and not "is it plausible?" but only the harder question — what do the actual trials show, and how strong is the evidence, really?

The answer, given in advance so that nothing here reads as either debunking or salesmanship, is a genuinely mixed one. Garlic is a wholesome food with a few real, modest, measurable effects that good studies support; it is also the object of a vast quantity of overclaim, in which laboratory findings and folk memory are dressed up as cures they have never been shown to be. Telling the two apart is the whole task. What follows takes no position on what any reader should swallow; it reports evidence, names its strength and its weakness, and stops there. This is a chapter about proof, not prescription.

It helps, before weighing the claims, to see why this particular vegetable attracts so many. Garlic is pungent, and the human mind has always half-believed that a food powerful on the tongue must be powerful in the body — the old doctrine of signatures in modern packaging. It is ancient, and antiquity lends authority. It is genuinely bioactive, so there is always some real laboratory finding to point to. And it is cheap, natural and familiar, which makes the promise of a hidden medicinal power especially welcome. Every one of those features recommends garlic to the mythmaker; not one of them is evidence. Keeping the reasons for a belief separate from the reasons to hold it is the whole discipline of this chapter.

The Heart — Where the Evidence Is Real, and Modest

The strongest case for garlic as more than food concerns the cardiovascular system, and it is worth being precise about how strong "strongest" is. Numerous randomised trials have tested garlic preparations — powdered tablets, aged garlic extract, oil — against placebo for their effect on blood pressure, and the trials have been gathered into several meta-analyses. The results converge on a real but limited conclusion: garlic supplements can lower blood pressure, and the effect is most reliable in people who already have hypertension. One 2020 meta-analysis pooling twelve trials in some 553 hypertensive participants found average reductions of about 8.3 mmHg in systolic and 5.5 mmHg in diastolic blood pressure — a reduction, in that group, of a size a physician would not dismiss. Other analyses report smaller figures; a review focused on aged garlic extract across nineteen trials found a systolic reduction of only about 2.5 mmHg. The honest summary is not a single tidy number but a range, and the range itself is the finding: the effect is real, it is modest, it is larger in people with raised pressure than in those without, and it varies considerably from study to study.

There is, at least, a plausible biological reason for the effect, which is more than many folk remedies can offer. Work published in the Proceedings of the National Academy of Sciences in 2007 showed that human red blood cells convert garlic's sulphur compounds into hydrogen sulphide — the same molecule that gives rotten eggs their reek and, in minute quantities, a natural signalling gas that relaxes the walls of blood vessels. In the laboratory, garlic-derived polysulphides relaxed rat arterial tissue in step with the hydrogen sulphide they produced. That gives the blood-pressure finding a mechanism to stand on: not magic, but a real vasodilator chemistry. The 2020 meta-analysis that reported the 8.3 mmHg figure noted more than a bare number, too, finding signs that garlic also improved arterial stiffness and shifted the make-up of the gut microbiome — early observations, but consistent with a genuine vascular action rather than a statistical mirage. Garlic also mildly discourages blood platelets from clumping, which is part of the same cardiovascular interest and, as a later section warns, part of its risk profile as well.

The cholesterol story runs on the same rails — a genuine effect, small and conditional. The largest synthesis, an updated meta-analysis by Karin Ried and colleagues published in Nutrition Reviews in 2013, gathered thirty-nine trials and found that garlic lowered total cholesterol by roughly 17 mg/dL and low-density lipoprotein — the "bad" cholesterol — by about 9 mg/dL, but only in people who began with raised levels (above 200 mg/dL) and who took it for longer than two months. In those with normal cholesterol, or over shorter courses, the effect shrank towards nothing. And here the honest caveat bites hardest: the most rigorous trials, the diet-controlled ones with the tightest methods, have tended to show the smallest reductions of all, some of them statistically indistinguishable from placebo. A modest effect that fades under the best scrutiny is not a cure; it is a hint that survives, weakly, into the good data. There is a mechanism here too: in laboratory studies garlic's organosulphur compounds inhibit HMG-CoA reductase, the very enzyme the body uses to manufacture cholesterol and the same one the statin drugs block — though the effect in a living person is a faint shadow of a statin's, not a rival to it. A separate strand of studies even finds a small reduction in blood sugar in people with type-2 diabetes, with meta-analyses reporting modest falls in fasting glucose and in the longer-term marker HbA1c; the cautious institutions allow only that garlic "may reduce blood sugar to a small extent" in such patients. It is the same shape of result once more — real-looking, modest, mechanism-backed, and dependent on who is measured.

That variation — heterogeneity, in the trade — is not a footnote to be skipped past. It is central. The trials differ in almost every way that matters: they use different preparations (raw garlic behaves differently from dried powder, and both differ from aged garlic extract), different doses, different durations, different populations, and different standards of rigour. When results scatter this widely, pooling them into one average conceals as much as it reveals, and it becomes correspondingly hard to say how much benefit a given person taking a given product would actually see. This is why the cautious institutions phrase it as they do. The American National Center for Complementary and Integrative Health, surveying the same literature, concludes only that garlic supplements may reduce blood pressure "to a small extent" in people with hypertension, and may reduce total and LDL cholesterol "to a small extent" in people with high cholesterol. "May," "small," "some": these are not the words of a cure. They are the honest words for a modest, genuine, and still-imperfectly-quantified effect. Garlic is not a substitute for a diagnosis, for a doctor, or for a prescribed medicine, and no responsible reading of the evidence suggests it is. What the evidence supports is narrower and quieter: that a food long eaten for the heart turns out, when finally put to the test, to have a small measurable action in the direction its reputation always claimed. That is a respectable result. It is not a headline.

A word is owed to the preparation that supplies much of this evidence, because it is not the clove on the chopping board. Aged garlic extract is made by steeping sliced garlic in dilute alcohol for many months, a process that strips away the harsh, unstable allicin and leaves behind stable, water-soluble compounds, chief among them S-allylcysteine, which serves as its chemical marker. Odourless and standardised, it is the most-studied garlic supplement in the whole cardiovascular literature — much of it produced under a single commercial brand — and some of the cleaner trials use it. In one Australian study published in Maturitas in 2010, seventy-nine patients whose high blood pressure had resisted their existing medication were given aged garlic extract or a placebo; those on the highest dose saw their systolic pressure fall by roughly 10 mmHg more than the placebo group. That trial is worth citing for a second reason, a detail that foreshadows this chapter's hardest problem: to keep the study honestly blind, the researchers slipped a drop of garlic-scented liquid into the placebo group's container, so that neither side could tell which capsule they held by its smell. It is a rare and admirable attempt to solve a difficulty that, as we shall see, quietly defeats most garlic trials.

The Antiseptic That Cannot Travel

If the cardiovascular claim is garlic's most solid, the antimicrobial claim is its most misunderstood — because here a real laboratory fact has been stretched into a clinical fiction. The fact is genuine and well established. In 1944 the chemists Chester Cavallito and John Bailey isolated allicin from crushed garlic and showed that it kills bacteria; it has since been shown to have real activity, in the test tube, against a broad range of bacteria, fungi and other microbes, including some that resist conventional drugs. In a dish, allicin inhibits Gram-positive and Gram-negative bacteria and yeasts alike: reported minimum inhibitory concentrations run to a couple of milligrams per millilitre against methicillin-resistant Staphylococcus aureus — the hospital "superbug" MRSA — and to single-digit micrograms per millilitre against Candida. The mechanism is exactly the promiscuous chemical violence you would expect of so reactive a molecule: allicin attacks the thiol (–SH) groups on which countless microbial enzymes depend, and interferes with the synthesis of their RNA. It is a scattergun, not a sniper's bullet, and at the right concentration in a Petri dish it is a genuine antiseptic. None of that is in dispute.

Nor is the reputation merely modern. The observation that garlic juice checks the growth of bacteria is often traced to the chemist Louis Pasteur in the nineteenth century, though the primary record for that precise attribution is surprisingly thin — an apt caution at the head of a section about telling fact from repetition. Garlic's use as a wound-dresser is, in any case, genuinely old, and it saw real service in the age of industrial war: in the First World War, with antiseptics scarce in the trenches, garlic juice diluted with water was daubed on dressings against the infection that killed so many of the wounded, and in the Second World War the Soviet Red Army is said to have leaned on it so heavily, when antibiotics ran short, that it acquired the popular nickname "Russian penicillin." That last phrase should be handled with care — it is a folk tag, repeated in a hundred articles and rarely traced to a firm military-medical source — and even the genuine wartime use proves less than it seems: it was a measure of desperation and scarcity, not a demonstration that garlic cures infection. What it does show is that the antiseptic action later pinned on allicin was real enough to be reached for when nothing better was at hand.

The fiction is the leap from the test tube to the sickbed — from in vitro to in vivo — and it is a leap the evidence does not support. A compound that kills bacteria in a Petri dish has cleared only the first and easiest hurdle on the long road to being a medicine. It must also survive being eaten, reach the site of infection in a meaningful concentration, act there without poisoning the patient, and prove itself in controlled trials against real disease. Allicin fails at almost the first step, for the same reason it is so pungent and so fleeting in the kitchen: it is chemically unstable, and — the very quality that makes it a broad antiseptic — it reacts with almost everything it meets. It is not present in intact garlic at all, forms only when the clove is crushed, and then rapidly decomposes; it does not survive digestion and circulate through the body as an intact antibiotic patrolling the bloodstream. A molecule that reacts with everything is neutralised long before it could reach a distant infection. Eating garlic, or swallowing a garlic capsule, is not the same as applying allicin to a wound in a laboratory, and there is no good clinical-trial evidence that garlic cures or reliably prevents any human infection. To call garlic "nature's antibiotic," as the supplement copy loves to, is to report the first line of the experiment and suppress the rest.

There is one honest exception worth granting, and it proves the rule by its narrowness. On the skin — where garlic's compounds can act at the surface, locally, without needing to survive digestion or circulate — there is a little genuine clinical evidence. A garlic-derived compound, ajoene, worked into a gel, cleared cases of athlete's foot in small trials, much as an ordinary over-the-counter antifungal would. That is a real result, and it is exactly the kind the laboratory data predict: garlic chemistry working on contact, in one place, applied directly. It is also the mirror image of what the supplement bottle promises, which is protection from within. The antiseptic that cannot travel can still, it turns out, do a little good precisely where it does not have to.

This same instability opens a second, quieter gap that every buyer of garlic pills should understand — the gap between what a label promises and what a body receives. Because allicin is too unstable to bottle, powdered garlic supplements do not contain allicin at all. What they contain is alliin, the odourless precursor, together with the enzyme meant to convert it — and manufacturers therefore print not an allicin content but an "allicin potential" or "allicin yield," a figure for the allicin that could form under ideal conditions. Whether it actually forms in the gut is another matter entirely. Careful analyses of commercial products, notably the work of Larry Lawson and colleagues, have found that the allicin a tablet delivers varies enormously from product to product, that enteric coatings meant to protect the enzyme from stomach acid often release far less than the label implies, and that taking a supplement with a protein-rich meal can cut the delivered dose further. A stated "allicin potential" on a bottle, in short, is a claim about a chemical reaction the pill may never complete, not a measured dose. When you cannot be sure how much active compound a product delivers, the trials testing such products inherit that uncertainty — and so does anyone relying on the result.

The Cold, the Cancer, and the Weight a Single Study Can Bear

Of all garlic's popular reputations, none is more firmly lodged than its power over the common cold — and none rests on thinner evidence. The definitive appraisal is the Cochrane systematic review Garlic for the common cold, whose authors searched the world's medical databases for randomised controlled trials worthy of the name. They found eight candidate studies. Of the eight, exactly one met the standard required to draw conclusions: a single trial of 146 participants — the 2001 study by Peter Josling, published in Advances in Therapy — who took either an allicin-containing garlic capsule or a placebo daily for twelve weeks and kept a diary of their colds. That one trial reported a striking result — 24 colds in the garlic group against 65 in the placebo group — and it is almost certainly the ultimate source of the whole modern belief that garlic wards off colds. But a single small trial, however suggestive, cannot carry that weight, and the Cochrane reviewers said so plainly: there is insufficient clinical-trial evidence on garlic for the common cold, and the popular claims of effectiveness "appear to rely largely on poor-quality evidence." Their verdict is the honest one, and it should be quoted whenever the cold claim is made.

Cancer is the most emotionally charged claim of all, and here the distinction between kinds of evidence matters most. There is a real body of epidemiological work — observational studies of populations — that finds people who eat more allium vegetables, garlic and onions among them, tend to have somewhat lower rates of certain cancers, particularly of the stomach and gut. The largest such effort is the European Prospective Investigation into Cancer and Nutrition, EPIC, which has followed some half a million people across ten European countries; a reanalysis of its data found that a higher combined intake of onions and garlic went with a lower risk of intestinal-type stomach cancer, an extra 10 grams a day associated with roughly a 30 per cent lower risk. It is the kind of number that makes a headline — until you read the authors' own word for it, which was that the association was only "borderline" significant. Such associations are worth taking seriously as prompts for research, but they are not proof of cause. People who eat plenty of garlic differ from those who do not in a hundred other ways — diet, wealth, habits, health — and any of those differences might explain the pattern.

The laboratory adds real reasons to take the idea seriously. In cell and animal experiments, garlic's organosulphur compounds — diallyl disulphide and diallyl trisulphide among them — can switch on the body's own detoxifying enzymes and drive cancer cells towards programmed death. There is even a tidy-looking thread to stomach cancer in particular, since garlic extracts inhibit Helicobacter pylori — the ulcer-causing bacterium that is itself a leading cause of gastric cancer — in the test tube. All of which made the question worth testing properly, and testing it means an intervention trial: give some people garlic and others a placebo, and follow them for years. That was done, most searchingly in a long-running trial in Shandong, China, in a population where stomach cancer and H. pylori are both rife — and the result was sobering. Garlic supplementation did not significantly reduce precancerous lesions, nor clearly reduce gastric-cancer incidence, even where the observational signal had been strongest; tellingly, in the same trial it was antibiotic treatment of the H. pylori infection, not the garlic, that eventually lowered cancer rates. The compound that killed the bacterium in a dish did not, when eaten, do the work a course of antibiotics did. Reflecting all this, the cautious institutions state flatly that garlic does not appear to reduce the risk of stomach cancer, and that whether it affects the risk of colorectal or other cancers is uncertain. The garlic-and-cancer story is therefore preliminary and unresolved — a hypothesis supported by population patterns and laboratory mechanisms, and not, on present evidence, an established human benefit. To sell it as one is to trade on people's fear at the exact point where the evidence is weakest.

Running beneath all these questions is a methodological difficulty peculiar to garlic, and it deserves naming because it quietly undermines even the good trials: garlic is almost impossible to test blind. The whole logic of a placebo-controlled trial depends on neither the participant nor the assessor knowing who received the real treatment — but garlic announces itself on the breath. In the very cold trial the Cochrane reviewers relied upon, more people in the garlic group than the placebo group noticed a garlic smell when they burped, which means the blinding may have been broken: some participants could likely tell which capsule they had, and expectation is a powerful thing. Add to this the strong commercial incentive for supplement makers to fund and publish trials that flatter their products — industry-funded nutrition studies do, on average, report rosier results than independent ones — and the mirror-image risk that trials finding nothing were quietly never published, the file-drawer problem that leaves the surviving record tilted towards good news. A meta-analysis can only pool the trials that exist; if the disappointing ones are missing, even an honest synthesis reads too kindly. You have, in short, a literature to be read with both eyes open.

Field Note — The One Trial That Carries the Cold

It is worth holding up a single specimen — not a clove this time but a study, the one small trial on which the entire popular belief in garlic-against-colds effectively rests. The Cochrane reviewers began with the whole published record and filtered it down to what could actually support a conclusion. From eight candidate studies, seven fell away for want of rigour. One remained: 146 people, twelve weeks, a garlic capsule or a placebo each day, a cold diary. Its numbers are genuinely eye-catching — 24 colds in the garlic group against 65 in the placebo group, and 111 sick-days against 366 — and taken alone they read like vindication.

Look closer and the picture complicates in the honest way real evidence always does. The recovery time, once a cold arrived, was much the same in both groups (about 4.6 days against 5.6), so even this trial gives no support to the idea that garlic shortens a cold you already have. More people in the garlic group noticed a garlic taste when they burped than in the placebo group, so the blinding that the whole method depends on may not have held. And the colds were self-reported in a diary, not confirmed. None of this makes the trial worthless — the reviewers judged its core methods reasonable — but it makes it a single, small, imperfect study, and the reviewers' conclusion follows inescapably: not enough to know. The lesson is not that garlic does nothing for colds. It is that we do not have the evidence to say it does, and that a belief held by millions can trace back, when you follow it to the source, to one modest experiment carrying far more than it was ever built to bear.

The Immune Claim, and the Shape of a Thin Result

Close beside the cold sits the broader promise that garlic "boosts the immune system," a phrase that adorns half the supplement shelves each winter. It is not baseless — but the base is thin, and its shape is instructive. The most-cited human trial is the work of Nantz and colleagues, published in Clinical Nutrition in 2012: 120 healthy adults took either aged garlic extract, about 2.56 grams a day, or a placebo, through the cold-and-flu season. After forty-five days the garlic group's immune cells — the natural-killer cells and the gamma-delta T cells that patrol for infection — multiplied more briskly in the laboratory than the placebo group's, a real and measurable difference. And over ninety days, the garlic takers reported milder illness: not fewer colds and flus caught, but fewer and less severe symptoms when they came, and fewer days knocked below par. That distinction matters and is easily lost in the marketing — the trial points, at most, to gentler colds, not to a shield against catching them.

One promising trial, however, is not a body of evidence, and the honest institutional reading is bracingly modest. Reviewing the whole field in 2022, the American complementary-health centre found that "very little research has been done on this topic," and that a systematic review that year "identified only two studies that suggest a possible benefit, and both studies included only small numbers of people and had weaknesses in the research." Two small studies, with a common thread worth naming: aged garlic extract is a proprietary product, and trials of proprietary supplements are often linked to the companies that sell them — not a disqualification, but a reason for the extra scepticism good readers bring to a manufacturer's own evidence. The immune claim, then, belongs firmly in the "promising but thin" tier: a plausible mechanism, one or two small trials pointing the same way as the lone cold trial, and nothing yet strong enough to build a health claim upon. It is exactly the sort of result that ought to prompt a large, independent trial — and exactly the sort that gets sold, long before that trial is run, as though the question were already settled. The proposed mechanism is at least coherent — garlic's sulphur compounds appear, in laboratory work, to sharpen the activity of several kinds of immune cell — but a coherent mechanism and one or two small, industry-linked trials are the beginning of a case, not its close, and the honest word for the immune claim remains unproven.

The Risks Nobody Advertises

The overclaim machine sells garlic as boundlessly benign, a thing that can only help. A food potent enough to act on the body, though, is potent enough to interfere with it, and the honest ledger has a second column that the labels leave blank. Three cautions are well enough documented to state plainly.

The first, and most striking, is a drug interaction. In a study by Piscitelli and colleagues in Clinical Infectious Diseases in 2002, ten healthy volunteers took garlic supplements for three weeks alongside saquinavir, an antiretroviral drug then used against HIV. Garlic cut the drug's blood levels drastically: average exposure fell by about 51 per cent, peak concentration by about 54 per cent, and the trough level between doses by about 49 per cent — enough to push the drug below the threshold at which it works, with the attendant danger of breeding resistance. Even after a ten-day washout with no garlic, saquinavir levels had not fully recovered. The likely reason is that garlic revs up the body's drug-processing machinery in the gut and liver, which handles a great many medicines besides this one, so the concern is not confined to a single obsolete antiretroviral. It is the clearest proof that "just a food supplement" can be a pharmacologically active substance with real consequences for anyone on prescription medicines.

The second caution is bleeding. Garlic has a mild antiplatelet action — it makes blood a little slower to clot — and while that is part of the cardiovascular interest, it becomes a hazard in the wrong company. The health authorities warn that garlic supplements "may increase the risk of bleeding," a genuine concern for anyone taking anticoagulants such as warfarin, or aspirin, and for anyone about to have surgery, which is why surgeons routinely ask patients to stop garlic supplements beforehand. The honest nuance is that controlled trials on warfarin itself have been reassuring — aged garlic extract did not significantly disturb the blood-thinning of patients in one such trial — so the risk from culinary garlic is largely theoretical, and even from supplements it is more a matter of prudent caution than of proven harm. But "tell your doctor, and stop before surgery" is sound advice, not scaremongering.

The third is the one the folk-remedy enthusiasts learn the hard way. Raw garlic crushed against the skin — applied to a wart, a spot, a suspected infection, as the internet advises — can cause a genuine chemical burn; a 2021 systematic review in the American Journal of Emergency Medicine collected a substantial run of reported cases, some of them second-degree. The same allicin that savages a microbe in a Petri dish savages human tissue too, given time and contact. Garlic is a recognised contact allergen as well: its diallyl disulphide can raise an itching, cracking dermatitis on the hands of cooks and food handlers who spend their days peeling and chopping it, a small occupational hazard with a long clinical pedigree. Fresh raw garlic used topically, the authorities note, "can cause severe skin irritation and chemical burns." (Eaten, garlic is safe in the amounts food supplies, and has been used safely in trials lasting years; the ordinary side-effects are the social ones — breath, body odour — plus, in some, heartburn or an upset stomach. Supplement doses beyond culinary amounts are of less certain safety in pregnancy.) None of this makes garlic dangerous to eat. It makes garlic a real substance rather than a magic one — and a real substance has to be reckoned with in both directions.

There is, finally, one garlic hazard that is genuinely lethal, and it belongs here because it inverts everything the mythology assumes. The real danger lies not in the clove but in how it is kept. Raw garlic left in oil at room temperature is a breeding ground for Clostridium botulinum and one of the deadliest toxins known, as Chapter 5 set out in full — the one garlic preparation that has actually killed and hospitalised people in living memory. The lesson is quietly damning to the cure-all story: the best-documented serious health effect of garlic in the modern record is not a disease it cured but a poisoning it caused through careless storage. Refrigerate garlic-in-oil and use it within days, or acidify it, and never leave it standing warm.

Behind all three cautions stands a problem this chapter has already met in part and should now state whole, because it undermines almost every confident claim made in garlic's name: the word "garlic" on a label can mean any of several quite different substances. There is raw garlic, all fleeting allicin; there is the dehydrated powder of the "allicin potential" tablets met earlier; there is steam-distilled garlic oil, rich in stable diallyl sulphides but containing no allicin at all; and there is aged garlic extract, built on odourless S-allylcysteine and likewise allicin-free. These are not four brands of one thing but four different chemical products that behave differently in the body — so a trial of one says almost nothing about another, and lumping them together into a single verdict on "garlic" is a category error the literature commits constantly. Layer on top of that the gap already described between a supplement's promised dose and its delivered one, and the fact that these products reach the shelf without a regulator's prior approval, and the "dose" behind any of garlic's effects — good or ill — becomes genuinely uncertain. The label rarely tells you which garlic you are holding, let alone how much of it will ever reach you.

<figure> <svg xmlns="http://www.w3.org/2000/svg" width="100%" viewBox="0 0 640 560" role="img"> <title>The Evidence Ladder — garlic's health claims ranked by strength of evidence</title>  <g fill="none" stroke="#43331f" stroke-width="1"> <rect x="10" y="10" width="620" height="540"/> <rect x="14" y="14" width="612" height="532" stroke-width="0.6"/> </g> <g fill="none" stroke="#6d5577" stroke-width="1.2"> <path d="M14 30 L14 14 L30 14"/> <path d="M626 530 L626 546 L610 546"/> </g>  <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#43331f"> <text x="320" y="42" text-anchor="middle" font-size="15" letter-spacing="4">THE EVIDENCE LADDER</text> <text x="320" y="58" text-anchor="middle" font-size="8.5" letter-spacing="2" fill="#6f5a40">GARLIC’S HEALTH CLAIMS, RANKED BY STRENGTH OF EVIDENCE</text> </g>  <g fill="none" stroke="#6f5a40" stroke-width="1"><line x1="52" y1="96" x2="52" y2="428"/><line x1="64" y1="96" x2="64" y2="428"/></g> <g stroke="#a9906c" stroke-width="0.7"><line x1="52" y1="116" x2="64" y2="116"/><line x1="52" y1="156" x2="64" y2="156"/><line x1="52" y1="196" x2="64" y2="196"/><line x1="52" y1="236" x2="64" y2="236"/><line x1="52" y1="276" x2="64" y2="276"/><line x1="52" y1="316" x2="64" y2="316"/><line x1="52" y1="356" x2="64" y2="356"/><line x1="52" y1="396" x2="64" y2="396"/></g> <path d="M58 96 L58 82 M58 82 L52 92 M58 82 L64 92" fill="none" stroke="#6d5577" stroke-width="1.4"/> <text x="40" y="262" transform="rotate(-90 40 262)" text-anchor="middle" font-size="8" letter-spacing="2.5" fill="#6f5a40" font-family="'Liberation Sans','DejaVu Sans',sans-serif">STRENGTH OF EVIDENCE</text>  <rect x="90" y="96" width="470" height="92" rx="5" fill="none" stroke="#6d5577" stroke-width="1.4"/> <line x1="252" y1="106" x2="252" y2="178" stroke="#a9906c" stroke-width="0.6"/> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#6d5577" text-anchor="middle"> <text x="171" y="134" font-size="12" letter-spacing="1.5">WELL</text> <text x="171" y="150" font-size="12" letter-spacing="1.5">SUPPORTED</text></g> <g fill="#6d5577"><circle cx="150" cy="168" r="3"/><circle cx="171" cy="168" r="3"/><circle cx="192" cy="168" r="3"/></g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#43331f"> <text x="264" y="130" font-size="9.5">Blood pressure — modest lowering in hypertensives</text> <text x="264" y="150" font-size="9.5">Cholesterol — small drop in total & LDL</text> <text x="264" y="170" font-size="7.5" fill="#6f5a40" font-style="italic">randomised trials & meta-analyses</text></g>  <rect x="90" y="214" width="470" height="92" rx="5" fill="none" stroke="#6f5a40" stroke-width="1.1"/> <line x1="252" y1="224" x2="252" y2="296" stroke="#a9906c" stroke-width="0.6"/> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#43331f" text-anchor="middle"> <text x="171" y="252" font-size="12" letter-spacing="1.5">PROMISING</text> <text x="171" y="268" font-size="12" letter-spacing="1.5">BUT THIN</text></g> <g><circle cx="150" cy="286" r="3" fill="#6f5a40"/><circle cx="171" cy="286" r="3" fill="none" stroke="#6f5a40" stroke-width="0.9"/><circle cx="192" cy="286" r="3" fill="none" stroke="#6f5a40" stroke-width="0.9"/></g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#43331f"> <text x="264" y="248" font-size="9.5">Common cold — a single weak trial</text> <text x="264" y="268" font-size="9.5">Antimicrobial — allicin active in vitro</text> <text x="264" y="288" font-size="7.5" fill="#6f5a40" font-style="italic">preliminary · in vitro is not clinical</text></g>  <rect x="90" y="332" width="470" height="92" rx="5" fill="none" stroke="#6f5a40" stroke-width="0.9" stroke-dasharray="5 4"/> <line x1="252" y1="342" x2="252" y2="414" stroke="#a9906c" stroke-width="0.6"/> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#6f5a40" text-anchor="middle"> <text x="171" y="370" font-size="12" letter-spacing="1.5">FOLKLORE</text> <text x="171" y="386" font-size="12" letter-spacing="1.5">UNPROVEN</text></g> <g fill="none" stroke="#a9906c" stroke-width="0.9"><circle cx="150" cy="404" r="3"/><circle cx="171" cy="404" r="3"/><circle cx="192" cy="404" r="3"/></g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#6f5a40"> <text x="264" y="366" font-size="9.5">Cure-alls, ‘detox’, infections banished</text> <text x="264" y="386" font-size="9.5">Panacea & miracle-prevention claims</text> <text x="264" y="406" font-size="7.5" font-style="italic">tradition & overclaim, not evidence</text></g>  <line x1="90" y1="458" x2="550" y2="458" stroke="#a9906c" stroke-width="0.6"/> <text x="320" y="478" text-anchor="middle" font-size="8.5" letter-spacing="1.5" fill="#43331f" font-family="'Liberation Sans','DejaVu Sans',sans-serif">MODEST PROVEN EFFECTS SIT ATOP A MOUNTAIN OF OVERCLAIM</text> <text x="320" y="496" text-anchor="middle" font-size="7.5" letter-spacing="0.5" fill="#6f5a40" font-style="italic" font-family="'Liberation Sans','DejaVu Sans',sans-serif">A wholesome food, not a medicine. Cancer links are associations only. No medical advice.</text> </svg>

<figcaption><em>Plate 13. The evidence ladder — garlic's health claims sorted by the strength of the actual evidence. Read from the solid ground up: firm footing in modest, real reductions in blood pressure and cholesterol for those who start high; a shaky middle of in-vitro antimicrobial action, one weak cold trial, and preliminary immune signals; and, at the thin-aired top, the cure-all folklore the evidence will not bear.</em></figcaption> </figure>

None of this is a verdict against tradition, exactly. Four thousand years of use tell us that garlic is safe to eat in quantity, that people everywhere have found it worth growing and reaching for, and that its reputation as a tonic is genuinely global — all real facts about human beings. What that long record cannot tell us is whether the tonic works, because people believed in bloodletting for just as long and with just as much conviction. Tradition generates hypotheses; it does not test them. The whole value of the last two centuries of chemistry and the last few decades of trials is that they can take a four-thousand-year-old belief and put to it the one question antiquity never could — against a placebo, in people who did not know which they had been given, does it actually do anything? For most of garlic's medical promises the answer so far is "we cannot yet say"; for a few it is "a little." That is not contempt for the old knowledge. It is the first time the old knowledge has been properly examined.

An Honest Verdict

Set the mythology aside and weigh only what the trials will bear, and garlic comes out looking like exactly what it is: a wholesome food with a few small, real, evidence-backed effects and an enormous surplus of overclaim. The cardiovascular case is the sound one — modest reductions in blood pressure, most dependable in people with hypertension, and small reductions in cholesterol, with the honest caveats that the effect is limited, the studies are heterogeneous, and the products are inconsistent. The antimicrobial case is a genuine laboratory fact stranded far short of a clinical one. The cold claim rests on a single small trial and, in the judgement of the best available review, on poor-quality evidence. The immune claim rests on one or two small, industry-linked studies pointing hopefully in the same direction. The cancer claim is a live research question and not a demonstrated human benefit. Set against these modest positives is a short, real list of cautions — a proven drug interaction, a bleeding risk worth minding around surgery and anticoagulants, a topical burn hazard — that the marketing never mentions. And the whole field is clouded by garlic's unblindable pungency, by the instability that separates a supplement's promise from its delivered dose, and by the commercial pressure to publish good news.

Where does that leave the eater? Precisely where a naturalist should wish to be left: free to love the thing for what it certainly is, and unwilling to pretend it is what it is not. Garlic is one of the great pleasures of the table and, by the modest measure the evidence allows, a food with a little quiet good in it for the heart. That is praise enough, and it has the singular merit of being true. This book will not offer it as a medicine, will not tell any reader to treat an illness with it, and will not dress a laboratory result or a folk memory in the costume of a cure. The extraordinary claims made for garlic have never been matched by extraordinary evidence — only, here and there, by ordinary, modest, welcome evidence, which is a finer thing than a myth because it can be trusted.

Having held garlic to the cold light of the trial and refused it the crown of a cure, we have, oddly, lost nothing worth keeping — and cleared the ground for the only praise that lasts, the kind that survives knowing exactly what the thing is. The final chapter gathers six chapters of understanding into affection: not the credulous love that believes garlic cures all ills, but the earned love of one who has learned the bulb entire, from its anatomy to its chemistry to the honest limits of its medicine, and adores it anyway.

บทถัดไป · 10 / 15Chapter 7 — In Praise of Garlic