Chapter 2 — The Chemistry of Pungency

"A number of curious sulfur compounds underlie the odor of garlic and the crying brought on by slicing an onion. The compounds also account for medical properties long ascribed to garlic and onions." — Eric Block, "The Chemistry of Garlic and Onions," Scientific American (1985)

T here is a fact about garlic that ought to be more disconcerting than it is, and the last chapter ended on it: a whole, sound clove barely smells of anything. Lift an unbroken clove to the nose and there is a faint, green, waxy scent, but nothing of the ferocity we mean by the word garlic — none of the pungency that clears a room, flavours a stew, and lingers on the breath until tomorrow. That flavour, the entire reason the plant has a book, is not sitting inside the clove waiting to be released. It does not exist yet. It is manufactured, in the space of a few seconds, at the moment the clove is cut, out of ingredients that were harmless while they were kept apart. Garlic's flavour is not a possession of the plant but an event — and, more precisely, an injury. What we taste when we taste fresh garlic is a wound.

The Odourless Bulb and the Wound

The plant keeps its weapon disassembled, in two parts that are stored, deliberately, in different places within the cell. The first part is a small, stable, odourless molecule called alliin — in full, S-allyl-L-cysteine sulphoxide — a sulphur-bearing compound that the garlic accumulates quietly in the fluid of its cells and that does nothing, by itself, to the nose or tongue. The second part is an enzyme, alliinase, held apart from the alliin behind cell membranes, in separate compartments, so that in the intact clove the two never meet. This separation is the whole point of the arrangement. Alliin is a stockpiled munition; alliinase is the detonator; and a living clove keeps them in different rooms.

Alliinase is no minor tenant of the cell, either. It is one of the most plentiful proteins the clove makes — a vitamin-B6-dependent enzyme held ready in such quantity that, once the walls come down, the conversion runs to near-completion within a minute. It has one notable weakness, and the kitchen can exploit it or blunder into it: it is switched off by acid. Crush a clove straight into lemon juice or vinegar and the enzyme is halted almost before it starts, so little allicin forms and the bite stays mild; the same acid-sensitivity means that a garlic tablet swallowed whole makes almost no allicin in the acid bath of the stomach unless it is specially coated to survive the passage — a point that will matter a great deal when this book comes to weigh the supplements.

Cutting, crushing, chewing or bruising the clove tears those rooms open. Membranes rupture, the contents of the cells run together, and alliinase falls upon alliin at last — whereupon it works with the speed that enzymes are built for, converting the odourless alliin, within seconds, into a new and highly reactive substance called allicin. Allicin (its chemists' name is diallyl thiosulphinate) is the true author of fresh garlic's smell and bite: the sharp, hot, unmistakable pungency of a raw clove just crushed is the smell of allicin being made in front of you. It is worth savouring the strangeness of this. The plant that has just been cut is not leaking a pre-formed flavour; it is, in self-defence, synthesising a caustic chemical on the spot, a deterrent meant to punish the insect or the fungus that dared to breach it. That the creature it most reliably attracts, rather than repels, is the cook is one of the happier accidents in the history of eating.

Allicin's second remarkable property is that it will not sit still. Having been made, it immediately begins to fall apart, and its breakdown is not a fading but a blossoming — a cascade of further sulphur compounds, each with its own character, unfolding over minutes and hours and accelerated by warmth. From allicin come the diallyl sulphides — diallyl disulphide and diallyl trisulphide chief among them — which carry much of the deep, cooked, meaty note of garlic; and, when garlic meets oil and gentle heat, the vinyldithiins and ajoene that flavour a slow-cooked confit and a warm garlic oil. The single sharp chord struck by the knife thus decays into a whole chromatic run of flavours, which is why raw garlic tastes of one thing at the first second and of something rounder a minute later, and why the smell of a kitchen where garlic is frying is not the smell of the clove at all but of allicin's numerous children. The pungency we prize is, chemically, a controlled demolition — a compound built to be unstable, doing exactly what it was built to do.

Here a small comparison earns garlic a compliment its cousin cannot share. The onion, cut, does something garlic never does: it makes the cook weep. The reason is a fork in the family chemistry. In the onion, the enzyme freed by the knife produces not a stable thiosulphinate but a small, volatile, tear-provoking gas — the lachrymatory factor, a fleeting sulphur compound that drifts up to the eyes and stings them to water. Garlic's chemistry takes the other branch: it lacks the additional enzyme step that manufactures that airborne irritant, and so, for all its ferocity on the tongue, it leaves the eyes alone. One may mince a mountain of garlic dry-eyed and be laid low by a single onion — a fair reminder that the alliums, though they share the same buried armoury of sulphur and the same trick of releasing it only when wounded, have each tuned the weapon to a different end.

Field Note — Allicin, the Wound That Makes the Flavour

Allicin was pulled out of garlic and named in the 1940s — the name borrowed straight from Allium — and almost everything peculiar about garlic in the kitchen follows from two facts about it: that it must be made, and that it cannot last. It does not pre-exist in the clove; it is generated on damage, the product of an enzyme meeting its fuel. And once generated it is doomed, decomposing steadily into the milder sulphur compounds downstream of it. Fresh crushed garlic is therefore never the same substance for two minutes running: it is most savagely pungent in the first moments, when allicin is at its peak, and it softens and deepens as that allicin spends itself.

Everything the good cook does with raw garlic is, without necessarily knowing the names, a way of managing allicin. To crush a clove to a paste is to make the most of it, rupturing nearly every cell so that nearly all the alliin is converted. To slice thinly is to make less. To leave the clove whole is to make almost none. To let the crushed garlic rest a few minutes before it goes near heat or acid is to give the enzyme time to finish its work before anything can stop it. And to understand that allicin is a wound response — a thing the plant would never have made had it not been hurt — is to understand the central paradox of the ingredient: that its generosity to us is, from the plant's point of view, an act of defence, and that we have spent four thousand years cultivating a plant for the express pleasure of provoking its alarm.

<figure> <svg xmlns="http://www.w3.org/2000/svg" width="100%" viewBox="0 0 640 384" role="img"> <title>Alliin to Allicin — the wound that makes the flavour</title> <g fill="none" stroke="#43331f" stroke-width="1"> <rect x="10" y="10" width="620" height="364"/> <rect x="14" y="14" width="612" height="356" stroke-width="0.6"/> </g> <g stroke="#6d5577" stroke-width="1.2" fill="none"> <path d="M14 30 L14 14 L30 14"/> <path d="M626 354 L626 370 L610 370"/> </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">ALLIIN → ALLICIN</text> <text x="320" y="58" text-anchor="middle" font-size="8.5" letter-spacing="2" fill="#6f5a40">THE WOUND THAT MAKES THE FLAVOUR</text> </g>  <g fill="none" stroke="#6f5a40" stroke-width="1.1"> <path d="M60 150 q-14 60 20 108 q46 40 100 20 q34 -14 30 -70 q-4 -70 -60 -88 q-64 -20 -90 30 Z"/> </g>  <g fill="none" stroke="#a9906c" stroke-width="0.9"> <ellipse cx="105" cy="205" rx="34" ry="30"/> </g> <g fill="none" stroke="#a9906c" stroke-width="0.5"> <line x1="150" y1="150" x2="190" y2="150"/><line x1="152" y1="165" x2="192" y2="163"/> <line x1="150" y1="180" x2="190" y2="180"/><line x1="150" y1="235" x2="188" y2="240"/> <line x1="146" y1="255" x2="182" y2="262"/> </g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#43331f"> <text x="105" y="203" text-anchor="middle" font-size="8" letter-spacing="1">ALLIIN</text> <text x="105" y="215" text-anchor="middle" font-size="6.5" fill="#6f5a40">(in vacuole)</text> <text x="163" y="270" text-anchor="middle" font-size="7.5" letter-spacing="1">ALLIINASE</text> <text x="112" y="300" text-anchor="middle" font-size="8" letter-spacing="1.5" fill="#6f5a40">INTACT CELL · ODOURLESS</text> </g>  <g fill="#6f5a40"> <circle cx="150" cy="255" r="2.2"/><circle cx="162" cy="248" r="2"/><circle cx="158" cy="262" r="1.8"/> <circle cx="140" cy="248" r="1.8"/><circle cx="170" cy="258" r="2"/> </g>  <g stroke="#6d5577" stroke-width="1.6" fill="none"> <line x1="235" y1="215" x2="320" y2="215"/> <path d="M320 215 L308 208 M320 215 L308 222"/> </g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#6d5577"> <text x="277" y="205" text-anchor="middle" font-size="9.5" letter-spacing="2">CRUSH</text> <text x="277" y="230" text-anchor="middle" font-size="6.5" letter-spacing="1">cells rupture</text> </g>  <g fill="none" stroke="#6f5a40" stroke-width="1.1"> <path d="M360 150 q-14 60 20 108 q46 40 100 20 q34 -14 30 -70 q-4 -70 -60 -88 q-64 -20 -90 30 Z" stroke-dasharray="6 5"/> </g>  <g fill="none" stroke="#43331f" stroke-width="1"> <path d="M410 200 L430 210 L450 200 L470 210"/> <line x1="450" y1="200" x2="450" y2="182"/> </g> <g fill="#43331f" font-family="'Liberation Sans','DejaVu Sans',sans-serif"> <circle cx="440" cy="205" r="3" fill="#6d5577"/> <text x="446" y="176" font-size="8" letter-spacing="1">S=O</text> <text x="440" y="235" text-anchor="middle" font-size="8.5" letter-spacing="1.5">ALLICIN</text> <text x="440" y="247" text-anchor="middle" font-size="6.5" fill="#6f5a40">(unstable thiosulphinate)</text> </g> <text x="425" y="300" text-anchor="middle" font-size="8" letter-spacing="1.5" fill="#6f5a40" font-family="'Liberation Sans','DejaVu Sans',sans-serif">RUPTURED CELL · PUNGENT</text>  <g stroke="#a9906c" stroke-width="0.7" fill="none"> <path d="M500 200 q10 -18 -2 -34 q-10 -14 4 -30"/> <path d="M525 210 q12 -20 0 -38 q-8 -14 6 -30"/> <path d="M548 225 q10 -18 -2 -34"/> </g> <g fill="#6f5a40"> <circle cx="500" cy="132" r="1.6"/><circle cx="529" cy="128" r="1.6"/><circle cx="546" cy="150" r="1.4"/> </g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#6f5a40"> <text x="565" y="120" text-anchor="middle" font-size="7" letter-spacing="0.5">diallyl</text> <text x="565" y="130" text-anchor="middle" font-size="7" letter-spacing="0.5">disulphide</text> <text x="565" y="148" text-anchor="middle" font-size="6.5">+ volatile</text> <text x="565" y="157" text-anchor="middle" font-size="6.5">sulphurs</text> </g>  <line x1="120" y1="332" x2="520" y2="332" stroke="#a9906c" stroke-width="0.6"/> <text x="320" y="352" text-anchor="middle" font-size="8" letter-spacing="1" fill="#43331f" font-family="'Liberation Sans','DejaVu Sans',sans-serif">ENZYME + SUBSTRATE, KEPT APART, MEET ONLY WHEN THE CELL IS BROKEN</text> </svg>

<figcaption><em>Plate 4. Alliin to allicin: in the intact cell the odourless alliin and the enzyme alliinase are kept apart; the crush mixes them, and alliinase converts alliin into pungent allicin, which then cascades into the volatile sulphur compounds of garlic's flavour.</em></figcaption> </figure>

How the Chemistry Was Uncovered

That the flavour was a manufactured thing, and not a stored one, was worked out only in the middle of the twentieth century, and by chemists who were, for the most part, chasing a medicine rather than a meal. Garlic had in fact lent its name to organic chemistry a full century earlier: in 1844 the chemist Theodor Wertheim, distilling garlic, isolated a reactive fragment he christened the "allyl" radical — from Allium — so that garlic named a whole class of compounds long before its own key substance was understood. Its power to kill bacteria, too, had been noticed before it could be explained; the observation is traditionally credited to Louis Pasteur in the 1850s. But the compound responsible went unnamed for the better part of a hundred years. The decisive step came in 1944, in the laboratories of the Winthrop Chemical Company at Rensselaer, in New York State, where Chester Cavallito and John Bailey isolated from garlic the unstable, oily substance that carried both its smell and its antibacterial bite, named it allicin — again, straight from Allium — and reported, in the Journal of the American Chemical Society, that it stopped both Gram-positive and Gram-negative bacteria in the dish. Within the same year Cavallito and his colleagues had settled its structure as a thiosulphinate.

What Cavallito could not explain was where the allicin came from, for he could find none of it in the sound, uncut clove. That half of the puzzle was solved a few years afterward and a continent away, at the Sandoz laboratories in Basel, where the chemists Stoll and Seebeck isolated the odourless precursor sitting quietly in the whole clove — the sulphur-bearing amino acid they named alliin — and identified the enzyme, alliinase, that fell upon it the instant the cell was broken. By 1951 the same pair had synthesised alliin from scratch, closing the account. The odourless bulb, the stockpiled munition, the enzyme detonator and the pungent product had all, at last, been named — and the wordless knowledge of four thousand years of cooks, that a clove must be broken to be believed, was handed its molecular explanation.

If one name stands above the rest in the modern chapter of this story, it is that of Eric Block, the chemist whose Scientific American line opens this chapter and whose decades of work mapped the sulphur chemistry of the whole onion genus in a detail no one had reached before. It was Block and his co-workers who, in the 1980s, isolated and named ajoene — the oil-formed compound met a moment hence — and who traced many of the branching routes by which allicin decays; his book Garlic and Other Alliums remains the fullest single account of the field. The chemistry a cook lets loose with the tap of a knife turns out to be deep enough to have held a serious laboratory for a working lifetime.

The Cascade and Its Children

Allicin's instability is worth following compound by compound, because the whole flavour-life of cooked garlic is written in its decay. Allicin itself is fleeting; what it becomes depends on where it finds itself. Left in the watery world of a crushed clove, or met with heat, it breaks down chiefly into the diallyl sulphides — diallyl disulphide and diallyl trisulphide — the compounds that carry the deep, cooked, faintly meaty reek of frying garlic and the pungent heart of garlic's essential oil; and, as a rule the chemists have borne out, the longer the chain of sulphur atoms in these compounds, the more potent and penetrating the smell, so that the trisulphide out-punches the disulphide. This is also why the commercial "garlic oil" of the essential-oil kind — the steam-distilled sort sold in tiny bottles — reeks so relentlessly of cooked garlic and not at all of the fresh clove: the heat of distillation drives the whole cascade to its end, leaving little behind but these deep diallyl polysulphides.

Dropped instead into ordinary oil at a gentle warmth — the moment a cook slips crushed garlic into olive oil — allicin takes another road entirely, rearranging into two compounds that form almost nowhere else. The first are the vinyldithiins, which account for much of the soft, mellow flavour of garlic steeped in oil; the second is ajoene, whose very name is built from ajo, the Spanish for garlic. These two are the signature of a confit and a warm garlic oil, and their near-total absence from a dry, fierce pan is one clear reason that garlic softened slowly in oil tastes so unlike garlic thrown onto bare heat. The single sharp chord struck by the knife thus decays not into silence but into a whole progression of them — each compound a different note, the sequence set by heat, by time, and above all by whether or not there is oil in the pan.

<figure> <svg xmlns="http://www.w3.org/2000/svg" width="100%" viewBox="0 0 640 430" role="img"> <title>The Sulphur Cascade — what the crushed clove becomes</title> <g fill="none" stroke="#43331f" stroke-width="1"> <rect x="10" y="10" width="620" height="410"/> <rect x="14" y="14" width="612" height="402" stroke-width="0.6"/> </g> <g stroke="#6d5577" stroke-width="1.2" fill="none"> <path d="M14 30 L14 14 L30 14"/> <path d="M626 404 L626 420 L610 420"/> </g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" fill="#43331f"> <text x="320" y="40" text-anchor="middle" font-size="15" letter-spacing="4">THE SULPHUR CASCADE</text> <text x="320" y="56" text-anchor="middle" font-size="8.5" letter-spacing="2" fill="#6f5a40">WHAT THE CRUSHED CLOVE BECOMES</text> </g>  <rect x="48" y="118" width="150" height="60" rx="4" fill="none" stroke="#43331f" stroke-width="1"/> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" text-anchor="middle"> <text x="123" y="146" font-size="12" letter-spacing="2" fill="#43331f">ALLIIN</text> <text x="123" y="160" font-size="6.5" fill="#6f5a40">(+)-S-allyl-L-cysteine</text> <text x="123" y="169" font-size="6.5" fill="#6f5a40">sulphoxide</text> <text x="123" y="193" font-size="7" letter-spacing="0.5" fill="#6f5a40">intact clove · odourless</text> </g>  <g stroke="#43331f" stroke-width="1.2" fill="none"> <path d="M206 148 H248 M248 148 l -7 -4 M248 148 l -7 4"/> </g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" text-anchor="middle"> <text x="227" y="139" font-size="7.5" letter-spacing="0.5" fill="#43331f">crush</text> <text x="227" y="164" font-size="7" fill="#6f5a40">alliinase</text> </g>  <rect x="250" y="118" width="150" height="60" rx="4" fill="none" stroke="#6d5577" stroke-width="1.5"/> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" text-anchor="middle"> <text x="325" y="145" font-size="12" letter-spacing="2" fill="#43331f">ALLICIN</text> <text x="325" y="159" font-size="6.5" fill="#6f5a40">diallyl thiosulphinate</text> <text x="325" y="170" font-size="6.5" font-style="italic" fill="#6d5577">unstable</text> </g>  <rect x="452" y="92" width="150" height="44" rx="4" fill="none" stroke="#6f5a40" stroke-width="0.9"/> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" text-anchor="middle"> <text x="527" y="112" font-size="8" letter-spacing="1" fill="#43331f">BLUE-GREEN PIGMENT</text> <text x="527" y="124" font-size="6.5" fill="#6f5a40">pyrroles · pickling, bruising</text> </g> <g stroke="#6f5a40" stroke-width="0.9" fill="none"> <path d="M400 138 Q 430 118 452 114 M452 114 l -8 0 M452 114 l -3 6"/> </g> <text x="424" y="127" text-anchor="middle" font-size="6" fill="#6f5a40" font-family="'Liberation Sans','DejaVu Sans',sans-serif">+ amino acids</text>  <g stroke="#43331f" stroke-width="1" fill="none"> <path d="M287 179 L 128 250 M128 250 l -5 -6 M128 250 l 5 -6"/> <path d="M325 179 L 320 250 M320 250 l -5 -6 M320 250 l 5 -6"/> <path d="M363 179 L 512 250 M512 250 l -5 -6 M512 250 l 5 -6"/> </g>  <g fill="#6f5a40"> <circle cx="208" cy="214" r="2.1"/><circle cx="322" cy="214" r="2.1"/><circle cx="438" cy="214" r="2.1"/> </g>  <g fill="none" stroke="#43331f" stroke-width="1"> <rect x="40" y="250" width="180" height="66" rx="4"/> <rect x="232" y="250" width="176" height="66" rx="4"/> <rect x="420" y="250" width="180" height="66" rx="4"/> </g> <g font-family="'Liberation Sans','DejaVu Sans',sans-serif" text-anchor="middle"> <text x="130" y="272" font-size="8" letter-spacing="0.5" fill="#43331f">DIALLYL DI- & TRI-SULPHIDE</text> <text x="130" y="287" font-size="7" letter-spacing="1" fill="#6f5a40">DADS · DATS</text> <text x="130" y="301" font-size="6.5" fill="#6f5a40">cooked-garlic aroma · garlic oil</text> <text x="320" y="272" font-size="9" letter-spacing="1" fill="#43331f">AJOENE</text> <text x="320" y="287" font-size="6.5" fill="#6f5a40">(E)- and (Z)-</text> <text x="320" y="301" font-size="6.5" fill="#6f5a40">oil & macerate · antiplatelet</text> <text x="510" y="272" font-size="8.5" letter-spacing="0.5" fill="#43331f">VINYLDITHIINS</text> <text x="510" y="287" font-size="6.5" fill="#6f5a40">2-vinyl-4H-1,3-dithiin</text> <text x="510" y="301" font-size="6.5" fill="#6f5a40">oil-macerated · distilled</text> </g>  <line x1="120" y1="356" x2="520" y2="356" stroke="#a9906c" stroke-width="0.6"/> <text x="320" y="376" text-anchor="middle" font-size="8" letter-spacing="1" fill="#43331f" font-family="'Liberation Sans','DejaVu Sans',sans-serif">ONE ENZYME CUT — A CASCADE OF SULPHUR: FLAVOUR, AROMA, MEDICINE</text> </svg>

<figcaption><em>Plate 5. The sulphur cascade: the crush lets alliinase turn odourless alliin into pungent allicin, which at once begins to fall apart — into the diallyl di- and trisulphides of cooked garlic and, in oil, into the vinyldithiins and ajoene of a confit; on a separate branch, where the wounded clove's sulphur compounds meet its own amino acids, it builds the harmless blue-green pyrrole pigments of pickled garlic.</em></figcaption> </figure>

A Chemistry Built for Defence

It is worth pausing on the oddity of an arrangement so elaborate. A plant does not build a two-part chemical weapon, and keep the halves in separate rooms at some real metabolic expense, for the benefit of cooks. The alliin-and-alliinase system is a defence, and a sophisticated one: a binary weapon, harmless while its two components are stored apart and caustic the instant they are mixed by the bite of a grub or the invasion of a soil fungus. The allicin that floods a wound is a genuine biocide — it was, after all, first isolated for its power to kill bacteria — and in the broken clove it goes to work on whatever did the breaking, punishing the herbivore and disinfecting the injury against the moulds and microbes of the soil in which the dormant bulb must survive for months. Garlic is not alone in the trick. The mustards, horseradish and wasabi run an almost identical two-component system with a different set of chemicals — the "mustard-oil bomb," in which an enzyme meets a stored precursor on wounding and releases the hot oils of the brassica — a case of two quite unrelated plant families arriving, separately, at the very same idea: keep the weapon in pieces, and assemble it only in the wound. That the flavours these defences throw off should be, to the human palate, not repellent but irresistible is one of the stranger jokes of the table. We have taken two of the plant kingdom's most determined chemical deterrents, the allium's and the brassica's, and built whole cuisines on the privilege of setting them off.

How allicin actually does its damage is worth a paragraph of its own, because the mechanism is the same whether the target is a fungus in the soil or a bacterium in a laboratory dish, and it explains the compound's very instability. Allicin is voracious for one particular chemical group — the thiol, or sulphhydryl group, the reactive –SH that sits in the amino acid cysteine and studs the working parts of countless proteins. Meeting a thiol, allicin swaps onto it in a quick exchange, hanging an allyl-sulphur tag where a bare –SH used to be; and because so many enzymes carry a crucial cysteine in their active site, that single kind of reaction can jam a whole array of them at once, strip the cell of its protective store of thiols, and tip it into what the microbiologists call thiol stress. This is why allicin is at once a broad biocide and a fleeting one: it reacts with almost any thiol it meets, the ordinary molecules of a cell included, so the very greed that lets it disable a microbe also ensures that it is spent quickly and cannot keep. What that reactivity is worth to human medicine — whether enough intact allicin could ever reach a living infection to matter — is a separate and far harder question, and the book saves it for the medicine chapter; here it is enough to see the chemistry of the wound for what it is, a real chemical defence that works by sabotaging the enzymes of whatever bites.

Why the Cut Matters

If flavour is made by injury, then the degree of injury sets the strength of the flavour, and this is the single most useful thing the kitchen can take from the chemistry. The rule is simple: the more thoroughly the cells are broken, the more alliin meets alliinase, the more allicin is made, and the sharper and more aggressive the result. A clove put through a press or pounded to a paste in a mortar has had almost every cell ruptured, and it is correspondingly the fiercest form garlic can take — raw garlic paste is a different order of intensity from raw garlic anything-else. A finely minced clove is strong but a shade less so. A sliced clove, its cells cut only along the faces of the slices, is gentler and cleaner. And a clove left whole, merely peeled, has scarcely been wounded at all, and will give up only a mild, sweet, background note however long it cooks. One bulb, therefore, is a whole spectrum of ingredients, and the cook selects along it not by choosing a different garlic but by choosing a different knife.

There is a second, subtler lever, and it is a matter of time. Because allicin has to be manufactured after the cutting, the freshly crushed clove is still, for a little while, a work in progress — the enzyme is mid-task. The kitchen-science advice to crush or chop garlic and then let it stand for about ten minutes before cooking rests on exactly this: the pause gives alliinase time to convert its full complement of alliin into allicin while the enzyme is still intact and active, so that a fuller, more complete flavour is developed before heat or acid can intervene. The point is not folklore. It emerges from work on garlic's biochemistry showing that heating garlic promptly, before the enzyme has finished, sharply curtails the allyl-sulphur compounds it can produce, whereas allowing a rest before heating preserves much more of them.[^1] The considerate cook, then, treats a crushed clove a little like a decanted wine — cut it first, let it breathe, and only then bring it to the pan.

This single spectrum, from the whole clove to the pounded paste, is quietly one of the most powerful controls in the kitchen, and the skilled cook plays both ends of it within one dish. A few whole or halved cloves dropped into a braise at the start will melt over the hours into a sweet, almost invisible background, their cells barely broken and their alliinase soon stilled by the heat; a clove of the same head, pounded raw to a paste and stirred through the finished dish at the last moment, will strike a bright, aggressive top note that the long cooking could never have produced. Nothing has changed but the timing and the force of the cutting — yet the two treatments deliver flavours so different that a blind taster might not guess they came from the same bulb. To cook well with garlic is, in large part, simply to decide how much of a wound to inflict, and when.

One thing no knife can control is how much alliin the clove holds to begin with, and that is set by the plant's age. Young "green" or "wet" garlic — the whole soft plant pulled in spring before its bulb has dried and divided — and the tender scapes of early summer carry far less of the stored precursor than a mature, cured clove, and are milder for it: sweet, grassy, only faintly garlicky, a different and gentler thing than the same plant will become by autumn. The pungency this chapter has been anatomising is, in part, simply the flavour of a bulb that has finished its year, banked its store, and gone dormant. Spring garlic is the same weapon, only half-loaded.

What Heat Does

Heat changes garlic more profoundly than any knife, and it does so by attacking the machinery itself. Alliinase is a protein, and proteins are undone by heat; a clove that is heated before it is cut, or crushed and then thrown at once into a hot pan, has its enzyme denatured before it can do very much, and the alliin it holds is largely left unconverted. The result is that little fresh allicin is ever made, and the savage pungency of the raw clove simply never appears. This is the whole secret of mellow, sweet, cooked garlic, and it explains the enormous distance between a spoonful of raw garlic paste and a head of garlic roasted slow and whole until it can be squeezed out like paste of another kind entirely.

For roasting does two further things as it silences the enzyme. It drives off moisture, concentrating what remains; and, more importantly, it turns the garlic's stored carbohydrates — the long fructan chains in which the bulb banks its energy — into simpler sugars, which then brown. That browning is the Maillard reaction, the same web of reactions between sugars and amino acids that gilds a crust of bread and sears a steak, and it is what gives roasted garlic its deep, nutty, caramel sweetness and its soft brown flesh. None of this involves allicin, which is why roasted garlic does not taste like a milder version of raw garlic but like a genuinely different food — sweet where the raw was hot, round where the raw was sharp, gentle enough to spread on toast where the raw would strip the paint. The naturalist should hold on to this as one of the deepest truths of the ingredient: raw garlic and cooked garlic are not two strengths of one flavour but two distinct substances, made by two different chemistries, from the same clove. To cook garlic is not to weaken it. It is to change what it is.

The Sugar in the Store

If sulphur is what garlic is famous for, sugar is what it is mostly made of. Set aside its water and the clove is, by dry weight, overwhelmingly carbohydrate — and not the starch that fills a potato but fructans, long chains of fructose of the inulin family, which make up the great bulk of the dry matter of the bulb. This is the "store" that the bulb exists to be: fructans are how an allium banks its energy against the dormant season and the sprint of spring, and the genes that build them were among those the 2020 genome finally spelled out. They account for more of garlic than one might guess. The slow sweetness that heat coaxes from a roasting head is these fructans breaking down under the warmth into simple sugars. And the notorious effect of garlic and onions on the human gut has the very same source: fructans are FODMAPs, carbohydrates that the small intestine cannot absorb and that pass on into the colon to be fermented by its bacteria — the honest, unglamorous chemistry behind garlic's reputation for troubling sensitive stomachs, and the reason a low-FODMAP diet strikes the whole allium family from the plate. The clove is a sugar-store armed with a sulphur-weapon: a pantry that bites.

Garlic Dried and Powdered

Between the raw clove and the cooked lies a third state the food industry leans on heavily: garlic dried to a powder. It sounds like a way of killing the chemistry, and done carelessly it is; but done gently it preserves it. If the cloves are sliced and dried at a low enough temperature, both halves of the weapon come through the loss of water intact — the stored alliin and, crucially, the alliinase enzyme — so that the powder is not spent but merely paused. Add water to good garlic powder and the rehydrated enzyme falls upon the alliin exactly as it would in a crushed clove, generating fresh allicin and the true pungency; this is why a pinch of proper dried garlic, or of garlic salt, can still assert itself, and why the better products tell the cook to let the moistened powder stand a moment before it meets heat, for precisely the reason a fresh clove is rested. Dry it too hot, though, and the alliinase is denatured before it is ever called on, leaving a powder that holds alliin but can no longer convert it — flat, faintly sweet, and missing the bite. The same fork governs the supplement shelf, where a tablet's worth turns entirely on whether its enzyme survived the making, a matter the medicine chapter will have to weigh with some care.

Why the Smell Outstays the Meal

Garlic is famous for outstaying its welcome, and the reason lies once more in the sulphur cascade — but in a compound that travels. A garlicky meal loads the breath with several odour compounds at once, and most are dealt with quickly; worked on in the mouth and the gut, they fade within an hour or two. One does not. Allyl methyl sulphide, or AMS, the body metabolises poorly, and rather than being broken down it is absorbed into the bloodstream and carried around the body, to be released slowly through the lungs, the skin and the kidneys over many hours — in a thorough garlic session, for as long as a day or two. This is why true garlic breath cannot be brushed away. A quantity of the odour after a garlicky meal is arriving not from residue on the teeth but from the lungs, exhaled steadily from within as the blood delivers its cargo of AMS to be breathed off; the mouthwash reaches the mouth, but the smell is coming from the chest. It reaches further even than the breath. A study by Mennella and Beauchamp in the early 1990s found that when nursing mothers ate garlic the odour passed into their breast milk — and, curiously, that the infants nursed the longer for it, a first, unwitting lesson in the family flavour.

The split — some of the smell in the mouth, some arriving from the blood — is why the folk cures for garlic breath only ever half work. In 2016 Rita Mirondo and Sheryl Barringer at Ohio State put several of them to the test and found the old remedies real: raw apple, raw lettuce and fresh mint each cut the garlic volatiles on the breath sharply, because the enzymes and the phenolic compounds in the raw foods chemically attack and destroy the odour molecules — the raw foods working better than the cooked, because heat had already spent their enzymes. But an apple can reach only the smell still in the mouth. The allyl methyl sulphide already circulating in the blood is beyond its help, and will go on being breathed off from the lungs for hours whatever the diner chews. Which is the plain and slightly comic truth of garlic breath: the mouth can be cleaned in a minute, but the rest must simply be waited out — exhaled slowly by a body that has, for a day, faintly become what it ate.

The Black Bulb

At the opposite temperament from that volatile afterlife lies the strangest thing garlic can be made to become: black garlic. Set whole heads of ordinary garlic in a warm, humid place and simply keep them there — at a gentle heat, commonly somewhere between about sixty and ninety degrees Celsius, at a high humidity of eighty to ninety per cent, for a matter of weeks rather than hours, anywhere from a fortnight to three months — and the bulbs slowly transform. The cloves turn soft and jammy, deepen from ivory through amber to a glossy jet black, and lose their heat entirely, taking on instead a dark, sweet, balsamic, faintly liquorice richness with a tang of tamarind or aged fruit. It is one of the most complete transformations in the whole of cookery, and it is routinely misdescribed. Black garlic is almost always called "fermented," and it is essentially not: no culture is added, and the change is driven not by microbes but by the same non-enzymatic Maillard browning that colours roasted garlic, run very slow and very long, converting the bulb's sugars and amino acids into the dark melanoidin pigments and deep flavours over its patient weeks.[^2]

As it darkens, it also grows — by the measures chemists use — markedly more antioxidant than the raw clove it began as. The same Maillard chemistry that blackens the flesh throws off dark melanoidin polymers and intermediate compounds such as 5-hydroxymethylfurfural, while the long warmth frees bound phenolic compounds from the softening tissue; the polyphenol content climbs, and with it the readings on the assays chemists use to gauge antioxidant activity, so that black garlic can test several times higher than the fresh clove — one of the few cases in which cooking a food at length appears to concentrate such activity rather than squander it. There is no fresh allicin here, of course — the enzyme was denatured on the first warm day — and so black garlic completes the argument of this chapter from the far side. Raw garlic is a wound; roasted garlic is that wound prevented and sweetened by an afternoon's heat; and black garlic is the bulb coaxed, over weeks of warmth, into a wholly new and gentle substance in which nothing of the original pungency survives at all.

<figure> <svg xmlns="http://www.w3.org/2000/svg" width="100%" viewBox="0 0 640 460" role="img"> <title>Black Garlic — the white-to-black transformation</title> <rect x="18" y="18" width="604" height="424" fill="none" stroke="#43331f" stroke-width="1.2"/> <rect x="24" y="24" width="592" height="412" fill="none" stroke="#6f5a40" stroke-width="0.6"/> <path d="M18 42 L18 18 L42 18" fill="none" stroke="#6d5577" stroke-width="1.3"/>

<g fill="none" stroke="#43331f" stroke-width="1.1" stroke-linecap="round"> <path d="M150 150 C130 168 122 210 132 252 C140 286 164 306 164 306 C164 306 188 286 196 252 C206 210 198 168 178 150 C172 140 156 140 150 150 Z"/> <path d="M164 150 C163 128 166 112 164 96" stroke-width="0.9"/> </g> <g fill="none" stroke="#6f5a40" stroke-width="0.7"><path d="M164 158 C162 212 162 268 164 302"/><path d="M142 172 C134 216 138 268 156 302"/><path d="M186 172 C194 216 190 268 172 302"/></g> <g fill="none" stroke="#a9906c" stroke-width="0.5"><path d="M152 196 C148 240 150 280 158 300"/><path d="M176 196 C180 240 178 280 170 300"/></g> <path d="M146 306 C154 316 174 316 182 306" fill="none" stroke="#43331f" stroke-width="0.9"/>  <g fill="none" stroke="#43331f" stroke-width="1" stroke-linecap="round"> <path d="M120 336 C110 344 108 366 114 384 C120 400 134 404 142 398 C150 390 150 368 144 350 C138 336 128 330 122 334 C121 334 120 335 120 336 Z"/> <path d="M130 342 C124 360 128 382 138 394" stroke="#6f5a40" stroke-width="0.6"/> </g> <g fill="none" stroke="#c8b79a" stroke-width="0.5"><path d="M118 356 C116 374 120 390 130 396"/></g>

<g fill="none" stroke="#43331f" stroke-width="1.2" stroke-linecap="round"> <path d="M470 150 C450 168 442 210 452 252 C460 286 484 306 484 306 C484 306 508 286 516 252 C526 210 518 168 498 150 C492 140 476 140 470 150 Z"/> <path d="M484 150 C483 128 486 112 484 96" stroke-width="0.9"/> </g>  <g fill="none" stroke="#43331f" stroke-width="0.8"> <path d="M462 168 C456 214 460 268 476 302"/> <path d="M474 158 C472 212 472 268 474 304"/> <path d="M486 156 C488 210 488 268 486 304"/> <path d="M498 168 C504 214 500 268 484 302"/> <path d="M456 200 C452 244 456 288 470 306"/> <path d="M512 200 C516 244 512 288 498 306"/> </g> <g fill="none" stroke="#6f5a40" stroke-width="0.6"> <path d="M468 188 C464 234 468 282 480 304"/> <path d="M492 188 C496 234 492 282 480 304"/> <path d="M480 176 C480 232 480 282 484 306"/> </g> <path d="M466 306 C474 316 494 316 502 306" fill="none" stroke="#43331f" stroke-width="1"/>  <g fill="none" stroke="#43331f" stroke-width="1" stroke-linecap="round"> <path d="M440 336 C430 344 428 366 434 384 C440 400 454 404 462 398 C470 390 470 368 464 350 C458 336 448 330 442 334 C441 334 440 335 440 336 Z"/> </g> <g fill="none" stroke="#43331f" stroke-width="0.6"><path d="M438 350 C434 370 438 388 448 396"/><path d="M452 346 C456 366 452 386 444 396"/><path d="M446 342 C444 366 446 386 450 396"/></g>

<g fill="none" stroke="#6d5577" stroke-width="1.2" stroke-linecap="round"> <path d="M256 226 L392 226"/> <path d="M392 226 L380 219 M392 226 L380 233"/> </g> <text x="324" y="214" text-anchor="middle" font-family="'Liberation Sans','DejaVu Sans',sans-serif" font-size="10" letter-spacing="2" fill="#6d5577">WEEKS OF GENTLE HEAT</text> <text x="324" y="250" text-anchor="middle" font-family="'Liberation Sans','DejaVu Sans',sans-serif" font-size="8.5" letter-spacing="1.5" fill="#6f5a40" font-style="italic">Maillard browning, not fermentation</text>

<g font-family="'Liberation Sans','DejaVu Sans',sans-serif" font-size="11" letter-spacing="3" fill="#43331f" text-anchor="middle"> <text x="150" y="428">FRESH · WHITE</text> <text x="484" y="428">AGED · BLACK</text> </g> <text x="320" y="42" text-anchor="middle" font-family="'Liberation Sans','DejaVu Sans',sans-serif" font-size="13" letter-spacing="4" fill="#43331f">BLACK GARLIC</text> </svg>

<figcaption><em>Plate 6. Black garlic: a whole bulb and a cut clove showing the change from ivory to jet, wrought not by fermentation but by weeks of gentle heat and the slow Maillard browning of the bulb's own sugars and amino acids.</em></figcaption> </figure>

A Quieter Store: S-Allylcysteine

Everything so far has followed the loud pathway — alliin, the enzyme, allicin and its volatile children — because that is the pathway of flavour and of the knife. But the clove keeps a second, quieter store of sulphur that never detonates, and it matters more to the medicine than to the meal. Alongside its alliin, garlic holds a set of stable sulphur-bearing peptides, the γ-glutamyl-cysteines, which the enzyme leaves untouched and the nose never detects. Leave crushed garlic to stand for months in a weak spirit, as the makers of standardised "aged garlic extract" do, or hold whole bulbs at a gentle warmth for weeks, as the makers of black garlic do, and these quiet peptides convert, slowly, into a compound called S-allylcysteine — odourless, stable, freely soluble in water, and, unlike allicin, well absorbed by the body when it is eaten. S-allylcysteine is the marker the laboratories reach for when they want to measure garlic in the blood, precisely because allicin itself is far too fleeting to survive the journey from gut to bloodstream; it is the compound that most of the aged-garlic health studies are, in truth, really about, and the reader will meet it again in the medicine chapter. For now it is enough to notice that the same bulb runs two chemistries at once — a fast, savage, volatile one that the cook wants, and a slow, mild, durable one that the pharmacologist does — and that ageing garlic, whether into a sweet black clove or a standardised capsule, is in large part simply the business of trading the first chemistry for the second.

The Blue-Green Bulb

One last transformation deserves its place here, because it alarms more cooks than it should and because it is pure sulphur chemistry caught in the act of making a colour. Peel and slice a clove of young garlic, drop it into vinegar or lemon, or cook it with something acid, and it may turn — sometimes within minutes, sometimes overnight — a startling blue or a clear jade green. Pickled garlic does it most reliably of all; whole cultures know the phenomenon and prize it, and the Chinese Laba garlic, steeped in vinegar for the new year, is valued precisely for turning a translucent emerald. It is entirely harmless, and it is not — whatever the kitchen myth insists — a sign of copper contamination from the pan; it happens as readily in glass. It is the wound chemistry again, taking a side road. When the clove is cut, the same reactive sulphur compounds that carry the flavour can react instead with the clove's own amino acids to build small nitrogen-containing rings — pyrroles — and these link up, several at a time, into larger pigment molecules: chemists who have taken the colour apart report that chains of three ring blue and chains of four tip toward green, a whole family of them together, so that greened garlic has been found to hold at least eight distinct blue pigments blended into its final colour. The reaction is favoured by young or immature garlic, richer in the precursors, and hurried along by acid and by a spell of cold storage, which is why the spring's pickled cloves green so eagerly and the autumn's dry supermarket head so seldom. It is the same kind of reaction, in the end, that reddens a bruise and browns a cut page — a colour built by chemistry out of an injury — and the cook who meets it has simply caught garlic doing in pigment what this whole chapter has watched it do in scent.

The chemistry, then, gives the cook a strange and powerful instrument: one bulb that can be made savage or sweet, fleeting or everlasting, coloured or clear, according to how it is cut and how it is heated, because its flavour is not stored but synthesised on demand and then allowed — or not — to decay. That such a precise and modern-sounding account of a clove should be the true one makes the next turn of the book all the more remarkable. For the plant whose wound-chemistry we have only lately learned to name was being crushed, prized, rationed and feared for four thousand years before anyone had heard of an enzyme — and it is to that long human history, older than the science by forty centuries, that the compendium goes next.

[^1]: The best-known demonstration is K. Song and J. A. Milner, "The Influence of Heating on the Anticancer Properties of Garlic," Journal of Nutrition 131 (2001): 1054S–1057S, which found that microwaving crushed garlic immediately destroyed its measured protective activity in a rat model, while allowing the crushed garlic to stand about ten minutes before heating restored much of it. The health endpoint was an animal-model result and should not be over-read; the underlying enzyme mechanism — that heat denatures alliinase, and that resting lets the enzyme finish first — is established biochemistry and is the load-bearing point here.

[^2]: A minor qualification, in keeping with this volume's evidence rule: some bacteria naturally present inside the bulb may assist in breaking down its fructans in the earliest stage, so "no microbial involvement whatever" would be too strong. But the browning and the characteristic flavour and colour of black garlic are non-enzymatic Maillard chemistry, not a true culture-driven fermentation, and the common label "fermented black garlic" is, strictly, a misnomer.

บทถัดไป · 6 / 15Chapter 3 — A History Told in Cloves