Direct answer: Chinese incense is a compositional craft before it is a shopping category. A 合香 (hé xiāng) formula hands each aromatic one of four jobs — 君 chief, 臣 minister, 佐 assistant, 使 guide — and those jobs were borrowed from herbal prescribing, which is why the oldest surviving incense recipes read like prescriptions instead of fragrance briefs. A compound is judged on whether the four jobs are filled, not on how lovely any single material is on its own.
Key facts
Direct answer: Four job titles, one dominant aromatic, one supporting dose, one signature note, and one binding step or process.
| Term | What it names | Why it decides how you read a formula |
|---|---|---|
| 合香 | Blending several aromatics into one designed scent | The unit of the craft is a design, never a single material |
| 君臣佐使 | Chief, minister, assistant, guide | Four jobs, taken over whole from herbal practice |
| 单方 | One aromatic on its own | The other pole: judged on material quality rather than arrangement |
| 两, 钱, 分 | Base weight, its tenth, its hundredth | The ladder a recipe has to be read from |
| 枚 | Counted pieces | The tell that a whole fresh component enters the mix |
Two rules follow. A material is picked for its job inside the arrangement, not for how it smells alone. And the arrangement is judged on whether the four jobs are filled, a different test from whether the ingredients are precious.
What the Word Commits You To
Direct answer: 合香 names a method, not a smell. Two formulas can share no materials at all and still belong to the same craft, because what they hold in common is the way the parts get chosen.
Burning one material asks a question with a short answer: is this piece good, and how good? The whole exercise is a ranking, and the better wood wins. A compound asks something that cannot be settled by ranking: do these parts hold each other up? A modest wood can outshine an expensive one when it sits inside a smarter arrangement, and an expensive resin can disappear into a compound that never gives it room. The craft is not the mixing. It is the arrangement.
That shift in the question is why the classical manuals read as cookbooks. The 香谱 collections, gathered across the Song and Ming, are full of formulas with doses and steps, because the thing worth writing down was the structure. A list of nice-smelling materials would have been useless to a reader who already knew which materials were nice. What the reader could not invent for himself was the proportion and the order of operations.
English-language incense writing has a vocabulary problem here. The market sorted its products by material name, one word to one smell: sandalwood, lavender, patchouli. That naming habit quietly assumes the material is the unit of meaning. 合香 breaks the assumption, because its unit is the formula. A reader trained on single-note names has no slot for the sentence “sandalwood is the minister,” so the compound tradition collapses into “sandalwood incense,” and the structure — the only part that was hard to get right — vanishes in translation.
The practical cost of that collapse is easy to state. A buyer who reads a compound as if it were a single material will sort it by its most expensive ingredient, which is the one thing the old makers did not do. Cost and rank run on different axes, and the rest of this page keeps coming back to that disagreement.
Four Jobs and the Question Each One Answers
Direct answer: Read a formula one role at a time, and each role reduces to a question. Chief: what carries it? Minister: what holds its edges? Assistant: what names it? Guide: what makes it burnable and keeps it stable?
| Role | The job | The question it answers | What fails when the role is empty |
|---|---|---|---|
| 君 chief | Supplies the aroma present from the first minute to the last | What carries the compound? | The blend opens brightly and dies inside ten minutes |
| 臣 minister | Rounds the chief at a fraction of its dose | What holds the chief’s edges? | The chief reads thin, or sharp at the corners |
| 佐 assistant | Adds the one note people remember | What names the compound? | The blend is pleasant and completely anonymous |
| 使 guide | Binds the powder, fixes the scent, or supplies a process step | What makes it burnable and stable? | The mixture crumbles, or the top note walks off in storage |
The four roles form a hierarchy, and the hierarchy is not a ranking of quality. The minister exists to serve the chief. The assistant exists to repair what the chief and minister together still lack. The guide exists to hold the result in a shape that will actually burn. Read the roles as a chain of service and a formula stops looking like a shopping list; it starts looking like an argument about what matters most in that particular scent.
The guide deserves a second look, because it is the role beginners skip. A guide can be honey, which binds powder into pills. It can be musk, which fixes volatile material and slows its escape. It can be 甲香, the operculum of a sea snail, used in the tradition as a fixative. It can also be nothing you can weigh at all: a steaming step, a resting period, a coat of dried flowers on the outside of a pill. That last possibility is the useful shock. A role in a formula does not have to be an ingredient.
The Doses Do the Sorting
Direct answer: Roles are not labels pasted on after the fact. They fall out of proportion, which means a formula can be read as a weight ladder before it is read as a story.
Follow the chain. A maker doses one material highest → that aroma arrives first and stays longest → the nose files it as the base of the compound → the tradition records that result with the name 君. The next-largest dose fills in the body the chief leaves empty → it never leads, but its absence would be felt as a hole → it gets recorded as 臣. A small, bright dose goes in where the first two leave a gap, and it registers precisely because it is scarce → it becomes the note the compound is known by → 佐. A trace of something sticky, or one more processing step, keeps the whole thing in a burnable shape → 使.
The doses do the sorting; the names only record it. Two formulas with the same four materials and different proportions are two different formulas, not one formula written twice. This is the single most useful idea on the page for anyone trying to read an unfamiliar recipe, because it means you can reverse-engineer the roles from the numbers before you understand a single ingredient name.
It also settles a beginner’s question about whether the chief must be the costly material. It does not, and the assumption causes real damage to how compounds get judged. Rank is a statement about proportion and function. Price is a statement about scarcity and trade. A precious resin can be dosed as a trace minister and a humble wood can carry the whole compound. When those two axes get confused, the buyer pays for the label and the nose gets a formula that was never built that way.
What the Guide Does to a Burn
Direct answer: The guide decides whether the compound survives contact with heat. A binder sets the physical shape, a fixative slows the escape of the volatile fraction, and a process step changes the material before a flame is ever involved.
Powder does not burn as a heap. A loose heap of milled heartwood and resin tears smoulders unevenly, channels along one wall, and dies. Something has to hold the granules in contact long enough for the ember line to travel the length of the piece: honey, 抹香 (makko, a bark powder used across East Asia as a combustible binder), or a resin that softens as it warms.
The chain runs like this: loose powder → voids between the granules → the combustion front stalls or races → a binder closes those voids → the stick or pill burns at a rate the maker can predict. Change the binder fraction and you change the burn before you change the smell, which is why a compound that misbehaves on the plate rarely has a scent problem.
Fixatives work on a different principle. Musk, 甲香, and amber appear in the guide position because the tradition holds that they slow the loss of the lighter fraction, so the opening note is still there after weeks of storage rather than evaporating out of the box. That is traditional reasoning about material behaviour, not a measured claim, and this page will not dress it up as one.
Process steps are the third kind of guide, and the hardest to see. Steaming, coating, and the long rest written as 久窨 do not add a material; they change the condition of the material that is already there. A rest period lets moisture equilibrate through the mass, so a hygroscopic powder behaves the same on the day it is extruded and on the day it is lit. Whether anything beyond that happens inside the jar is a question this page leaves open.
Three Formulas, Opened Up
Direct answer: Three worked examples — one classical, one Song-attributed, one modern — show the same four jobs filled by different materials, and show that the assistant, not the chief, is the line that supplies the name.
Proportions survive in several versions and use historical units. Read the structure rather than the gram count, and treat the material below as cultural history, never as instruction for a physical effect.
| 江南李主帐中香 | 雪中春信 | 二苏旧局 | |
|---|---|---|---|
| Attributed to | 李煜 (Li Yu), last ruler of the Southern Tang; Five Dynasties–Song, recorded in 香乘 in several versions | 苏轼 (Su Shi) — traditional belief, not documented; Song, recorded in 香乘 | 陈云君, 燕居香语; contemporary revival |
| Variants | Proportions differ between editions | Some versions substitute 棧香 and 杉木炭 | A modern composition, not a 香谱 text |
| Chief 君 | 沉香 agarwood, one 两 | 檀香 sandalwood | 沉香 agarwood |
| Minister 臣 | 檀香 sandalwood, one 钱 | 香附子 cyperus with 鬱金 turmeric | 檀香 sandalwood |
| Assistant 佐 | 鵞梨 goose pear, ten pieces | 樟腦 camphor | 乳香 frankincense with 琥珀 amber |
| Guide 使 | 蒸三溜 steaming, 久窨 resting | 麝香 musk, 羊脛灰 bone ash, honey | 蜂蜜 honey, 茉莉 jasmine coat |
| Method | Hollow the pears, pack in the powders, cap, steam three rounds, peel, grind, rest | Grind fine, bind with refined honey, rest, then burn | Dice the woods, mill the resins, bind in jujube-flower honey, roll into pills, dust with jasmine, dry |
The pear is a vessel. Follow the process and the design becomes obvious: powder packed into a cored pear → three rounds of steam → the fruit softens and its juice wets the powder through → the skin is peeled and the mass reground → the sweet note ends up inside the wood instead of sitting on top of it. Set the doses beside that process and the economy is the point: ten pears and one 两 of agarwood against a single 钱 of sandalwood. The chief outweighs the minister by an order of magnitude, which is the difference between agarwood-with-a-lift and a muddle of two woods.
The camphor is the name. The same reading works in reverse: camphor dosed at a trace → a cool flash at the opening → the nose keeps the flash longer than it keeps the wood underneath → the compound is remembered and sold as Snow in Spring rather than as sandalwood-with-herbs. The smallest line produced the identity, and the largest line produced the body.
The assistant can be split. In the third formula frankincense opens the scent while amber settles it, two halves of one job pulling in opposite directions. No other example makes the point as cleanly: a role is a function, not a headcount, and “assistant” means whatever the opening needs at that moment.
Three forms, three centuries, one shape: heartwood carries, a lighter material rounds it at a fraction of the dose, a scarce note names the piece, and a binder or a process holds it together. A learner who sees the one shape behind the three examples has the craft, not just three facts.
Units Before Numbers
Direct answer: The unit words carry more information than the digits. They tell you whether a material was weighed, counted, or never put on the scale at all.
The inventory of units is short. 两 (liǎng) is the base weight, 钱 (qián) its tenth, 分 (fēn) its hundredth. 枚 (méi) counts whole objects — pears, plums, pieces of fruit — which tells you the component entered fresh rather than as powder. 份 (fèn) marks a portion, a looser instruction that assumes the maker knows the batch size. The absence of any unit, or a phrase such as 久窨 (“rest a long time”), marks a process: a step measured in hours rather than grams.
Chain that forward and the unit becomes a sorting signal: the unit word is read first → it places the line into weighed, counted, or processed → a 两 dose outranks a 钱 dose before you have parsed a single digit → the skeleton of the formula appears before the ingredient list does. A reader who jumps straight to the numbers reads the formula half-blind, because a line reading “十枚” is not a small quantity; it is ten whole pears.
The gram question is where modern readers get stuck. The 两 drifted across dynasties — on the order of 30–40 g in the later ones, with about 37 g often cited for the Song–Ming period — so the figure is a convention, not a certificate. What survives intact is the relationship: one 两 of agarwood against one 钱 of sandalwood is a 10:1 ratio, and that ratio is the meaning of the recipe. Weighing the ink instead of reading the sentence is the failure mode; the proportion is the argument, and the gram weight moved with every change of dynasty.
Reading a Formula You Cannot Translate
Direct answer: The language is not the barrier it looks like. Four moves get most of the way: find the largest number, find the smallest distinctive line, find anything that is a step rather than a material, and compare a second surviving version.
The largest number belongs to the chief with almost no exceptions, because the highest dose is what lands first and lasts longest. The smallest line that still carries a vivid character — a cool resin, a fruit, a single flower — is usually the assistant, and it usually explains why the piece has the name it has. Lines that read like verbs rather than nouns (steam three times, rest long, coat the outside, bind with honey) are the guide, and they are easy to mistake for footnotes because they carry no material to look up.
The fourth move is the one that separates reading from guessing. Classical formulas survive in several editions, and the editions disagree about proportions. When two versions agree on which line is dosed highest and disagree on the exact figure, the agreement is the structure and the disagreement is the transmission. When they disagree about the rank as well, you are looking at two different formulas that share a title, which happens more often than a modern reader expects. The formulation library gathers recorded versions for exactly this reason, and the manual tradition explains why the editions diverge.
One caution about names. Old texts use literary and trade terms that do not map cleanly onto what a supplier lists today; a word that meant one grade in one century can mean another later, or a different material. A reader who finds nothing under a name has not necessarily mistranslated it. Naming instability is a feature of this material culture, not a failure of the dictionary.
Component Family to Likely Role
Direct answer: Component families map onto roles in a repeating pattern: woods lead, resins round, bright things name, binders hold. The pattern is a heuristic and it earns its keep because it fails in instructive ways.
| If the material is | It usually takes | Because |
|---|---|---|
| Agarwood or sandalwood heartwood | 君 chief | Slow, steady combustion and a base note that lasts the session |
| A lighter wood or a resin | 臣 minister | Rounds the chief at a fraction of the weight without competing |
| Camphor, a fruit, a bright floral, a spice | 佐 assistant | Supplies the one note that becomes the compound’s name |
| Honey, musk, makko 抹香, 甲香, a process step | 使 guide | Binds powder, slows volatile loss, or sets the material’s shape |
| Herbs and roots such as spikenard, angelica, cyperus | 臣 or 佐 | Herbal middle tones, rarely strong enough to carry alone |
The heuristic cuts both ways. Given a role to fill, you can guess the shelf: a chief from the woods, a signature from florals or fruit, a binder from the fixatives. That is what turns a grab-bag of pleasant smells into a sorted inventory where each shelf supplies a job.
Where it fails, it fails informatively. A resin can lead a resin-led compound, since frankincense and myrrh will carry a scent if the maker wants them to. A flower can be dosed as an assistant and still be the only thing anyone remembers. The pattern describes the common case, not a law, and a formula that breaks it is usually making a point.
The Inventory a Formula Draws From
Direct answer: The materials fall into six families, and the family predicts the job far more reliably than the individual scent does. Function comes before fragrance when a line is chosen.
| Material | Chinese | Family | Job it usually takes | What it does at the burner |
|---|---|---|---|---|
| Agarwood | 沉香 | Resinous heartwood | 君 chief | Burns slow and long, holds a deep base for the whole session |
| Sandalwood | 檀香 | Heartwood | 君 or 臣 | Creamy, steady, and quieter than agarwood, so it rounds rather than leads |
| Frankincense | 乳香 | Resin tear | 佐 assistant | Lifts the opening, adds a citrus-balsamic brightness |
| Myrrh | 没药 | Resin tear | 佐 or 使 | Bitter, dark, and fixing; used sparingly or it dominates |
| Benzoin | 安息香 | Resin | 使 guide | Sweet and sticky; binds and slows the escape of lighter notes |
| Clove | 丁香 | Spice bud | 佐 assistant | Sharp, warm, and unmistakable in tiny doses |
| Cassia | 桂皮 | Bark | 佐 or 臣 | Warm sweetness with a peppery edge, closer to cinnamon than to spice-heat |
| Fennel | 茴香 | Seed | 佐 | Green aniseed lift, easy to overdose |
| Spikenard | 甘松 | Root | 臣 minister | Earthy, woody-herbal body; a classic supporting tone |
| Angelica | 白芷 | Root | 臣 or 佐 | Dry herbal mid-note, bitter at the edges |
| Cyperus | 香附子 | Rhizome | 臣 minister | Warm, faintly nutty middle; the body of Snow in Spring |
| Turmeric | 鬱金 | Rhizome | 臣 | Bitter-warm, yellow, and more structural than it smells |
| Camphor | 樟腦 | Crystalline | 佐 assistant | Cold, sharp flash that can name a whole compound |
| Borneol | 龍腦 | Crystalline | 佐 | Cool and piercing like camphor, with a resinous tail |
| Musk | 麝香 | Animal fixative | 使 guide | Fixes and lifts in trace amounts; the classic finish |
| Amber | 琥珀 | Fossil resin | 佐 or 使 | Stillness and depth; pulls a bright blend downward |
| 甲香 | 甲香 | Marine operculum | 使 guide | Traditional fixative, often processed before use |
| Honey | 蜂蜜 | Binder | 使 guide | Binds powder into pills and holds the mass together |
| Makko | 抹香 | Bark powder | 使 guide | Combustible binder for sticks and coils, near-neutral in scent |
| Goose pear | 鵞梨 | Fresh fruit | 佐 assistant | Juice and sweetness, worked in by steaming |
| Jasmine | 茉莉 | Dried flower | 佐 assistant | Bright floral window at the top, usually as a coat |
| Plum | 白梅 | Fruit | 佐 | Sour-sweet top note, historically paired with woods |
| 棧香 | 棧香 | Agarwood grade | 君 or 臣 | A lesser agarwood used where a variant formula calls for it |
| 杉木炭 | 杉木炭 | Charcoal | 使 | Burns clean and hot in small proportion |
| Bone ash | 羊脛灰 | Mineral | 使 guide | Bulks the powder and keeps a pill from collapsing |
Read the far-right column and the pattern appears: the deep, slow materials sit at the top of the ladder, the bright and volatile ones supply identity, and the sticky or inert ones hold the shape. A maker shopping for a formula is really shopping for four behaviours, not four smells; the ingredients glossary covers the same shelf material by material.
Where the Scheme Came From, and Where It Stops
Direct answer: 君臣佐使 is not an incense invention. It came from herbal compounding, and incense borrowed the organizing logic while keeping a different purpose.
In a herbal formula the chief addresses the main condition, the minister strengthens that action, the assistant evens out an excess or covers a shortfall, and the guide steers the whole or smooths it into harmony. The foundational materia medica, 本草纲目 among them, record the scheme as an organizing principle rather than as a recipe.
The manual tradition that carries these formulas spans the Song (960–1279) and Ming (1368–1644) compilations, with the bedchamber formula sitting at the Five Dynasties (907–960) boundary and the materia medica scheme itself recorded far earlier.
Incense took the scheme because the two fields ran on the same premise: a composition is built by choosing each part for the job it performs inside the whole, never for its isolated effect. That shared premise is why old incense formulas come across as constructed objects. They were assembled, not improvised, and the assembly is the craft.
The borrowing is directional and narrow. The organizational logic came from medicine. The purpose did not. A herbal formula is arranged to act on the body; a compound is arranged to act on the nose. Same skeleton, different target. Holding that line is not pedantry: the closer the parallel looks, the easier the slide from “organized like a medicine” to “works like one.” A balanced compound is not thereby a healthful one, and the tradition’s own vocabulary never claimed otherwise. This is the author’s editorial position, and it is the boundary this page will not cross.
单方 and 复方: Two Disciplines, Not Two Grades
Direct answer: The tradition holds two poles. 单方 is one material appreciated for its own quality; 复方 is several materials composed into a structure. Neither is the advanced version of the other.
Single-material practice centres on grading and on heat. A chip of agarwood is judged for resin density, species, origin, and the way it behaves when warmed indirectly (隔火焚香), and the skill lies in the material and the technique. Compound practice centres on composition, and the skill lies in proportion and role. A person can be excellent at one and a beginner at the other.
The two poles are also two questions, and confusing them is the most common misread in the whole subject. 单方 asks: how good is this material? 复方 asks: how well is this compound built? A reader who only owns the first question will judge a compound by its most expensive line and miss everything the compound is doing. Learning to ask the second question is the actual work of this page.
Adjusting a Formula Without Breaking It
Direct answer: Change one line at a time, and change it in the direction the role allows. Raising the chief thickens the base; raising the assistant renames the compound; changing the guide alters how it burns, not what it smells like.
Some working rules, offered as craft judgement rather than measurement:
- More staying power. Raise the chief, or add a fixative to the guide position. Raising the minister instead usually muddies the middle without extending the tail.
- A brighter opening. Nudge the assistant, and warm the material more gently at the same time. The same dose turns harsh under more heat, and harsh is not the same as sharp.
- A cleaner burn. Work on the guide. Binder fraction, sieve aperture, and drying schedule decide whether a piece holds together; aromatic material cannot rescue a stick that will not stay lit.
- A substituted chief. Match the physical behaviour first — heartwood density, resin load, how the chips behave on a heated plate — and the aroma second. A substitute that smells plausible but burns twice as fast changes the formula even when the role name stays the same.
Watch what a single change does to the whole ladder. Raise the assistant by a noticeable margin → the opening sharpens → the nose attaches to that flash → the compound is now known by its new signature → the formula has been re-identified by an adjustment to one line. That is the least intuitive part of the craft: the smallest dose carries the most identity, so the smallest edit moves the name furthest.
Keep the ledger in ratios rather than grams. A ledger entry that says “chief 10, minister 1, assistant 0.3, guide to bind” survives a change of batch size; an entry in grams survives nothing.
Reading a Label That Claims 合香
Direct answer: A label that invokes the tradition should be readable by the tradition’s own questions. If it cannot say what carries the scent, what names it, and what holds it together, the word is decoration.
| What the label says | What it actually tells you | What it leaves out |
|---|---|---|
| “Agarwood blend” | A chief family, probably | Species, part of the tree, and whether agarwood leads or merely appears |
| A genus or species name | A checkable claim about material | Whether that material is the chief or a trace at the bottom of the list |
| “Hand-blended” | A claim about labour | Nothing at all about structure or proportion |
| A fragrance oil in the ingredients | A different product category | The material tradition the word 合香 was borrowed from |
| “Long-lasting” | An adjective | Burn time, binder share, and what the tail actually smells like |
| No ingredient list | Nothing | Everything that would let you read the compound |
The test is not whether the label is poetic. A short list that names a wood, a resin, and a binder, and says which is which, tells a reader more than a paragraph of atmosphere. The word 合香 describes a method, and a method leaves traces a buyer can check. A product that keeps the word and drops the traces is using the vocabulary of composition to sell a single note.
Six Ways Readers Misread a Formula
Direct answer: The four roles attract four specific mistakes, and two more come from the numbers. A table is the fastest way to unlearn them.
| Misreading | What it assumes | The correction |
|---|---|---|
| Cost equals rank | The priciest line is the chief | Rank is dose. A precious resin can be a trace minister |
| More materials means better | A longer list is a richer compound | Count roles, not components. Four clear answers beat thirty vague ones |
| The assistant is minor | The smallest dose matters least | The assistant is usually the name and the memory of the compound |
| The guide is scented | Every line contributes aroma | The guide often smells of nothing; it is what makes the rest burnable |
| The title names the chief | “Plum incense” is mostly plum | Titles come from the assistant. The chief is usually an anonymous wood |
| Old grams are exact | A classical formula can be reproduced to the gram | The ratio survives; the unit weight drifted by dynasty |
Two of those mistakes come from reading a composition as a collection, which rewards adding more and paying more. A composition rewards neither. The title mistake is the expensive one: a compound called “plum blossom” takes its name from the bright assistant while a plain wood does the carrying, so weighing the plum heavy produces a different design that keeps the name and loses the structure.
The Working Glossary
Direct answer: A dozen terms carry the whole subject, and each one earns its place because the English alternatives are vaguer.
| Term | Say it | What it means here |
|---|---|---|
| 合香 | hé xiāng | Blending several aromatics into one designed scent |
| 单方 | dān fāng | A single aromatic, valued on its own quality |
| 复方 | fù fāng | A multi-material compound, the strict sense of 合香 |
| 君 | jūn | Chief: the highest dose, the carrying aroma |
| 臣 | chén | Minister: the supporting dose that rounds the chief |
| 佐 | zuǒ | Assistant: the scarce, bright note that names the piece |
| 使 | shǐ | Guide: binder, fixative, or process step |
| 两 | liǎng | Historical base weight; the unit a formula is built on |
| 钱 | qián | One tenth of a 两 |
| 分 | fēn | One hundredth of a 两 |
| 枚 | méi | Counted pieces, used for whole fruit |
| 久窨 | jiǔ xūn | A long resting period after blending |
| 香谱 | xiāng pǔ | The manual tradition that records formulas |
Read pinyin and character together once and old recipes stop reading as code. The lexicon is deliberately small; the tradition worked it hard.
Checks Before You Buy a Compound
Direct answer: Six questions separate a formula from a fragrance with a story attached, and none of them requires expertise.
| What to ask | A good answer sounds like | A weak answer sounds like |
|---|---|---|
| Which material carries it? | A named wood, with the species and the part | “A blend of precious woods” |
| What names it? | A specific signature note and its source | “Inspired by ancient courts” |
| What binds it? | Makko, honey, or a named resin | Silence, or “natural binders” |
| Is the aroma material or fragrance? | A direct answer either way | “Premium fragrance oils” hidden in small print |
| How was it processed? | Steamed, rested, coated, with a timescale | “Handcrafted using traditional methods” |
| What does it weigh and how long does it burn? | Numbers | Adjectives |
The pattern in the middle column is disclosure. A seller who has built a compound knows what carries it, because the carrying material is the expensive decision; a seller who has bought a scented base and packaged it does not, and the language slides toward atmosphere. The three questions that matter most — carrier, signature, binder — are also the three the four roles already asked, which is the practical payoff of learning the scheme: it turns into a shopping script without anyone having to mention 君臣佐使 at the counter.
Why Two Readers Disagree About the Same Recipe
Direct answer: The recipe is stable; the reading is not. Units, editions, material names, and sensory vocabulary all introduce variation, and knowing which one is causing a disagreement usually settles it.
Start with the units. A 两 that weighed one thing in the Song and another in the Ming makes every gram conversion approximate, so two readers converting the same line can end up with different weights and both be faithful to their source.
Then the editions. Formulas survive in multiple versions that disagree on proportions, and a reader working from one printing will build a different compound from a reader working from another. When the rank of two lines swaps between editions, the disagreement is not about reading at all; it is about which text you are reading.
Material names drift as well. A term that meant one grade of wood in one century can mean a lesser grade later, or a different material in a different region, so a modern substitution made on the strength of a shared name can miss by a wide margin.
Sensory vocabulary is the least tractable source. “Cool”, “sweet”, “deep”, “bright” are comparatives, not measurements; two noses trained on different material will rank one compound differently along those axes. Such disputes yield to side-by-side tasting against a shared reference, never to argument.
What the Four Questions Cannot Tell You
Direct answer: The scheme reads architecture, and architecture is not quality. A well-built arrangement of mediocre material still smells of mediocre material, and no amount of role-reading will change that.
Four specific blind spots are worth naming, because each one is where confident readers get caught.
The first is material identity. Knowing that a line is the chief says nothing about what the chief actually is: species, part of the tree, age since harvest, resin load, whether the chips were stored dry or damp. A label can be structurally perfect and materially worthless. The framework describes how a maker distributed attention; it does not audit the ingredients that attention was spent on.
The second is burn behaviour. Whether a piece smoulders, channels, or goes out depends on things the role names never mention: the sieve aperture the powder passed through, the binder fraction, the drying schedule, and the relative humidity of the room on the day it is lit. Two batches with identical proportions can behave like different products.
The third is air. A formula tells you nothing about how much particulate it throws into a room, and the tradition’s own preference for warming over burning is partly a preference for less of it. Anyone reading old recipes for their own use should treat ventilation as a separate question with its own answer, not as something a formula settles.
The fourth is effect. The four questions sort a composition. They do not predict what it does to a body, and this page has no interest in pretending otherwise. That is the boundary, and it holds in both directions: a compound being well organized does not make it therapeutic, and a compound being therapeutic is not what the roles were for.
Storage, Rest, and the Limits of Aging
Direct answer: The rest written into old recipes is a real instruction with a modest, practical effect, and the claims made for long aging are mostly tradition rather than measurement.
The physical part is easy to accept, and none of it needs a laboratory. Freshly milled powder is hygroscopic and holds moisture unevenly; a sealed rest lets it equilibrate, so the batch behaves the same on the day it is rolled and on the day it is burned. A pill bound in honey also firms up as it dries, which changes how it sits on a mica plate and how slowly it gives up its scent. Both surface within a few batches, with one variable changed per batch.
The stronger claims are where caution belongs. Aging is said to round a compound, mellow its edges, marry its parts. Some of that is plausible: the lightest fraction of any aromatic is also the most mobile, and a sealed jar gives it time to spread through the mass instead of flashing off in the first week. How much rounding happens, and whether a given blend improves or merely changes, is not something this page can assert. Treat the improvement story as traditional belief and test it yourself.
What storage actually asks of you is unglamorous. Keep the batch in a sealed jar with a desiccant sachet in the headspace, out of daylight and away from radiators, clear of anything strongly scented, and write the date and sieve aperture on the lid. A floral coat, like the jasmine dusting on the third formula above, is the first thing to fade, because it sits on the surface with nothing above it and nothing to anchor it. A pill that has lost its coat has not spoiled; it has simply become a quieter version of itself.
The test is worth running once. Split one batch into two small portions, burn one now and one after a few weeks, and compare them side by side rather than from memory. Whatever the tradition says about patience, that comparison is the only evidence you will have about your own material, in your own room, at your own humidity.
A First Attempt Worth Making
Direct answer: Start with three materials and one variable. The first blend is a lesson in adjusting one thing at a time, and a large first project teaches nothing except how to hide mistakes.
The smallest experiment that still has a structure is a carrying wood, a supporting material at a fraction of its weight, and one bright note. Warm the three together in a rough proportion, then change one amount and notice what moved. Reproducing a classical formula is not the goal at this stage. Feeling what “balance” means in your own hands is.
Two habits make the practice stick. Keep the materials ordinary enough to waste, so that caution does not replace curiosity. And run comparisons in pairs rather than in sequence, because the nose holds a contrast far better than it holds an absolute. A ledger does the rest: what changed, by how much, whether the compound read as one smell or as several, and what the sieve and the hygrometer said that day.
The equipment question is smaller than the market suggests. A heatproof dish, a gentle heat source, and a small scoop cover a beginner’s needs; presses, extruders, balances, and dedicated burners are conveniences for people who already know what they are adjusting. The craft lives in the balancing, not in the hardware.
One test settles most disputes about a batch. Before lighting it, write down which material carries, which one names the piece, and what holds it together; then light it and check whether all three answers survived contact with heat. If the carrier vanishes inside five minutes, the ladder is wrong. If the signature never arrives, the assistant was buried under the base. If the piece will not stay lit at all, the guide is wrong, and no amount of scent work rescues a structural failure of that kind.
Summary
- The unit of this craft is a formula, not a bottle of scent.
- Four jobs need filling: something carries, something holds the edges, something names the piece, something makes it burn.
- Proportion assigns those jobs. The role names are only a record of what the proportions did.
- Across the worked examples, heartwood carried every compound and the assistant supplied every title.
- Read the unit words first, and the ratio before the gram figure.
- The herbal parallel is architectural. A composition for the nose is not a treatment.
FAQ
Q: What does 合香 (hé xiāng) actually mean?
Literally “combined fragrance,” and it names an assembly method, not a product category. A blender hands each aromatic a defined job, so the result is one scent with internal structure. Two formulas qualify while sharing nothing in the jar. The contrast case is 单方, where a single aromatic stands alone and the skill lies in grading and heat.
Q: Does the chief have to be the most expensive material?
No, and the two rarely coincide. Rank follows the dose; price follows scarcity in trade. A plain wood at the top of the ladder can hold a compound in which a costly resin whispers at the bottom. Assume otherwise and you pay for rarity while judging a design you never read.
Q: Why whole pears rather than pear juice?
The fruit doubles as a container. Powder goes into the cored pear, the cap goes back on, and steam drives juice inward before the mass is peeled, milled, and set aside. Juice poured over powder wets the outside and leaves the core dry. The counting unit 枚 flags a whole fresh component rather than a measured powder.
Q: Can a classical formula be rebuilt with different materials?
Often, if the proportions survive the swap. Another heartwood can take the top of the ladder provided it burns with comparable patience; the supporting dose stays at its fraction; the signature must remain unmistakable. Change the ladder and you have written a new formula wearing an old name, which is how traditions drift without anyone deciding to change them.
Q: Is 合香 the same idea as a perfumer’s accord?
Closer cousins than either trade admits, and not the same. An accord is composed to be smelled from skin or air and can be rebuilt from isolates. A 合香 formula must survive heat, hold a physical shape, and stay stable in a jar, which is why a binder and a fixative earn roles of their own. Balance is shared; combustion is the constraint that separates them.
Sources
- 周嘉胄《香乘》(Ming) — where these formulas survive. Wikisource.
- 洪芻《香谱》(Song) — a Song manual that treats blending as a body of knowledge. Wikisource.
- 本草纲目 — the materia medica behind the organizing scheme borrowed here. Wikisource.
- 陈云君《燕居香语》 — the source for the modern 二苏旧局. Print edition, not consulted online.
Related Resources
- Chinese Incense Recipes (香方) — surviving proportions.
- The Incense Manuals (香谱) — the primary texts, and the work of reading them untranslated.
- Incense Ingredients Glossary — every material above, with its Chinese name.
- How Incense Is Made — from balanced powder to a piece that burns.
- What Is Chinese Incense? — the wider tradition around it.