You probably do not know this, but in Jinxian County, Jiangxi, there is a small town called Lidu. Any random jar of white liquor pulled from its shelves can be traced back, year by year, for more than seven hundred years.

I want to be honest with you about how I ended up writing this piece. I was working through a stack of county annals from Jiangxi, the kind of municipal-level zhì volumes that nobody outside a handful of historians and a few stubborn local writers ever reads. Most of them are dense, repetitive, and frankly a little boring. They list officials, list flood years, list tax reforms, list which magistrate repaired which bridge in which year of which reign. I was skimming fast, looking for something unusual. And then I hit the Lidu section.

The first time I stumbled across the seven-hundred-year fact, deep inside the Annals of Jinxian County, Jiangxi Province, I sat still for a long while. Before the 2002 excavation, most people, even people who drink a lot of baijiu, assumed Jiangxi’s distilled spirits really only took off in the Ming and Qing. The standard story you hear in bars is that the big baijiu regions are Sichuan, Guizhou, Shanxi, and a few pockets in Henan. Jiangxi is not on that list. But Lidu’s cellars, hearth pits, cooling halls, and an entire Yuan-dynasty brewing toolkit were quietly buried in the soil on the eastern bank of the Fu River, waiting a thousand years to be found.

It is a strange feeling. What you swallow is liquor. What you taste is an entire dynasty.

That is the only way I can describe the way a county annals can hit you, if you read it slowly enough. The numbers are not even the most interesting part. The most interesting part is the way the same place keeps showing up across seven centuries, holding the same trade, selling the same product, under completely different governments.

The “Jiaoshi Lijia Du” on the Fu River

To talk about Lidu Liquor, you have to talk about the Fu River first.

Lidu Town was known in ancient times as “Jiaoshi Lijia Du,” a name that comes from the rocky shallows (焦石, “scorched stones”) that made the fording point famous long before the town itself existed. It sat at the strategic choke point between Linchuan, Fengcheng, Nanchang, and Jinxian counties [1]. From the stretch of the Fu River beside Lidu, you could pole upstream to Nancheng, Jinxi, and Fuzhou, and float downstream to Nanchang and Jiujiang. Boats and passengers came and went, cargo loaded and unloaded, and the place worked, for centuries, as a natural water crossroads [1].

A geography like that decides, almost by itself, how far a town’s liquor can travel. Land transport in pre-modern Jiangxi was slow, expensive, and broken up by toll stations. Water transport was a tenth of the cost. If you were a distiller in 1350, the Fu River was the difference between selling to your own county and selling to the entire middle Yangtze corridor.

And Lidu, locally, was not just a transit point. There has always been a saying there: “In Jiaoshi Lijia Du, they sell wine and make tofu” [1]. In other words, for a very long time, the town was already a “shop in front, workshop in back” cluster of distillers. Every household knew how to diào shāojiǔ (literally “hang distilled liquor,” the local term for low-pressure distillation) and how to brew rice wine. The Annals: Industrial Records makes the point bluntly, almost dismissively: “liquor production in Lidu was the most prosperous in the entire county” [2].

That sentence, by the way, is not idle flattery. It is a record.

By the 37th year of the Minguo, that is 1948, the whole of Jinxian County had 47 registered distilleries on the books, and 7 of them, the largest and most time-honored, sat inside Lidu Town alone [2]. For a county-level unit, that is an absurd density. In a town with maybe a few thousand households, seven flagship distilleries means distilling was the local economy, full stop.

What this tells me, reading it years later, is that Lidu was never a craft town. It was an industry town, with a craft face.

I should mention one more piece of geography that is easy to miss. Lidu sits where the Fu River’s floodplain narrows. The river is wider to the north, narrower here, with a high bank on the east side that historically stayed dry even in the worst flood years. That high bank is exactly where the cellars were dug. When I first read that, I thought, okay, that explains the cellar geography. The distillers did not pick the site by accident. They picked it because the geology was already right.

A set of numbers you have to remember

If you only tell stories about history, you end up with folklore. Real history needs numbers.

I pulled a few key rows out of the Annals: Finance and Taxation volume, the ones that span the late Qing into the early Minguo, and laid them out as a simple table. All figures are in silver dollars unless otherwise noted [4].

  • 1914 (Minguo 3). One tael of Land-Head Silver converted to 2.2 silver dollars, with an additional 0.07 handling fee.
  • 1927 (Minguo 16). One tael of Land-Head Silver converted to 3.0 silver dollars, with surtaxes merged into the provincial surcharge.
  • Same year, 1927. One dàn of converted rice tax, 4.0 silver dollars, rounded to a unified value.
  • 1939 (Minguo 28). County-wide total, 271,550 silver dollars, with an additional 97,997.2 in surtaxes.

Lay these next to each other, and do the math.

In 1914, Jinxian’s Land-Head silver-dollar collection was 33,000 dollars and its rice-conversion collection was 43,500 dollars, totaling 76,500 dollars (excluding the handling fee) [4].

By 1939, the two principal lines alone, land tax and surtax, summed to 369,547.2 silver dollars [4].

I did the arithmetic. The 1939 total is 4.83 times the 1914 total.

Twenty-five years. Nearly five times.

That is the kind of growth rate that, when it lands on a row of small distillers, breaks things. To survive, many of the workshops either quietly raised their alcohol content or shaved aging time off the schedule. When you sip an “aged” bottle today, understand that, in that era, aged liquor was a luxury reserved for the few.

I want to dwell on the math for a second, because it is the kind of number that does not look dramatic at first glance and then suddenly does. A 4.83x increase over 25 years is about 6.5% annual growth in real terms, which is not, in a modern economy, particularly scary. But this was happening on top of a base that was already extracting everything the local economy could produce. Land tax in a poor county is essentially a fixed pie. When the surcharge rate jumps from a small multiple of the principal to 1.8 times the principal, you are not collecting more efficiently. You are just pushing the taxpayer closer to the wall. And for the distillers, who had to pay land surtax on top of the liquor surtax, the wall was very close.

Four people by the river

If you only stare at the numbers, history feels cold. A town that survives a thousand years is held up by specific people.

I picked four of them.

The first is Shu Fen, zhuangyuan (top scholar) of the palace examination in 1517, the 12th year of Ming Zhengde. His courtesy name was Guoshang, his literary name was Zixi, and he was a son of Jinxian County, recorded in both the Annals: Chronology of Events and the Annals: Literary Records [5, 6]. This man was, by all accounts, stiff-backed his entire life. He advised the Zhengde Emperor against a southern tour and got caned for it. But back home, teaching in Zixi just a few miles from Lidu, he really did love his drink. The annals preserve one of his poems, Shi Ao: “Do not say a trickling stream has no surplus moisture; the Great River is eventually filled from here” [7]. Read those two lines and you understand, immediately, that what this Ming-dynasty top scholar was drinking was not liquor. It was the water of his hometown.

The second is Cao Hantang, a potter from Daye in Hubei, who arrived in Lidu Town during the Qing Xianfeng era [8]. He moved there to fire kilns. Distilled liquor needs cellaring, and cellaring needs well-sealed ceramic jars. The moment this man settled, Lidu’s “vessel” problem was solved. The Cao family rules were interesting. Strict, almost absurd: “The craft can only be passed to sons, not to outsiders” [8]. That kind of family-level technical lockdown is what let the Cao family’s pottery hold a long-term monopoly inside Lidu’s liquor industry. By 1984, the family’s ceramic business in Lidu had run for five generations [8].

The third is Magistrate Nie Dangshi, who took office as Jinxian’s county magistrate in 1664, the 3rd year of Qing Kangxi [9]. When he arrived, Lidu and the surrounding polder lands were getting flooded year after year, the grain needed for brewing kept failing, and liquor prices lurched every season. He did one thing. He “supervised the repair of levees, and the Qinglong and Danfeng bridges at the east gate of the county seat” [9]. The key projects were the Zixi and Fengle levees, which protected tens of thousands of acres of farmland along the Fu River and, in turn, kept grain flowing into Lidu’s distilleries.

The fourth is Tang Zuochen, a jiànshēng (imperial college student) of the Qianlong era, named in both the Annals: Finance and Taxation and the related court records [10, 11]. He could have lived a quiet life studying for the exams. Instead, on the question of inflated tax deeds, he went all the way to Governor Tie Bao and formally impeached Magistrate Ding Yingzhou, bankrupting his own household in the process. The chaos he was fighting looked like this: “for every tael of principal silver, an extra 0.14 silver dollars of pure surcharge was being demanded” [11]. Tang’s case was not, strictly speaking, a liquor-tax case. But the line of merchants standing behind him was, in large part, the Lidu distillers.

Look. Inside one jar of liquor: a top scholar, a potter, a magistrate, a would-be official.

And there is a fifth group I did not get to write in, the porters, the bargees, the grain dealers, the sauce-shop owners. Every autumn, when the new grain came in, the queue along the Fu River stretched from Lidu all the way downstream to Wenjia Town. These people have no names in the annals. But they were the real reason Lidu’s liquor could be sold, year after year, for a thousand years.

I should pause on Cao Hantang’s story for a moment longer, because the family-closed craft model is, in my opinion, the most important technical fact in the whole annals. When a county-level annals records something as granular as “the craft can only be passed to sons, not to outsiders” [8], that is the editor telling you, in a very deliberate way, that this rule had real economic consequences for the rest of the industry. And it did. The Cao family monopoly on storage jars meant that the price of a properly sealed aging vessel was effectively set by one extended family in one town. Other potters in the area could make jars, sure, but for the kinds of micro-pore-ceramic seal that long-term sorghum liquor aging requires, the Cao jars were the only ones that worked. That is a kind of industrial power that no modern brand would believe it could get away with. The Caos got away with it for five generations.

The microorganisms three meters underground

So why has Lidu Liquor lasted seven hundred years?

It comes down to the cellar.

Archaeology and the chronicle records, read side by side, say the same thing: Lidu’s fermentation pits sit three meters underground, in soil dense with diverse microorganisms, kept at constant temperature and humidity [2]. That is exactly the kind of environment that lets sorghum ferment all the way down. In the early Minguo years, the alcohol content of Lidu Liquor held steady at 56 degrees, which is, frankly, the sweet spot of this kind of traditional pit process [2].

The Annals: Industrial Records has a single line of evaluation for Lidu Sorghum Liquor that I think is the thesis of the whole article:

“Lidu Sorghum Liquor evolved from folk brewing and is said to have a history of over a thousand years. Its characteristics are: clear color, mellow taste, and fragrant aroma, with functions of invigorating blood and sterilization.” [2]

“Clear color” is the signature of mature distillation. “Does not cause headaches or thirst” is the signature of long cellar fermentation, which lets fusel oils separate out [2].

Put the two sentences together and you realize something: seven hundred years ago, the people of Lidu were already running a brewing engineering project that was more refined than most of their contemporaries.

A small aside on what “three meters underground” actually means in practice. The Lidu cellars are not basements. They are pits dug into the water-table layer of the alluvial soil, with a sealed brick or stone cap on top. The temperature at three meters below grade in this part of Jiangxi stays close to 18 degrees Celsius year-round, regardless of the air temperature above. The humidity stays high, but the cap keeps the worst of the seasonal swings out. That stable environment is what lets a slow-fermenting grain like sorghum actually go through the full set of microbial transformations that produce a clean, high-yield mash. Surface-level fermentation, which is what most folk distillers used, gives you a thinner, more variable product. The Lidu cellars were, in industrial terms, a deliberate quality bet, made seven hundred years before anyone had a name for “terroir.”

A county that has seen everything

To understand why Lidu’s cellars kept running, year after year, when so many other craft centers in China got hollowed out, you have to understand the county they sat in.

Jinxian County was established in the 2nd year of Song Chongning, which is 1103 [17]. The original purpose, according to the Annals: Overview, was administrative consolidation of the lower Fu River basin, including a famous visit by the Confucian scholar Dantai Mieming, whose tomb is still in the county [17]. So even at the founding, this was not a backwater. It was a deliberately chosen seat of local administration, picked because the river made it governable.

And then, of course, the disasters came. The Annals: Chronology of Events volume runs through centuries of floods, droughts, and military campaigns. In the late Yuan, during the Red Turban rebellion, the county seat was overrun and the local magistrate, whose name was Xin Jing [18], held the administration together through a period of extreme disorder. I do not want to over-romanticize this. Magistrate Xin was not, in any obvious sense, a liquor hero. But the fact that the Annals record him at all, in a chronicle that is mostly about tax rates and bridge repairs, tells you that the county’s administration was, even in the worst moments, still functioning. A functioning administration is the prerequisite for a functioning market. And a functioning market is the prerequisite for seven hundred years of cellar fires.

The list goes on. Through the Ming, the Qing, the Minguo, and into the early People’s Republic, the Annals record repeated disasters [5, 9, 18-22] and, just as importantly, the responses to them. The levees that Nie Dangshi supervised in 1664 [9]. The bridges rebuilt after floods. The town walls re-erected after sieges. Each of these entries is, in its own small way, the reason Lidu’s distilleries could keep working. You cannot run a brewery in a flooded town. You cannot sell sorghum liquor across a war zone. The county’s basic survival, project by project, is what kept the industry’s supply chain alive.

Liquor surtax, “one-plane-per-county donations,” and the weight of the Minguo

You can have the cleanest pit in the province. It still has to survive the tax collector.

I pulled out a few numbers from the Annals: Finance and Taxation volume that scared me a little.

In 1871, the 10th year of Qing Tongzhi, Jinxian’s annual collection of Land-Head Silver and Canal Grain Silver came to 36,961 taels [3]. That is the late Qing baseline, and it is, by the standards of the time, restrained.

Then the Minguo arrives. The tax base is converted to silver dollars. Surcharges multiply.

In June 1934, the 23rd year of Minguo, the land-tax surcharge in Jinxian reached 5.4 silver dollars per tael, which is 180% of the principal tax [14].

Read that again. For every 1 dollar of principal tax, the surcharge was 1 dollar and 80 cents.

For Lidu’s merchants, on top of the liquor surtax, they also had to pay the “one-plane-per-county donation” (a fundraising line for buying military aircraft), the “raincoat donation,” and the “winter clothing donation” [15].

By 1947, the 36th year of Minguo, Jinxian’s local tax revenue alone accounted for 66.1% of the county’s total budgeted revenue, and the per-capita tax burden had reached 4,184 yuan [15].

The chronicle never uses the phrase “draining the pond to catch the fish.” It does not need to. The numbers say it for them.

Under that pressure, the famous Lidu Liquor slid, on the eve of liberation, into a state the chronicle politely calls “having a license but no business” [16].

The phrase “having a license but no business” is one of those Chinese phrases that sounds bureaucratic but is actually devastating. It means, literally, that the distillery still had a government-issued permit to operate, but there was no commercial activity left to operate. The cellars were quiet. The jars were empty. The staff had been let go. This was a town that, in 1948, still had seven flagship time-honored workshops [2], and by 1949 it was a license with no one holding it.

I want to flag, too, the bizarre specificity of those “donation” surtax names. The “one-plane-per-county donation” [15] was, in real money, a county-level fundraising drive to help the Nationalist Air Force buy warplanes. The “raincoat donation” and the “winter clothing donation” were the same idea applied to military supplies. They are not, in any real sense, taxes. They are emergency wartime levies dressed up in civilian clothes, and they got collected on top of all the actual taxes. When you read the Annals list, the layered structure is almost artistic in its cruelty: principal tax, administrative surcharge, education surcharge, road surcharge, military-aviation donation, military-clothing donation, and on, and on, until the merchant has no margin left.

From “convert to silver” back to “collect in grain”

If you only look at the numbers, you would think Lidu’s liquor was simply crushed by tax.

But there is a window in the middle, a real boom, that was rolled back.

In 1568, the 2nd year of Ming Longqing, the land tax was permitted to follow the “One Whip Law” [5]. Complex physical tribute was converted, in one stroke, into silver payments. Grain became a commodity at scale. A town like Lidu, whose whole business was processing grain, hit an explosive growth phase.

That growth lasted, in real terms, almost four hundred years. Then came the reversal.

In 1941, the 30th year of Minguo, the land tax was returned to central jurisdiction, and the collection method was changed from fiat currency back to physical goods. Every yuan of tax was converted to 2 dǒu (pecks) of rice [12].

From “convert to silver” back to “collect in grain.” One step forward, one step back. It is, in two lines, a microcosm of modern Chinese fiscal history.

By 1945, the 34th year of Minguo, total tax collection across the county had reached 4,180,290 yuan in fǎbì (legal-tender currency) [13]. The Lidu distilleries kept that thousand-year thread of smoke alive, but only just, only barely, under a tax load that was nearly predatory.

The word “barely” is the one I kept coming back to. Seven time-honored workshops in a single town [2], surviving year after year in the cracks of the tax system, not by luck but by craft, by personal networks, and by the stubbornness of pouring another jar and sealing another pit.

The 1941 shift from fǎbì to physical rice is also worth a moment of your attention, because it is not just a fiscal story. It is a sign of how bad things had gotten. A government that has to switch from collecting paper money to collecting actual rice is a government that has lost faith in its own currency. The price of rice was, at that point in 1941, doubling every few months. So when the authorities said “two dǒu of rice per yuan of tax” [12], what they were really doing was locking in a payment in a commodity they trusted more than their own notes. That is, in any era, a sign of a state in serious trouble. The distillers, of course, were on the other end of that same rice trade, so they got squeezed twice: they had to pay in rice, and they had to buy grain inputs in the same inflated rice market.

The kiln fire, the river, and a story that does not end

I closed the Annals one night in June, with rain coming down outside the window.

One thought kept circling.

Lidu Liquor survived to today not because of any miraculous microorganism, and not because of any one miraculous generation of craftsmen. It survived because of a river. The Fu River.

A river can route around wars, around land checkpoints, around bandit bands. In every dynastic transition, the Fu River’s water carried Lidu’s liquor downstream to Nanchang, downstream to Jiujiang, and further on to wherever a paying mouth was waiting.

In 1959 the state-owned Lidu Distillery was formally established [2]. In 1962, at the province’s first wine-tasting conference, it was officially listed as a famous Jiangxi spirit [2].

You see the arc. A Yuan-dynasty cellar. Ming and Qing surtaxes. Republican fǎbì. “Having a license but no business” on the eve of liberation. And yet the cellar still runs.

I want to leave you with one last detail from the Annals. In 1823, in the Daoguang reign, the magistrate Zhu Bai wrote a new preface to a reprint of the county chronicle [26]. The preface is, by his own admission, mostly an apology for the many things previous editions had gotten wrong. But in passing, he also writes a few lines about what Jinxian actually was, in the early nineteenth century. He does not talk about tax rates. He does not talk about flood years. He talks about the river, the cellars, the smell of fermenting sorghum drifting up from Lidu on autumn nights, and the fact that, in his county, even a poor household could usually find a jar of decent liquor if a guest came by.

That is a small thing. But it is, I think, the thing that explains the seven hundred years. Not the dynasties, not the tax collectors, not the warplanes. The fact that, for a long stretch of time in this corner of Jiangxi, having a guest over and pouring them a serious drink was a normal, ordinary, expected thing to do. A place gets to keep a craft alive for seven centuries when the craft is woven into the texture of daily life, not just into the ledgers of a few large workshops.

I realize, putting the article together, that I keep going back to the same image. The cellar. Three meters underground. A few centimeters of brick cap between the fermentation and the sky. Every batch of Lidu Liquor that has ever existed has been made in some version of that geometry, by people whose names we mostly do not know, in a town that has been renamed, re-taxed, re-administered, and re-flooded more times than any of us can count. The geometry, though, has not changed. That is the part that gets me.

It is also the part that makes me skeptical of any clean, top-down story about Chinese craft. Lidu Liquor did not survive because the Ming built a strong state, or because the Qing built a weak one, or because the Republic was chaotic, or because the early People’s Republic was disciplined. It survived in spite of all of those. It survived because, generation after generation, somebody in Lidu decided that the cellar was worth digging back open after every flood, every tax hike, every war, and every currency reform. The data points I have walked you through, the 4.83x surcharge jump, the 180% rate, the 66.1% share of the budget, the 4.83 times increase, the wartime levies in the name of raincoats and aircraft, are all the reasons the cellar should have been abandoned. The cellar was not abandoned. That is the part no table can capture.

So when I tell people, in person, about the Annals of Jinxian County, this is the section I always end with. Not the numbers. The cellar. And the small group of distillers, in a small town on a small river, who kept it running.

The next time you lift a glass of Lidu Liquor and that cellar fragrance hits your nose, try this:

What you are smelling is not liquor.

It is seven hundred years of Fu River water.

Research Data Sources

  • [1] Annals of Jinxian County, Jiangxi · Geographical Records: Lidu Town’s location and the “shop in front, workshop in back” layout.
  • [2] Annals of Jinxian County, Jiangxi · Industrial Records: Lidu sorghum liquor quality, cellar craft, and historical evaluation.
  • [3] Annals of Jinxian County, Jiangxi · Finance and Taxation: 1871-1874 (Qing Tongzhi 10-13) annual combined Land-Head and Canal Grain total of 36,961 taels.
  • [4] Annals of Jinxian County, Jiangxi · Finance and Taxation: Detailed Minguo 3 to Minguo 28 land tax, rice conversion, and surcharge data.
  • [5] Annals of Jinxian County, Jiangxi · Finance and Taxation: 1568 (Ming Longqing 2) “One Whip Law” reform.
  • [6] Annals of Jinxian County, Jiangxi · Literary Records: 1517 (Ming Zhengde 12) Zhuangyuan Shu Fen’s origin and life.
  • [7] Annals of Jinxian County, Jiangxi · Literary Records: Shu Fen’s Shi Ao, “Do not say a trickling stream has no surplus moisture.”
  • [8] Annals of Jinxian County, Jiangxi · Industrial Records: The Cao Hantang family’s ceramic craft and the “passed to sons only” family rule.
  • [9] Annals of Jinxian County, Jiangxi · Chronology of Events: 1664 (Qing Kangxi 3) Magistrate Nie Dangshi’s appointment and the levee works.
  • [10, 11] Annals of Jinxian County, Jiangxi · Finance and Taxation: Tang Zuochen’s impeachment of Magistrate Ding Yingzhou and the “0.14 surcharge per tael” practice.
  • [12] Annals of Jinxian County, Jiangxi · Finance and Taxation: 1941 (Minguo 30) return to physical collection of land tax.
  • [13] Annals of Jinxian County, Jiangxi · Finance and Taxation: 1945 (Minguo 34) fǎbì tax collection of 4,180,290 yuan.
  • [14] Annals of Jinxian County, Jiangxi · Finance and Taxation: June 1934 (Minguo 23) 180% surcharge rate.
  • [15] Annals of Jinxian County, Jiangxi · Finance and Taxation: 1947 (Minguo 36) per-capita burden of 4,184 yuan and 66.1% share of total budget.
  • [16] Annals of Jinxian County, Jiangxi · Industrial Records: “Having a license but no business” state on the eve of liberation.
  • [17] Annals of Jinxian County, Jiangxi · Overview: Founding of the county in the 2nd year of Song Chongning.
  • [18] Annals of Jinxian County, Jiangxi · Chronology of Events: Yuan-dynasty Magistrate Xin Jing and the Red Turban campaigns.
  • [26] Annals of Jinxian County, Jiangxi · Appendix: Magistrate Zhu Bai’s 1823 (Daoguang reign) preface to the reprinted county chronicle.