← explore connections

One story leads to another

Read history

Discover the events behind the dates. Start with the landmarks, then follow the stories that catch your attention.

970 stories · Page 12 of 81 · Research and review status

AI-generated illustrations · Each story shows its review status.

AD 529

Monte Cassino

The man credited with founding the Benedictine Order did not, strictly speaking, found an order — he founded autonomous monasteries.

AI-generated illustration for Monte Cassino View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

A monastery, not an order

Benedict of Nursia left the Subiaco area and established a single monastic community on Monte Cassino. The abbey itself dates the foundation to about 529.

That foundation did not create a centralised Benedictine order. Benedict established a monastery; the wider family of houses following his Rule developed later.

The year is traditional and approximate rather than a securely documented foundation date.

Gregory’s Apollo story

Gregory the Great’s later account says Benedict dismantled a sanctuary of Apollo at the site and built chapels dedicated to Martin of Tours and John the Baptist.

In the same narrative, the neighbouring priest Florentius had tried to undermine Benedict and corrupt his disciples, prompting him to leave Subiaco with several monks.

Gregory wrote Dialogues Book II around 593, roughly six decades afterwards, as hagiography — so the Apollo episode should be attributed to him, not reported as archaeologically certain.

What was and was not there in 529

Gregory’s tradition attributes twelve separate monasteries in the vicinity of Subiaco to Benedict before the move.

The Rule cannot securely be dated to the foundation year: its exact time and place of composition are unknown, though about 530 and Monte Cassino are considered plausible.

Monte Cassino stands above Cassino in central Italy, roughly between Rome and Naples — not simply a site outside Naples.

Researched 23 Aug 2026 5 sources not yet audited Traditional date

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Traditional date. Conventionally dated here; the research places it at ~0529.

What the sources leave uncertain

  • The year 529 is traditional and approximate, not a securely documented foundation date.
  • Gregory's Dialogues, written about six decades later, is the principal narrative source and is hagiography rather than a contemporary administrative record.
  • The Rule may have developed over time; neither its completion nor its promulgation can be assigned specifically to the foundation year.

Checked against

March AD 534

Vandalic War

Before surrendering, Gelimer asked his besieger for three things: one loaf of bread, a sponge for his swollen eye, and a lyre.

AI-generated illustration for Vandalic War View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

March 534, down from the mountain

In March 534, after approximately three months of winter blockade on Mount Papua in Numidia, the last Vandal king, Gelimer, agreed to surrender himself, his relatives and the Vandals still with him.

The blockade had stopped supplies reaching the mountain. Procopius says hunger, illness and the suffering of the children with Gelimer broke his resolve.

An earlier attempt to storm the mountain had cost the blockade commander Pharas 110 men.

Sworn guarantees first

Belisarius authorised Cyprian and Pharas to swear guarantees for Gelimer’s safety and honourable treatment before Emperor Justinian.

Only after receiving them did Gelimer descend to the foot of the mountain. He was escorted to Carthage and appeared before Belisarius in the suburb called Aclas.

Procopius says that when he came before Belisarius as a captive, Gelimer burst into uncontrollable laughter.

What March did not include

Procopius marks the capture as the end of the Vandalic War; modern scholarship dates it to March 534 and describes the campaign as having ended in about six months.

Belisarius then requested permission to bring Gelimer and the captive Vandals to Constantinople. That voyage and the triumph came later, not on the surrender day.

No exact day in March is preserved, and the identification of ancient Mount Papua remains uncertain.

Researched 23 Aug 2026 5 sources not yet audited

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

What the sources leave uncertain

  • No exact day in March is preserved; the month is reconstructed from the campaign chronology and the statement that the three-month blockade ended as winter was closing.
  • The precise identification of ancient Mount Papua is uncertain.
  • Procopius is the indispensable near-contemporary narrative source, but his dramatic speeches and the cake episode are literary history and cannot be independently corroborated.

Checked against

AD 536

Annals of Ulster

The entire Irish notice is three words in translation: “Failure of bread.”

AI-generated illustration for Annals of Ulster View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

Three words under 536

Entry U536.3 of the Annals of Ulster contains only the terse statement “Failure of bread.”

It supplies no month, no location, no cause, no named participants, no death toll and no description of what happened next.

“Early in 536” is therefore unsupported: the entry carries no month or season at all.

Even the year wobbles

The U536 heading itself carries the chronological rubric “AD 535”, which illustrates how complicated it is to convert the annals’ internal chronology to modern dating.

A parallel annalistic tradition, the Annals of Inisfallen, places the same “Failure of bread” notice under 537.

The surviving Annals of Ulster manuscript tradition is medieval, not an autograph written in 536, so the age of this individual notice cannot be demonstrated from the entry alone.

No volcano in the text

The annal gives no cause. It does not say the sun darkened, that snow fell, or that people resorted to cannibalism.

The volcanic explanation is a modern inference drawn from ice cores and tree rings, not something the annalist stated.

Modern research treats the early Irish annals as valuable climate evidence and finds that they credibly reflect the severe downturn beginning around 536 — which still does not turn this entry into a contemporary report of an eruption.

Researched 23 Aug 2026 6 sources not yet audited Traditional date

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Traditional date. This date is the conventional one, not one fixed by a contemporary record.

What the sources leave uncertain

  • “Early in 536” is unsupported: U536.3 has no month or season.
  • The U536 heading itself contains the chronological rubric “AD 535,” illustrating the complications in converting the annal's internal chronology to modern dating.
  • The parallel Annals of Inisfallen assigns the bread failure to 537.
  • The surviving Annals of Ulster manuscript tradition is medieval, not an autograph written in 536; the age and transmission of this individual notice cannot be demonstrated from the entry alone.

Checked against

August AD 537

Volcanic winter of 536

The familiar date is probably off by a year: the traceable snow entry belongs to 537, not 536.

AI-generated illustration for Volcanic winter of 536 View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

The entry is under 537

The listing dates the Chinese summer snow to August 536. The identifiable entry belongs to the following year: the History of the Southern Dynasties records that in the autumn, seventh lunar month of 537, snow fell in Qingzhou and damaged seedlings and grain crops.

The Chinese seventh lunar month does not map exactly onto the Gregorian August either, so “August 537” is itself a conventional approximation.

The same chronicle places severe frost at Qushan in Qingzhou during the preceding lunar month, which makes the snow part of a sequence of damaging summer cold.

Damaged, not delayed

The chronicle says the snow harmed the crops. It does not say the harvest was delayed, and it does not quantify the loss.

The Nan Shi was compiled later — credited to Li Yanshou, more than a century after the reported weather.

It supplies no exact day, no temperature and no snowfall amount.

The cause is a modern inference

The chronicle gives no physical cause at all.

Modern ice-core and tree-ring research makes volcanic aerosol forcing from an eruption in 535 or early 536 a strong explanation for the broader cold anomaly, but it cannot prove the cause of this particular snowfall.

A recent scholarly review concludes that summer frosts and snowfall affected China in 536–537, contributing to crop failures, drought and famine — while not supporting treatment of every Chinese report as an August 536 observation.

Researched 23 Aug 2026 5 sources not yet audited Date disputed

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date disputed. Sources disagree; the research places it at 0537-08.

What the sources leave uncertain

  • The listed date appears to be wrong by a year: the identifiable Nan Shi entry belongs to 537, not 536.
  • The Chinese seventh lunar month does not map exactly onto the modern Gregorian month of August; 'August 537' is a conventional approximation.
  • The Nan Shi was compiled later and provides no exact day, temperature or snowfall amount.
  • Chinese records from 536 describe other events—hail, great cold, famine and drought—but should not be conflated with the Qingzhou snow entry.

Checked against

AD 540

Rule of Saint Benedict

The famous motto “pray and work” never appears anywhere in Benedict’s Rule.

AI-generated illustration for Rule of Saint Benedict View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

No day, and a thirty-year window

No surviving source identifies a day — or even the year 540 — on which Benedict wrote or completed the Rule. A Benedictine scholarly reference dates the composition broadly to between 530 and 560.

Monte Cassino is the traditional and probable place of writing, but the exact place and date remain unknown, as does whether the work was written all at once.

Much of Benedict’s biography comes from Gregory the Great’s hagiographical Dialogues, written decades later, which supplies no dependable composition date.

A prologue and 73 chapters

Benedict composed a prologue and 73 chapters governing a cenobitic community — monks living in a monastery under a rule and an abbot.

The Rule calls the monastery a “school for the service of the Lord”, and expressly says it is laying down a rule for cenobites, one of the four kinds of monks it identifies.

It was not wholly original: its most important textual source was the earlier anonymous Rule of the Master, from which Benedict copied extensively, especially in the prologue and first seven chapters.

Wine, two dishes, and no knives in bed

The Rule permits a daily measure of wine, while ruefully observing that monks of Benedict’s day could not be persuaded to abstain altogether.

Two cooked dishes were placed on each table — not for luxury, but so that a monk unable to eat one could eat the other.

Monks were told to sleep clothed, but without knives at their sides, so they would not cut themselves during the night.

Researched 24 Aug 2026 10 sources not yet audited Date disputed

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date disputed. Sources disagree; the research places it at ~0540.

What the sources leave uncertain

  • The exact date and place are unknown. “540” is a convenient approximation inside the better-supported 530–560 range, not a verified event year.
  • Whether the surviving Rule was composed as a unit or developed gradually in response to the community's needs cannot be determined. Source: https://www.newadvent.org/cathen/02436a.htm
  • Much of Benedict's biography comes from Gregory the Great's hagiographical Dialogues, written decades later; it does not provide a dependable composition date.

Checked against

AD 550

Babylonian Talmud

A work of roughly 1.8 million words has no datable author and no completion notice anywhere inside it.

AI-generated illustration for Babylonian Talmud View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

550 is a model, not a record

No contemporary evidence records the Babylonian Talmud being completed in 550, and the work supplies neither a date nor a datable author. Modern scholarship gives ranges instead of a completion day or year.

The mid-sixth-century “final redaction” belongs to older scholarship, based heavily on medieval rabbinic pedigrees.

Recent proposals extend from the mid-seventh century into the second half of the eighth.

Not two editors with a manuscript

Bava Metzia 86a calls Rav Ashi and Ravina the “end of instruction”, or hora’ah. It does not say they finished a written Talmud.

A major modern study rejects the traditional reading that the two comprehensively edited the Talmud, citing internal contradictions and material that postdates them.

Ravina’s death date cannot be combined casually with 550 either: the competing traditional dates are 475 and 499/500.

Edited in memory

The evidence points to prolonged oral transmission, compilation, supplementation and stabilisation, with no securely identifiable closing session or publication date.

David Weiss Halivni’s term Stammaim designates the scholars responsible for much of the Bavli’s anonymous Aramaic framing, argument, summaries and conclusions. The Savoraim are traditionally credited with later refinement, but their precise contribution is debated.

“Redaction” can mean organising attributed teachings, creating anonymous dialectic, oral stabilisation or written fixing — which makes any claim of a single final redaction especially hazardous.

Researched 24 Aug 2026 7 sources not yet audited Date disputed

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date disputed. Sources disagree; the research places it at ~0550.

What the sources leave uncertain

  • The date 550 represents an older approximate model, not a recorded completion year.
  • The identity of the Ravina paired with Rav Ashi and Ravina II's death date are disputed; sources give 475 or 499/500.
  • Scholars disagree over how much anonymous material is amoraic, Savoraic, Stammaitic, or even later.
  • The word “redaction” can refer to different stages—organization of attributed teachings, creation of anonymous dialectic, oral stabilization, or written textual fixing—so claims of a single final redaction are especially hazardous.

Checked against

AD 581

Emperor Jing of Northern Zhou

The child emperor handed over his throne in March and stayed alive as a duke until July.

AI-generated illustration for Emperor Jing of Northern Zhou View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

Two events, four months apart

The listing merges them. On 4 March 581, Yang Jian had the child emperor yield the throne, ended Northern Zhou, took the imperial title and founded the Sui dynasty.

The deposed Yuwen Chan was not killed at the accession. Yang Jian made him Duke of Jie.

He was secretly killed approximately four months later, on 9 or 10 July 581 — the official histories disagree by one day.

A purge that began before the throne changed

The elimination of the Yuwen imperial clan unfolded over time. Yang Jian executed princes and households before the accession, continued removing close male relatives afterwards, and killed the former emperor last.

“Fifty-nine Zhou princes” confuses an often-repeated count of royal-family members with titled princes, and compresses killings spread across that whole span into a single date.

No strong source consulted supports the precise claim that Yuwen Chan and 58 relatives were executed together at Chang’an.

Nine sui, and imperial mourning

After Yuwen Chan’s death, the court announced mourning and buried him with the honours due an emperor — notwithstanding the account that Yang Jian had secretly ordered the killing.

The Book of Zhou records his age at death as nine sui, the traditional East Asian age count.

Born in August 573, he was seven years and about eleven months old by modern completed-year reckoning: nearly eight, but not yet eight.

Researched 24 Aug 2026 4 sources not yet audited Date disputed

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date disputed. Sources disagree about when this happened.

What the sources leave uncertain

  • This is not one event with one date. The dynastic accession occurred on 4 March; Yuwen Chan's death followed in July.
  • The Book of Sui and Northern History place the killing on 9 July, while the Book of Zhou and its associated chronology give 10 July.
  • The chronicle calls him nine sui at death. Born in August 573, he was seven years and about eleven months old by modern completed-year reckoning—nearly eight, but not yet eight.
  • No strong source consulted supports the precise claim that Yuwen Chan and 58 relatives were executed together at Chang'an.

Checked against

AD 581

Sui dynasty

The eight-year path to reunification began with a child emperor’s staged abdication.

AI-generated illustration for Sui dynasty View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

4 March 581

On 4 March 581, Yang Jian replaced Northern Zhou with a new dynasty named Sui and assumed the imperial position later known by his posthumous title, Emperor Wen.

The transfer was staged as an abdication: the child Emperor Jing yielded the throne after Yang had controlled the Northern Zhou government as regent, and was made Duke of Jie rather than killed that day.

The new reign adopted the era name Kaihuang, and the dynastic name came from Yang Jian’s former Sui title.

What made it possible

Emperor Xuan’s death in 580 had left a child on the Northern Zhou throne, and Yang Jian became regent.

Yuchi Jiong and other regional commanders resisted; Yang’s forces defeated them.

He then eliminated important Yuwen rivals, accumulated the title and ceremonial privileges of a prospective emperor, and compelled the abdication.

Not yet an empire, not yet a capital

Sui began in 581, but political reunification of northern and southern China was not completed until 589 — the southern Chen dynasty remained independent for another eight years.

Daxing, the capital the dynasty is remembered for, did not yet exist: construction began in 582 and the court moved there in 583.

Yang Jian’s Sui reign ran from 581 to 604.

Researched 24 Aug 2026 5 sources not yet audited Date corrected

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date corrected. The research places this at 0581-03-04.

What the sources leave uncertain

  • The 4 March date is a modern conversion of the traditional Chinese calendar date and is supported by Cambridge's chronology.
  • The abdication and Yang Jian's accession may be assigned adjacent civil dates in some converted datasets, but Cambridge identifies 4 March as the dynasty's establishment.
  • The former Emperor Jing's later death date is disputed by one day—9 versus 10 July—but that does not affect the founding date.

Checked against

AD 595

Khadija bint Khuwaylid

The earliest surviving biography gives Muhammad’s age at the wedding — and says nothing about Khadija’s, despite the famous pairing of 40 and 25.

AI-generated illustration for Khadija bint Khuwaylid View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

A conventional year, not a recorded one

Modern reference works conventionally place the marriage around 595, but no contemporary record supplies an exact year or day, and the surviving narratives were written or compiled generations later.

In the early Islamic biographical tradition preserved by Ibn Ishaq and Ibn Hisham, Khadija was a wealthy Meccan merchant who hired Muhammad to conduct a trading venture with her goods in Syria.

The same tradition says she subsequently expressed her desire to marry him, and that the marriage was arranged in Mecca.

The 40 is disputed

Ibn Hisham and al-Tabari give Muhammad’s age at marriage as 25.

The earliest surviving Ibn Ishaq narrative does not securely establish Khadija’s age. Ibn Sa’d and al-Baladhuri transmit 40, while other reports and one modern source-critical study argue for about 28.

Accounts also differ over Khadija’s prior marriages, and over which of Muhammad’s uncles — Abu Talib or Hamza — assisted with the proposal.

Employer first

The early account describes Khadija hiring men on profit-sharing terms, with Muhammad initially working as her compensated agent.

Calling the couple a “successful merchant partnership” projects a modern business arrangement onto the evidence: the sources document an agency arrangement before the marriage, not a formal joint enterprise afterwards.

The listing also compresses employment, the Syrian venture, the proposal and the wedding into a single dated event, though the tradition describes a sequence whose intervals are unknown.

Researched 23 Aug 2026 5 sources not yet audited Traditional date

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Traditional date. Conventionally dated here; the research places it at ~0595.

What the sources leave uncertain

  • The year 595 is a conventional reconstruction, not a contemporary recorded date.
  • The surviving narratives were written or compiled generations after the marriage.
  • The familiar claim that Khadija was 40 and exactly 15 years older than Muhammad is not in the earliest extant Ibn Ishaq narrative. Ibn Sa'd and al-Baladhuri transmit 40, while other reports and modern source criticism favor a younger age.
  • Accounts differ over Khadija's prior marriages and over which male relative represented each party.

Checked against

AD 601

Merovingian dynasty

In the actual year 601 the supposed Carolingian Empire was not merely weak or fragmented — it did not yet exist.

AI-generated illustration for Merovingian dynasty View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

No event to date

No source consulted identifies a transfer from Merovingian to Carolingian rule, a Frankish conquest of most of Europe, or a continent-wide rise of “feudal lords”, as an event in 601.

In 601 the Frankish kingdoms were under the Merovingian dynasty.

The year-only date supplies no day or month and does not correspond to a recognised occurrence.

The Carolingian dates are much later

The Carolingians did not replace the Merovingian kings until Pepin’s accession in 751.

The Carolingian imperial phase conventionally begins with Charlemagne’s coronation as emperor in Rome on Christmas Day 800.

An Oxford study frames the Carolingian emergence as a period running from about 718 to 800.

Neither three peoples nor one system

Merovingian and Carolingian are dynastic labels for ruling families within Frankish political history — not three unrelated peoples successively taking power.

Neither realm straightforwardly controlled “most of Europe”, and Frankish territories and power changed repeatedly, so the phrase is too vague to verify.

Modern medievalists caution that “feudalism” is a post-medieval historiographical construct, not a single system that suddenly arose across Europe in a datable year.

Researched 23 Aug 2026 5 sources not yet audited Date disputed

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date disputed. Sources disagree about when this happened.

What the sources leave uncertain

  • The year-only date supplies no day or month and does not correspond to a recognized event.
  • Frankish territories and power changed repeatedly, so 'control most of Europe' is too vague to verify.
  • The meaning, chronology and usefulness of 'feudalism' remain contested among medievalists.

Checked against

March 12, AD 604

Pope Gregory I

Gregory was buried on the day he died, in the portico of Old St Peter’s rather than inside the basilica.

AI-generated illustration for Pope Gregory I View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

12 March 604

Gregory I died in Rome on 12 March 604, ending a pontificate that had begun on 3 September 590.

A historical account reports that he was buried the same day, in the portico of Old St Peter’s Basilica, in front of the sacristy.

He died after prolonged ill health, including severe gout that his late correspondence says sometimes left him barely able to speak; the sources do not establish a specific fatal pathology.

The chair stayed empty

The Roman see became vacant.

Sabinian, the 65th pope, follows Gregory in every papal list — but his pontificate did not begin until 13 September 604.

Imperial confirmation delayed the consecration, leaving the see empty for roughly half a year.

What the day did not carry

Gregory’s mission to the Anglo-Saxons, led by Augustine of Canterbury, was dispatched in 596 — earlier policy, not an act of his death date.

He helped develop Roman claims of primacy, but “laid the foundations of papal absolutism” is anachronistic and contested: his correspondence also rejected the title “universal bishop”.

Even the tenure is rounded. Fourteen years is conventional inclusive counting; the dated interval was about thirteen and a half.

Researched 23 Aug 2026 6 sources not yet audited

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

What the sources leave uncertain

  • The death date and pontifical dates are strongly attested.
  • Calling the tenure '14 years' is conventional inclusive rounding; the dated interval was about thirteen and a half years.
  • The exact medical cause of death is not securely recorded.

Checked against

AD 610

Grand Canal (China)

The famous “completion in 609” dissolves into three separate construction dates: 605, 608 and 610.

AI-generated illustration for Grand Canal (China) View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

Three orders, not one completion

The major Sui canal works were a multiyear sequence rather than a single completion in 609. Emperor Yang ordered the Tongji Canal in 605, work on the northern Yongji Canal began in 608, and he ordered the southern extension toward Hangzhou in 610.

No consulted scholarly chronology identifies 609 as the year the whole system from the north to Hangzhou was finished.

Some summaries use 605–609 for Emperor Yang’s principal works, while detailed chronologies place the southern extension in 610 — and one recent Cambridge excerpt describes the broader project as completed in 611.

What each stage joined

The 605 Tongji project linked Luoyang to the Huai River and joined an older route running south to the Yangtze at Yangzhou, establishing the Yellow–Huai–Yangtze connection before 609.

The Yongji Canal begun in 608 ran north-east from near the confluence of the Yellow and Luo rivers to the vicinity of modern Beijing.

The 610 order carried the network from the Yangtze opposite Yangzhou to the head of Hangzhou Bay — about 270 miles, in the Cambridge history’s figure.

Dug by levy, and partly already there

Corvée levies did the work. For the 608 project, women were compelled to labour alongside men.

The Sui shu reports one million or more workers for the longer 605 northern stretch and 100,000 or more for the Huai-to-Yangtze stretch, though the historian cautions those figures may count twenty-day corvée periods rather than people working simultaneously.

The resulting system used artificial canals, canalised rivers and older waterways rather than one wholly new trench.

Researched 23 Aug 2026 6 sources not yet audited Date disputed

How this was checked

Researched from the web into a fact sheet and rewritten from that sheet. This article has not yet had an independent model audit. How the pipeline works →

Date disputed. Sources disagree; the research places it at ~0610.

What the sources leave uncertain

  • No consulted scholarly chronology identifies 609 as the year in which the entire Beijing-to-Hangzhou system was completed.
  • Some summaries use 605–609 for Emperor Yang's principal works, while detailed chronologies place the Jiangnan extension in 610; one recent Cambridge excerpt even describes the broader project as completed in 611.
  • 'Grand Canal' can denote the Sui-Tang network, the later Beijing-Hangzhou main stem or the larger UNESCO heritage system. Their routes and lengths are not interchangeable.
  • The Sui sources' enormous labor totals may count cumulative corvée service periods rather than people working simultaneously.

Checked against