Things fall apart; the centre cannot hold … A post somewhat about China

By Lorenzo

Historically, taxing land (rents) and trade have been the dominant income sources of rulerships not reliant on labour service (not to be confused with taxes on labour income, which have a different dynamic).* Trade was a particularly attractive source of income because it often involved taxing outsiders. But trade was also mobile–too much tax for too little benefit could see it fall away or move elsewhere–and is subject to fluctuations that might well be beyond the control of the rulership. As Tome Pires, a C16th Portuguese apothecary and adventurer, famously said: 

Whoever is lord of Malacca has his hand on the throat of Venice.

If trade dropped dramatically, then the rulership could suddenly find itself with significantly less income than its structure of government presumed. The effects of this could be unfortunate …

The Han dynasty collapsed in 220 AD, the Parthian Empire was overthrown by the Sassanids in 224 AD and Rome’s Crisis of the 3rd Century began in 235 AD. Is this sequence a coincidence? It would be a remarkable one if it were, but that is not at all likely.

Rome in pink, Parthians in brown, Han in yellow. How much control the Han had over the Tarim basin (that big westward bulge) is debatable.

Rome in pink, Parthians in brown, Han in yellow.
How much control the Han had over the Tarim basin (that big westward bulge) is debatable.

The Han dynasty ruled, with a brief interruption, for 4 centuries (206 BC – 220 AD). The Parthian Arascid dynasty ruled for close to five centuries (247 BC – 224 AD). During the first two centuries of Han rule, the highly commercial Roman Republic–having conquered the highly commercial Carthaginian Empire–proceeded to conquer the highly commercial Hellenistic kingdoms founded by the successful Diadochi (successors to Alexander). Probably snuffing out the first Scientific Revolution while doing so. In accordance with economist David Friedman’s argument that states reliant on taxes on land (rents) will tend to be small (i.e. principalities), those reliant on taxes on trade will tend to be large (empires) and those reliant on taxing labour will tend to be based on [ethno-]linguistic boundaries (nations), the Sino-Parthian anchoring of the Silk Road system led to the Mediterranean world being ripe for unification under a single commercial jurisdiction; particularly one as favourable to commerce as Roman Law.

The Han, Parthian and Roman empires came to span most of Eurasia East to West. This was an excellent basis for the first great flowering of the Silk Road system. Silk flowed westwards from China, buying horses from the various nomad confederations, spices via India, silver and glassware from the Mediterranean. Incense (frankincense and myrrh) flowed from Arabia northwards. Mediterranean copper, lead, tin, red coral and wine flowed via the sea route to India, which had a substantial Roman trade colony up until about 200 AD. (The claim that Thomas the Apostle went to India is not inherently implausible; that religions follow trade routes is a well-established pattern, so–given the extent of Roman-Indian trade–the early spread of Christianity to India is not surprising.)


One of these would swallow all the others …

The Parthians were well-placed to profit even further from these trade flows when the pax Romana in the West and the pax Hanica in the East provided poles of trade–stable empires ruling over tens of millions (a likely peak population of 50-60m each) with vast internal trade and wealthy elites to provide markets for luxury goods. This was the Rome of the Principate, when a lean Imperial bureaucracy (perhaps 300 full-time officials in the central administration) supported rule by a cosmopolitan elite who provided their own aides when appointed to governorships, with cities being largely self-governing.

The first great blow to this mutually supporting trading system linking Western and Eastern Eurasia was the Antonine plague, which killed perhaps a tenth of the population (up to a third in some regions) of the Roman Empire (so about 5-6m people) striking hard at the Roman Army. The Han dynasty had increased taxes to pay for fortifications along the Silk Road system, to better protect and profit from trade that was now significantly reduced. This aggravated an agrarian crisis due to famine. The subsequent Yellow Turban rebellion was defeated, but at the cost of millions of deaths, warlordism and the subsequent collapse of Han rule.

Domino effects

Shapur celebrating in stone

Shapur celebrating in stone

One pole of this trading system had collapsed into war and rival kingdoms, the other had suffered a significant demographic blow. The Parthian Empire–in the middle of this trading system–was suddenly vulnerable due to the loss of trading income. One of their vassal kings, Ardashir I (r.224–241), took this moment to strike, overthrew the Parthian Arascid dynasty and established the Sassanid Empire. The new Empire, with a record of military success behind it, was much more aggressive than its Parthian predecessors, putting the Roman Empire under serious military pressure. The low point for the Romans of this pressure being the capture of Emperor Valerian and his army by Shahanshah (King of Kings) Shapur I (r.240/42 – 270/72) in 260. This was while another serious plague was raging within the Roman Empire, with probably similar mortality levels to the previous outbreak.

Zenobia being wrong but romantic

Zenobia being wrong but romantic

The Roman client king Odenthasus of Palmyra saved the Roman Empire in the East by vigorous and effective campaigning. But he was assassinated, and his widow Zenobia took over, carving out the brief Palmyrene Empire from Rome’s Eastern provinces before being defeated and captured by Emperor Aurelian (r.270-275), the first emperor to be officially styled dominus et deus (“master and god”) in official documents. (Zenobia was taken back to Rome, where she apparently married a Roman senator.)  The unsuccessful Roman wars against Ardashir and Shapur likely cost the Romans about 150,000 soldiers in a 35-year period, when the Roman Army numbered 350,000 to 450,000 for the entire Empire.

Meanwhile, in the West, two centuries of contact with the Romans meant that the Germanic tribes had also become part of a trading system that was now delivering less; it is likely the decline in trade also adversely affected various nomad peoples, possibly putting some eastward pressure on the Germanic peoples. The Roman Army had been disrupted by disease and was concentrated against the Sassanids. Germanic peoples began to raid and then invade the Empire, which abandoned two border provinces (Dacia and Agri Decumates) entirely.


Under internal and external pressure, the Roman Empire survived only by changing profoundly. The dramatic drop in population and (especially) trade meant that the grand estates became much more self-contained, leading to the rise of the manorial economy. That made their interests much more local.

With less trade to tax, the imperial administration moved more to direct acquisition of goods to support the Roman Army. This required a much larger bureaucracy (numbering perhaps 30-35,000 officials). The Roman state had to extract more from less while providing less benefits. Citizenship–which had become universal for free residents of the Empire–came to mean much less. Peasants were tied to the land, turning them into serfs. Slavery largely disappeared–serfdom and slavery not being compatible (pdf) systems. A larger bureaucracy required much more management, leading to the permanent division of the Empire in Western and Eastern halves. (It was not the Empire which became too large for one person to manage–it was actually geographically smaller than it had been, with a lower population overall–it was the expansion of the imperial bureaucracy and the rise of localism which drove the division.) This was the empire of the Dominate; bureaucratised autocratic manorialism.

An Iranian form, spreading West

An Iranian form, spreading West

Where the Roman Army had been an infantry force with cavalry supports, it now became much more based around a cavalry elite in standing field armies with lower-status garrison forces on the borders, whose interests also became more local. It is arguable that the Romans were copying their Sassanid opponents, a process that continued under the later Christian Graeco-Roman (“Byzantine”) Empire. Especially with the development of the theme system.

There was a period of stabilisation in the C4th that saw the Christianisation of the Empire–helped by the death of Julian the Apostate (r.361-3) in battle against the Sassanids–and the building of Constantinople, the New Rome. Late in the C4th, Hun migrations put external pressure first on the Germanic peoples and then on the Empire directly. The continuing Germanic pressure after the Hunnic Empire collapsed was likely partly driven by the end of the Roman warm period in the C5th that also further undermined the demographics and economy of the Empire. Even without the whips of hunger, the more trade retreats, the more land and violence become the dominant sources of income.

The incorporation of Germans into the Roman Army, and the prolonged contact with the Empire, meant that Roman organisational advantage over the Germans waned dramatically. The various Germanic peoples also had much more internal cohesion than the hierarchical and socio-economically localised Empire. The geographically much more exposed and lower-population density Western Empire was unable to withstand the decades of raid and invasion and eventually collapsed as a coherent entity; the fatal blow being the Vandal conquest of North Africa–the last unravaged tax base for the Western Empire.

Romans and Iranians

In the C6th, the surviving Graeco-Roman Empire under Justinian I (r.527-565) attempted to re-unify the Mediterranean under Roman rule. Enjoying considerable success under generals such as Belisarius and the Romanised Armenian Narses–who claimed descent from the Arascid dynasty (a branch of which had ruled Armenia)–Roman armies overthrew the Vandal and Ostrogothic kingdoms and took a coastal province from the Visigothic kingdom. The Empire was clearly the dominant Mediterranean power. Then along came the plague of Justinian, and any chance of further such expansion stopped.


Khosrau II on his favourite charger

In the C6th, the Sassanid Empire had become much more centralised under Khosrau I Anushirvan (“Undiminishing Soul”, r. 531–579), becoming a form of autocratic bureaucratised manorialism; presumably copying their Roman opponents but with the provision of heavy cavalry much more explicitly tied into the manorialism. With the realm’s Central Eurasian borders secured by decisive victories against the Turks, Khosrau I Anushirvan’s grandson, Khosrau II (r. 590-628), attempted to re-establish the Achaemenid Empire, an ambition that dated back to the beginnings of the Sassanid dynasty.

Khosrau II Aparvez (“the Undefeatable”) achieved dramatic successes, which were eventually reversed by Emperor Heraclius, who also made Greek the official language of the Graeco-Roman Empire. Khosrau was assassinated and the Sassanids suffered a prolonged period of internal revolt and dissension. (When considering contemporary Iranian perspectives, it is well to remember this long imperial and cultural history–in recent centuries, conjoined with Shi’a universalism.)

Islam arises

Khosrau II had abolished the Lakhimid Arab kingdom, placing it under direct Sassanid rule, depriving the Sassanid realm of the Arab buffer allies and early warning system that had protected its Arabian border. With both empires exhausted by their decades of struggle–the Romans had likely lost about 200,000 soldiers in 25 years of war and the Iranians probably a similar number–and Roman rule in Egypt and Syria only recently re-established, this was the moment when Arabia became unified for the first time and, inspired by their new unifying religion, with the flexible and meritocratic leadership structure common to successful new movements, overthrew the Sassanid Empire (much of the Sassanid imperial family escaped to Tang China) and took about half the territory and two-thirds of the population from the Graeco-Roman Empire, establishing a vast Arab Empire. Islam, the only major religion founded by a merchant, was generally highly favourable to trade (when raiding the infidel did not get in the way). The Islamic Empire became a pole and a conduit of trade in its own right.

(As an aside, there is a line of jihadi thought which equates the Soviet Empire with the Sassanids and the American hegemony with the Roman Empire; Washington’s neoclassical architecture even provides some aesthetic reinforcement to the analogy. In this reading, the Afghan War is what brought down the Soviet Union.)

In a further display of excellent timing, the Arab Empire was established a few decades after China was reunited under the Sui (581-618) and Tang (618-907) Dynasties. The other pole of the Silk Road trading system was once again united under single rule, with the population and commercial upsurge that went with that. Possibly, the increased trade had made it easier for Khosrau II to fund his ultimately disastrous campaigns of conquest.

Ab? ?Abdall?h Mu?ammad ibn M?s? al-Khw?rizm? (c.780-c.850) Iranian mathematician whose name gave us algebra, algorism and algorithm.

Ab? ?Abdall?h Mu?ammad ibn M?s? al-Khw?rizm? (c.780-c.850) Iranian mathematician whose name [legacy] gave us algebra, algorism and algorithm.

With the Graeco-Roman Empire reduced to a Balkan and Anatolian rump, this was the period of Classical Islam under the rule of the Caliphs residing first in Mecca under the early Caliphs (632-661), then Damascus (the Umayyads 651-750) and finally Baghdad (the Abbasids 750-1258), the new imperial city built not far from the former Parthian and Sassanid capital of Ctesiphon. As the “nexus” civilisation–the civilisation in direct or near-direct contact with the other major civilisations of Eurasia–ideas and techniques from these civilisations mixed together: Chinese inventions (notably the compass and paper); Iranian art, architecture and forms of government; Indian mathematics [and metallurgy] and Greek science and philosophy. Creating the spectacular intellectual, economic and cultural flowering of Classical Islam.

Islam localises

Then the Tang dynasty declined after the devastating An Lushan rebellion (755-763) and eventually fell (907). (The Abbasid Caliphate sent mounted warriors to help suppress the rebellion; suggesting they understood their interests in Chinese stability and prosperity despite earlier clashes with the Tang.) The East Asian pole of the Silk Road trading system declined and divided. The decline in trade having its normal de-stabilising effects, the authority of the Abbasid Caliphs decayed, being usurped by local rulers. Rather than warriors paid out of centralised taxes, iqta or tax-fiefs were handed out, creating local loyalties. The Fatimids conquered Egypt (969), setting up a rival Shi’a Caliphate based in their new capital of Cairo. The fading power of the Sunni Abbasid Caliphate and the resolution of the iconoclast struggles led to a resurgence of the Graeco-Roman Empire under the Macedonian dynasty (867-1056).

China was reunited under the Song Dynasty (960-1127). During this period the Ghaznavids (963–1186) were able to set up an Iranian-based empire stretching into Northern India, before being supplanted in Iran and Iraq by the Turkish Seljuq Empire (1037–1194).

The conversion of the Turkish peoples to Islam (ironically fostered by the Abbasid policy of accepting all Muslim converts, the basis of their success against the Arab exclusionism of the Umayyads) had removed ghazis from that border, paving the way for Turks supplanting Arabs as the dominant military force in Islam. The Seljuq dynasty was able to sweep a Graeco-Roman Empire suffering divisions between court official and thematic generals out of most of Anatolia after the Roman military disaster at Manzikert 1071, creating the Sultanate of Rum (“Rome”). But the Seljuqs proved unable to set up a stable system of unitary rule, likely not helped by [trade effects from] the Song losing control of Northern China to the Jurchin (i.e. nomad) Jin Dynasty (1115–1234) and considerable conflict among nomad peoples, leading to the rise of Genghis Khan.

Not the first folk to look or fight like that to visit the neighbourhood

Not the first folk to look or fight like that to visit the neighbourhood

The collapse of Caliphal political authority led to medieval Islam, the low points for which were the conquest of the Levant by Crusading knights (1096-9) and the Mongol incursions which saw the destruction of the Abbasid Baghdad Caliphate and the imposition of Mongol rule over Iran, the Ilkhanate (1256–1335). The Mongol Empire (1206–1368) incorporated China, the Eurasian steppes, the Iranian plateau and Russia into one vast imperial structure, though its rule over China lasted under a century (1271–1368). The growth in trade also spread the last and greatest of the great [Eurasian] plagues, the Black Death.

Chinese unity and Middle Eastern stability

But China was entering the longest period of consecutive unified rule in its history, under the Yuan (1271–1368), Ming (1368–1644) and Qing (1644-1912) dynasties. Attempts to create and maintain centralised rule in the Middle East were no longer at the mercy of serious and prolonged disruptions of trade created by extended periods of Chinese disunity. Enter the Ottoman Empire, the longest-running Muslim Empire (1299–1923) which managed to push Muslim rule deep into Central Europe until its disastrous defeat in the Battle of Vienna (1683) ushered in a quarter of a millennia Muslim retreat. (In 1944, the only independent Muslim states were Turkey–under secularising rule–Afghanistan, and Saudi Arabia; all the rest were colonies, protectorates or occupied: including Iran, which was under Anglo-Soviet occupation.)


From the late C16th, the Spanish silver peso (the original dollar) became the silver coin of choice in coastal China, as neither the Ming or Qing states minted silver coins while the hyperinflation collapse of the Yuan paper notes discredited Ming attempts to issue their own paper notes. There were periods when Chinese trade declined due to the disruptions which accompanied the transitions from the Yuan to the Ming and from the Ming to the Qing and periodic attempts by the Ming and Qing regimes to block maritime trade.

The success of such attempts is easy to exaggerate–one of the traps of Chinese history is taking the official records too much at face value. Late-medieval European states regularly had more reliable internal information and control mechanisms than imperial China, which was infected from the beginning by a concern for form that easily overwhelmed substance–such as passing off the silk-for-horses trade as a matter of horses-as-tribute and silk-as-gifts–and other pathologies and limitations of bureaucraticised autocracy. The Spanish Empire could enforce reliable coinage over a much more far-flung realm notably better than the Ming or Qing Empires.

Nevertheless, despite such disruptions, the overall level of trade with China was clearly generally much higher, notably during (pdf) the period the Ottoman Empire was expanding most vigorously in the last half of the C15th and first half of the C16th. Chinese silk and porcelain went westwards as European glassware and silver from the expanded Central European silver production went eastwards, a trade highly profitable in both directions. (It is amazing how much historical and anthropological commentary does not seem to grasp the elementary point of gains from trade–no wonder comparative advantage is beyond so many highly educated folk.)

Chinese unity and European expansion

But a unified China also profoundly assisted the post-1500 expansion of the Atlantic economy. Not only did the late C15th economic expansion make it easier to fund voyages of exploration, the Portuguese, the Spanish, the Dutch and the British were all drawn to the trade opportunities the Chinese economy offered, albeit as part of a global smorgasbord of trading opportunities. Hence the phrase “all the tea in China” signifying great wealth; at one stage the opium trade alone provided a fifth of the revenues of British East India Company rule in India (the power of modern corporations is a pitiful thing compared to that of the Honourable East India Company, whose Army was often bigger than the British Army).

A substantial amount of the flood of silver from the Americas went to purchase (pdf) Asian, particularly Chinese, goods. Indeed, the Spanish Empire was built on that flood of silver, silver whose value was based on apparently endless Chinese demand for silver (pdf), such that:

China’s immense demand-side “silver sink” had supported Spain’s rise as a world power (p.10).

It has been argued that silver declining to ordinary levels of profitability in the 1640s, led to a decline in the Ottoman and Spanish Empires and the collapse of Ming rule (all of whom collected their taxes in set quantities of silver [pdf, Pp10-11]).  More likely, it was the large (and economically disrupting) drop in silver flows to China from the cessation of Japanese silver production and exports as well as falling American silver production, aggravated by interruptions to supply lines and Phillip IV of Spain (r.1621-1640) attempting to crack down on corruption (pdf, p.33) which led to a sharp monetary (and thus commercial and revenue) contraction in the three Empires.


The Spanish and Chinese Empires operated in mutually supporting tandem; flourishing together in the C16th, both weakening in the C17th while the C18th revitalisation of Spain was based on a new cycle of silver exports to a stable and expanding Qing Empire. The two Empires then weakened in sequence in the C19th. The collapse of central authority in the Spanish Empire when Napoleon seized the Spanish king and crown prince and invaded Spain led to a loss of standardisation in Spanish American pesos, which worsened after Spain’s mainland American colonies gained independence, leading to the economic destabilisation (pdf) of the Qing Empire by serious silver coin deflation and copper coin inflation when the shortage of reliable silver coins began to seriously bite in the 1820s and 1830s.

Two millennia of mutual benefit

From the beginnings of the Silk Road system around 220 BC until the beginning of Industrial Age globalisation in the 1820s, a unified and prosperous China has been a boon to Eurasian trade. In more ordinary times, China’s long travail from the start of the Opium Wars (1839) to the start of Deng’s reforms (December 1978)–incorporating the demographic disasters of the Taiping Rebellion (20m+ dead), Japanese invasion (17m+ dead) and the Great Leap Forward (18m+ dead) as well as the more ordinary chaos of the Warlord Era, the Chinese Civil War and the Cultural Revolution–would have been a serious blow to global trade. As it was, beginning in the 1820s, the massive expansion of global trade from dramatic falls in transport and communication costs plus the dramatic and continuing rise in productivity more than compensated. But it also provided a deeply distorting perspective on China’s longer run role in the Eurasian, and later global, trading economy.

The economic resurgence of China since 1979 does not represent some threatening new development, despite its dubious internal political economy, it represents the return of a stabilising influence in human affairs. With the major Western Hemisphere Power and China once again linked by trade and monetary flows; except it is China’s dubious saving options, rather than dubious monetary structures, which make US Treasuries–the new “silver pesos”–so attractive (especially now the shine has gone off gold).


ADDENDA REVISED *Since I came across David Friedman’s paper after writing most of this post, the melding was not as smooth as it could have been. To expand on this slightly cryptic distinction, taxes on labour income applies to (monetary) taxes in highly monetised societies. Labour service is, of course, historically a common way to tax farming, but such extraction from coerced labour–particularly for local production–functions in terms of economies of scale as if it was a tax on land. Indeed, a classic reason to coerce labour is to compensate for downward pressures on rents due to labour scarcity/land plenty; so income from coerced labour can be regarded as a substitute for rents. Particularly as such coercion is all about landlord power.

There is also a distinction between early autocracies where water transport (typically a large river) permitted a centralised autocracy to levy labour service (e,g, Pharaonic Egypt, the Khmer Empire) and later periods when mounted warriors could control, protect and extract income from a village. In a situation where trade income was minimal, the ability to effectively aggregate such villages together was limited–i.e, there were diseconomies in revenue collection and no significant economies in taxing land–hence the tendency to small political entities.


  1. þorfinn
    Posted June 10, 2013 at 3:28 pm | Permalink

    A co-incident natural disaster could be a cause such as a volcanic eruption. Taveuni in the Fijian islands is recorded as erupting in 220.

  2. Jonathan D
    Posted June 11, 2013 at 10:18 am | Permalink

    Algebra doesn’t come from Al-Khwarizmi’s name, but from al-jabr ‘restoration’, one of his two operations for simplifying equations (replacing a subtraction on one side with an addition on the other).

  3. Posted June 11, 2013 at 5:37 pm | Permalink

    [email protected] I stand corrected 🙂
    [email protected] Krakatoa was a famously large eruption, but had little effect. While David Keys has an interesting theory. it does seem to a very specific case. The Santorini eruption, one of the largest known of, also seems to have limited effects.

  4. tony o'brien
    Posted June 11, 2013 at 5:56 pm | Permalink

    A corruption of the name of Al-Khawarismi gave us the term “algorithm”, the spelling a result of a medieval assumption that it must have come from ancient Greek.


Post a Comment

Your email is never published nor shared. Required fields are marked *