I'm reading Fukuyama's "The Origins of Political Order", and on pages 92-93 he makes the following claim (emphasis my own):

But most important, the state that emerged in China was far more modern in Max Weber's sense than any of its counterparts elsewhere. The Chinese created a uniform, multilevel administrative bureaucracy, something that never happened in Greece or Rome.

With regards to precise time-period, I admit that it isn't entirely clear to me either whether he is referring here specifically to the Qin dynasty, or if he is also considering later administrations.

A bit later, he also states that the Roman empire was much more heterogeneous than the Chinese (which I agree with), but I don't think that is specifically what he is referencing here in terms of the differences in administration.

I've read little about imperial China, while I've read Gibbon's work on the Roman Empire and "The Roman Revolution" by Ronald Syme, as well as some other, varied reading from more disparate sources. To me, it seemed that the Roman administration was quite sophisticated and, given the Empire's longevity, relatively successful.

My question is thus two-fold: (a) what exactly should I understand under "uniform, multilevel administrative bureaucracy" (especially in comparison to, say, the Roman administration), and (b) is it true that no administrative structure comparable to that of early imperial China (and potentially later imperial China) ever existed in the Ancient Greek and Roman worlds (particularly the Roman)?

  • 1
    Can you describe the research that led you to conclude that "the Roman administration was quite sophisticated"? I confess that I don't think I can point to the Roman Imperial administration. "Roman magistrates should not be thought of as bureaucrats within a unitary governing institution. Rather each magistrate is an independent actor, granted certain powers to oversee the public interest in a specific field. " ACOUP
    – MCW
    Commented Oct 31, 2023 at 10:40
  • 4
    I don't know enough about Chinese history to answer the question, but I do know Roman history, and the nature of Roman government changes radically over the centuries: Republican Rome had no formal bureaucracy -- all was done ad hoc. Something resembling bureaucracy developed slowly during the Principate, but even after 150 years was still rudimentary. (The Principate was been described as being as much a federation of cities as an empire -- Rome relied on city governments for most civilian matters.) The late Empire may be better seen as an army with attached civilians than as a government.
    – Mark Olson
    Commented Oct 31, 2023 at 12:14
  • 2
    @MarkOlson - Unless someone comes with a much more complete answer soon, I'd honestly like to see that comment made into an answer. Its easily more than half an answer (on the theory that China could hardly be doing it less organized).
    – T.E.D.
    Commented Oct 31, 2023 at 13:15
  • 3
    S.E.Finer's 'History of Government' Vol.I Book II has interleaved chapters on the Roman Republic & Empire and China from the Qin to the Han, with the aim of comparing their development. I can't remember (read this >2 decades ago) his stance on this point, but if you compared the relevant sections of each chapter (each state is analysed using a standard format), then you would find a answer to the question there.
    – Matthew
    Commented Nov 3, 2023 at 21:31
  • 2
    @Matthew This looks like an exceptional piece of scholarship; thank you for bringing it to my attention. It is brutally expensive though, so this might be a "library-only" type of work for me in the foreseeable future.
    – Trisztan
    Commented Nov 4, 2023 at 2:34

3 Answers 3


I thought MCW's answer had some very good points, and I upvoted it. But I think I will try to expand some more on the evolution of the Roman government "system". (The scare quotes express my doubt that it could be called a system for most of the Western Empire's history.)

To start with, sources are sparse and pretty much all biased, usually in favor of the senatorial class -- the top aristocrats -- because they did most of the writing. By the mid Principate, though, Equestrians -- the much larger rank of rich people who were not Senators -- did a lot of the writing, but even then they were clients of Senators or were expecting to enter the senatorial class as their careers progressed.

For the mid-Republic (say from the First Punic War to the Third Punic War) we have various fragmentary histories and not much more. The late Republic is rich in sources -- Caesar and Cicero, of course plus historians, again -- but they still reflect a limited view of the world. The Principate (Augustus through perhaps Vespasian/Titus) and the Dominate (Domitian through the late Antonines) are pretty well documented and then things get really complicated with the mess of the Crisis of the 3rd Century and then the restoration under Aurelian and Diocletian, which turned to mush not too long after Constantine and the whole thing died a hundred years later.

I wish I could recommend a good book on the evolution of Roman government, but if there is one, I have not found it.

In the mid-Republic, government of the Empire was by the city council of Rome -- the Senate -- and provincial governors typically served for a single year. There was no permanent staff in the provinces, and governors brought members of their own households and maybe a few young men of allied families and made do. The governors, basically, had two duties: act as chief judge in the province, and command any military in their province. (Tax collection was done by private enterprise who bid an amount and then went to squeeze that plus a profit out of the province.)

The best way to understand Republican Rome is if the Mafia ran a country. Everything was custom and legalism (when convenient) and who had the most clout. The late Republic demonstrated this through fifty years of civil wars including Caesar successfully becoming capo de capo, but ending only when Augustus did it with more finesse. It was all very, very personal, and pretty ad hoc.

When Augustus started the Principate, he began an evolution of appointing governors for longer terms -- e.g., Pilate was Procurator (basically, a junior grade governor) of Judea for around ten years -- but does not seem to have done much more than that. By the mid-late Principate we start to see powerful long-term administrators in Rome. (I, Claudius provides a pretty good picture of this, with Pallas and Narcissus and others acting as ministers over parts of the government.) But it was still seen as a Family affair (e.g., Pallas' affair with Nero's mother...)

The early Dominate certainly saw further evolution: Most of the Empire was run by legates appointed by the Emperor and their terms of office were often long enough for them to develop a process, but it was still very, very personal. The best example of this is the letters of Pliny the Younger, a senator who was in Trajan's second circle of buddies. Pliny was governor of Bithynia and published his correspondence with Trajan and it has survived. Pliny writes many dozens of letters to Trajan asking for detailed instructions on how to do this or that, and Trajan responds personally. (He had a secretarial staff, and probably dictated parts and had the staff fill in boilerplate for others, but still. There were dozens of governors at Pliny's level, all wanting the Emperor's ear. Overwork was the Emperor's constant companion.) The picture we get is still of an Empire which has very, very little routine bureaucracy.

The post-Constantine Empire had developed something like a bureaucracy. We have an excellent picture of the state of affairs around 400 AD in the Notitia Dignitatum, a work describing a couple thousand civil and military offices around the Empire. But this organization was too little, too late and the Western Empire fell 75 years later.

Some general comments:

First, I've ignored the military, and this is probably a mistake. While the war-fighting function of the military is probably irrelevant to this question, there is evidence that in times of peace, the army did a lot of miscellaneous work in the provinces, especially what we'd call public works. They also probably provided police functions, to the extent that such existed. I suspect that a lot of it was informally organized between the governor and the top army officers in the province.

Second, the Empire was an intensely urban thing -- even near Rome, the countryside could be lawless and Rome didn't care a whole lot about it. Rome relied on its cities to provide local services for each city and its hinterland. (Service as a city magistrate brought Roman citizenship and other Imperial bennies.) Rome really didn't want to mess about running cities and as long as they were good citizens (i.e., paid their taxes and didn't revolt too much) it usually didn't care.

Third, it's very important to understand the Roman familia. This not our nuclear family, but would include dependent clients and a large number of slaves and freedmen. All were people who owed support to the head of the familia. I'm still not sure I really understand the relationship, but it seems, more than anything, to be very Mafia-like. But everyone from the Emperor down used their familia as their staff. (Which meant that continuity was not terribly great.)

Fourth, for most of the time, Rome did not have a written law code. It had laws, to be sure, and some were recorded, but it was all very messy and complicated. In the high Republic, when a Praetor (essentially a top judge) took office, he would publish a list of the laws, procedures and customs he intended to follow. If he messed about too much, of course he'd get push-back, but it was still very much a personal thing.

So, that's a broad outline of Roman government. It's only that, but I think the general picture of an always-evolving, terribly ad hoc, very personal approach to government is accurate.

I wish I knew enough about ancient China and could make a comparison. I hope someone does.


I'm currently reading Mary Beard's new book Emperor of Rome (an excellent book which goes into a lot of detail about Roman administration). In it she writes:

Leaving aside the army, no other empire in history of the world has operated with so few official boots on the ground (the Chinese empire had proportionately twenty times the number of senior administrators than did Rome).

pp 232-233. This is not original research, but Beard certainly counts as a pretty authoritative source.

Added Even Later:

If you can find a copy, Ronald Symes' 1939 book The Roman Revolution goes into the development of the bureaucracy and government by Augustus in considerable detail (chapters 25-27, particularly).

It's consistent with what I wrote, above.

  • Why do you start Dominate from Domitian? AFAIK Dominate is typically considered to have started following Diocletian's restoration from the Crisis of the 3rd Century, or following Aurelian at earliest. While Principate spans from Augustus til the beginning of the Crisis of the 3rd Century. And this is how Wikipedia describes it.
    – gaazkam
    Commented Nov 1, 2023 at 19:41
  • 2
    @gaazkam Domitian made a major break with the Principate which I think quite important, including an increasing reliance on Equestrians in the imperial service. We're slicing up a continuum here and I have no trouble at all breaking the Domitian-Diocletian++ era into several increasingly autocratic sub-eras, but for this narrative that much subdivision seemed excessive. YMMV, of course! Perhaps there's a better term for that era.
    – Mark Olson
    Commented Nov 1, 2023 at 20:04

A bureaucracy is a form of work organization. The historical meaning of the term refers to a body of non-elected government officials, but is nowadays understood as an administrative system used by corporations and public institutions. Cornell.edu

Compare with

"Roman magistrates should not be thought of as bureaucrats within a unitary governing institution. Rather each magistrate is an independent actor, granted certain powers to oversee the public interest in a specific field. " ACOUP


"In view of the slowness of communication, the administration of the empire was centralized to a fantastic degree ("Jones Α. Η. Μ. [1964], The Later Roman Empire, Blackwell: Oxford.)"

It has been a couple of years since I read Fukuyama Origins of Political Order; I loved the book, but frequently felt like there was subtlety to the language that I might not fully appreciate. I think that is what is happening here.

Fukuyama is comparing Roman governance and administration to that of the Chinese. The periods don't matter; Fukuyama is not comparing China of year X with Rome of year X; Fukuyama asserts that at some point China developed a sophisticated multilevel administration and Rome did not. (Imagine instead that the assertion was "The United States developed the Atom Bomb while Fiji did not". That comparison would not be helped by "The United States developed the Atom Bomb in 1945 while Fiji did not" - the point is that the US developed the bomb at some point in history; Fiji did not develop the bomb at any point in history)

I've read the data, but the total number of people in the Roman administration was only a handful (particularly prior to Diocletian's reforms) Rome's governance was entrusted to magistrates with assistants. When the magistrate's term ended, they left and a new magistrate arrived. There was almost no bureaucracy. Rome governed through a combination of client/patron relationships, a powerful military and (custom/tradition).

n keeping with his move from an ideology of republicanism to one of autocracy, Diocletian's council of advisers, his consilium, differed from those of earlier emperors. He destroyed the Augustan illusion of imperial government as a cooperative affair among emperor, army, and senate.[219] In its place he established an effectively autocratic structure, a shift later epitomized in the institution's name: it would be called a consistorium, not a council.[221][Note 13] Diocletian regulated his court by distinguishing separate departments (scrinia) for different tasks. Wikipedia:Diocletian's reforms

Prior to this Roman administration had (essentially) no departments, no officers, no professionals. (The cursus honorem effectively prevented the development of a bureaucracy. To the extent that there were professional bureaucrats they were slaves attached to the office. (You could make an argument here, but I think it would miss the point).

I'm also not an expert on Chinese governance, but

The Chinese government during the Qing was an integrated bureaucracy — that is, political power flowed from the top to the bottom through a series of hierarchically ordered positions that extended down to the county level, where a local magistrate headed a county office, called the yamen. This hierarchically integrated bureaucracy was remarkable because the people who had positions as officials within the bureaucracy were not there because they were members of a hereditary aristocracy. Rather, they had acquired their positions according to a system of merit. This system of meritocracy — perhaps the first of its kind in the world — was established on the basis of government examinations. Columbia.edu


The Qin dynasty (221–207 BCE) established the first centralized Chinese bureaucratic empire and thus created the need for an administrative system to staff it. Recruitment into the Qin bureaucracy was based on recommendations by local officials. This system was initially adopted by the succeeding Han dynasty (206 BCE–220 CE), but in 124 BCE, under the reign of the Han emperor Wudi, an imperial university was established to train and test officials in the techniques of Confucian government. Britannica

Rome never had anything resembling a civil service system.

The distinction is subtle, and in part relies on very different conceptions of Empire. Rome ruled Romans and "admitted" other peoples to the Empire. It took wars to grant even close allies civil rights (a struggle which was never achieved). Rome's administration of non-Roman cities was to establish a treaty/relationship (this is far more complex, but I'm trying to write an answer not a book. ACOUP covers this in much more detail, but for complex reasons is not the ideal citation; that would be a separate answer), that established military and financial obligations, and a form of "profit sharing", plus the illusion of mutual defense. Rome didn't care what happened within the city/nation, so long as it didn't happen to a Roman citizen.

China (Chinese scholars, please forgive and educate me if I'm wrong) never had such relationships; they accepted tribute, but China ruled all the land within the empire and had a strong, multilevel administration that survived the death of the emperor (and even often swaps in the Mandate of Heaven).

This is a subtle distinction that is difficult to pull a single quote.

OP assumes that size, wealth, standard of living, etc. are associated with bureaucracy; I don't think those assumptions are valid in the context in which Fukuyama uses the word.

Not my particular area of expertise, so I'm happy for corrections.

  • 2
    @njuffa The reference given in the source I pulled the citation from is "Jones, Α. Η. Μ. [1964], The Later Roman Empire, Blackwell: Oxford." So, I'd say that's probably it. The source itself is this: jstor.org/stable/40751852.
    – Trisztan
    Commented Oct 31, 2023 at 22:25
  • 5
    The issue with the Columbia.edu quote is that it talks about the Qing Dynasty (1636–1912 CE) while a comparison with Ancient Rome should be with its contemporaries, the Qin (221–206 BCE) and Han Dynasties.
    – dROOOze
    Commented Nov 1, 2023 at 5:21
  • 4
    @dROOOze Why? As far as I can tell, Fukuyama's quote that's being questioned doesn't compare contemporaries, but, rather, the development of a society as a whole. Whether that's appropriate or not is a separate question. Commented Nov 1, 2023 at 12:37
  • 8
    @DanilaSmirnov It's really bad history to compare a modern state to an ancient one, to the point of meaninglessness. Compare such an absurdity as "The Americans developed nuclear weapons, which neolithic man never did." Moreover, the work in question is The Origin of Political Order and he mentioned "the state that emerged." I'd say it's reasonable the comparison should highlight differences among those that are alike.
    – cmw
    Commented Nov 1, 2023 at 14:00
  • 5
    @DanilaSmirnov IMO the Qin/Han dynasties are unilaterally a better point of comparison than the Qing dynasty because they are contemporaries which share a temporal and physical connection (they had an active trade route with Ancient Rome, faced similar challenges with them, and thus could adopt each other's models of administrations if there was a clear benefit). Under other circumstances (e.g. comparing Pre-Columbian Mesoamerica with BCE Egypt on a thesis which compared the role of pyramids in these societies), I would agree that not being contemporaries wouldn't be important.
    – dROOOze
    Commented Nov 1, 2023 at 19:55

The quote from Fukuyama is explicitly reporting the opinion of Max Weber, one of the founding figures of modern sociology. To understand what Fukuyama means, one has to look to Weber and to what others have said about him. This is arguably more relevant to the question posed then what historians today have to say about China or Rome.

Weber was not primarily a sinologist or expert in China and was writing over a century ago. Historians of China today can argue against his view on many points. Weber was generalizing about the overall patterns of the Chinese state in order to explain the relative durability of its empire over two millennia. The details of those patterns varied significantly from dynasty to dynasty, emperor to emperor, and region to region. But the key point that Weber was first to point out is that compared to other ancient states, administration of the Chinese empire was relatively rationalized.

For a little more on all of this, see "Max Weber on China" by Van Der Sprenkel (1964). This article summarizes Weber's key ideas about China as follows:

(1) Weber's correct assignment of the beginnings of "rational" policies in internal administration, military organization and the like, to the Warring States period; (2) the importance he gives to water- control as the factor mainly responsible for the growth of centralized political authority; and (3) his unerring identification of the "literati" as the key status group in Chinese society, and of the bureaucracy as its creation and creature.

One could elaborate on each of those points but perhaps the most emblematic aspect of the rational character of Chinese bureaucracy was the recruitment and promotion system. As Creel (1964) explains in some detail, at least a century before the common era, there was a process, which included dedicated exams and universities, allowing even common people to become state bureaucrats. Moreover, there advancement through various ranks were determined by relatively objective criteria which were tracked with written records. While elements of bureaucracy in this sense also existed to a lesser extent in ancient empires to the west, such practices were never as developed or as persistent anywhere else as they were in China.

Regarding the Roman empire, Weber did not deny that bureaucracy existed there, but he emphasized the extent to which military and aristocratic power shaped and limited it when compared with China. According to Antonio (1979):

Roman social structure was characterized by extreme verticality and low levels of social mobility (see MacMullen, 1974:88-120). Two aristocratic orders monopolized the most powerful and lucrative social, economic and political positions. It has been estimated that the senatorial order constituted approximately two-thousandths of one percent of the Roman people.

This is perhaps the clearest point of contrast with China's massive and meritocratic bureaucracy. And for Weber, this is important to explaining why the Roman empire collapsed in the way that it did, producing European feudalism.

Without a doubt, Weber was making a broad generalization, but this is the central context of the Fukuyama quote. If you want to get deep in the weeds of how bureaucracy worked in different periods of the Chinese and Roman empires, check out the edited volume, State Power in Ancient China and Rome (2015).

  • 1
    "Roman society was more unequal, and also less urbanized." Source for this, particularly the latter claim? I looked it up and found varying estimates for the Roman urbanisation rate, from a minimum of 10% (though more likely 12%-13%) to around 20%. In comparison, the only thing I found for the Chinese urbanisation rate
    – Trisztan
    Commented Nov 3, 2023 at 1:30
  • 1
    was this, which claims that the Tang dynasty had the highest urbanisation rate of all previous dynasties at only 10%. So I'm confused on how the Roman Empire was less urbanised. With regard to inequaliy, are you referring to wealth inequality or general social mobility? It seems to me that the wealth inequality in both was similar
    – Trisztan
    Commented Nov 3, 2023 at 1:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.