Comparing sizes of the armies of the medieval East and West


Ad Honorem
Feb 2011
I don't know what you mean by "archaeological evidence". It's a road, I already provided pictures of the road. Providing anything else would be archaeological evidence for things that's NOT the road. I don't know what more people could possibly be thinking about.

As for the map:

Jan 2015
Some unexcavated ruined sections of the roads crossing the country.

For scale,
Comparison with a 12 meter wide road.

and a 2.5 meter sidewalk (right) with a 3.2 meter wide cycling path (left) for comparison.

Last edited:
Jun 2007
Before insulting your fellow posters you could at least check previous posts
For someone who resorts to cheap debating tactics to overwhelm your counterpart, that was light for him.

Let me assure you that I am one of those who enjoy reading very much East and Southeast Asian history, but I have combatted much against Sinocentrism in my own reading times and online discussion. Large army of roughly 100,000-200,000 was not an exception to this part of the world. Elsewhere, we saw this figure as well. The last powerful and effective Seljuq sultan Sanjar Khan had at his disposal 100,000 horsemen according to Ibn al-Athir on the eve of the fateful battle of Qatwan. For a grand ruler like him, this number itself shouldn't be inherently implausible, though we should still trust it with care. His force was, of course, a composite one, gathering from princes and amirs of his eastern domain, such as those from Sistan, Ghur, Ghazna (still under direct control of Ghaznavid rulers), Mazandaran and Khorasan (virtually all domains under his effective control apart from Iraqi region such as Mosul or Syria, which at this moment was formed by independent kingdoms under powerful atabegs such as the Burids or the famous Zangids.
Jun 2007
I will give you guys two accounts in Chinese historical records that represents what is likely reliable and what is utterly unreliable.

The first account is taken from a description found in Jinshi (History of Jin, a state that was found by the Jurchens in XII century) of the battle of Sanfeng mountain.


Zhang Hui (he was an officer in charge of the infantry units of the Jin army, the original size of this army was 150,000 men, 130,000 infantry and 20,000 elite cavalry which belonged to the so-called Zhongxiao army) and An Demu stood on the mountain top and saw that the Northern army (the Mongols) must have numbered 200,000-300,000 troops, they must have covered some 20 li. An Demo said to Zhang Hui: "If we do not engage them here, what else should we do now?" (This is in reference to the fact that the Mongols refused to give battles to the Jin army, in order to exhaust them on their march to the capital Bianjing).

As you guys have seen here, the number 200,000-300,000 is clearly just an impression of Zhang Hui and An Demu. The actual number of the Mongol army at this moment was only roughly 40,000 led by Tolui. He originally commanded only 30,000 troops, later Ogedei sent Kun Buqa (口溫不花) and Alchidai (按赤台) along with 10,000 cavalry to aid them. Ogedei later would join the battle with his main force. Thus the figure reported Jinshi was clearly unreliable.

This was the last great pitched battle in which the Jin would ultimately lost almost all of her talented generals and best army. Not all 150,000 men perished on this mountain however, but a significant number of them was massacred here and along the way to Junzhou (now in Henan province).

This was height of the Mongol art of warfare, yet few historians paid enough attention to reconstruct this battle.

The second account is taken from the Yuanshi (History of Yuan dynasty), chapter on Annam (Northern Vietnam today), the record was related to the second Mongol failed conquest of Vietnam in 1288.


In the twenty fourth year (1287), the first month, 1000 Xinfu (men who were former subject of the Southern Song dynasty, now turned to serve the Mongols) troops were assigned to Abaci to conquered Annam, then [the court] ordered 70,000 Mongol, Han and Kuan (troops who were paid on duty) from three provinces Jianghuai, Jiangxi, Hunan, 500 boats, 6000 Yunnan troops, 15000 men from Hainan Island.

Adding all this together, we yield the total number of this army at: 92,000 men. With such a clear breakdown of this army composition, one could hardly doubt its veracity. The total force was later confirmed by other sources which put it at 100,000 men.

So, these are clearest example. In rarer cases, you could even end up having the date when these forces were called to mobilize, and where should they meet together, along with other more interesting details such as prostitutes who followed the army.
Last edited:
Mar 2019
For example, the medieval kingdom of Vijayanagara is said to have fielded more than 200k men by both European and Persian-language sources. Many battles that are contemporary of it, speak of similar numbers of men. No matter how skeptic you are, you can not scale down a number of 200k which is claimed to be true by multiple sources, to mere 7 or 8 thousand.
vijayanagar was probably much bigger than france, it covered nearly entire south india, and vijayanagar was a very wealthy state, it was trading diamonds when the rest of the world hadn't any, it was using turkic soldiers as well.

Mar 2019
In case of India, for example, we read accounts of much larger armies coming out of kingdoms which were much smaller than say France of 100 years war era and equally feudal/decentralised.
decentralized doesnt mean feudal to be honest, i dont know why such european meanings have been imposed on smaller indian kingdoms, im not much learned on the political systems on india after gupta era, but i can give two contrasting examples of maurya and gupta empires and their management, mauryans have been thought to have run a very centralised govt. As opposed to that gupta empires who ran a very decentralised govt, so even though both were big empires and covered nearly entire indian subcontinent, i think the system of govt in one of them was decentralised, so i dont think it means feudal.