The Question

By the latter part of the 13th century, the Mongols' frantic territorial expansion was losing momentum. In the east, their empire reached the Pacific, and their efforts to conquer Japan were thwarted by the famous "Divine Wind". In the south, their empire reached both the Himalayas and China's historic southern border. In the west, the expansion of the empire's Middle Eastern fiefdom, the Ilkhanate, was checked at the Battle of Ain Jalut in 1260, while the Golden Horde, the empire's Eastern European component, had its armies annihilated during the Second Mongol Invasion of Hungary in the winter of 1285-86.

Map of the Mongol empire

But what about the north? Why did the empire not expand all the way to the Arctic Ocean? On one level, the answer is obvious and straightforward: the territories to the north of the Mongol empire were extremely inhospitable and, from the Mongols' point of view, did not contain much in the way of resources worth extracting. But this answer does not explain why the boundaries were set where they were set specifically, and not, say, a hundred miles to the north or to the south.

A Possible Answer

The Mongols, of course, were great horsemen. Their armies employed vast numbers of horses, and the only way practical means of feeding so many horses for more than a single campaign would be to pasture them in massive tracts of grassland. Such grasslands are easy to come by in the southern portions of Siberia, but venture any further north and one runs into the taiga: dense coniferous forest with little vegetation that could sustain a horse. Is it possible that the taiga turned back Genghis Khan and his sons? Comparing the northern boundary of the Mongol empire with a map of Russia's biomes, it does seem that way; although the Mongols' control over the region around Lake Baikal remains to be explained by such a hypothesis.

Map of CIS biomes

Supplementary Questions

Amur River

The various maps of the Mongol empire which I consulted - no two are quite the same, even for the same date - are almost unanimous in stating that at least some portion of the Amur river formed a part of said empire's northern border. This suggests to me that, at some point, the Mongols made a conscious decision not to cross the Amur. But is there any other evidence for that point of view?

Khanate of Sibir

The Khanate of Sibir emerged from the collapse of the Mongol empire, and this state did extend as far north as the Arctic Ocean. (I am basing that claim wholly on the Khanate's Wikipedia page, which itself quotes page 148 of Riasanovsky's History of Russia (1999).) Was this Khanate truly Mongolian in character? That might seem like an unacceptably vague question. What I mean to ask, more specifically, is: was the Khanate of Sibir ruled by a corps of horsemen wielding composite bows? If so, that would seem to refute my hypothesis about the Mongols' advance being halted by dense forest, since the Khanate extended deep into the taiga; if not, then it might confirm it. The accounts of the Battle of Chuvash Cape that I've read seem to imply that the bulk of the Khanate's forces were foot archers, and not cavalry. Is that correct?

  • 11
    What do "boundary" and "set" mean in the context of this question? – kimchi lover Apr 19 at 20:14
  • 1
    @kimchilover As I understand it, medieval Siberia was divided up into tribes rather than strictly geographical districts. I would count such a tribe as belonging to the Mongol empire if it accepted the sovereignty of the Great Khan, or if it paid him significant tribute. By the term northern boundary I mean the line south of which most of the tribes in the area belonged to the Mongol empire, and north of which most did not. – Tom Hosker Apr 19 at 20:59
  • 1
    @kimchilover By setting the boundary I mean the course of action pursued by the Mongol leadership, which, by commission or omission, caused the boundary to advance so far north, but no further. If you require more precise terminology than the above, then please feel free to use whatever definitions you find most helpful in answering the question. – Tom Hosker Apr 19 at 21:01
  • 8
    I don't think the actual "boundary" is quite the crisp line pictured on the map. – chepner Apr 20 at 13:04
  • 1
    @chepner Of course. But the empire did not extend all the way to the Arctic Ocean, and therefore there must have been a line on the map past which the Great Khan's authority was less and less tangible. – Tom Hosker Apr 20 at 13:30

There always is a trade-off when occupying someone else's territory: the resources you can extract plus any strategic value versus the resources you need to invest to maintain control.

The wooded areas north of the steppe belt were relatively hard to control for the Mongols - e.g. even the 1207 campaign against the "Forest Peoples" had somewhat mixed results (see below) - and had little valuable resources. The Mongols of course did need some wood, but certainly not all wood that Siberia has to offer. The other commodity that these areas offer would be everything that can be made out of reindeer plus some luxury items like harem girls, fur, and falcons, but apparently none of those were important enough for them to try to conquer those areas. Especially when, to the east and west, there were China, Korea, and the empires Central Asia - areas that could be plundered and/or taxed, and that might be able to pose a military threat some day if left unchecked.

Generally, I believe it is a good idea to be rather skeptical towards those northern borders on your maps. I do not think they are based on more than guesswork and maybe some geographical features (rivers, mountains, vegetation zones) that may or may not have made a convenient border.

As for your observation regarding the Amur river, this Wikipedia article makes it appear as if the Mongols did have (some) control over areas north of the river at some point.

As for horses, the larch forests of eastern Siberia are actually not that dense. Are you sure it would be impossible for a large number of horses graze there? Forests have an obvious disadvantage compared to open pasture in that it is easier to lose sight of animals, but are there many problems beyond that?

Since another post has brought up the carrying capacity of steppe vs. woodland and because that is a subtopic that lends itself very well to sources that are available for free on the internet, I have done some spreadsheet calculations with statistical data from Mongolia. The source is called Хөвсгөл аймгийн статистикийн эмхэтгэл 2018 and can be found via www.1212.mn/. The area data is on p.13 and the lifestock count on p.55 and p. 61.

That report gives following lifestock counts and areas:

District Area(sq.km) Sheep Goats Cattle Yaks Cattle/yak hybrids Horses Camels
Tosontsengel 2042 83903 62101 8184 243 0 3510 13
Ikh-Uul 2023 64630 61380 7865 57 4 3505 40
Burentogtokh 3768 107356 74506 9134 3940 161 4759 40
Erdenebulgan 4694 15262 17489 8010 76 9 2235 0
Tsagaan-Uur 8735 3081 2175 6883 5 0 1306 0
Chandmani-Ondor 4487 8218 12337 8275 4544 914 1171 0

The first three districts are mostly open steppe, the last three have lots of forests (according to google maps). None of the districts has much farming, except Erdenebulgan on around 50 to 55 sq.km (p.46 of source mentioned above). The large differences in numbers of yaks and hybrids lead me to suspect that these are sometimes counted separately from cattle and sometimes not.

Calculating with 40 kgs per sheep or goat, 500 kg per head of cattle or hybrid, 400 kg per yak, 300 kg per horse and 800 kg per camel, I get around 5000 kg per sq.km for the steppe areas and 500 to 1500 kg per sq.km for the forest areas. Given frequently mentioned problems with overgrazing and overhunting in Mongolia, I think one can assume that a) those steppe areas are close to or above carrying capacity, and that b) there is no significant additional mass from wild mammals in the steppe areas. Limitations of this approach are that the selected districts might not be representative (e.g. districts further south might be able to carry more lifestock), that the masses for the different types of lifestock may be off (the whole calculation is somewhat sensitive to the weight ratio between cattle and sheep or goats), and that I am not entirely sure how well those statistics conform to the real lifestock numbers on the ground.

For a sanity check, one might look into recommendations for standard cows per acre natural pasture in South Dakota (p.2 figure 1) or Montana (p.4 table 1). Montana and South Dakota have roughly the same latitude as Mongolia, but annual rainfall is slightly higher. The South Dakota document gives a capacity of less than one standard cow per 16 acres for most (> 90%?) of the state, and the Montana document a capacity of less than one cow per 25 acres (2.2 acre per standard cow per month = 12*2.2 acre per standard cow for the whole year). That is 10 to 15 cows per sq.km and (at 450 kg per standard cow) 4.5 to 7 tons per sq.km.

Anyway, I think it is possible to conclude that forests are not very suitable to keeping large numbers of sheep and goat, and that the carrying capacity of wood areas is really significantly lower than that of steppe. Although the difference also seems to be much less pronounced than claimed in another answer to this question.

Regarding the the 1207 campaign against the Forest Peoples: my interpretation from the relevant passages (240 and 241) from Erich Haenisch's German translation of The Secret History of the Mongols is that two emissaries were sent to one particular forest tribe (the Khori Tumat) to demand tribute in the form of thirty harem girls. The tribe refused and took the emissaries prisoner. When Chinggis sent his troops, the Khori Tumat managed to kill one of his generals. His replacement managed to outmaneuvre the forest tribe, but only after constructing a road that the army could march on. In the end, a hundred Khori Tumat were killed in revenge for the general and the Mongols got their thirty girls. Note that all of this happened rather close to the steppe.

  • 2
    I'd be interested to hear more about the 1207 campaign. In my Googling, I hadn't found any mention of the Mongols actually going to war with peoples to their north. I assumed that they acquired their northern territories in a similar fashion to the Russian conquest of Alaska: they arrived! But the 'mixed results' of the 1207 campaign might answer my question perfectly. What difficulties did the Mongols encounter? Was a lack of fodder for the horses mentioned as an issue? – Tom Hosker Apr 19 at 20:34
  • 3
    Upvoted. Given that the tundra above was mostly only useful for reindeer herders, the Mongols most likely didn't feel like bothering with it, and the native herders were few enough on the ground that the easiest and wisest course of action would be to just avoid the pastoralist Mongols where they were. There's probably a lot of wiggle in that northern border as a result, it might even be a control continuum, but who really cares? – T.E.D. Apr 20 at 16:09
  • 2
    @T.E.D. '...but who really cares?' Ha ha. Well, if it wasn't for people with an unreasonable interest in the minutiae of world history, there wouldn't be a History SE in the first place! – Tom Hosker Apr 20 at 19:43
  • 3
    @TomHosker - Fair. – T.E.D. Apr 21 at 0:21

To answer this I would first answer the question, why did Orda retreated from Hungary and central Europe. As the guy named Lord of Hosts explained in his "articles" there were, and still are, economic reasons that contributed to mongol success and their failure. And by economic I don't mean financial only; in fact I don't mean them at all. Economy is not only about currency and money. Just like one economist calculated that Jurassic sauropods needed to move 100 times a day to feed themselves, he utilised economy to do that. Same applies to horses. So to the point - economic laws apply to war as well. As explained in mentioned articles Orda was thriving on steppes, and grass fields, not in the forests. And the main reason for it is simple.

The key to Mongolian victory was speed, range and mobility. Without these the ordu would have been bogged down, encircled and obliterated by weight of numbers of the scope which only settled agriculture could support. The Mongols’ unparalleled success in steppe warfare required a string of remounts, often as many as eighteen. Horse-power required pasture, superabundant in the great pastoral belts of Mongolia, China, Russia and the Middle East, but almost non-existent in the dark forests and arable lands of Europe.

The Alföld — the great Hungarian plain — was the largest unbroken pasture on the continent of Europe, in fact the only one of any extent. It was therefore the main target of Batu’s invasion. Prior to urbanisation in the twentieth century, it contained some 16,371 square miles of pasture, barely sufficient to sustain 150,000 wiry steppe horses. So even had the herders restricted themselves to ten remounts, the Alföld couldn’t even have sustained two tüman: only 15,000 nomads against an active Western military reservoir of millions.

In theory, by accepting that five-horse limit, which still enabled 65,000 Mongols to spread their terror through the deserts of the Middle East, Batu Khan could have maintained on a permanent basis in Europe only three tüman or 30,000 troops, half the number that he had brought with him on his initial foray into Poland and Hungary, and only a fifth of the number with which he had crossed the Volga and overcome Russia, an empty steppe country with a pastoral economy suitable for nomadic warfare, and with only a tenth of the population and military potential of the West.

So, Mongols needed steppes, vaaaast steppes, otherwise they couldn't stay for long, which means if they couldn't conquer a territory in swift manner they couldn't conquer it at all. They were unable to besiege fortified settlements which were often in Europe, even Russia, and after first wave, people in Europe learnt to build fortifications and not fight in open fields. That's the reason why second invasion failed either in Hungary and in Poland - people were not willing to fight in open and defended in towns and castles, that were out of reach for the Mongols. Here, enough to say that on a later campaign it would take Hülegü three years to transport a thousand crews of Chinese artillerymen and their siege equipment two and a half thousand miles from the steppes of western Mongolia in 1253 to Khurasan in 1256, and another two years before they could topple the walls of Baghdad a thousand miles farther west.

And now back to Northern Russia - by the time, it was heavily forested country, that would not offer enough fodder nor food for long campaign. In his book Das Rätsel der Menschwerdung: Die Entstehung des Menschen im Wechselspiel der Natur, Josef Reichholf, who is biologist talks about role of steppes (grass fields) in evolution of grass eaters, like horses are. It turns out that 1km^2 of grass can sustain 30 tons of mammals, when 1km^2 of forest can sustain only 300kg of animals. That's 100 times less. Even, if the forests were not dense, there still is nothing to eat for the horses.

As if it was not enough, lately it's been discovered by scientists, that at the time of first invasion there was significant climate change, when temperatures dropped almost one degree for several years making pastures less productive and making nomadic army, totally dependent on them, in even more difficult spot. This is minor however in comparison to the economy presented above.

And the last accord is forest itself in warfare. It renders cavalry and archery useless no matter how sparse it is. It was one of the reasons why Charles Martel took battle of Tours in the forest against superior muslim forces. It disabled cavalry for a great part and Moores had to fight "on European" terms, which meant, no speed, no mobility leaving light army forced to fight against armoured troops.

The more to the north, the less settlements, less fields and more forests, making the sustainability of the Mongol army problematic and conquest questionable.

  • +1, but I just did a some entirely unscientific spreadsheet calculations with statistical data from Mongolia, and I get 500 to 1000 kg/sq.km for woody areas and something like 5000 kg/sq.km for steppe areas.While I do not entirely trust the data for wood areas, the data for steppe areas is probably quite OK. I also think that those steppe areas are quite close to carrying capacity or even above, – Jan Apr 20 at 20:21
  • Of course such calculations are somewhat sensible to weight estimates for the respective animals, but still I think Mr Reichhoff may be overstating the difference in carrying capacity of wood vs. steppe. – Jan Apr 20 at 20:26
  • 1
    My family lives close to forests and there are no more than 300kg/km^2 of grass eaters (namely roe deers) in them. There is nothing to eat for them - leaves are too high and understorey is not very abundant in grass. Even if it was 1000kg/km^2, which is overestimation, it's still 1 horse/km^2. Just one. To sustain 50 thousand horses, which is gross underestimation, for 1 tüman (10'000 men), one would need 50'000 km^2. In reality way more, like two, three times more. You get the idea, numbers don't lie. – tansy Apr 21 at 0:24
  • 1
    Forests in Novgorod oblast (region), that's what we're talking about, are moderate/northern taiga, are behind the verge of deciduous trees like oak, consist mainly spruce (Picea abies), pine (Pinus sylvestris), sometimes birch and look like this: Korobozha Lake at Valdai, Novgorod Region. – tansy Apr 21 at 13:31
  • 1
    About overestimation and the area he refers to, it was verge of ice during ice age. On this map everything in green, was covered with grass, and it was further south, warmer during summer and "todays Central Europe winter", so it's possible to be able to feed 30T/km^2 of grass eaters.Here they calculate 1.7 acre/cow (450kg), which gives 65T/km^2. So it is possible to grow 30T/km^2 and 6T/km^2 in Mongolia quoted by @Jan is really low. – tansy Apr 21 at 13:59

Mythical answer: Ignach Cross stopped them from going north.


  • 2
    Welcome to HSE. Are you aware of any historical evidence for that? – José Carlos Santos Apr 21 at 9:04

Post number 14 at Historum, and the posts which follow it, suggest that the Mongol Empire extended much further north into Siberia, and was much larger, than it is usually depicted on historical maps, and perhaps was actually larger than the British Empire.

So possibly the correct answer might be that the northern border of the Mongol Empire was not where you think it is, but much farther north, possibly reaching to the Arctic Ocean everywhere in Siberia.

  • 1
    I'd be really interested to look through the posts you mentioned, but the link you've supplied just sends me to a 'Lost in this World' page, even after I create an account. – Tom Hosker Apr 20 at 19:45
  • Link has a defective [1] at its end which bricks the pagination part of the URL, suggested a edit fixing that – masterX244 Apr 21 at 15:17
  • The link has been fixed, at least for now. – MAGolding Apr 21 at 16:34
  • 2
    I think that post says that some Mongol expeditions got close to the arctic circle, but that is different from exercising any control. – Jan Apr 21 at 17:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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