August, 1945 Soviets in Japan

August 11, 1945

The Soviet general watched the reverse Dunkirk with pride. The few landing craft were reserved for T-34s. Meanwhile, a mix of destroyers, minesweepers, fishing boats, passenger ferries, towed barges and even surfaced submarines was also unloading. Intelligence was dead accurate, enemy forces on Hokkiado were a jumble of distant second and third string units. The political and military confusion after the two atomic bombs was not helping either.

One small, poorly emplaced Japanese unit was blasted point blank by a crusier. A short distance away, another was being mopped up by Soviet Marines. A kilometer inland, a jeep mounted (thank you Uncle Sam) recon unit radioed that they had mowed down a militia unit that could not even make a straight Banzai charge.

The General turns to his aid “Japanese language broadcasts about the regional Soviet will start at once” “Prisoners and civilians need to be well treated and then immediatly released into the interior with propaganda packets", he then adds "The progressive POWs from Manchuria are very valuable, particularly the educated ones..."

-Will the Soviets set up a separate occupation zone and then a separate state on Hokkiado?

- How will the western allies react?
 
Last edited:
Amphibious invasions are hard. Not only do you need to work out the logistics of landing troops, protect the landing forces, you also need to be able to move inland rapidly and have follow up landings for reinforcements and resupply faster than the enemy can marshal their own forces to wipe out the beachhead.

Hokkaido is much further away than Sakhalin, Kuriles, or Korea. It is much riskier for the Soviets to land there than elsewhere. The Japanese fleet may be essentially destroyed, but they do have some ships and submarines. Furthermore, I don't think the Soviets have land based aircraft that could cover the invasion the fleet, and they don't have carriers.

When the Soviets invaded Southern Sakhalin in August 11, it took them 5 days to break through despite outnumbering the Japanese 3 to 1 and having a land border.

If the US is not involved in this and completely surprised, it means the Soviets have to do this entirely on their own with their Pacific Fleet.

So we have a situation where a limited amount of Soviet troops have landed. They have no aircover. Reinforcements and resupply will take a while. For the Japanese, they have second rate troops, but they have high morale since they are fighting for their home islands, interior lines, and plentiful supplies. They have complete control of the air although their pilots are probably not good.

Not knowing more details about the composition of Soviet and Japanese forces, I think at best the Soviets will have a besieged beachhead and not make much progress. The Japanese can quickly mass what forces they have on Hokkaido and contain the landing. Worst case scenario is that they destroy it after taking heavy casualties themselves. Still, it's only 4 days before the surrender - perhaps the Soviets can hold out.

In any case, I think the Soviets are acting very premature in your posts. They aren't going to make any unilateral decisions about an undefined occupation zone. The Allies and Soviets worked very closely on their postwar arrangements. The Cold War is hindsight. Everyone was hoping or expecting that there would be close cooperation between the three major powers. It took 2-3 years before it became clear that Anglo-American and Soviet interests (as defined by Stalin) were incompatible.

In any case, the US can quickly land troops in Hokkaido (probably Sapporo) if needed to secure the most important areas. Occupation zones will quickly be decided by agreement between Stalin and Truman. Perhaps the northern half of Hokkaido will become a Soviet occupation zone. Most likely, the US will insist that all of Japan is a US occupation zone and that the Soviets are to withdraw from the islands just like the US withdrew from the parts of the Soviet Occupation Zone in Germany once peace was declared.

Because the US has overwhelming naval superiority, there is no way that the Soviets can do anything in Hokkaido unless the US agrees. If the US gets angry, it jeopardizes gains that the US has already agreed to in the Pacific - the Kuriles, Sakhalin, and interests in Manchuria not to mention plans in Europe.

There are no vital Soviet interests at stake in Hokkaido. The US can prevent any gains in Hokkaido if it wants to. The Soviets want cooperation with the US in other areas more important to them. So the Soviets leave.

In any case, there will be no plans to set up a separate state in Hokkaido. There were no plans to do so in Korea or Germany. That happened by default as an unexpected measure. If for some reasons the Soviets are allowed to stay in part of Hokkaido, it is such a small and insignificant territory, it has no capability of becoming its own state. It's likely the Soviets give it up in some deal at some point. Perhaps in exchange for recognizing the PRC, an ancillary agreement after successful peace talks in Korea, or something else.
 
Soviet Invasion of Japan

If you're going to have a Russian assault on Japan, I think you have to butterfly away the A-Bomb or have it delayed until spring 1946.

Stalin obviously knew about the A-Bomb and once the Los Alamos test had been successful, he would have been aware of the weakness of the Soviet position in the future.

The second priority for Moscow was securing the Eurasian landmass and that meant in effect driving the Japanese from Sakhalin, Manchuria and Korea and either annexing these territories directly (reversing the Treaty of Portsmouth) or establishing pro-Moscow Governments.

Much of the Manchurian campaign took place AFTER the Japanese had formally surrendered on August 15th. The Soviet forces went on to achieve their objectives against an enemy which had given up. Without the A-Bomb, it's possible the Russians would have faced more serious and prolonged resistance.

The other point to consider is that the Japanese knew that a Russian invasion would essentially destroy every vestige of Japanese culture including of course the position of the Emperor. Once the Americans had conceded that the person of the Emperor would be respected, it became better to surrender to the Americans than to risk a Soviet intervention.
 

Cook

Banned
Hokkaido is much further away than Sakhalin, Kuriles, or Korea. It is much riskier for the Soviets to land there than elsewhere.
Only a couple of miles further than the southernmost Kurile Island in fact and the Russians were initially going to land the 33 Motor Rifle Division on Hokkaido, but the mission was cancelled sometime around the 27 August 1945.
 
Much of the Manchurian campaign took place AFTER the Japanese had formally surrendered on August 15th. The Soviet forces went on to achieve their objectives against an enemy which had given up. Without the A-Bomb, it's possible the Russians would have faced more serious and prolonged resistance.

The fighting in OTL after Japan's effective surrender was pretty savage, I doubt it could be more so. Nor by this point did the Japanese have many resources to hold off the Soviets outside Gas.
 
In the event that the Soviet Union manages to occupy the entirety of Hokkaido, the island will likely become their occupation zone, and later become the People's Republic of Japan or some other Communist entity. At the same time though the USSR could use it as a bargaining chip to gain additional Occupation territories elsewhere in return for handing it over to the United States. It entirely depends upon the value placed upon the island by both the USSR and the US.​
 
Amphibious invasions are hard. Not only do you need to work out the logistics of landing troops, protect the landing forces, you also need to be able to move inland rapidly and have follow up landings for reinforcements and resupply faster than the enemy can marshal their own forces to wipe out the beachhead.

Hokkaido is much further away than Sakhalin, Kuriles, or Korea. It is much riskier for the Soviets to land there than elsewhere.

Oddly enough from just about all maps I've seen of Soviet operations in the area the Soviets landed in southern Sakhalin and the southern Kuriles from the Sovestkaya Gavan region (so they would have had to bypass Hokkaido to get there) and all landings in the northern and central Kuriles occured from Petropavlovsk. So from a distance perspective it seems the Soviets actually covered much larger distances from Sovetskaya Gavan to Rumoi (their planned landing site on northwestern Hokkaido).

The Japanese fleet may be essentially destroyed, but they do have some ships and submarines. Furthermore, I don't think the Soviets have land based aircraft that could cover the invasion the fleet, and they don't have carriers.

They wouldn't need carriers as I do believe they did have land based aircraft that could cover any invasion. I know in one of the books I read there was this claim, but looking through various Soviet WWII aircraft and their ranges and the distance between the Soviet Far East and Sakhalin and Hokkaido, it's impossible to believe that the Soviets didn't have an entire inventory of aircraft from fighters to bombers which could cover Hokkaido. They used these aircraft to carry out attack in Manchuria.

When the Soviets invaded Southern Sakhalin in August 11, it took them 5 days to break through despite outnumbering the Japanese 3 to 1 and having a land border.

5 days isn't that long. It sounds rather typical actually as the US took about a week to secure Tinian despite extensive bombardment of the tiny island and outnumbering the Japanese defenders by 3 and 1/3 to 1. Then there was Iwo Jima which took months despite similar force proportions. From what I understand they got held up because they invaded from land first and came up against a defensive line across one of the narrowest parts of the island (so the Japanese could concentrate their forces there). The Japanese resistance collapsed when the Soviets landed in their rear area from Sovetskaya Gavan.

If the US is not involved in this and completely surprised, it means the Soviets have to do this entirely on their own with their Pacific Fleet.

The US couldn't be surprised by this because they had been trying to get the Soviets in for over a year. In fact Admiral Halsey apparently gave one of the reasons in his memoirs for the July 1945 bombing of Kure as being that it would ensure that the Japanese could not disrupt the planned Soviet invasion of Hokkaido....so if Admiral Halsey expected it, it would be surprising if the US leadership did not. They may not know exactly when the Soviets would plan to do it, but I really don't think they would be surprised when it does occur.
 
The Japanese can quickly mass what forces they have on Hokkaido and contain the landing. Worst case scenario is that they destroy it after taking heavy casualties themselves. Still, it's only 4 days before the surrender - perhaps the Soviets can hold out.

It is worth noting that the U.S. took 7,000 prisoners in the last days of the Okinawan campaign. Though most of these were Okinawan militia, it did show that even the iron will of the Japanese was collapsing due to starvation and casualties. I believe all of Japan's remaining first line and effective second line troops were in Kyushu or around the Tokyo area in Honshu to face the looming U.S. invasion.

In short... I dont think it is beyond the realm of possibility that the remant Japanese on Hokkiado would lack both the skill and the will to decisively stop the Soviets.

Because the US has overwhelming naval superiority, there is no way that the Soviets can do anything in Hokkaido unless the US agrees. If the US gets angry, it jeopardizes gains that the US has already agreed to in the Pacific - the Kuriles, Sakhalin, and interests in Manchuria not to mention plans in Europe.
Good point. The U.S. had also advanced into Czechloslovakia, a Soviet sphere country. I imagine that any Soviet sponsored "Soviets of Workers and Peasants" on Hokkiado would see U.S. sponsored elections in Czechloslovakia. Likewise, Albania may be pulled away from the future Soviet orbit as well.
 
Last edited:
The Soviets can pull it off rather well. I actually discussed this before in a thread on polichat, bringing in a thread on tanknet...

The Soviets carried out five amphibious assaults in three different locations on the same day. There were three landings in North Korea (note: the overland forces never made it all the way to the Yalu before the supply lines gave out), one in the Kuril's, and one on Sakhalin. The Soviets actually had a extensive merchant marine that could be (and was) modified into a improvised transport fleet. Still vastly inferior to the American fleet and even the Japanese invasion styles in the early war, but possibly enough to do the job.

The initial landing on Hokkaido was scheduled for August 21, but was delayed to the 24th or 25th because of unexpected resistance on Sakhalin, although a further delay into early September seems likely. The Japanese have very little prospect for significant reinforcements from further south. The boats that ran between Honshu and Hokkaido had been decimated by the Americans. The transfer of additional aircraft also appears to me unlikely for two reasons:

1. It would have burned fuel that was very scarce for the Japanese.
2. American air dominance means a lot of those aircraft would simply be shot down en-route.

This means the Japanese will have to make do with what they already have in Hokkaido. The biggest question is: how many aircraft do the Japanese have in Hokkaido in August of 1945? The Japanese ground forces are to ill-trained and to spread out to effectively counter a Soviet naval invasion by themselves, but if the Japanese aircraft can attack the Soviet transports, even as Kamikaze's, the landing will definitely fail.

Of course, there is nothing preventing the Soviets from trying again at a later time, but it would be months before the USSR could scrape up the transport capacity to try again. A second attempt after a failed first one would probably come after Olympic had begun, but before Coronet.

And for the bit on tanknet... both posts on the second page of this thread.

I can not speak for what may be present in "Downfall". But about 3 divisions certainly fits what I would expect of the Soviets.

Without any reference to specific Soviet plans, but only with the background of having read of several of their other operations crossing large bodies of water, here is how I would expect a Soviet operation to play out:

First 24hrs: Nights are long at this lattitude during October. During nighttime hours 3 or perhaps 4 re-inforced battalions conduct landings at disparate locations. Probably re-inforced with extra artillery (probably 120mm mortars) and AT assets. Perhaps one of the landings is an air drop. (The Soviets lost their taste for this in combat with the Germans in 1943, but appear to have re-gained it in 1945, as there were several air drops in their campaign against the Japanese.) Each landing force moves to establish a perimeter of about 1 Km radius from the landing point, building a hard but hollow shell with only a very small reserve of 1 or 2 platoons in the center (probably combat engineers, who are engaged in improving their landing site when/if not called on for combat).

The Japanese have no experience with Soviet methods in this kind of operation. The first counter-attack they conduct meets a very stiff defense. As the local Japanese commander, you are likely to interpret these as raids rather than full-scale invasions. They are tough fighters, but in positions offering no mutual support. You will probably seek first to contain them so that they don't grow into a larger threat, and so you work to put screening forces around them as you concentrate a sizeable force to counter-attack the Soviet enclaves one-at-a-time. Divide and destroy them in detail -- sounds like a good plan to most professional military men.

Second 24hrs: The battalions that face stiff opposition are given some fire-support by Frontal Aviation units, but are otherwise on their own. The 2 of the battalions that have succeeded in establishing 1km perimeters are re-inforced during the nighttime hours to re-inforced regimental size, and push their perimeters out to 2 or 3km radius. More artillery and air-defense assets are provided to these expanded beachheads, but they are still primarily a stiff crust, with little filling.

The Japanese have probably finished isolating either the airhead or one of the beachheads, and are busy reducing it. Their attempts to probe at the rest of the beachheads find them to be even tougher than the first, and so they will probably seek to concentrate even more resources as they move on to the next target.

Third and fourth 24hrs: The two largest and most successful beachheads again expand their perimeters out to about a 4km radius. The one beachhead which appears to be facing the least resistance will be re-inforced by two divisions during nighttime hours. As the forces concentrate on shore they are carefully camoflaged and remain still and silent during daytime hours. Each unit that lands carries with it all necessary fuel, ammo and food for 3 days. The second "large" beachhead, the one which does NOT have multi division forces gathering within, will actually become the more active in terms of offensive patrolling, vehicle traffic, and radio transmissions.

The Japanese will likely not appreciate the risk. Given how tough the Soviets are in defense, and how slowly they seem to be building up, they will continue with their one-at-a-time isolation and reduction work. Doubtful they will manage to complete the elimination of even 2 of the Soviet enclaves by the 4th day.

Fifth 24hrs: The floodgates are opened. Two divisions break out of the perimeter, one towards the closest other beachhead in a maneuver which envelopes a sizable portion of the local Japanese counter-force, and eventually clears a significant amount of coastline, the other driving multiple spearheads in an "expanding torrent" deep into the Japanese rear-area, rolling up artillery, logistics and HQ units. At least one airfield will be included as an objective in this maneuver. Also now, for the first time, landings (of the 3rd available division, as well as supplies) will take place during daylight hours.

The Japanese are now past the point of effectively defeating the Soviet invasion. Nothing they have on the ground can stop a Soviet mobile advance. Their Army forces do not have sufficient mobility to get in front of and contain the "expanding torrent", and their top-down decision making is confounded by the high pace of operations of the Soviet advance. Where and when they do manage to assemble a meaningful defense prior to the arrival of a Soviet spearhead, the Red Army simply turned the succeeding waves of the advance in another direction, and the Japanese defense is by-passed. If air resources are concentrated against the daylight landings, there will be no support for slowing the advance of the ground forces. But if the further daylight landings are not slowed/stopped, the torrent just grows and grows.

Could the Japanese have stopped a Soviet advance once it started? I doubt it. In China and Korea the model worked against them nearly perfectly. And the Japanese military, while capable of assessing failures and modifying doctrine to some extent, was certainly not fast at doing so.
The Japanese willingness to use Honshu-based resources against the Iwo Jima landings in February are a poor benchmark for the likelyhood they would use Honshu-based resources against a Hokkaido landing in October.

I believe available documentation of Japanese plans show that they were hoarding their Kamikaze aircraft (and Kamikaze boats) for the grand "final battles".

I believe in the Japanese view that Honshu was seen as the "final" arena of combat. Even Kyushu did not qualify for the status of "final", and so only a minority of resources were placed there, and only those local forces would have been used against the US in Operation Olympic.

If you were the Japanese High Command, and you knew that the US was eventually coming with a mighty fleet of thousands of ships, and that they were likely to land on Honshu to assemble a mighty army on the Kanto planes, and so to take Tokyo itself, how likely would you have been to release your last reserves against Soviet "raiding forces" on Hokkaido? The problem is that if you didn't, by the time you realized it was more than a series of un-connected raids, it would be too late.

That is why I believe the Soviets could have succeeded in invading Hokkaido.

You fail to understand the Soviet doctrine, as demonstrated in multiple advances across large water obsticals.

The USN emphasized control of the sea. The Americans approach was to amass a huge fleet, park it offshore, and occupy the ocean as they contucting their invasions against a single defended target. That doctrine was developed during 1942 and 1943, when the USN was tasked with supporting invasions a thousand miles or more from their fleet bases.

The Soviets would probably not have pursued such an approach. Their navy had no doctrine of sea-control, no experience in reaching out to hemispheric distances, and they shunned large amphibious landings against prepared defenses.

If they intended to deploy 3 divisions in an invasion, they would likely have started with 3 or 4 battalions (10 to 15% of their force) to obtain their foothold. They would have established their foothold by initiating several "reconnaissance in force" landings. They then would have chosen to re-inforce where they had achieved maximum success, allowing the less successful landings to whither on the vine.

By the time the main body began to move, the invasion landing and establishment of a beachhead would already have been completed. The movement of the divisional-sized forces would be more of an administrative/logistical move rather than an opposed landing.

The Germans came to understand this. If one reads the post-war writings of German officers, they often repeat the formula they developed for defending any river or coastline:
1 ) Make sure every foot of the line is under observation
2 ) Always keep strong mobile reserves
3 ) Immediately and ruthlessly counter-attack against ANY and ALL beachheads, no matter how small or insignificant
they appear to be. Even a platoon-sized enclave can become a major threat if left un-molested for 1 day.

None of this had been shared with the Japanese, and there is no reason to believe that the Japanese had developed this understanding on their own.
 
Last edited:
Japan is prepared at Kyushu and on Honshu outside Tokyo, not so much on Hokkaido. The Soviets should do well while the Americans will likely be having an explosive inner debate over the new intelligence showing how strong Kyushu is.

I presume MacArthur would be unwilling to change his mind and pursue an alternative plan.
 
I presume MacArthur would be unwilling to change his mind and pursue an alternative plan.

He might be okay with using atom bombs as tactical support weapons...

The problem is the Soviet landing on Hokkaido might cause the Americans too panic and inadvisedly try to move-up their attack date. On the other hand it might also push the Japanese (who were scared shitless at the prospect of communists taking over) too surrender to the US before any more of their territory falls into Soviet hands...
 
Manchuria and Hokkaido?

Nuker's magisterial post well-nigh convinced me the Soviets could have executed an invasion of Hokkaido in 1945. But would allocating necessary resources, especially aircraft toward a landing on Hokkaido have interfered with the invasion of Manchuria?
My impression is that Stalin et al decided invading Manchuria was the most efficacious use of available troops, tanks, motor transport and aircraft.
Even viewed from a Cold War perspective, helping the Maoists consolidate their position in N. China is more useful than holding Hokkaido as a bargaining chip.
 
Nuker's magisterial post well-nigh convinced me the Soviets could have executed an invasion of Hokkaido in 1945. But would allocating necessary resources, especially aircraft toward a landing on Hokkaido have interfered with the invasion of Manchuria?

In terms of the over-land advance? I would think it would only have a minor impact, if at all. But when it comes too the amphibious assaults on Korea, Sakhalin, and the Kurile Islands... sacrifices would have to be made that would likely very much impact these operations. It was heavier-then-expected resistance on Sakhalin that delayed the invasion in the first place after all.

If we could slide in an earlier POD that changes something (like the Japanese weakening the Kurile or Sakhalin garrisons, the Soviets coming out of the European theater stronger, and/or Operation Hulu* going faster) then we could conceivably get an August 11th invasion date. Otherwise, early or mid-September is a much more realistic date, and that necessitates Japan holding on longer.
 
Cryptic, Okinawa offered very little evidence that Japan's fighting morale was at any risk of collapsing.

The number of Japanese regulars taken as POWs on Okinawa, even including those taken after being rendered unable to resist by injuries, was so low as to be of no importance. Only the fact that this was the first landing on Japanese soil, bringing Japanese militia into play, led to the US taking a significant number of prisoners(by Pacific War standards) and the casualty rate among the Japanese militia and Okinawa's civilian population left the Americans with some extremely unpleasant assumptions as to what the invasion of Japan's Home Islands would involve in losses on all sides.
 
Cryptic, Okinawa offered very little evidence that Japan's fighting morale was at any risk of collapsing.

The number of Japanese regulars taken as POWs on Okinawa, even including those taken after being rendered unable to resist by injuries, was so low as to be of no importance. .

Considering that in all previous battles, nearly all Japanese troops fought to the death (including Korean para military troops), 7,000 Japanese troops of any composition surrendering is a very good sign.

Also, please note that even Japanese regular units in Manchuria had apparently not resisted Soviet advances to the death. Likewise, the Soviets took relatively high ratios of Japanese prisoners in the Kuriles (by Pacific war standards). At the end of the day, the Japanese willingness to commit mass suicide in August 1945 was not as strong as the hard core militarists imagined.
 
Nuker's magisterial post well-nigh convinced me the Soviets could have executed an invasion of Hokkaido in 1945. But would allocating necessary resources, especially aircraft toward a landing on Hokkaido have interfered with the invasion of Manchuria?
My impression is that Stalin et al decided invading Manchuria was the most efficacious use of available troops, tanks, motor transport and aircraft.
Even viewed from a Cold War perspective, helping the Maoists consolidate their position in N. China is more useful than holding Hokkaido as a bargaining chip.

In terms of the over-land advance? I would think it would only have a minor impact, if at all. But when it comes too the amphibious assaults on Korea, Sakhalin, and the Kurile Islands... sacrifices would have to be made that would likely very much impact these operations. It was heavier-then-expected resistance on Sakhalin that delayed the invasion in the first place after all.

If we could slide in an earlier POD that changes something (like the Japanese weakening the Kurile or Sakhalin garrisons, the Soviets coming out of the European theater stronger, and/or Operation Hulu* going faster) then we could conceivably get an August 11th invasion date. Otherwise, early or mid-September is a much more realistic date, and that necessitates Japan holding on longer.


One good thing about Glantz's work is that he likes to put in maps. Two maps of his works which I overlayed on each other really shows where the Soviets were in Manchuria in relation to where they expected to be. When I copied these overlays onto a Manchukuo map this is what I got:


The green lines show the Soviet front lines at the time Japan started the surrender process (August 14/15) and about 6 days after the Soviets began operations in Manchuria. The blue lines show the expected advances of Soviet forces by their high command with the days of operations numbered beside the lines (so for instance a 10 beside a line indicates the expected front line after for that area after 10 days of operations). Given how well the Soviets seemed to have prepared for Manchuria (such that in some cases they had advanced within 6 days well beyond where they had expected to be 15 days into the operation), I think the side operations in South Sakhalin, the Kuriles and Hokkaido would not have had much of an impact on the main operation in Manchuria. Besides which, having had a 3 month rest period between the fighting in Europe and the fighting in Asia and being able to bring in rested reinforcements from Europe and the European USSR (probably on a temporary basis in the expectation that the survivors from these reinforcements would be shipped back to Europe to continue occupation duties after the war or demobilized to get back to work in the planned economy) should allow the Soviets to bring in enough forces for Manchuria, Sakhalin, the Kuriles and Hokkaido (though I wouldn't expect the Stalin and his generals to consider advancing much further than in those theatres).

Manchuria campaign actual Soviet advances in green with days and planned advances in blue with d.jpg
 
Cryptic, that the percentage of Japanese regulars surrendering on Okinawa was a tiny fraction of the total was an extremely bad sign and the fact that more than 80% of the Japanese militia died fighting, with a majority of the prisoners first becoming casualties, was hardly reassuring. Then there was the death toll among the civilian population...
 
Top