Kagemusha: Fire Without Smoke (or: How to End an Epic)

The guns used in Kagemusha trouble me for a few reasons.

For one, I don’t know if I should call them arquebuses or muskets. These terms seem often to be used interchangeably, while at other times there is some sort of a distinction between them, but one that remains unclear to me. Wikipedia’s entries on the subject only make me more confused. What say our resident history and gun experts — what should we call those things that go ‘boom’ in the last battle scene?

Then, and more importantly, the guns don’t seem to function realistically in the film. For one, shouldn’t there be far more smoke? And secondly, isn’t their actual range far shorter than the film suggests? I am almost completely ignorant of weapons, but I noticed that also Prince mentions these things in his commentary, so I thought I might actually be right.

And with so much of the film historical, down to extremely small details, it is puzzling that the guns are so unrealistic, especially since they really are a central character in the final scene. Do you think that Kurosawa did this deliberately, and if so why? The guns now remind me of westerns, a genre that Kurosawa was of course very much into, but that hardly seems like an explanation.

Also, if I may be creative for a while, had more smoke been produced by the guns in the final scene, it would have opened up a new way to end the film. Namely, we could have had the gun smoke drift over the small valley through which the Takeda troops are attacking. The smoke would cover the action after the first shots, effectively pulling a curtain over the battle and preventing anyone, including us, from seeing what is happening. Since the film doesn’t want to show us the action itself, this way the camera could still show us the place of action but no action itself, rather than being confined to showing us the reactions of the observers (an effective but I feel somewhat forced directorial choice). After the last wave of attacks, silence would fall and the smoke begin to dissipate, revealing the mass murder that has taken place.

But I suppose it would actually have been similarly unrealistic this way, not to mention somewhat too theatrical, as well as perhaps being less clear about the fact that the generals knew exactly what they were doing when they were sending their troops to death?

I’m no expert, but that never stops me from pretending to be. 😉

I’m not sure what the correct name should be, the common term nowadays is just black powder muzzle loader as similar guns are still in usage today for hunting. Perhaps the difference of arquebuse and musket are simply related to the time period versus any mechanical difference. As both do load from the muzzle, are smooth bore, and used a balled 0.5″ projectile.

Indeed smoke should be horrendous, in fact even up to America’s civil war(1861-1865) despite major advances in firearms and gun powder, it was reported that the shooters were blinded by the first shot due to the lingering smoke. In war, after the first volley of fire, it was reported the entire battle field was hidden from a thick lingering smoke.

I would not at all think unrealistic to see nothing but thick smoke clouding everyone’s view, especially from the massive amounts of gun fire that happened towards the end of Kagamusha.

Accuracy, giving the rifles were smooth bore, and the projectile was crudely rounded into a ball shape, the result was a projectile that didn’t expand larger then the barrels diameter, so to dig grooves in the bullet from the rifling in the barrel, both sealing off gas pressure behind the bullet, and creating a stabilizing spin as seen in modern weapons. Instead the bullet bounced around hitting against the walls of the smooth barrel due to loose tolerances, as too the gas pressure from the gun powder escape though gaps in between projectile and barrel, both creating a unstable projectile. The accuracy that commonly would not result in a hit in a human size target past 50-75 yards, due to previous mentions, and poor ergonomics and no sights. Ranges further were not achieve till bullets took a more conical shape, and rifling in the barrel was introduced, in the early 1800’s (not 100% on the date).

Whatever the effective range is, the gun simply had no repeatability, where one round would hit, the next would not, the further the distance the greater the change. And even today, there is no such thing as being able to place bullets into the same exact hole as the last(although damn close with very specialized weapons and trained shooters, at large however, no way possible or even war practical)

Part of the lack of accuracy at long ranges with weapons of this time period, was due to the resistance of wind upon a large, relativity slow moving projectile, over coming forward movement and redirecting in a way that often results in a curve. As well as the bullet not coming out the same exact position within the barrel due to previous mentions bouncing around. If you shoot paintball guns, or BB gun, or even played golf(golf balls now have dimples to increase stability now), you can see the result of a round object being overcome by air’s resistance and taking a curve which is really more of a hard deceleration, after a short distance of straight flight. Hence while modern rifle rounds are elliptical to lower coefficients(boat-tail), still however at great distances they are greatly effected by air resistance and drift to some degree away from the straight line they once went( ignoring speed reduction and gravity reducing original height.)

The point is, the part in which the sniper hits the tree, would be next to impossible, due to distance, and general physics at which the gun of the time period operated under. And even would be difficult with modern weapons giving the tree’s diameter and distance.

The distance that Takeda supposedly was when he was hit(same as tree), even with some incredibly accurate pre-measurement, there is no way to be certain you would hit anything at all(it was dark). Although certainly possible, and even a theory to the real-life death of Takeda, although no information is known of where or how far about the sniping took place to my knowledge.

Nevertheless the accuracy shown is well above what could be expect of guns of the time period, and some degree even today in typical war arsenals.

This is all very forgivable to me, after all no movie, even the ones trying to be historically accurate ever show guns as they are in real life. And with movies enforcing that a person can simply shoot like crazy while running, at a moving target and still get perfect shots. I do wonder if a movie ever produced real-life gun works if the audience would deem it fake or stupid. Theatrical too, it doesn’t look good, if the hero picks up a gun, and can’t hit a damn thing, as would most likely happen giving the typical situations.

What I did find very annoying, and what disconnected me from the movie was the sound effects.

Specifically the lame Western movie sounds that are suppose to represent ricochets, whizzed by,and various other non-sense. To my knowledge the rifles of Kagamusha’s time period were not capable of launching a projectile faster then the speed of sound. So minus the deep cannon like sound of the gun powder igniting, the bullet itself is largely silent. So none of that “Pshhooo” should be had, nor any high pitched ricochet giving the bullet was big ball of lead. When impacting a object at worst there is a deep thud sound and certainly would not ricochet off wood and the light gauge metal of the time period. Which ricocheting is really rather uncommon for any bullet new or old, minus a round hitting at an extreme angle a hard object, and BB guns, and some shotgun shot.

All these Western-like sound effects do, is make the shooting scenes somewhat comical and less serious. If the shooting scenes had very deep blast and lots of smoke. I think it would deepen the impact of the

visuals. Instead it seems to lighten them up, turning drama into action.

Even if we ignore the science behind it, or even that hypothetically the bullets do produce these sounds. The effects once again still play the same negative role. It is this I think Kurosawa made a poor choice, the dramatics of death in the last scene being rather painful for Kagamusha, and the generals to watch, need not have a bunch of action like effects to diminish their agony.

Vili’s idea of having a very large presence of smoke. As previously mention are not at all unrealistic to think the battle field curtained in smoke. Vili does make a great point however, that the smoke would in effect hide the battle, to the point that it may not be clear the generals knew they were sending people not to fight, but to die.

Nothing that could not be clarified with some acting.

The cinematic value of a smokey battle find could of been use to great effect. If done as Vili suggest I do agree would strengthen the impact of the tragic deaths, and Kagamusha being crushed by the result.

Great post and great reply.

The portrayal of gunfire ts something that has always puzzled me about his films. I think Ran is, if anything worse, in that the battle scene where the horses run the gauntlet of gunners is particularly jarring, the rate of fire is vastly greater than would have been achievable in reality and the sound effects just feel very wrong (not having heard a musket fire I’m not sure what it should sound like). For me its an embarrassing rejoinder to my frequent comment that I love the realism of the battle scenes in AK’s films.

The thought has occurred to me that either consciously or unconsciously, AK saw guns as ‘foreign’ and so portrayed their use in a less realistic but more cinematic (as in, portrayed as they are in film) manner than traditional Japanese weapons. So that in a reversal of Richies ‘representational/presentational’ theory of Japanese film, swords are portrayed in a presentational manner, guns in a representational way. Certainly, in Yojimbo, the use of a six-shooter seems straight out of a western, even down to the shooting of a bell and Nakadai’s cowboy strut.

I like the idea of the battle obscured by smoke, but I think it would have been very difficult to portray. I really liked the way that the battle is seen in the faces and reaction of the watching warriors. I know its not a terribly original scene, but I found it brutally effective.

By the way, in the second from last shot, where Nakadai staggers into the river after being shot, is it my imagination but is his face the colour of the rotting corpse in the jar (and the warrior in his dream)?

Hey, Vili, Jeremy, Ugetsu, such good and interesting observations! I can both visualize an effective “smoke” ending to Kagemusha, and simultaneously feel extremely satisfied with the ending as it is. In another post Ryan suggests it is too long-all those horses struggling-(and we all know the Criterion version is longer than the original international release) but, I dig it, and watch with fascination. (Teruyo Nogami tells us the horses snored very loudly, and even when awakened after, and led off the field, they were walking and snoring!)

Guns are a “character” and plot point in Kurosawa films! From the stolen gun of Stray Dog to the guns of Kagemusha, guns play as large a role (even when not seen) as some of the other human characters. And, mostly, that role is just plain “badness”. At the very least guns are a big problem when stolen as in Stray Dog, and in Seven Samurai guns are used by the bad guys, not the heroes. The gun is the weapon of choice for Kurosawa’s villains-and, wrapped up in all of that is the loss of the samurai way.

Ugetsu said, “

By the way, in the second from last shot, where Nakadai staggers into the river after being shot, is it my imagination but is his face the colour of the rotting corpse in the jar (and the warrior in his dream)?”

I certainly hope so! (By the way, turning the color of deep jade, and sitting in perpetuity in a bundled position in your armor-is really spooky to contemplate, but, wow, that’s a dang cool image! Enough to give you technicolor scary dreams for sure!)

Thanks for the replies! Really interesting stuff.

So interesting in fact that I just spent about an hour typing a reply to a few points raised here. However, when I wanted to take some screenshots to illustrate a few things, my computer (which, as I have mentioned, is falling apart) decided to crash. As I had not bothered to save the post anywhere and was simply typing into the reply box here, everything was lost. My own fault, really. Anyway, it’s quite late already and since I’ve been ill and everything, I don’t want to start rewriting the post now. I’ll reply in more detail tomorrow.

Ugetsu, maybe still the rate of fire was too fast, rotating volleys of fire was common practice since the invention of gun in warfare. And I do believe it was this Battle of Nagashino, were Japan introduced European methods of fighting.

The only thing I do know, that while I’m sure like modern guns, muskets too had different sounds based on powder and gun model. They at large are best described as a deep and short “boom”, similar to a cannon. This opposed to the more higher decibel elongated percussion of modern rifles, largely from high gas pressure in the gun, and the bullet cracking the sound barrier.

I think Coco, and I might of mentioned this before, but Kurosawa while perhaps falsely, smartly betrayed guns as the weapon of the enemy, solely that the gun does remove any honor. From what was once wars of skill, bravery, and honor, in which the victor, was truly the victor. Turns into the winner being the one with the largest numbers, or greatest equipped.

It would be hard to correctly show the greatness of the main characters in AK films, if they simply used a gun to which any idiot is largely equal to any great person. This doesnt really apply in Kagamusha, but Kurosawa did however play this well in Sanjuro and to some degree Stray Dog, were we see brute force is no match for intelligence, and the gun while surely decaying honor at mass, will never overtake individual honor.

And sure we see that Kagamusha might of shown guns unrealistically, but their character importance is brilliantly done for the most part. And this in films such as AK’s is far more important then pure truth.

Also I was speaking to some more knowledgeable in weapons of the age, so according to them. It was very rare to engage an enemy further then about 50 yards. As past this, both accuracy, and ability to penetrate armor was greatly reduced. Most didnt think it unrealistic that a hit could occur from a sniper, from what I guess was less then 75 yards were Takeda sat. However the likelihood of a shot in the dark hitting the intended target was of course very slim but within possibility.

(The tree thing they laughed at)

Also the bullet wound by today standards would be rather minor if hit within the chest from that distance. According to them, the bullet characteristics after impacting armor and then flesh, resulted in very little shock to the organs, or penetration into the body, due to deformation of the bullet, and it’s size and relatively slow speed allowed the bullet to very quick lose it’s energy.

However the bullets size did create large sized if although shallow wounds, in which death via bleed out was most common. And while Takeda didn’t bleed out, medical science was virtually non-existed, and all minor wounds were major. If Takeda did die from a gun shot, it would be most likely due to it creating infection, and not directly from the gun itself. After all even into the American Civil war, the great majority of deaths were result of infection from gun shot, rather then the actually gun shot.

This starts to get irrelevant to discussion, but nevertheless, I think the death of real-life Takeda, via sniper is the less popular theory anyways, so I’d just tossed that in there.

Jeremy, if you had made Kagemusha, I think the guns would probably have been portrayed differently. There is a rhapsodic quality to your discussion-you find the logic, power and beauty possible in guns, and you are knowledgeable about the physics involved. Your discussion has the poetry of the famous Harold Edgerton images of an apple being pierced/exploded by a bullet: http://web.mit.edu/edgerton/

I agree with Ugetsu that Kurosawa portrayed guns less realistically than he did traditional Japanese weapons. In Seven Samurai the timing of the sound effects for the guns is always off, at least in the copy I have; if I remember correctly the sound effect comes after the person has been shot instead of before. And for the reasons adduced guns tend to be used by the villains. Not only it this a signal of the decay of traditional Japanese (read: samurai) values but of the accessibility of lethal weaponry without the need to study and perfect one’s use of the weapon as was true with swords. Now any idiot could be a killer.

Now any idiot could be a killer.

Right on, lawless.

I am thinking of Seven Samurai when Kikuchiyo steals the gun and shoots it as he runs back to the samurai/farmer defensive point. They are disgusted that he left his post to steal the gun, and, maybe disgusted at his using it? Just a thought.

His utter glee in shooting the gun! They think he really is an idiot with so much to learn!

Vili: I’ll reply in more detail tomorrow.

So, “tomorrow” turned into “next week”, but here I finally am reconstructing what I meant to communicate. Only, I can’t quite remember what it is that I wanted to say a week ago. 😕

Well, I’ll reply in bits and pieces.

Jeremy: The distance that Takeda supposedly was when he was hit(same as tree), even with some incredibly accurate pre-measurement, there is no way to be certain you would hit anything at all(it was dark).

Indeed. Note also that while the method used by the sniper may give him the height in which to hold the weapon and possibly even its vertical angle, I can see no way in which he can determine the horizontal angle in the dark. Although the window through which he shoots is not very wide, there is still ample room for him to shoot either left or right of his target.

Actually, if you have looked at the drawings Kurosawa made for the film, you’ll notice that when he was doing the illustrations, the method used by the sniper was slightly different. Instead of the rope and stone mechanism, he employed what looks like a two-legged tripod (dipod?). If you take a look at the “Image: Kurosawa’s Continuity” feature (it should be on both the Fox and Criterion releases), you should see the drawing at around the 8 minute mark. I wanted to make a screenshot of this but thanks to copy protection and a broken computer, that won’t happen.

Jeremy: What I did find very annoying, and what disconnected me from the movie was the sound effects.

I think that you are absolutely right. Indeed, unlike I first thought it really isn’t the smoke or the range that bothers me and makes me think of westerns, but the sound effects. The effect really is comical and less serious, and totally out of place.

Jeremy: To my knowledge the rifles of Kagamusha’s time period were not capable of launching a projectile faster then the speed of sound. So minus the deep cannon like sound of the gun powder igniting, the bullet itself is largely silent. So none of that “Pshhooo” should be had, nor any high pitched ricochet giving the bullet was big ball of lead.

I’m not sure if I understand you, guns or the physics involved correctly, but the way I thought it works is that with a supersonic object pretty much all you should get is a loud bang (which you hear after the bullet has already passed your point), while subsonic projectiles like musket bullets give that hissing sound as they move towards and away from your point of observation. I do agree about the high pitched ricochet sound, though.

This actually made me think about a problem that has to do with the observer’s (our) position. Since we as the audience actually aren’t in the same physical space where the action takes place, how do laws of physics apply to us here? Supposedly our position is always the position of the camera, but when you start to think about it, and this is going way off-topic now, you actually run into theoretically interesting (although probably practically uninteresting) paradoxes considering that the camera can instantly move from one place to another, unlike anything in the natural world.

Jeremy: I do wonder if a movie ever produced real-life gun works if the audience would deem it fake or stupid. Theatrical too, it doesn’t look good, if the hero picks up a gun, and can’t hit a damn thing, as would most likely happen giving the typical situations.

I think that Kurosawa actually did this, only with swords. I’m thinking of the woodcutter’s second account of what happened in the forest in Rashomon. That’s what I would imagine a realistic sword fight to look like between amateurs.

Ugetsu: I like the idea of the battle obscured by smoke, but I think it would have been very difficult to portray.

I’m not actually sure if it is so. While Jeremy probably knows more about the logistics involved in setting up a scene like that, there are at least two scenes in Kagemusha that utilise smoke somewhat similarly. We have it once when Shingen is buried into the lake, and then a second time in the sequence that I think is the battle of Takatenjin, where at one point an army moves into view from camera right, emerging from a dissipating smoke.

Ugetsu: By the way, in the second from last shot, where Nakadai staggers into the river after being shot, is it my imagination but is his face the colour of the rotting corpse in the jar (and the warrior in his dream)?

That’s well spotted! I went through the scenes and you are right, the make-up is very similar.

Coco: we all know the Criterion version is longer than the original international release

Just a bit of a comment/correction here: the Criterion version is the original Japanese release, while the shorter international release was a later cut.

Jeremy: If Takeda did die from a gun shot, it would be most likely due to it creating infection, and not directly from the gun itself. After all even into the American Civil war, the great majority of deaths were result of infection from gun shot, rather then the actually gun shot.

That’s a very good point, Jeremy!

lawless: In Seven Samurai the timing of the sound effects for the guns is always off, at least in the copy I have; if I remember correctly the sound effect comes after the person has been shot instead of before.

But isn’t this only realistic? Light travels faster than sound, therefore we should see the shot before we hear it. It does of course depend on where we as observers are situated in relation with the action. See also my earlier comments.

Coco: I am thinking of Seven Samurai when Kikuchiyo steals the gun and shoots it as he runs back to the samurai/farmer defensive point. They are disgusted that he left his post to steal the gun, and, maybe disgusted at his using it? Just a thought.

That’s an interesting idea!

I assume you mean bipod, rather then dipod, but yep, that’s what they’re called, or if they are nothing more then 2 sticks crossed with the gun placed in the pivot point, they are called shooting sticks. Bipods or shooting sticks, however are just for stabilization, and have nothing to do with the actually aiming process, so even that would be rather ineffective.

At close proximity you have the chance of hearing anything coming by you, typically however only supersonic rounds, because of the whistling sound created as it compresses the air(the compression of air can too boil the moisture in the air, leaving behind a faint vapor trail-sort

like a Matrix(movie) effect. This is only visible from directly in front or behind the bullet.)

If there is a hissing sound from a subsonic round I can’t say in certain, but giving so many variables that can be involved, most likely so. Having shot subsonic rounds with a suppressor(silencer) I can only say they make no easily heard noise while in the air, but I too never had a round come near my ear.

Having too shot supersonic round in a suppressor(thus containing the gun powders concussion) a supersonic round, makes a loud high pitched short-traveling crackling sound, not too unlike lighting. Although without a suppressor, the shooter rarely gets to hear this sound, as the concussion of the gun powder igniting overbears the sound from their point. A short distance from the shooter, a person hears a very quick two stage sound, the first is short and shallow gun

powder concussion, and the second slightly elongated high pitch sound is the sound barrier breaking.

The untrained ear likely will not hear the two, but have them merge together. From a long distance, the two sounds do in sense become one quick and mild toned, this is too the distance were a person can start to really determine the gun being shot by it’s report.

Up close to the gun, you hear nothing but your ears ringing, I can’t express how loud a gun is when the blast is directed towards you. Indoors, if your ears aren’t bleeding, you’re lucky.

Subsonic rounds are only nowadays found in some pistols, and at the ranges a pistol is used to engage, the shooter and the person being shot at, are far too close to the gun to not have the round’s sound be masked by the gun powder igniting, and even with a suppressor(which are nowhere near movie quiet) there is too much noise from the gun’s action for it to be heard.

Are we still talking about Kagamusha? 😕

If you suggesting, that the sound effects should reflect the different location the movie cuts in a battle. It’s very interesting, but I wonder if that would be really weird for the audience, if the cuts were in a action scene-like speed. If it’s just a long cut from close to far, then I would expect the sound to be reflected more muffled. Sound however is my weakness, I’m not sure what is wisest.

The greatest flaw in many movies is missing the dramatic quality of two people using weapon, to whom have no expertise. The scene mention in Seven Samurai, is indeed a fantastic usage of real people fighting, and so much better then two masters of everything going about.

As for the smoke, I can’t think of any problems from a logistic standpoint. You could always dump a rubber tyre on a fire 😉

About Kikuchiyo and guns in Seven Samurai – I think one reason they’re disgusted with him is that by shooting the gun off he’s enabled the bandits to better determine their position.

Vili – Good point about the difference between the speed of sound and light. I never took physics. It’s just that when watching a movie I expect the visual and sound to coincide, not to be off the way they are. It’s disconcerting and counterintuitive.

Jeremy: I assume you mean bipod, rather then dipod, but yep, that’s what they’re called

If that’s what they are called, then that’s what I meant. 😆

Thanks for all the information about weapon sounds, Jeremy! As always, it’s pleasure to learn from you.

By the way, some related discussion and sound files seem to be available here. It’s a little bit beyond my comprehension though, as the bullet sizes and weighs say nothing to me.

Jeremy: If you suggesting, that the sound effects should reflect the different location the movie cuts in a battle. It’s very interesting, but I wonder if that would be really weird for the audience, if the cuts were in a action scene-like speed.

It could well be very weird for the audience, as indicated already by lawless’s comments. It’s a bit like showing fireworks on the screen. If you do it right, the rocket explodes first and the sound follows a little bit behind. But it may feel strange for the audience, just like the belated gun sounds in Seven Samurai bother lawless.

You could actually go as far as into Einstein when you start thinking about these issues (believe me, I did), since they have much to do with relativity between observers (for more discussion, see for instance Wikipedia). That would, of course, be total overkill, and quite far removed from Kagemusha.

I wonder if the sound delay used in Seven Samurai was an attempt to physically ‘distance’ the audience from the action? Sort of like the Gods eye view in Ran. The thinking may have been that the time delay would create the perception of viewing the action from a great distance. Why he would do this, I have no idea.

A really neat link Vili.

Noise coming from a subsonic round comes as bit of surprise, but does make a lot of sense now that I think of it. The high coefficient(especially the flat noised wad cutter, intended to punch clean round holes in a target for practice, oppose to the tiny puncture normally had) round pushing against the drag of air, would certainly make noise. But as the article mention you really have to be so far away that the discharge of the gun doesn’t mask the rounds sound. And while subsonic rounds can travel great distances, the amount of drop that have over long range, and remaining energy at distance makes them rather impracticable to engage with.

So, we find that indeed a subsonic round makes noise, but unless your in a non-war enviroment, and far away from the gun. I still have to say having the bullet make noise in the movie is a bad idea, and really unlikely to be heard in a practical war enviroment and the ranges such encounters happen.

Now, a modern gun battle, indeed the supersonic round passing by, is very loud and easily heard in the loudest of fights. And should be a used to some effect, but even then I think you risk reducing the dramatic effect of a war, and instead offer a sort of light feeling, non-threat action scene.

Unless your movie is to have a intro by Carl Sagan or Michio Kaku, it may be best to avoid special relativity, much less general. But of course if the movie really centered around the idea, giving sort of a Rashomon series of events, that would make from something really cool.

For Kagamusha, and fireworks sake, while the timing will be reported different on varying points of view, sound still will never be reported before light. These are two principals widely separated-The only time for fireworks, I can think of when sound would be reported before light, would be, if say the time it takes magnesium to heat, and put off light after ignition is slower then the speed of sound coming from that same ignition. Anyone in close proximity may report sound first. And at some distance light would pass by sound, and have anyone at distance report light before sound.

I’m not one to underestimate audience intelligence, but overestimating is just as dangerous 😛