Japanese defensive strategy...

[mdiehl]

Not out to lunch. The people who are out to lunch are the ones who think that any other result was likely. Frankly, I view the "sh1t happens" approach to understanding anything to be remarkably useless, since it eschews the development of a good understanding of cause and effect. You could say perhaps that my 99.9% pov is not statistically substantiatable because we have only one event on which to base an opinion, and I'd agree. But the fundamental underlying causes of the defeat were there, in detail, as a result of deliberate planning on the part of both combatants based on their detailed knowledge of the likely circumstances surrounding such an engagement. The most d@mning proof is, in my opinion, that both the IJN operational planners and USN operational planners took a good look at the IJN battle plan and identified the same weaknesses. The difference was in choice. The IJN op planners decided to wish the problem away. The USN decided to act upon the opportunitity.

Rendova, I do not dispute the claims attributed to Hartmann. I could care less whether or not he's got the high score. I'm saying that pretty much every "pilot claim" or even "confirmed kill" amounts to little more than a crude estimate, more likely to be wrong than correct, given the few available instances where units can be matched up against each other. It would not surprise me if Hartmann made half as many kills as he is credited with, and for those who care about such things, it would not surprise me if he still turned out to be the "high scorer" even given such a downward revision.

I think early war US claims are pretty unreliable, especially pre wing-camera estimates provided by the USAAF. It has already been documented that the USN tended to overestimate "confirmed" kills by a factor of 2, which might put USN pilot claims somewhere in the factor 4-6 overestimate range. On the whole, IMO, that'd still make USN estimates more reliable than official IJN/IJAAF ones. But none of them would be sufficiently useful for designing a game.

[mdiehl]

One wonders how Midway would have turned out if as little as a single key search plane had been shot down ...

Another search plane would have made the identification?

We know how it would have turned out if Tone No.4. had not flown its delayed recon mission. The Yorktown would never have been spotted.

[jnier]

ORIGINAL: mdiehl

Not out to lunch. The people who are out to lunch are the ones who think that any other result was likely. Frankly, I view the "sh1t happens" approach to understanding anything to be remarkably useless, since it eschews the development of a good understanding of cause and effect.

The "sh-t happens" approch is a fundamental to understanding nearly everything. It's trendy formulation is chaos theory, but the ideas have been around forever. The reason it partially eshews traditional cause and effect (i.e. determinism), is because it is impossible to understand all the potential causes of a given outcome in a complex system (like, for instance, a battle involved 1000's of individuals and machines - in which each individual and machine is itself a complex system whose actions cannot be predicted with certainty). Since it is impossible to model or understand all the potential causes and how these causes interact with one another, you can never really know the outcome of a complex system with certainty. That is just a fact of life. To think that you can predict this outcome with certainty is hubris.

[Mr.Frag]

ORIGINAL: mdiehl

One wonders how Midway would have turned out if as little as a single key search plane had been shot down ...

Another search plane would have made the identification?

We know how it would have turned out if Tone No.4. had not flown its delayed recon mission. The Yorktown would never have been spotted.

Thats what makes it interesting, the sighting report of a single aircraft potentially alters the outcome [X(]

[LargeSlowTarget]

ORIGINAL: Sabre21

There was plenty of luck on the US part at Midway. Yes..we laid a trap for the Japanese..but we still had to destroy their carriers and stop the invasion of Midway.

It was bad luck on the Japanese part that the one seaplane that had spotted the US had radio problems...it was good luck on the overall US part that the Devestator torpedo squadrons arrived first on the scene with little to no fighter escort..although bad luck on their part. It was bad luck for the Japanese that most of the fighter cap got involved in chasing down the torpedo planes. It was good luck on the US part that at the moment of the devestators demise..two dive bomber squadrons from two seperate task forces arrived nearly at the same time over the top of the Japanese. It was bad luck on the Japanese that they had just begun launching their own strike and had left munitions lying all over the place at the when the Dauntless's arrived overhead. It was bad luck..although not critical at the time, that the Hornet dive bomber squadron couldn't find the Japanese and were forced to divert to Midway.

So there was plenty of luck to go around that day.

Sabre21

It was also good luck for the US that McClusky, the commander of Enterprise's SBDs, when searching in vain for KB, sighted destroyer Arashi from 19,000 feet altitude hurrying to catch up with KB after depth-charging Nautilus. McClusky decided to proceed in the direction of the DD's course, correctly assuming that this would lead him to KB.
Overall, I think that the strategic planning, i.e. setting up the trap, was not a matter of luck but of good intel and planning, but that the operational and tactical levels of the battle were indeed decided by a great deal of luck and fortunes of war. I've read that the Midway battle has been re'played' several times in staff exercises, and that the historical outcome has never been duplicated in those wargames. On the contrary, the USN side always lost heavily. The whole chain of cause and effect based on several coincidences and chances apparently was just too bizarre and well, lucky. Unfortunately I can't find my source for that claim at the moment.

[mdiehl]

The "sh-t happens" approch is a fundamental to understanding nearly everything. It's trendy formulation is chaos theory, but the ideas have been around forever.

1. I dispute your contention that the ideas have been around 'forever.'

2. Chaos theory has absolutely nothing to do with it and anyone who thinks so has no understanding of chaos theory or has based their entire notion of chaos theory on a couple sentences uttered by Jeff Goldblum in Jurassic Park.

The reason it partially eshews traditional cause and effect (i.e. determinism), is because it is impossible to understand all the potential causes of a given outcome in a complex system (like, for instance, a battle involved 1000's of individuals and machines - in which each individual and machine is itself a complex system whose actions cannot be predicted with certainty).

Horse hockey. Chaos theory absolutely does NOT eschew cause and effect. It does argue that small permutations in initial conditions result in radically different evolutionary trajectories in any given phenomenon in which the number of linkages between decision nodes is very great. Even there, it is generally recognized that whole classes of events can result from a decision space that resemble each other enough to generalize.

Also, don't presume to lecture me on the subject. I did a doc fellowship at the Santa Fe Institute for a year and know more than enough about chaos theory to be unimpressed by such a trivial and contentless invocation of a complicated and largely mathematical subject.

Since it is impossible to model or understand all the potential causes and how these causes interact with one another, you can never really know the outcome of a complex system with certainty. That is just a fact of life. To think that you can predict this outcome with certainty is hubris.

No one, I mean, NO ONE who is directly involved in mathematical treatments of chaos theory, or direct applications in evolutionary ecology, cpu design, or behavior, assumes that (1) it is impossible to understand any potential cause, (2) that it is impossible to model complex adaptive systems, or (3) that you can never know the outcome with certainity. Indeed, with respect to demonstrated chaotic interactions (which at present can only be confirmed to exist in mathematically described Mandelbrat sets) if you feed in the same initial conditions every time you get EXACTLY the same output, every time. On a more general level there exist a whole suite of epiphenomena called "attractors" that seem to regularly develop under rather predictable circumstances when conditions promote their occurence. A whole lot of the science to chaos theory at present is directed at the mathematical and experimental effort to identify complex adaptive systems and to identify attractors within those systems.

Invoking "chaos theory" as though it has anything to do with the subject at hand is at best a spurious assumption and, given what you seem not to know about it, a rather unimpressive attempt to intimidate through the dissemination of jargony b.s.

P.S. You may have heard of the 'butterfly effect;' before you trot that one out, I caution you that not only has it not been ever emperically demonstrated, it's not even clear that it is theoretically possible.

[TIMJOT]

ORIGINAL: mdiehl

Incorrect. Consider that the IJN at Midway lost 2 CVs to Yorktown in the initial strike. Had Enterprise launched a surprise airstrike on 7 December it is completely plausible that Japan would have lost two or three CVs without ever locating Enterprise. Frankly, I'd trade Enterprise for 2 Kido Butai CVs on 7 December any day. It means that the force ratios in subsequent engagements are going to be more favorable for the USN.

What you neglect is that Enterprise was no where near full strike capable on Dec 7th. Its VS-6 Sqd. of 18 SBDs arrived at PH during the attack and was badly shot up loseing 4 a/c. VF-6 loss 4 out of 6 a/c by friendly AA later that day. That left 18 SBDs of VB-6, 18 TBDs of VT-6 and just 8 F4Fs of VT-6. Now concervatively at least 1/2 of the SBDs would be needed for scouting. That leaves you with just 9 SBDs and 18 TBDs perhaps 4 F4Fs for a first strike depending how audatious Halsey felt that day. Those numbers do not portend a Midway-esque outcome. Neither would Halsely have the luxury or inclination to wait to see if he could catch the CVs in the middle of recovering planes and rearming. IMHO, About all you could hope for is a 1 to 1 exchange and that would be extremely lucky at that. Add to that the whole Navy was looking in the wrong direction, it is highly unlikely that the Enterprise would find Kido Butai at all on Dec 7th.

[mdiehl]

You have no idea what Halsey might have been inclined to do. As for the general context of the discussion, the point, which you seem to have missed, is that if one can imagine such a deviated set of historical circumstances that puts 2 US CVs at PH on 7 December then it is entirely an equally appropriate to imagine that 2 US CVs had their full compliments of aircraft and were in strike range, undetected, of Kido Butai, in the early dawn hours of 7 December 1941.

[Ol_Dog]

What's Santa Fe Institute? Not to be a smart ass or anything.

Is it part of a state University in New Mexico or a separate school?

[TIMJOT]

ORIGINAL: mdiehl

You have no idea what Halsey might have been inclined to do. As for the general context of the discussion, the point, which you seem to have missed, is that if one can imagine such a deviated set of historical circumstances that puts 2 US CVs at PH on 7 December then it is entirely an equally appropriate to imagine that 2 US CVs had their full compliments of aircraft and were in strike range, undetected, of Kido Butai, in the early dawn hours of 7 December 1941.

Actually I have a pretty good idea, but call it an educated guess inpart to his well documented aggressive nature and that simple fact he would be faceing 6 to 1 odds and could not once locateing the TF afford to be cute at a time when a possible strike could be heading toward that at any moment.

As for context I thought the context was "IF KB had hung around for a 3rd strike" and or remained in the area to search for the missing CVs. In that context the Lex does not appear nor does the Enterprise get full strength capable.

Regards

[mdiehl]

Ol_Dog -

Neither. It's an NFP founded by and routinely host to people who are generally much better trained and conversant at mathematics than I. Radically altered the way I think about evolutionary ecology, evolutionary biology, and human/primate behavior.

http://www.santafe.edu/

As for context I thought the context was "IF KB had hung around for a 3rd strike" and or remained in the area to search for the missing CVs. In that context the Lex does not appear nor does the Enterprise get full strength capable.

Not my understanding of the context, but I can see how your pov would differ if that is the operating alt-history assumption.

[jnier]

As usual, mdeihl you have failed to read what a poster has actually said. I did not say that my arguments were based on chaos theory - I just said it was trendy to associate chaos theory with the "sh-t happens" perspective - even if these associations are erroneous.

Instead of reading what I really said you have launched in a long irrelevant diatribe designed to attack me since I disagree with your assertions.

My argument is not about a chaos theory. It is based on the practical limits associated with understanding causes and effect in a complex system. Since you are dealing with an incredibly complex system, in which you cannot , in practice, model all of the causes and effect, how can you accurately predict an outcome? It is not possible - so all predictions must be probabalistic - and the more poorly you understand the causes and effects and their interactions, the more probabalisitic the outcomes will be. By saying that only .1% of the time would the outcome have been less favorable to the USN is a remarkably cocky assertion (and completely unsupported by any real evidence), because in saying so, you are saying that you understand nearly all of the potential causes in the system and how all these causes interact with one another. Again, hubris.

Unless you have found a way to accurately understand and model nearly all the potential causes and their interactions then you can not predict the outcome with the degree of accuracy you claim. So again this not about choas theory, but simple principles of determinism, which ironically leads to the "sh-t happens" conclusions, at least in practice. YOU do not know the causes YOU cannot predict the outcome with certainty - no one can.

[Ol_Dog]

I looked at your link. They have some interesting research under way.

[mdiehl]

I did not say that my arguments were based on chaos theory - I just said it was trendy to associate chaos theory with the "sh-t happens" perspective - even if these associations are erroneous.

Oh, you just happened to introduce a wholly irrelevant arena of research as a response to my remark to the effect that '"sh1t happens" is a useless approach to phemonenology,' and in so doing you managed to expound in detail about your notion of chaos theory even though your notion is profoundly wrong.

My argument is not about a chaos theory.

Then you probably should not introduce chaos theory to your argument.

It is based on the practical limits associated with understanding causes and effect in a complex system. Since you are dealing with an incredibly complex system, in which you cannot , in practice, model all of the causes and effect, how can you accurately predict an outcome?

How's that? It's a matter of knowing what is relevant and what is not relevant. I can illustrate by example. If one were to create a dynamic model of the solar system at a scale down to, say, 100 meters, then plug in all the locations of all the known objects 100m or greater in diameter, one would have a very complicated and dynamic system of gravitational interactions that explained a whole lot of behavior with respect to the motion of masses, but not all behavior and not perfectly. Still, add one more theoretically known 100m object to the system AT PRESENT and things don't change much. Add one a long time in the past (say) and things might change somewhat. Hard to tell. Delete Jupiter from the model and the whole thing goes nutso regardless of when you invoke the event.

The US trap at Midway, and the problem of IJN mission overtasking during the battle, were both Jupiter scale phenomena, not 100m epiphenomena. Delete the USN CVs and intel (the trap) and the IJN wins at Midway by default. Delete the mission overtasking in the IJN strike force and the IJN's chances greatly improve.

YOU do not know the causes YOU cannot predict the outcome with certainty - no one can.

Of COURSE I know the causes. You know the causes. Nagumo knew the causes before the battle even began, and it was coldly demonstrated to him in IJN simulations prior to the battle. The difference is that unlike Nagumo, who knew why he was beaten rather quickly after the fact, you are still in the pre-Midway state of denial.

I'll repeat it for you. Mission overtasking.

[Mr.Frag]

Probably the best example of this is the "failure" of the search planes radio.

When one looks at it as a single event, one draws the conclusion that had it not failed, things would have turned out quite differently.

When one widens the scope to look at all radio communications, it becomes quite apparent that radios of that time period failed quite often coupled with the fact that communications were commonly disrupted even when the radios were working correctly.

Suddenly, the radio failure becomes not a triggering event, but actually something that should have been expected and planned for. Knowing this and not accounting for it in the plan resulted in the results, not the actual individual failure.

They should have expected the plane to not have a functional radio and sent more then one to prevent a single point of failure of such a critical nature. It could have just as easy been engine problems or dirty gas or carb icing. The actual cause looses meaning when looking at the overall model.

[jnier]

ORIGINAL: mdiehl

Oh, you just happened to introduce a wholly irrelevant arena of research as a response to my remark to the effect that '"sh1t happens" is a useless approach to phemonenology,' and in so doing you managed to expound in detail about your notion of chaos theory even though your notion is profoundly wrong.

Please read my posts - I wasn't talking about chaos theory. The reason I cited chaos theory is because people often associate "sh-t happens" with chaos theory. They have not heard about the more relevant work that I have studied in my post-doctoral work - the application of dynamic systems theory to explain the behavior of complex social systems (which, by the way, cannot be predicted accurately, at least in practice). People at have at least heard of chaos theory - had I realized that you would have used it as an opportunity to attack me, I would not have mentioned it.

It is based on the practical limits associated with understanding causes and effect in a complex system. Since you are dealing with an incredibly complex system, in which you cannot , in practice, model all of the causes and effect, how can you accurately predict an outcome?

The US trap at Midway, and the problem of IJN mission overtasking during the battle, were both Jupiter scale phenomena, not 100m epiphenomenon. Delete the USN CVs and intel (the trap) and the IJN wins at Midway by default. Delete the mission overtasking and the IJN's chances greatly improve.

[/quote]

As usual the core of your argument is just an assertion - not a fact. You assert that these are the only two important causes - and that you can predict the result had these causes been absent. Just an assertion. And as usual, not a shred of evidence to support it.

Of COURSE I know the causes. You know the causes. Nagumo knew the causes before the battle even began, and it was coldly demonstrated to him in IJN simulations prior to the battle. The difference is that unlike Nagumo, who knew why he was beaten rather quickly after the fact, you are still in the pre-Midway state of denial.

I'll repeat it for you. Mission overtasking.

I'll repeat for you - other factors that cannot, in practice, be modelled influenced the outcome. And could have possibly produced a different outcome. Which is why there could have be substiantial distribution of outcomes, if boths sides were to line up and fight Midway 1000 times. The difference between me and you is that I am not cocky enough to presume that I know what that distribution would look like. If you are right and you can accurately model military conflicts with the degree of precision you claim, you are the greatest military mind of our times. But I would guess you're wrong.

[Sabre21]

Miedhl

I can see that you are pretty one sided in your views. I brought up the one possible option of an alternate history with the 2 US carriers in Pearl at the time of attack...my statement was that had this occured, most likley both carriers would have been sunk...that is IMO a highly likely outcome.

There are many variations of alt history..I simply cited one. Another possible scenario could have been that the US acted upon intelligence that they had been receiving and correctly assumed that the Japanese were enroute to make a strike. This is a plausible scenario too since wargames had been carried out on several occasions in the years leading up to the war with this type of attack in mind. Had the US placed not only the 2 carriers available at pearl, but also the Saratoga in a position to attack the Japanese as they had at Midway, it is possible that the Japanese could have received a devasting blow to their strike force.

I'm sure each of us on this forum could think up many possible alt history variations without talking down to one another.

As for what took place at Midway..I will have to disagree with several of your assumptions. It was bad luck on the Torpedo Squadrons part that they arrived early and received the attention of the cap, but it was fortunate that the cap had come down to engage the torpedo planes and that the dive bombers arrived when they did. This was pure luck! We all know the end results. Had the squadrons all arrived together...no one can say what the outcome would have been. To say that 3 carriers would have been sunk regardless has as much foundation as saying that no carriers would have been sunk...no one can know.

As for the radio on the Japanese scout plane..if they were so notoriously bad as you suggest..then why didn't the Japanese launch more scouts?

The US operational plan was to strike the carriers when they were in refuel/rearm..that I agree with. Had they flown directly to the correct position of where the Carriers were, it is likely they would have arrived early, it is difficult to say for sure. As it turned out, the US squadrons were forced to search the area running extremely low on fuel..I don't think that was part of the plan. Nor was it part of the plan for the Hornet squadron to abort the search and land at Midway. Spotting that one destroyer when and where he did was lucky on McKlusky's part placing his squadron over the enemy carriers at the best time possible.

As for your contention that the Japanese were task overloaded..I don't agree with that either. IMO, it was pretty much Nagumo's indeciciveness that caused much of the problems. Conducting a strike and placing a cap overhead would not cause a workload problem. Conducting recovery operations, preparing a strike, and maintaining a cap should not have been a problem. Nagumo had the assets for it had he made the contingency to do so.

Military operational planning is a very difficult task. Even the best planners can't think of every contingency. There are so many variables to consider, especially in air operations, that there will almost always be unforseen problems...weather, maintenance, actual disposition of the enemy...all these play a significant role. In those days especially, navigating over the water was done by time/distance/heading and what few instruments there were like the ADF were highly unreliable. The operational commander must rely on his subordinates to make the right decisions when these problems arise so as to successfully accomplish the mission. The US prevailed at Midwat as a result of good intelligence, commanders that acted on that intelligence, a good overall plan, good leadership at the lower levels, a lot of courage...and a hell of a lot of good luck.

[sven6345789]

completely agree with you on all points here. [:)]

[mdiehl]

Probably the best example of this is the "failure" of the search planes radio.

When one looks at it as a single event, one draws the conclusion that had it not failed, things would have turned out quite differently.

When one widens the scope to look at all radio communications, it becomes quite apparent that radios of that time period failed quite often coupled with the fact that communications were commonly disrupted even when the radios were working correctly.

Suddenly, the radio failure becomes not a triggering event, but actually something that should have been expected and planned for. Knowing this and not accounting for it in the plan resulted in the results, not the actual individual failure.

They should have expected the plane to not have a functional radio and sent more then one to prevent a single point of failure of such a critical nature. It could have just as easy been engine problems or dirty gas or carb icing. The actual cause looses meaning when looking at the overall model.

Excellent point. I agree that that is a good example. In some ways I suppose you could say that I have a preternatural faith, even irrational faith, in Murphy's Law. If a person sets up a bad plan to begin with, or builds an object such as a ship or building using a bad design, with no redundancy in the important processes or critical components, such designs often or even usually fail. In the case of Midway, both the Japanese operational planners and the US op planners attacking strike force knew that there were certain intervals, certain critical moments, in flight operations and in operational plans, that provide opportunities for Murphy's Law to be invoked by circumstance.

There was a pretty good book, recently, analyzing the components of diasasters, such as the French Concorde crash, a bunch of other crashes, the Challenger explosion, teh Apollo 1 fire, WTC and a few other events in which the author made exacyly the same point that I've made here. Disasters do not happen because of luck. Virtually every disaster that has ever occurred, including the Japanese defeat at Midway, had moments where someone stood forth and said "You know, if X happens as we expect it to, then all is well, but if X and Y happen, we could be in deep sh1t." And in every major disaster where humans had ANY opportunity to alter the outcome, that person has been ignored.

I see idiots who believe in luck every day. They're usually weaving through traffic at speeds 10 mph faster than everybody else and jerking back and forth between lanes, racing down the suicide lane (a reversible center lane) or racing along the breakdown lane. These guys speed, they drive dangerously, and they leave themselves "no options" in the event that a problem comes up (like a car stopped in the breakdown lane ahead of them, or a kid chasing a ball). If you are the sort of person who thinks that "luck," or whatever, just happens, you're not going to understand my point about Midway. If you're the sort of person who often makes plans thinking "Okay, if I do this and something happens, how can I get out of it?" then you will understand my argument. At Midway, the IJN left itself NO WAY OUT in the event that things did not procede in accordance with their optimistic script.

Jneir, Sabre21, I'm just going to leave the discussion with the definitive statement that you don't know what you're talking about in re Midway. And, Jneir, I don't buy your wierd explanation for invoking chaos theory. You tried to baffle me with bullsh1t, and you failed because I know enough about chaos theory to know that it doesn't apply. Even if you meant it as a metaphor, it's a rhetorically risky move for you to invoke explanations or analogies when you don't know what you're talking about.

[mogami]

Hi, All I know is when I am Japan. I will win Midway. I won't move there for any reason other to sink USN vessels beginning with CV. I'll know how many USN CV exist and plan for them all to be waiting. The only way the USN will win is if they are commanded by one lucky SOB.

(of course I could be wrong, I've lost CV battles in UV where I thought right up to the moment the combat ended "Boy is he going to get it")