Monday, August 29, 2016

Third Offset Strategy

The Third Offset Strategy (TOS) is America’s great hope for future warfare, according to our current leaders.  We’ve consciously and deliberately ceded the advantage of numbers (quantity).  We seem to have no interest in bigger and better explosives (witness our stubborn refusal to upgrade or replace Harpoon until just recently), supersonic missiles, or wide area effect weapons like major caliber naval guns.  Our quality edge is eroding at an incredible pace with the Chinese having either caught up or on the verge of doing so in the areas of stealth, armor, ballistic missiles, etc.  The Russians have leapfrogged past us in the area of electronic warfare as evidenced in Ukraine.  That only leaves us a possible advantage in networking, data sharing, and autonomy, according to our leaders.  Hence, the Third Offset was born.

What is the Third Offset Strategy?  According to its principle architect, Deputy Secretary of Defense, Bob Work,

“The Third Offset is really kind of simple at its core,” Work said. (The first offset was nuclear weapons, the second smart weapons). “It basically hypothesizes that the advances in artificial intelligence and autonomy [are] going to lead to a new era of human-machine collaboration and combat teaming.”

Shades of Battlestar Galactica!!!

Somehow, in some unexplained way, this human-machine collaboration is going to overcome numerical and explosives deficits. 

What does human-machine collaboration mean, anyway?  Haven’t we had human-machine collaboration for decades?  For example, the Aegis system is a computerized battle system with human interface and oversight in which the human-machine system can operate through the entire spectrum of 100% machine (fully automated) to 100% human controlled.  Another example would be the flight control systems of nearly all modern combat aircraft.  The aircraft are not flight-stable so computer systems take the pilot inputs (control stick) and translate them into control surface movements that a human could not manage.

So, what does human-machine mean?  Honestly, I have no idea.  It seems to be more of a marketing thing than any actual technology.

Taking away all the hype and marketing, human-machine and Third Offset seems to be mainly an emphasis on unmanned vehicles and networks.  Of course, we’ve already discussed the problems with assuming that networks will work flawlessly in the face of peer level ECM and cyber attacks.  We’ve also noted that unmanned vehicles have a disturbing tendency to lose communications and wander off, never to be seen again.

Why we would base our entire future military strategy on such unreliable technology is baffling.

Let’s look a bit closer at the Third Offset Strategy as described by its chief proponent, DepSecDef Bob Work, in a speech he gave in early 2015 (2).

Work notes the temporal aspect of the TOS.  The first two offsets gained us advantages that lasted decades until the rest of the world caught up.  This is unlikely to happen this time around.   Unfriendly countries are hacking our secrets as fast as we generate them.  Essentially, we’re doing the R&D for our enemies.  Thus, any advantage we might gain from the TOS is likely to be very short lived.  As Work states,

“First, it's going to have a much more trying temporal component. In 1975 and in the 1950s, we knew our adversary and we said, "We can pick something where we will have an enduring advantage." We don't think we're in that type of environment right now.” (2)

So, by Work’s own admission, the TOS benefits will be short lived if they ever even exist.  His solution to this challenge is to focus on technologies that can be developed quickly – he suggests a five year period.  Unfortunately, any new technology that can be developed in five years almost by definition can’t have much of an impact.  Look how long it’s taken us to develop a straightforward aircraft, the F-35.  Does anyone really think we can develop a world-changing technological capability in five years?  Heck, it takes us nearly five years just to generate the requirements list that we can give to industry in a request for proposals!

To be fair, he offers a nod to longer range projects but having noted that we can’t seem to maintain secrets, there appears to be no possibility of reaping any long term advantages.

Work offers a fascinating observation about the source of new technology now as compared to during the previous offsets.

“And the third big difference is that in the 1950s and the 1970s, generally these advances were military capabilities that were brought along by military labs. But now with robotics, autonomous operating guidance and control systems, visualization, biotechnology, miniaturization, advanced computing and big data, and additive manufacturing like 3D printing, all those are being driven by the commercial sector.”

We’ve noted this phenomenon before.  Rather than the military driving development, the military is abdicating its responsibilities and depending on the private sector to come up with advances.  There are two problems with this. 

First, industry is driven by profit and will offer those technologies that can generate the most profit rather than those that generate the greatest military benefit. 

Second, industrial technology is highly susceptible to foreign capture.  Even setting aside hacking, industry runs on public disclosure of patents, statements of technologies for the benefit of shareholders, floods of technical papers by company scientists, informal networks of shared data, and so forth.  Anything industry develops will be known to foreign countries as soon as it happens.  It will be impossible to gain any lasting advantage.

Still confused about what the TOS is?  Let’s see what DoD Live website has to say.  They describe five key aspects (3).

  • Deep Learning Systems – this involves data pattern discovery;  the ability to discern patterns from less directly relevant data; an example would be radar detection of stealth aircraft using collated data from multiple sources

  • Human-Machine Collaboration – this is the use of machines to aid human understanding and decision making;  an example would be fused sensor displays

  • Human-Machine Combat Teaming – this is the pairing of UAVs with human platforms;  an example would be the Triton UAV paired with the manned P-8 Poseidon

  • Assisted Human Operations – this is wearable technology that enhances human capabilities;  an exoskeleton would be an example

  • Network-Enabled, Cyber-Hardened Weapons – this is the production of weapons that are immune to ECM and hacking

That list is fine but it’s already nearly obsolete as far as providing an enduring advantage.  Many countries are already engaged in Deep Learning Systems.  Witness the anti-stealth radar detection technology that is based on collating indirect data from multiple sources.  By all accounts, unfriendly countries are well advanced in this effort.  All countries are deep into Human-Machine Collaboration efforts, developing their own versions of fused sensor displays, for example.  Human-Machine Combat Teaming is more advanced in Russia than it is here.  Russian robotic-human pairings on the Ukrainian battlefield are reportedly well beyond anything we currently have.  Assisted Human Operations are a technology that has not yet been significantly applied by any country, as far as we know.  Network-enabled, cyber-hardened weapons are being pursued by all countries.

Clearly, we have no current significant advantages in any of these areas and, in some, we appear to be behind our enemies.  Whatever we develop will provide no lasting advantage and, quite likely, will not even provide a momentary advantage.  If fact, we may well find ourselves scrambling to catch up to other country’s Third Offset advantages!

As you’ve read this, do you see the military’s focus on technology over training, maintenance, tactics, etc.?  The US military has an almost myopic focus on technology – a belief that technology is the solution to every problem.

Consider this radical alternative …  What if, instead of focusing on non-existent technologies, most of which will never pan out, we were to focus on these aspects of the military:

Maintenance – What if we had an absolute dedication to ensuring that every piece of equipment works and that we have the best trained technicians in sufficient numbers to ensure that the technology and machines that we have work, work flawlessly, work reliably, and can be instantly and locally repaired when they do fail.  Hand in hand with that would go adequate and readily available inventories of spare parts.  For example, what if instead of a 50% readiness among our F-22/35 fleet we had a 98% readiness.  Or, what if instead of depots full of non-functioning F-18 Hornets, we instead had those 200 or so Hornets out in the fleet?  Our air wings would have around 20 extra combat aircraft each!  Or, what if instead of having to retire entire classes of ships early due to neglected maintenance and resulting physical deterioration, we took care of those ships?  We’d have many dozens more combat ships.  Our fleet wouldn’t be shrinking, it would be growing.  Imagine a force with that kind of physical readiness – it would be a force to be reckoned with.

Training and Tactics – What if our soldiers and sailors were absolute masters of their craft.  What if our Captains and Admirals actually knew how to tactically utilize a multi-carrier battle group?  What if we conducted regular and frequent live fire exercises?  Not only would our personnel be better trained but we’d uncover the flaws in our weapons before we have to use them in combat!  What if we conducted actual and realistic amphibious training assaults?  We’d see the flaws in our doctrine and correct them without paying the price in blood to do so.  What if we conducted realistic training rather than the silly set-piece training exercises that we do today?  Maybe our sailors would be prepared to fight instead of surrendering to any Iranian that wanders by.

Readiness – What if our readiness was always at peak levels?  What if our non-deployed pilots weren’t limited to bare flight certification hours like they are now?  What if we had actually usable doctrine for surface action groups, multi-carrier operations, amphibious assaults, and so on?  What if our non-deployed units were actually surge-ready, unlike now?  Readiness ties back to maintenance and training.  What if all that was good to go, all the time?

Force Structure – What if we actually had mine countermeasure assets in sufficient numbers to be effective?  What if we had offensive mine warfare delivery capability?  What if we hadn’t wasted money on the LCS and, instead, had a dedicated ASW vessel?  What if we had some SSKs?

Imagine if, instead of wasting time, money, and resources on some nebulous and idiotic TOS, we focused on maintenance, training and tactics, readiness, and force structure.  What an enormous and enduring advantage that would give us over our enemies.  That alone could be our Third Offset Strategy and it would be infinitely more beneficial than what we’re attempting now.  Compare such an impeccably maintained, trained, and ready fleet to the Russians who have major reliability problems.  Their fleets of submarines, for example, are barely seaworthy.  The level of their ship commander’s training is almost non-existent since they so rarely can afford to put to sea.  Their enlisted ranks border on untrained.  The Chinese suffer from some of the same shortcomings as the Russians, particularly in the quality of their enlisted ranks and the level of training and experience of their commanders.  We could enjoy a perpetual, staggering advantage just from the factors I’ve described. 

Maintenance, training and tactics, readiness, and force structure should be our Third Offset Strategy.  In essence, make what we have work perfectly before we buy more stuff that won’t work and make the quality of our personnel unmatched in the world.  There’s your advantage.  There’s your offset.


(1)Breaking Defense website, “Iron Man, Not Terminator: The Pentagon’s Sci-Fi Inspirations”, Sydney J. Freedberg Jr., 3-May-2016,

(2) website, “The Third U.S. Offset Strategy and its Implications for Partners and Allies”, transcript of speech delivered by Bob Work, 28-Jan-2015,

(3)DoDLive website, “3rd Offset Strategy 101: What It Is, What the Tech Focuses Are”, Katie Lange, 30-Mar-2016,


  1. A better start would be superior people, leadership, training, and tactics/strategy.

    At least BSG had Adama...


    1. You're absolutely right that people are the key to any attempt. I thought about including that but left it out as utterly impossible. The other things I mentioned, however incredibly unlikely, have a theoretical possibility of being implemented. Asking Navy leadership to replace itself isn't even theoretically possible. That Navy leaders would one day wake up and suddenly say, "Hey, we suck. Let's all resign and replace ourselves with better people.", is not even in the remote realm of possibility.

      Recognize that I'm not criticizing your suggestion. Quite the contrary. It's the better solution. If we had better people, all the rest would take care of itself. I just tried to limit myself to something that had at least a one in a gazillion chance.

      So, absent a mass turnover of personnel, the things I suggested are things we already do and it would just be a matter of re-prioritizing. Hope you understand my reasoning?

      As usual, you're spot on.

    2. The Navy has, and continues to get, good people; the trick is clearing the deadwood and empowering them.

      Sadly, it increasingly appears that real reform will only happen after a massive mission failure accompanied by major loss of men and ships.


  2. Now the Flags/SESs can get cushy jobs with Silicon Valley Companies, at least until the Entrepreneur's figure out they can't do anything but powerpoint and politic.

    Mission Accomplished - we expanded our cushy post service employment opportunities.

    Meanwhile LCS 1 is back in the shop with a rusted Diesel engine. Really who would hire these bozos running/ruining the Navy?

    1. Okay. Nice rant - nothing wrong with that, we all do it.

      Now, do you have any comment about the Third Offset? Is network/unmanned the right approach? My approach? Something else? Having consciously ceded numerical superiority what can we do to maintain a decisive combat advantage? What is the relative importance of explosives to data? And so on...

    2. well now that the rant is out of the way - yes I do.

      Like with all of these new gee whiz toys, get them working in the lab FIRST, then do a rigorous OPEVAL and THEN and ONLY THEN program them for production platforms.

      But on the tech side, the internet of things is so unsecure as to be laughable. Furthermore, the Navy does not have the discipline to harden these systems with the few published steps. TURN IT ON, I WANT THIS is the Mantra of the day. I know I have worked on projects with IA requirements, and been overruled by the higher ups.

      And that is just the subtle approach of hacking the systems, as you point out the brute force EW approach is even more scary.

  3. You should review some of the work going on in the Battlespace Awareness Division on the Navy Staff. It appears they've worked on some groundbreaking initiatives that address some of the areas you've been discussing. You appear to be of like minds.

    1. I've looked at lots of battlespace awareness documents and I've yet to see anything that wasn't just a collection of buzzwords, platitudes, and vague, unsupported wishlists of capabilities.

      Here's the kind of stuff I see,

      "XXXX will allow us to synergistically maximize information dominance via networked collaboration resulting in optimal warfighting decisions and decisive execution of strategic objectives across the spectrum of kinetic, non-kinetic, and cyber domains."

      That's just garbage. It says nothing. It's just a string of buzzwords.

      If you can point me at something worthwhile, I'd love to see it!

    2. Sure, what's the best way outside of these comments?

    3. Just include a link in the comments. Either insert it via HTML or, if you're not familiar with that, just copy it in plain text.

  4. I Do hope we continue with our development of unmanned vehicles, because they may be the way we can restore our quantity advantage. I don't see our country every being willing to deploy large numbers of drafted troops in any but the most desperate situations (invasion of the continental US for example). But we need to change our focus to drones that are cheap, reliable, and most importantly, expendable in large numbers, instead of all the bells and whistles (like the current UCLASS requirements).

    Randall Rapp

    1. So your vision of a Third Offset is drones in overwhelming numbers? Quantity over quality. Okay, that's potentially valid.

      By implication, then, you're in favor of automated combat drones, I assume. Lots of drones are no good unless they can destroy stuff. Further, the ability to individually control huge numbers of drones, especially in the face of high level ECM, is not possible. Therefore, the drones will have to have artificial intelligence aimed at destruction. Setting aside the fact that we don't have that level of programming sophistication, are you comfortable with turning loose huge numbers of killing machines that make their own decisions? Are you comfortable that those drones can't be hacked and repurposed against us? How will these thousands of drones be programmed for individual targets?

      Most importantly, what do you think such a drone will cost given that the lowest level of drone, a Tomahawk missile, costs around $2M?

    2. An autonomous unmanned system has significant advantages in a quality strategy over manned systems. First, the autonomous systems can be used in riskier operations. Second, the personnel and training requirements would be a lot less than manned systems - no training pilots. The maintenance personnel would also be a lot lower if the systems are designed to be expendable with a low survival rate in combat. I seems to me that a weapon designed under this philosophy would be a lot closer to a Tomahawk than a Predator.

    3. A massive investment in Tomahawks is a potentially valid approach. Now, take the other side of the argument. What are the drawbacks to a massive drone approach?

    4. I struggle with that. I've tried looking at the other side, but I can't see why we would need a UCLASS to do the strikes over a moderately smart Tomahawk in swarms. Even if you bump up the price to $3million to have enhanced AI and a stealthy case, it would give you a ton of capability without the added complexity needed to bring the platform home.

    5. You don't need stealth if you have sufficient numbers. To be ridiculous, if you launch a thousand Tomahawks at a single target, you can have them emit giant honking beacons with their exact co-ordinates. So what if a hundred get shot down. The remaining 900 will do the job. The enemy hasn't got enough anti-weapons.

      Adding stealth and greater AI is going down the path of fewer and more expensive which, in turn, leads to still fewer and still more expensive which ... is exactly the route we've gone and it's failed! And now you're proposing to do it again! You see how easy it is to fall prey to the lure of a "little more" added to each platform or weapon?

      Instead of one supremely capable, uber Tomahawk that costs a gazillion dollars and we can only make one of, why not make a thousand that are dumber than dirt, easily detected, cost next to nothing, and totally expendable, are utterly reliable, and will simply overwhelm any possible defense. K.I.S.S.

    6. Valid point on my falling prey to the upsel. I kind of like the simple strategy in terms of bang for the buck. It still leaves the issue of maybe targeting, but as an assault option I like it alot.

      The only thing I can see being a major issue at that point is getting enough launchers.

    7. "The only thing I can see being a major issue at that point is getting enough launchers."

      Which is why retiring the 4 SSGNs with their 150 or so missiles, each, is a very questionable decision.

    8. I'm on board; though to be fair how much more life do the Ohio's have in their hulls? (It may be a lot, I don't know, but they aren't young platforms).

      But if there isn't alot I don't see a problem with adding 4 SSGN's to the follow on SSBN that they are planning; or maybe, to keep things cheap, that is where buying foreign AIP subs with VLS could help.

  5. The first and second offset strategies were successful because they increased the offensive lethality of a military unit. The use of nuclear weapons in the first offset strategy increased the destructive yield of a single weapon and enabled a lone plane to destroy a city. The use of precision guided weapons in the second offset strategy significantly increased the accuracy of each weapon and enabled a single plane to destroy multiple targets in one sortie. Therefore, it seems to me that a successful third offset strategy must considerably advance offensive lethality. In contrast, the nebulous concepts currently put forward as a third offset strategy seem unlikely to generate a breakthrough in offensive abilities.

    We have revolutionized weapon yield and weapon accuracy. What more is there to do? Moreover, what technologies are mature enough to become useful?

    Perhaps advances in directed energy will afford a greater increase in the rate of fire and reduce the costs of destroying enemy PGM's. As of today, the technology is still not mature enough to be the basis of an offset strategy.

    Perhaps the answer lies in drone and sensor technology to improve target spotting.

    This is taboo, but the only mature and underutilized technology that would dramatically increase our offensive lethality is cyber weapons. The use of cyber weapons to attack physical and computer infrastructure could be devastating and deny an enemy the use of their industrial base to replace combat losses.

    1. Outstanding comment! I love your assessment of the purpose for an offset and the apparent lack of that purpose for the third offset.

      Cyber ... Interesting. How do you propose injecting the cyber code (I assume that's what you're referring to) into an enemy's infrastructure? The first thing an enemy would do when entering into a war is to isolate their country from the Internet (actually, relatively simple to do). Injection of cyber code into electronics that are not built for electromagnetic reception is challenging. Tell me more about how you see this happening.

    2. CNO,

      I like the notion of robotics in the TOS but the services are loathe to release ATR to AI and make the targeting autonomous.

      The quad-rotor UAV drone swarms concept are keeping many in the USN awake at night who are paying attention. If one were to use them as part of a layered offense against USN assets they can easily overwhelm the munitions capacity of every defensive kinetic weapon on the ship.

      Imagine wave after wave of incoming that cost the threat hundreds of dollars to build and deploy for hours or days.

      Then if part of a complex attack, a follow-on flight of far more lethal PGMs could cripple or sink the ship. The naval mandarins have already claimed any harm to a carrier and they will reverse course and head home. The enemy is listening.

      The drones also do something no other airborne munitions can do right now. They can loiter at the back of the pack in selected numbers to provide UAS and BDA feedback for follow-on missions.

      The drones could also employ non-kinetic cyber packages. Take a close look at the Russian invasion of Ossetia in 2008 and the follow-on conflict in Ukraine and study the Russians using cyber to set their conditions and prep the battlefield.

      DE weapons are a long way off until the thermal bloom problem is resolved.

      I think the five key aspects you cited from TOS are a pipe dream in the current military environment and acquisition mess. They can't seem to master shallow learning much less deep learning.

      CNO, please expand on the ECM resolution for drone swarm defenses.

      Bill Buppert

    3. Good comment. I think, though, that you may be looking at offensive drones in isolation. In other words, you've described a scenario in which a swarm of drones is, out of the blue, attacking a carrier and you're concluding that that would be a problem to defend against.

      What you're failing to include in the scenario are factors like,

      1. How did the drones get there? Carriers operate hundreds to thousands of miles at sea. If the drones are as numerous as you suggest, they must be very cheap. In order to be cheap, they can't be big. If they aren't big, they can't have much range. If they were transported by a drone "carrier" ship, it would be susceptible to being detected and sunk on the way and all the drones would be lost.

      2. What kind of munitions would the drones carry? Again, if they're cheap, they're small and would have a very small payload. A small rocket? An RPG-ish warhead? A Maverick (probably too big for a small drone)? Munitions that small won't do more than superficial damage to a ship the size of a carrier.

      3. How are these drones controlled? There would be too many to individually control.

      4. What kind of sensor would these drones have? Again, small size means small sensors. Small sensors would be susceptible to ECM, decoys, obscurants, etc.

      An offensive drone swarm would certainly be a concern but a practical drone swarm is not yet feasible. Still, it's good to be thinking about such things now.

      I'm not sure what your last request was talking about?

    4. 1. B2, instead of a bomb truck, its a drone truck. How many 10's of tons can the B2 deploy?

      2. Use the Harrop as a model. It is the munition. Explosives coupled with kinetic energy. Drill 30 drones one after each other into the same point on a carrier, so 20kg charge multiplied by 30 and and cycled release, you'll drill a hole clear to the bottom of the ship. Nothings going to recover from that.

      3. Thats the trick isn't it. You dont have pilots, humans i mean, its a multi thread multi core processor. ITs exactly what a computer can do. Your home PC has a CPU that has at least 2 cores on it, and each one likely has at least 4 threads on it. Thats 8 concurrent calcs at incredible speed.
      Thats nothing, your crappy old video card has a GPU, not CPU, it can do hundreds of concurrent calcs. Pure number crunching is its thing.

      4. Again, Use Harrop as model. Its got plenty, both EW and radar/thermal. Swarm, networked to satellite and awacs telemetry, etc etc etc etc

      Not trying to be obstinate. Playing devils advocate to your negating the drone swarm.

    5. "Use Harrop as model"

      Did you happen to notice the price tag of $10M each? To deliver a 50 lb warhead on a one-way trip?

  6. I am not sure if cyber should be the third offset. I brought it up because cyber appears to be the only technology at this time that both mature and lethal enough to use as an offset strategy. Cyber has significant downsides and is potentially destabilizing precisely because cyber could shift the offense-defense balance far in favor of attackers.

    [1] Challenges in attributing cyber attacks offer a degree of stealth. A cyber "Pearl Harbor" attack could precede kinetic action.

    [2] Many valuable communications, financial, and industrial assets are connected to the internet which is a vector for inserting malicious code. The potential for inflicting damage is high.

    [3] The costs of an unsuccessful attack are relatively low. The primary disadvantage of a cyber attack is that the enemy can learn from your method of attack.

    To answer your question there are ways around those defenses. You could launch the first attack before the internet is shut down. Also some systems need to be in computer networks to remain functional. Getting an adversary to close off their internet and electronic communications lines would be a major victory in itself. It would be pretty difficult for Chinese factories to find raw material inputs or export products if its internet and phone lines to the outside world are shut down.

    You could enter the enemy's internet through Bluetooth, cell towers, or wireless networks. North Korea is having an issue with foreign information entering across its borders by these means. You could import infected machines into the adversary's country before the conflict and tell the to launch the attack if they lose satellite or internet connection to the US after so much time. You could send infected machines into the enemies country during the conflict. You could also introduce malicious code from spies or a fifth column who work inside the facilities.



    1. Don't apologize! You've offered a great analysis and a fascinating alternative. Explore it. You may conclude it's not suitable for the basis of a third offset but that's okay. I have no problem with any of that. Whether cyber alone offers enough capability to constitute an enduring and sufficient advantage to be the focus of a third offset is the question. At some point, cyber warfare will transition to explosive warfare and then where are we?

      Cyber warfare also offers relatively little in the low level, terrorist war since terrorists don't depend on cyber and electronics to a potentially crippling degree. This, however, is a lesser issue.

      How do you inject code into ships, aircraft, and missiles?

      Also, important military installations would be on isolated networks although there's always ways in.

      Good observation about forcing an enemy to shut down systems. Still, eventually, we have to destroy stuff and no amount of cyber will do that.

    2. Think what cyber has the capability of doing.
      Should it work and be effective enough to knock out a cities utilities, and transport infrastructure, thats an awful lot of devastation and expenditure afflicted on your enemy, with near impunity, and virtually no one to retaliate against. We're so reliant on all our megainfrastructure which is completely reliant on networked computer systems, that it is possible to make a city almost uninhabitable.
      That sounds like nearly the threat of a nuclear strike. With the added bonus that you leave the infrastructure intact. Which makes it even more of a deterrent than nuclear weapons.

    3. Think what a penny could do to an entire city if, through some miracle of chance, it were to placed in exactly the right location to cause a cascading series of power failures resulting in a catastrophic shutdown of an entire city and, possibly, an entire country. I think we should make pennies our Third Offset Strategy.

      Look, no one is arguing against pursing cyber efforts. They can only help. This post is about the wisdom (or folly) of making network/cyber/unmanned the basis of our military for next few decades to come.

      Unlike nuclear weapons which were a sure outcome when detonated or smart weapons which were a pretty sure outcome when launched, cyber is very much a hit or miss affair. It depends on just the right vulnerabilities being present, is very hard to deliver, may or may not work, is unpredictable in its results, and is relatively easy to recover from. Again, not the basis for an entire military strategy.

      Setting all that aside, as was made clear in the post, an offset strategy should result in a significant and enduring advantage. This does not. Arguably, our enemies are more advanced than we are in this realm, having already demonstrated widespread and significant hacking and cyber attacks against us. There is neither a significant advantage (they can do the exact same thing to us that we can to them - no advantage) nor an enduring one.

      Reread the post and focus on the main premise not one small aspect.

  7. CNO I don't think machine learning and whatnot is that far off the mark, but you are right that the Military should experiment with actually maintaining it's equipment and keeping it operational because well that would be a big start.

    But data mining is used widespread in the commercial industry, for instance data mining is probably the biggest thing driving economics of fracking right now. It has resulted in roughly halfing the number of stimulation needed for each well because all that data has enabled better models for exactly where we should be stimulating the field.

    Obviously the DOD/DARPA should stay away from developing the tech so to say, but utilizing commercial standards could improve performance of existing equipment, i.e. it could improve radar performance by collecting all the data and optimizing the search and tracking patterns, that sort of thing. Improve ASW by improving the models used to determine where we dip the sonar buoys.

    And augmented decision making is already being rolled out, for instance when fire control soloutions are calaculated that is an augmentation. I think weapon swarms could be potent future weapons too, the conformal weapons pod on the ASH can carry 10 SDBs, I can imagine a fighter (i.e. F15/Su34) or a drone carrying the equivelant of two maybe three such pods. In this case we are talking about 20-30 of these small munitions, we are going to need a degree of co-ordination to allow for redundancy in the form of ensuring targets are still being hit even if interception occurs or one does insufficent dammage, and to take care of targets of opportunity and high value targets. Avoid threats like radar, etc..etc..

    In this case I could imagine a soloution which might see a great number of moderately intelligent SDBs that operate in a slave mode to an intelligent SDB, so you put the Radar Warning Receivers or other expensive equipment on only a small portion. Still enough for redundancy. The rest can remain relatively cheap and stupid...

    But we can't conduct a complex strike utilizing dozens, maybe even hundreds of such bombs if they don't have the brains. I mean the whole point of the SDB is to have a stand-off weapon capable of targeting individual targets, and providing more stowed kills per sortie than larger bombs right? We need the intelligence to make that work, so the SDBs don't double up and hit the same targets.

    1. And I can imagine Drones working in concert with fighters, I mean lets look at the Avenger 2 canceled bomber, or it's derivitive and cancelled UCAV now tanker. Sure it is roughly equivelant in size to a F18 and you pay for planes by the pound or so the saying goes.

      But when we are building a drone (something without a person in it) like this, when it is a bomber as opposed to a Fighter. Then we can start stripping capability.

      Do we need the afterburner capability in the engine? Do we need the high thrust to weight ratio. No, we can cut back there. Do we need the manouverability or the sensors or the redundancy for battle dammage survivability, the life support systems, etc..etc..

      We can strip the plane back to the bare minimum, strengthen it to have one or two decent weapon bays (for big missiles and many smaller bombs). Add back the comms equipment, targeting computer and sensors including an IRST and a RWR. And build a much cheaper bomber. Give it limited intelligence. You are going to get a much cheaper delivery platform than our current fighters that need to do everything, and they can accompany your Fighters or conduct strikes autonomously.

      They can fly in formation and in advanced with IRST on, and you can have your actual fighter behind, not emitting, improving your kill chain... Id they see it, all they have shot down is a drone with the equivelant of an IRST/Target pod and RWRs built in. On a potentially even comercially derived engine with no after-burning capability...

      They can jam comms, but jamming point-to-point laser or radar comms at short-medium range is hard.... It is an intermediate step between remote control, and fully autonomous.

    2. I hope you're not confusing my disdain for electronics and networks AS THE THIRD OFFSET STRATEGY that we'll bet our entire military future on with my opinion about machine "assists" in general, are you? I have nothing against using computers and networks (as long as we don't develop a crippling dependence on them as we've done for GPS) to enhance our combat capabilities. The post was not about combat enhancements. The post was about the Third Offset Strategy that will be the basis for our entire future military. Enhancements do not constitute an offset and certainly not enhancements that are so questionable in the face of ECM and cyber attack.

    3. "But when we are building a drone (something without a person in it) like this, when it is a bomber as opposed to a Fighter. Then we can start stripping capability."

      No, you can't. Or, rather, you can but you'll fail the mission. If you're going to build a long range, penetrating bomber, which is what you seem to be suggesting, it'll have to be big enough to have good range and big enough to carry a worthwhile payload. Now, how will it survive to reach its target? If it's going to be anything other than a flying target drone, it will need speed, maneuverability, and stealth - now we're adding capabilities, not stripping them out! It'll need an ECM suite and sensors. It'll need a ground/surface targeting system and probably a laser designator. In short, it's going to be an F-35 without a pilot which means it's going to be expensive - hideously so. It won't be expendable. If you do strip out all the things I mentioned then it won't survive long enough to reach its target. People seem to associate some kind of almost magical properties with an unmanned aircraft. The reality is that they're just aircraft - as good or bad as any other aircraft and just as expensive. In fact, they'll be so expensive that they won't be expendable and if they're cheap they won't be survivable. That's the UAV quandry.

  8. I am increasingly of the opinion that in some regards, a state run defense industry may very prove to be the lesser evil.

    It seems that the private sector obsession for profit at all costs, has become a threat to the long-term viability of the US, not just in terms of fighting ability, but to deliver a good standard of living to the middle class.

    Like it or not, only governments seem to have the resources to invest in a viable military. They have to invest in their people, in training, in research, and in areas where the pay-off may not happen until decades later.

    1. Speaking of the LCS, in case you haven't heard, the USS Freedom "will need an engine rebuild or replacement after a seawater leak" affected her diesel engines, according to

    2. I disagree. The problem is not the defense industry, per se, but the fact that the military has allowed the defense industry to dictate products and prices. It used to be that the military designed ships and aircraft, either directly or via very specific requirement lists, and then simply used industry to do the construction under military supervision. Now, however, the military has eliminated their internal design capabilities and depends on industry to tell them what they need and what it will cost.

      We need to reconstitute our in-house design capability and return industry to the status of a pair of dumb construction hands.

      Related to this, the military has also mismanaged force numbers resulting in a catastrophic reduction in the number of defense industry companies. This has stifled competition and increased prices.

      There is nothing wrong with the private sector's profit motive. Properly used, it will drive down costs and increase quality.

      Had we halted LCS contracts after the first few when it became apparent that LM and Austal were producing crap, and awarded the contracts to some other company, cost and quality would have improved. As it is, LM and Austal have no incentive to produce a good product. Quite the opposite. If the ships need extensive post-delivery work, that's just more money for the manufacturers because we have no viable alternatives.

    3. US private arms industry is heavily subsidised and patronised by US government, and you're right. Middle class sees no discernible benefit from it.
      Conversely, the German private industries are subsidised by government, but not so heavily patronised, and lots of their sales are foreign. German middle class is fairing better than the US one. So, its not universally bad, just, US execution and methodology is flawed. Perhaps a study of the german model is warranted.

    4. "Related to this, the military has also mismanaged force numbers resulting in a catastrophic reduction in the number of defense industry companies. This has stifled competition and increased prices."

      I think that this is a massive issue. We have 1 main surface combatant, one small surface combatant that is supposed to pick up 3 roles, and 1 'fighter' for 3 services.

      If you are a defense contractor you win that contract or realistically lose the ability to produce that type of product.

      When we had an interceptor, long range attack aircraft, short range/light attack aircraft, COD, ASW, and some rebuilds of existing airframes for tanking, there were multiple contracts that companies could bid on. Don't win the F-14 gig? Go for the F-18. Don't win that? Try for the S3....
      That allowed companies like Grumman, Vaught, Lockheed, etc. to have a shot. And having multiple companies allowed them to realistically compete.

      Same with surface combatants. We used to have DDG's, FFG's, CG's, MW, .... etc. Now we have only DDG's and LCS's being built.

      Now, once the F-22 got cancelled, it was win the F-35 contract or.... don't build fighters anymore. Didn't win the LCS competition? Oh well, because there isn't going to be an MW ship or ASW ship contract you can go after if you are a smaller yard. CVN's or SSN's? So big or complex only a couple companies can do them.

      Add defense industry lobbying and general officer level complicity on top of all that (I get a little sick in my mouth every time I hear Brad Byrne talk about the LCS like its the Iowa, or read about an admiral who went to work for LM) and you have a badly broken system. But its not surprising because if you have that ONE CONTRACT to go for, you'd better pull out all the stops to get it. 'Exaggerate', Lobby, promise jobs to general officers, whatever. Billions are at stake.

      I'm not sure how to fix it. My worry with government owned defense is that you've completely eliminated any competition at all. I think in part the government is responsible our current situation. I've read that in the 90's there was a concerted effort on the part of the government (2 administrations) to pare things down so we could 'enjoy' the 'peace dividend'.

      How do we fix all this? Maybe do a bell type break up of the defense contractors, coupled with changing the force composition ( going back to purpose built platforms)? I do like CNO's idea, but I think it has to be combined with breaking up the current contractors and finding a way to reign in lobbying.

      Alot easier said than done, but something has to be done if we are going to go down the path of sustainability.

    5. @CNO

      The defense industry will always try to dictate prices and products. Doing so is economically rational for them because they maximize their profits (at least in the short to medium term).

      The military brass have an inherent interest in pleasing a private for profit industry in exchange for a lucrative career after they retire. It is a major conflict of interest in the brass - are they agents of the defense industry or of the people?

      It's the same problem that politicians face really.

      Also, if it makes sense to have in house design, then why not take the next step and have in house manufacturing, at least for some items that are procured? It's quite common for functions in the military that are outsourced to cost more after they are outsourced, not less.

      @Nate Dogg

      A thorough study of the German and Nordic models are warranted. They seem to have a much more vibrant standard of living than we do here in North America.

      They have a much more vibrant manufacturing sector as well in the civilian world. I've said it before - it does no good to have a powerful military if America's manufacturing heartland becomes the Rust Belt. That's a serious threat to national security.

      To be honest, I have become increasingly convinced that a Nordic Social Democracy is superior in terms of average living standards for the typical middle class citizen. Actually given the declines in living standards, there are articles claiming the US is no longer a middle class.

  9. This is starting to make the rounds on the financial sites.


  10. Somehow, in some unexplained way, this human-machine collaboration is going to overcome numerical and explosives deficits.

    Yes, correct, despite your sarcasm, thats exactly what they're planning.
    As an example, during dessert storm, 3% of all munitions expended were smart bombs, yet in post war assessments, they were found to have caused 76% of effective damage/casualties caused.

    Expect theyre counting on similar asymmetric capability.
    Whether you believe theyre capable of doing it, well, they made nukes and smart bombs work. Have a little faith.

    1. Have a little faith??!! Do I need to go through the litany of incredibly poor decisions the military has made over the last few decades? There's not much basis for faith there!

      Setting aside the absurdity of the "have a little faith" appeal, the reality is that the game changing nature of nuclear weapons were obvious to everyone as was the ability to utilize smart weapons. The game changing nature of network/cyber/unmanned is far from obvious and, quite the contrary, has many serious flaws and questions. That's not what you want to bet your military future on!

      There is an overwhelming amount of historical and logical evidence that this is a poor decision.

  11. Ah, thanks for the explanation. Ive been seeing the term. But failed to get a good grip on the details.

    Now i know a bit about data mining as its my job. Intellegence wise this is becomming defining. As we have more data now than we know what to do with.

    But hardware wise are they maybe thinking about concepts like phalanx and full auto aegis ?

    Both of which enhance weapons capability in terms of speed of responce and action well beyond that of a human. And offset numbers as it were.

    One means less AA guns per ship required. The other offsets a saturation attack by many adversaries.

    Is this possibly what they mean ?

    1. I suspect you're still missing the main point of an offset strategy. Another commenter put it better than me when he said the previous offsets, nuclear weapons and precision weapons, provided a huge and enduring boost in unit lethality. That's the definition of an offset strategy: a huge and enduring increase in lethality that provides an enormous advantage for us.

      Thus, the question is not whether some technology is better than what we have (data mining, for example) but whether it will provide a huge and enduring increase in lethality.

      A Phalanx, to make up an example, that is improved a bit, or even quite a bit, is great to have but does not constitute a huge increase in lethality for our overall military.

      So, does networking, data mining, and unmanned vehicles which are some of the manifestations of the Third Offset, offer a huge and enduring increase in lethality sufficient to make other countries hesitate to engage us or make them cannon fodder if they do engage? If so, we have our Third Offset. If not, we may have nice improvements but not an offset advantage. Frankly, I don't think those things provide even a bit of advantage and certainly not an enduring one. Arguably, China and Russia are already ahead of us IN OUR OWN CHOSEN OFFSET!!!!

      Hope this made things even clearer.

  12. Im not sure lethality is really the term.

    Its about winning wars not killing people.

    The two are not necessarily synonamose.

    Beyond that. Yes as you say, its about a provable verifyable capability increase providing deterrance.

    And hopefully stopping wars before they start.

    I would argue that all weapons require a target. Targeting requires intelligence and intelligence gathering requires data.

    Now back in the day hundreds of guys with magnifying glases used to paw over photos.

    Today with ISTAR. AEWAC, SIGINT, drone and saterlite data streaming in real time.

    Flexable and fastmoving battlespaces. EW and manipulation of the EM environment. And the key precepts of command and control in a blitz krieg arena.

    'Just' data mining and information processing COULD make all the difference.

    If you can collate and summerise. Strain out superflous detritase you can strip away the fog of war and increase the speed and accurasy of the desision making process.

    You cant get enought guys with magnifying glasses, working fast enought, to do this in top end modern warfare.

    We are talking about prosecuting an entire warfront live here not just a battlefront. Potentially globally. THAT is the advantage !

    Now dont get me wrong. Maintenance/ training and the rest is 100% critical at all times. But this really is about THE NEXT STEP not the current status que.

    Staying ahead of the game not just mastering the current one.

    1. Ben, you're generating two or more duplicate posts, consistently. Any idea what's happening?

    2. "Im not sure lethality is really the term.
      Its about winning wars not killing people.
      The two are not necessarily synonamose."

      War is absolutely about killing and destroying. Even the case of a mammoth country simply annexing a small one is about the threat of killing. Give me one historical example to support your contention that winning wars is not about killing people.

    3. You're still missing the main premise! An offset, as defined, has two characteristics:

      1. A huge advantage (in lethality)
      2. An enduring advantage

      An advantage that is not huge (a better bullet, for example) is nice but does not constitute an offset to build an entire military around.

      A huge advantage that lasts a weeks accomplishes nothing because after that week, you're right back to par.

      It is highly doubtful that networks and unmanned vehicles are going to produce a huge advantage. At best, they might produce some useful benefits.

      Networks and unmanned vehicles are absolutely not going to produce an enduring advantage and, likely, will not produce any advantage. Russia and China are certainly ahead of us in electronic warfare and cyber warfare. Russia is well ahead of us in battlefield robotics. Both are as advanced as we are in general unmanned vehicles. WHERE'S THE ADVANTAGE LET ALONE AN ENDURING ADVANTAGE?

  13. Sorry. Im in Peru right now not GB. networks seem unrealiable.

    Please dont think im being "hippy".
    Death is an unfortunate by product of warfare. But not the point of it.

    Historically very few wars have been won by genocide. I.e. killing everyone.

    Its in-efficient.

    In state on state warfare you concentrate on removing an enemy states ABILITY TO MAKE WAR. This can mean infrastructure. ( Hardware ) and / or political will ( removal of the governing body ).

    The last war focused on "lethality" as a resolution would have been the first world war. And we all know how that went.

    Im a little surprised we are having this convosation ?


    NO Advantage is permenant. 20th century warfare has been a continous evolution.

    Your continous underlying assumption ( or perhaps just my interpretation of you tone ) that we can in any way "stay still" is misplaced.

    As leaders in the field we must speculativly inovate. And as best as possible STEAR the direction of this evolution. Not allow others to dictate the nature of the change. ( too much )

    Yes states will adapt and overcome. This is natural.
    But that is not the point.

    This is an eternal struggle. Not an end game.

    If you want happy ever after. Try Disney not warfare ;)

    Please dont misenterprate me as argumentative. Im not. Its just a debate. Ben.

    1. Very few wars are won because one side asks for the other's territory and they willingly give it. Wars are won because the threat of death is there if you don't get what you want. You can't win a war without killing or, at least, the realistic threat of killing. That doesn't have to include total genocide.

      We've forgotten what real war is. War is killing in order to achieve one's goals.

      Again, give me one war that was one by means other than killing or threat thereof.

    2. "NO Advantage is permenant."

      Who claimed that?

    3. "Your continous underlying assumption ( or perhaps just my interpretation of you tone ) that we can in any way "stay still" is misplaced."

      Where did I say that?

      I said that this Third Offset is wrong. I didn't say that we should do nothing. I don't mind disagreement but disagree with something I've actually said!


Comments will be moderated for posts older than 7 days in order to reduce spam.