Gun Control


Post-Newtown, like post-Aurora, post-Gabrielle Gifford, post-Virginia Tech, post-Columbine, ad nauseum, will bring the usual hue and cry for more “gun control” laws. The arguments on both sides will be typically irrelevant, having nothing to do with decreasing these tragic incidents nor reflective of the Second Amendment.

“Gun control” has more to do with the 10-ring than it does with protecting citizens.

Both the Left and NRA are missing the point. If what the Left really wants is to make society “safer,” whatever they mean by that; and if the NRA really wants to protect a citizen’s right to “keep and bear arms,” whatever they mean by that, they should be able to work together. But that’s not, and has never been the case. The Left wants to ban guns. Period. The NRA wants any weapon desired to be obtainable by citizens. Period. That’s a debate that will be won by the Left (because more voters are against automatic weapons for sale at WallMart than want to own one), but they will have rendered a more dangerous society in “winning.”

A point intentionally overlooked by the Left is that, in overwhelmingly most cases of horrific gun attacks, several existing laws were violated. The crux of the problem is enforcement, not inadequate legislation – but professional legislators will always be eager to leap into the spotlight with the newest “remedy.” A point intentionally overlooked by the NRA is that it is inexcusable for not making personal weapons that only respond the rightful owner. We can do that now, but it would make guns more expensive, “hurting the poor disproportionately.”

The Left should be livid at the apparent lack of interest in enforcing existing gun laws, but you never hear that from them. The NRA should be insisting that gun shows be tightly regulated (on-site), but you never hear that from them.

Reality check:

Assault weapons have always been illegal – and still are, even though the misnamed assault rifle ban sunsetted in 2004. An assault weapon is one that is designed to be used on automatic, and those weapons cannot be sold to the public. This is a situation where the Left’s propensity to relabel things so as to mislead has been very successful. They should admit that the weapons they are railing against are not automatic weapons, rather auto-loading semi-automatics. If they want to ban those, be honest about what they talking about.

AR-15s, Bushmasters, and the like can be converted to fully automatic relatively easily (breaking existing law), and those kits are illegal to sell or buy in the United States. Yet they are widely sold at gun shows and on the internet. The NRA should admit this and work toward tightening regulations regarding the open sale of the capability of making a gun illegal if it were sold as modified. If the NRA wants fully automatic weapons legalized, they should admit it and be honest about they are talking about.

We have the ability to make a weapon recognize the registered owner and only function when being used by that person. Sure, these systems can be hacked, but it would immediately reduce the number of malcontents able to use a stolen gun (like those used at Newtown).

Last but not least, there are around 270 million private guns in the United States – excluding those owned or issued by a government entity, federal, state and local. How many generations do you think it will take for attrition to remove those from the public if perfect gun bans were instituted tomorrow? Unless, of course, the Left is talking about rounding up existing guns, and if they are, how do they propose to do that without shredding the Bill of Rights?

But don’t hold your breath. Like all other emotional issues in this hyper-divisive age, this one will argued with disingenuousness, hyperbole, made-up numbers and outright lies. It makes for better television and newspaper sales.

Mayan Calendar Properly Understood


OK, so they missed it by ten days, not bad after twenty-five hundred years … that’s less than six minutes a year. Again, not bad for a culture that couldn’t tell time at night.

Actually, the joke has been on us all along. We’ve got a Congress that so mistrusts itself that it issued a poison-pill to prevent them from inaction on debt reduction, and now is calling its own bluff! Incredible.

Please stop us before we act again!

These past four years have been the most pathetic exercise in governance of which I’m aware – and that includes some Third World nations. At least they can claim ignorance. All we’ve got is utter stupidity.

We’ve got a president who thinks “million” equals 250,000, Democrats who can’t see anything other than race, gender and class, and Republicans who can’t agree on much of anything. We’ve got a Senate Majority Leader who says he won’t bring anything the House produces up for a vote and a Speaker of the House who keeps sending things to the Senate he knows won’t pass. Our president says he is willing to compromise, but when pressed on any given feature, his answer is always “my way or the highway.”

There was a particularly funny poll taken sometime last year that reported 43% of New Yorkers would prefer replacing Congress with random names taken from the Manhattan phone book. May be time to try that.



Well, here it is, Saturday December 22nd, and we’re all still here. Turns out that Neolithic fortune-telling is no better than Gypsies or horoscopy.

The “Galactic Alignment” – when the Earth is aligned with the “equator” of the Milky Way – that was supposed to trigger the end of the world, we were told, is important because it only happens every 26,000 years. Completely forgotten in the hysteria is that this has happened 176,923 times since the Earth was formed, and we are still here. It happened again this month, and we are still here.

Turner Classic Movies showed a marathon of Doomsday films yesterday, and it was amusing to watch. From the 1950s came nuclear war and alien invasion films; from the 1960s, we got the biological disaster and alien invasion stories; the 70s added computers-run-amok; the 80s and 90s gave us more of the same. In the 2000s, however, we began zeroing-in on the Mayan calendar. Seeing them serially juxtapositioned proved interesting. The early films, generally, had gritty scripts and shoddy special effects, while the more current you got, the sloppier became the scripts and slicker got the effects. Hollywood, like us, had become all hat and no cattle. All show, no go. Style over substance.

For all their trying – and writers have tried everything from giant ants to cometary strikes to body-snatching aliens to natural disasters – we are still here.

Frequency is Directly Proportionate to Humaneness


Technological developments have changed the ways in which wars are fought – gunpowder, powered flight, tanks, submarines, and so forth – and the net effect has been an ever-increasing force multiplier. So it is with remote, semi-autonomous and fully autonomous combat platforms. The difference here, however, is that these machines also have the potential to change the very nature of warfare.

The drastic reduction in total casualties involved in small wars conducted using remote platforms is making war “cleaner[1].” The ratio of combatants killed to non-combatants killed will remove some of the disincentive for going to war among some populations. One of the net effects of remote platforms of increasing autonomy might be to make war easier to politically tolerate. One or both sides of an asymmetrical conflict, or those on the losing side of a conflict, may opt for the Hamas strategy of placing crew-served weapons and artillery rockets next to hospitals, schools, in residential areas, etc, to drive civilian casualties back up. If the world treated this practice as the cowardly act that it is, it would be less desirable as an option. But public acquiescence seems to have legitimized the practice, so I’m sure we’ll see more of it. These tendencies work against each other – increased accuracy and autonomy driving civilian casualties down while placing civilians in the line of fire drives their casualties up – I think, long term, the former will overpower the latter.

Another aspect of these machines that will increasingly come into play is the number nations bringing them into inventory – 51 countries operate UAVs right now, made by over 200 different airframers worldwide. Will they all deploy them with same degree of caution as we do? Undoubtedly not. But the overall effect should net fewer civilian casualties, even if not by as much as could otherwise be the case. Mixed in here, of course, will be those tyrannical governments who will use them against their own citizens, but again, worldwide, the effect should be on the humanitarian side of the ledger. What won’t balance toward humanitarian ends is the fact that these systems are far less expensive than man-rated analogs (fighters, attack aircraft, etc). While they may be out of the reach of small cells of terrorists, they could easily be supplied, for example, to Hizbollah by Iran. This would further complicate the West’s anti-terror efforts. One conciliation is that only state-sponsored groups would be able to afford to use AG-114 Hellfire missiles, at ~$68,000 apiece (to legitimate customers) they would be a bit pricey to train on and use with any regularity. More likely, sub-national groups might try to mate FIM-92 Stinger-like heat-seekers (about half the price and far less training involved) to UAVs and use them in the anti-air role[2].

If war becomes politically easier to wage, we will have more of them – that’s human nature. As remote, semi-autonomous and autonomous combat platforms become more widespread, war becomes tactically easier to wage. This, too, will tend to make war easier to politically wage, as the promise (fulfillable or not) of quick victory seems plausible to lay audiences. What the widespread use of UAVs will accomplish, however, is to complicate the air superiority task.

Air superiority has been enjoyed by the United States in every conflict since the early days of Korea, and its importance is lost on generations born since (largely because they’ve never had to consider our own troops coming under a withering vertical attack such we visit on our enemies). Due to their relatively small size and stealthy design, second- and third-generation UAVs are very difficult to detect by radar, giving them increased opportunity to operate in hostile airspace in the ground support, interdiction or attack roles. This will accelerate the fledgling field of anti- and counter-UAV technology and tactics. This back-and-forth will affect the efficacy of UAVs over short periods, but not their absolute rates of use.

The probable result of this contest between UAVs and defenses will lead to increased cyber- and directed energy weapons[3], both of which have multitudinous “off-label” uses – direct cyberattacks on adversaries (already well underway), ballistic missile defense (BMD), tactical missile defense, classical air defense, communications and navigational disruption, and so on. These things, too, will make war politically easier to wage. To the casual observer, this presents a “cleaner” impression of combat than television images of bodies in the streets and on the field of battle.

The net-net of this whole field is that it will tend to reduce the total number of casualties in a given conflict, and drastically reduce the number of non-combatants killed and wounded in a given conflict. This is a good thing, unless, of course, it also reduces the political threshold for going to war in the first place.

[1] There has been much said about “innocent” people caught up in UAV strikes in the tribal areas of Pakistan, but in the strictest sense, those who harbor enemy combatants can’t claim non-combatant status.

[2] This could put rotary-wing platforms at much greater risk than is now the case.

[3] The cyber and directed energy environments will be examined under separate cover.

UAVs as Threshold Technology


Air Force Colonel John Boyd, who literally wrote the book on dog fighting[1], was a brilliant strategist and tactician, and the brains behind the “Lightweight Fighter Program” which produced the F-16 Fighting Falcon and F/A-18 Hornet families of fighters. At his insistence, the Air Force included maneuverability in the design call for the F-15 Eagle – the first aircraft specifically designed with maneuverability as a chief criterion. The result was the invention of the air superiority fighter[2]. His most profound legacy is his study of how organisms react to their environment, and why that is important to know.

He produced what he called the OODA-Loop – the process of Observation, Orientation, Decision, and Action – the process by which complex organisms (including humans) process sensory input in order to adapt to new conditions. You notice something new, orient it to your current gestalt, decide if it’s relevant enough to require action, and if it is, what to do about it, and then do what you’ve decided is appropriate. Observe, orient, decide, and act. This has import to military operations because is it explains the natural tempo of such undertakings – generally consisting of engaging the enemy, followed by an operational pause, followed by renewed engagement. The original contact is driven by a recent OODA process, but as we know, no battle plan survives first contact with the enemy, so the initial contact is evaluated (the operational pause) and the next round of contact is driven by that evaluation. OODA.

The strategist’s (theater command’s) OODA-Loop tends to be long-wave – we need to stop fighters from infiltrating from outside the theater, while the field commander’s/shooter’s OODA-Loop is far more tactical and short-wave – “can somebody get that machine gun out of that window!” But in both cases, the ideal is to operate inside of your enemy’s OODA-Loop – ambush the enemy patrol before it can set up and ambush you. This goes a long way to explain exactly why small units are far more agile and mobile than large units – they take local observations and act on that information far quicker than a cumbersome chain of command can process similar information and stage a response. This phenomenon is also why jointness works better than the pre-Goldwater-Nichols “stove-piped” architecture of military organization[3]. The OODA-cycle is greatly shortened when all assets are known a single decision-maker, rather than waiting for the various owners of the individual assets to “get back to you.” Even if they do so promptly.

In the first known example of network enhanced operations, two American aircraft carrier commanders coordinated the actions via eMail when the interception of radio communications proved probable. Their rapid response to deteriorating conditions in the Taiwan Strait defused the situation before it could spin out of control. This was in 1997 and gave birth to the Network Centric Warfare school of thought[4]. This, again, is an attempt to shorten our OODA-Loop by providing combat commanders with increased situational awareness. A product of that theory was also demonstrated during the First Gulf War when General Schwartzkopf had a “dashboard” that displayed current conditions around the theater. Live feeds from satellite, surveillance aircraft, ground-based cameras and real-time updated status charts on units in the field were displayed in windows on a large screen in the general’s operations center. It was fairly rudimentary, but a large jump in real-time information available from the field to theater command, and it proved to be invaluable during ground operations as Desert Shield became Desert Storm.

UAVs are an extension of this ability, as a Predator, for example, can orbit overhead and send real-time imaging back to a combat or theater commander of an area of interest, or can track a known bad guy back to his operational headquarters or home cell. Such ability was exploited along the Afghanistan-Pakistan border, and resulted in watching Taliban and al Qaeda patrols cross the border, allowing us to intercept them at a point of our choosing. It allows us to operate inside their OODA-Loop. Arming UAVs again shortened the reaction-time between spotting a bad guy and getting ordnance on his location. Often scrambling aircraft took longer than it took for the bad guy to disappear into the population. If the observer can shoot, the reaction-time is vastly shortened.

It must also be noted that the mere processing of the individual nodes in an OODA-Loop is far faster for an algorithm than it is for the conscious mind. The trouble is, an algorithm has no subjective filter – it does what it is told to do. Period. This is supposed to be the realm of AI – “fuzzy programming” that gives the algorithm some leeway to evaluate qualitative, rather than quantitative, information. Going from concept to code is proving problematic.

And it is this aspect of autonomous systems that the public is going to insist be resolved (to the public’s satisfaction) before confidence in these systems renders them acceptable.

We’ve been concentrating on the micro, but there are also some necessary macro concerns that autonomous systems must satisfy. That resort to war can only be just under certain conditions goes back at least to Cicero. Augustine of Hippo, Thomas Aquinas and Hugo Grotius later codified a set of rules for a just war, which today still encompass the points commonly debated, with some modifications. This is called jus ad bellum, or “Just War Theory.” Equally important are the questions about the right way to use force when justified (the concern of jus in bello). This commentary goes back to Deuteronomy, where God informs Joshua the proper way to raise forces and conduct violence against an enemy – including the warning of a town before an attack begins, something Israel practices to this day. The philosophical discussions of jus ad bellum and jus in bello have been reduced to more legalistic language in the Geneva Conventions, which spell out the international laws of war. Autonomous combat systems will be required to operate within these constraints.

The trickiest aspects of these restrictions will have to do proportionality and protection of non-combatants. A nuclear strike on a state that interdicted an American-flagged ship at sea would be considered disproportionate, for example. While that’s an extreme, there are concerns about, not only retaliatory actions, but also initial attacks. Many of these concerns overlap with being able to differentiate between combatants and non-combatants. Autonomous combat systems will have to quantify a response to an observed provocation, and must be able to distinguish non-combatants. Neither of these abilities are currently codable using today’s state-of-the-art computer languages[5]. An autonomous UAV or UUV will need to be able to recognize the sovereign airspace of non-belligerent nations and their national waters. I have an idea that these systems will remain semi-autonomous – human-on-the-loop – long after they are capable of true autonomy.

Once satisfied that the laws of war can be taught to autonomous combat systems, there are the rules of engagement, which are tailored to particular conflicts. This brings two parameters into focus: being able to distinguish between levels of response to variable stimuli; and the necessity to essentially reprogram each platform to reflect the particular rules-of-engagement environment within which it will be used – and those ROEs may well change mid-conflict, so the ability to seamlessly alter those directions will have to be available on short notice, in-theater. This introduces a new paradigm into the equation. The final code allowing a combat system to function with military utility while observing the laws of war will necessarily involve millions of lines and be complex beyond current operating systems. Now, they must be able to be altered in the field without disrupting unaltered code, and without spawning unintended consequences that could cost lives, both friendly and hostile.

That’s a rather high hurdle about which I haven’t seen even conceptual discussions. I suppose that top-down modular programming architecture could allow for a certain amount tinkering, but the possible random behavioral results would keep me up at night were I responsible for such alterations.

And, of course, these chip sets will have to be hardened to withstand electromagnetic interference, heat and cold, damp and dry, dust and mud, and will have to be encrypted beyond even rather sophisticated attempts to jam or spoof the platforms. They will be networked, which means the networks will, likewise, need to be robust and redundant, and resistant to jamming and spoofing.

These systems are coming, but probably not as fast as the popular press makes it seem at times. The software will be enormously complex, and testing will have to be exhaustive. The questions raised herein must be addressed and satisfied. These discussions can’t begin too early, both to assist in the public’s understanding of the technology, and to help guide those who will bring it about.

[1] Colonel Boyd wrote the tactics manual for the Air Force’s Weapons School at Nellis AFB [NV].

[2] For an overview of Colonel Boyd’s work, see John Boyd and John Warden, Air Power’s Search for Strategic Paralysis, School of Advanced Airpower Studies, Air University, Maxwell AFB [AL].

[3] The Goldwater-Nichols Act prompted the services to work more closely together toward the common end of getting support to the shooter. This was exemplified in the First Gulf War when we had a single Air Tasker, who knew what aircraft were available and could respond much more quickly to requests for air cover than sending the request up the Air Force, Marine and Navy “stove pipes” to see what was available. It was not uncommon for Army units to get Navy Tomcats or Hornets overhead in support of an ongoing firefight.

[4] See, for example, Paul Murdock, Principles of War on the Network-Centric Battlefield: Mass and Economy of Force, in Parameters [US Army War College], Spring 2002, pp. 86-95; Clay Wilson, Network Centric Operations: Background and Oversight Issues for Congress, Congressional Research Service, March 15 2007; Vice Admiral Arthur K Cebrowski and John J Garstka, Network-Centric Warfare: Its Origin and Future, in Naval Institute Proceedings, January 1998.

[5] Some very interesting work has been using the computer language LISP, which can be coded so that the machine can, essentially, reprogram itself – alter its code to fit new situations. Needless to say, this raises a whole new set of potential nightmares, and probably won’t figure into these machines.

Classical Warfare morphs into Asymmetric Warfare


General war, total war, medieval war – whatever you wish to call it – is rendered unacceptable by the presence of fission and fusion weapons. And nuclear weapons, we find, are not operationally scalable. Sub-KT warheads have tactical utility, but the attendant radioactive residue outweighs that utility. There are niche uses for nuclear explosives – the deploying of an electromagnetic pulse (EMP) to disable the electrics and electronics over a wide area, for example – but exploding a nuclear weapon in the airspace of a sovereign state would most certainly be viewed as a hostile act. Not to mention the tracking of an inbound missile. It could only be taken as the first strike of a nuclear war. The same for using orbital nuclear-pumped lasers to blind communications and GPS satellites.

That’s a sketch of the nuclear profile for sovereign states. It’s quite different for non-state and sub-state actors. If al Qaeda, for example, could get their hands on a nuclear warhead (and the codes for arming it), they would use it for blackmail, and I wouldn’t put them above detonating it. But that’s the top of the pyramid. They can use spent nuclear fuel rods, or cobalt (Co60) from medical equipment, or tailings from uranium reprocessing – any radioactive materials – to jacket an explosive charge in creating a “dirty” bomb. Any of these stolen materials could be used to spike a water supply, or a corn field. The point is, a tactically insignificant amount of radioactivity released in public is a terrorist act – it creates panic rather than damage, and demonstrates just how powerless a government can be in protecting its citizens. KGB used polonium (Po109) to assassinate dissidents, rogue scientists and foreign assets for whom they had no further use.

And speaking of spent fuel rods, we – and by “we,” I mean humanity, not just Americans – still have no idea how to securely dispose of them, so they sit in Čerenkov pools at power stations around the world, waiting for someone to provide a solution (or figure out how to steal them). Needless to say, this is a precarious situation. The point is, nuclear materials are an ongoing problem when terrorism is rampant, and much of this is due to naïveté during the Atoms for Peace program of the 1960s. The situation in Iran today is a product of the Nuclear Nonproliferation Treaty (NPT) – allowing all activity up to manufacturing a nuclear weapon. It’s not illegal under the Treaty to enrich uranium to 93%, only to then use it as a bomb core. It’s not illegal to machine a bomb core out of U235 and Pu239, only to then place them into a bomb. A state can get “a screw away” and be legal under NPT. This is why Iran can enrich their uranium hexafluoride (HEX) to any degree of purity they wish under the Treaty, or work on the micro-shaped-charges needed to symmetrically implode the uranium jacket onto the plutonium core, or investigate the world of neutron initiators, or conduct research on the carbon-laced ceramics needed for re-entry, on and on, and not violate NPT. The naïveté was allowing the processing of uranium outside the five original nuclear states – the United States, the United Kingdom, Russia, France and PRC.

But we’re here now, and the lesson is to stop being naïve.

This could begin by thinking hard about the “No Nukes” mentality. Is the world better with or without general war between great powers? History shows us that the Cold War represents the longest period during which the dominant powers did not engage one another in general war. What’s different about the Cold War period? Both dominant powers were in possession of fission and fusion weapons.

Speaking of naïveté, “No Nukes” presupposes that no nuclear power will cheat during or after the nuclear disarmament process. Does anyone really think that DPRK and Pakistan will forego what they see as the only thing making them relevant? Will Beijing or Moscow? A childish supposition. Beyond that, once the world is truly nuclear-free, there is no disincentive for great powers to engage in total war. What started out as our genie and bottle problem is now the world’s.

Proliferation is a problem, as we have seen, for which there exists no reliable safeguard. What could be done, however, is for the West’s nuclear powers to say out loud that any nuclear detonation will be traced to the party whose enriched uranium was used, will be considered a nuclear attack by that nation, and retaliation will be visited on that nation.

The asymmetrical warfare being waged against the West changes the way in which the West fights as well. Insurgents and terrorists know that they cannot win in force-on-force engagements, so they avoid them. They instead hit-and-run, plant improvised explosive devices (IEDs) along roadsides, assassinate personnel, and so on. Guerrilla warfare. These insurgencies can be won by the dominant power, but it takes time – more time than a democracy typically allows. It’s a bottom-up exercise – convincing the population that you are a better risk than the fighters from their own homeland whom they harbor. That we will answer the problems that drove the insurgency’s beginnings. That we can deliver to them a responsive government. All of this must be done while militarily fighting the insurgents (and protecting non-combatants). Britain was very good at this throughout their Empire, we did it in the Philippines, France did it in Algeria. We were starting to do it in Viet Nam when Congress pulled the plug. David Kilcullen and then-Major General David Petraeus laid out a textbook counterinsurgency (COIN) operation in the “Anwar Awakening,” where the Sunnis of Anwar Province were talked into backing coalition forces against the Sunnis of al Qaeda.

But the cat is out of the bag now. The Arab Spring has unleashed multiple insurgencies of varying degrees of legitimacy, all of which are infused by terrorist opportunists. Tunisia, the index case of the Arab Spring, produced a genuinely elected leader to replace a despot, but the situation has since deteriorated and is being co-opted by al Qaeda in the Islamic Maghreb (AQIM); Egypt’s popular ousting of Hosni M’barak was co-opted by the Muslim Brotherhood (who were pronounced as “moderate” by our State Department, in spite of eighty years of terrorist activities). Libya was botched by us to the point of losing our ambassador and three State Department people in the terrorist sacking of our consulate. Syria is rapidly unraveling. We are reaping the fruits of not having a cohesive foreign policy. No one knows what America stands for anymore, and that includes our foreign allies and adversaries. Our friends no longer trust us and our enemies don’t fear us. That is a dangerous mix.

Most revolutions are not won by the political idealists who start them. They get co-opted by the best organized and most brutish element of the dissidents, and this is what has happened in the Greater Middle East. One of the things that makes America truly exceptional is that our revolution was won by the political idealists who started it, and they were able to establish the government they envisioned. The Founders knew that equality is a snapshot of society in which liberty is a dependent variable, and liberty is condition of society in which equality is an independent variable. They opted for liberty. Government was to establish and protect a geopolitical space within which the people could be free. That’s what “Life, Liberty and the Pursuit of Happiness” means.

But I digress. The ways in which insurgent tactics changes the ways in which the superior power fights are of interest here, and one of those changes is the replacement of the cavalry by ISR in supplying the combat commander with situational awareness. COIN is an intelligence-driven experience. Guerrillas are vastly more nimble and mobile than are the formal forces of a state. Their leaders blend into the background, their fighters emerge from the noise and disappear back into it. They strike from apparently random directions and at unpredictable targets. Enter UAVs. In 2001, CIA deployed General Atomics’ MQ-1 Predator drones – Unmanned Aerial Vehicles (UAVs), in official parlance – to watch the movements of individual and groups of insurgents beyond the visual range of our combat commanders. Basic intelligence on the whereabouts and movements of enemy forces. But UAVs have brought a profound upgrade to that ability by being able to loiter for up to 14 hours over an area of interest; streaming real-time video back to shooters in the field as well as to the UAV operator (who could be close by or in a trailer in Arizona); can follow persons of interest to see who they report to; and so forth. The Air Force brought more MQ-1s into the game and took them into Iraq in 2003. This cracked open the door on remote capabilities in combat.

On March 4 2002, a CIA-operated MQ-1A armed Predator fired an AGM-114 Hellfire missile into a reinforced Taliban machine gun bunker that had pinned down an Army Ranger team whose CH-47 Chinook had crashed on the top of Takur Ghar Mountain [Afghanistan]. This action took place during what has become known as the “Battle of Robert’s Ridge”, a part of Operation Anaconda. This appears to be the first use of an armed UAV in the close ground support role. This kicked the door the rest of the way open on remote weapons in combat. We all have seen these things carry out strikes on television since. It’s not here that the problems lie. These strikes are carefully vetted, court tested and carried out with precision and professionalism. The problems lie in where industry goes from here.

Once turned on, Israel’s Iron Dome will automatically engage incoming Hamas rockets that it deems headed for populated spaces. Our Navy’s Aegis system automatically engages incoming low-skimming aerial targets (cruise missiles or aircraft), and can be pointed at medium- and high-altitude aerial targets. There are autonomous machines that can pull armed sentry duty. These are all task-narrowed machines that are carefully programmed to perform a limited range of actions in response to a limited range of stimuli. This is also true of LockeedMartin’s RQ-170 Sentinel UAV, one of which wound up in Iranian hands.

An experimentally modified Northrup Grumman RQ-4 Global Hawk was flown from Edwards AFB to an airfield in Sydney with human intervention only to start the engine in California and turn it off in Australia. It taxied, took off, flew its route, landed, rolled off to the tarmac and parked, totally autonomously. That’s all this Global Hawk can do – fly back and forth between that Australian airfield and Edwards AFB – but it’s a start on fully autonomous activity. All that is needed is the software to make decisions along the way. That’s the stuff of artificial intelligence (AI), and they’re working on it in AI labs from MIT to Carnegie-Mellon to CalTech.

DARPA (Defense Advanced Research Projects Agency) has been running Robotics Challenges for some years now, the most widely known being their annual competition for fully autonomous cars running obstacle courses. There are several of these “Challenges,” open to universities, corporations or individuals, that foster R&D and practical application of sensor-fusion and AI programs to produce dependable, accurate and creative decision-making by machines to unexpected stimuli.

It’s not a question of “if” we can produce fully autonomous combat platforms, but “when.” There is a current discussion regarding armed drones being used in Afghanistan (and Iraq before our departure), and that’s a good thing. Most of the discussion is non-germane to ill-informed to agenda-driven, but it is important that discussions are occurring because as the capabilities of these machines increase, the more important these discussions are going to be. For example, what happens when an autonomous combat platform engages a wrong target? Who will be held responsible? The operator? The field commander? The theater commander? The programmer? The manufacturer? The perpetrator is a machine – can’t place responsibility there.

Industry has defined combat platforms into three categories: human in the loop (which is what we’ve got now); human on the loop (where an operator has veto power over what the platform is doing); and human out of the loop (where the platform is unsupervised). The Predator is a prime example of “human in the loop,” in that it is operated by a human via video and sensor feedback. The experimental Global Hawk is an example of “human on the loop,” in that after its engine was started, it went about its business. Within its narrow task-set, Iron Dome is an example of “human out of the loop.”

The first class – in the loop – we’ve pretty much got a handle on. The aspect that still needs exploring is the spread of ISR UAVs into the civilian sector, and this is already happening as police departments and lower government agencies are acquiring them and the FAA is considering how to license them and assign them altitudes and routes, etc. The concerns yet to be resolved have to do with how they interface with the public. Will they be regulated differently than, say, police helicopters or cruisers? Will a wiretap warrant cover sensor-captured electromagnetic intelligence gathering (ELINT)? What about private investigators? What legal problems will arise from private use of remotely piloted platforms that mount cameras and transmitters? Other sensors?

The second class – on the loop – isn’t as mature a technology as Class One, although several mature systems of this type are in use. Essentially autonomous platforms that are overseen by operators will more problematic than human-in-the-loop systems because some of its activity (most, actually) will be under software control, and could execute actionable behavior before human intervention is possible as a practical matter. These situations will yield to the same accountability questions as fully autonomous systems, although the operator, even if unable to stop the questionable behavior, will be the prime target for these accountability questions. This will be exacerbated by the fact that the operator will probably be the lowest-ranked individual in the chain of command associated with semi-autonomous combat platforms.

A sub-class of semi-autonomous platforms that are currently being tested is referred to as a “swarm,” whereby a single operator “controls” a number of similar platforms on a common mission – “flying” the “flock,” taking control of a single machine only if necessary to normalize its behavior. Once on-station, these platforms are programmed to execute a common activity, communicating among themselves to accomplish the overall task. If perfected to the point of deployability, swarms will give rise to a whole new set of accountability problems associated with operations outside of the laws of war, rules of engagement, proportionality of attack, differentiation of combatants from non-combatants, and so forth.

And of course, Class Three platforms are fully autonomous, human-out-of-the-loop machines that operate on their own. Northrup Grumman is making strides in that direction with its X-47B UCAV (unmanned combat aerial vehicle), a UAV designed from the start to be armed and to test new realms of autonomy.

It has undergone initial flight testing at Edwards AFB, and is now aboard CVN75 USS Harry S Truman for semi-autonomous carrier launch and trap testing, and then autonomous launch and trap testing. It will demonstrate autonomous air-to-air refueling (using a KA-6D Intruder tanker variant). The turbofan- powered UCAV has two internal weapons bays for up to 4,500 pounds of ordnance. UCAVs are sexy, and get most of the press, but true autonomy will probably come first to land-based platforms – Boston Dynamics’ Big Dog robot, for example, is being tested to carry equipment and supplies for patrols, the quadruped can semi-autonomously matriculate over rough terrain while carrying up to 340 pounds of supplies.

Automated systems already deployed include a HUMVEE-mounted device that hears sniperfire, pinpoints its origin, and returns fire with an M-60 7.62mm machine gun. All without human input, although it is a human-on-the-loop system that can be switched over to human-directed behavior. It has been very effective under actual tactical conditions. The Israelis have deployed an automated sentry system that watches over remote areas of the Gaza-Israel border and can challenge intruders, firing on them under a strict set of rules of engagement. We’ve already covered the Navy’s Aegis system and Israel’s Iron Dome, which is based on our Patriot anti-air battery. These are all examples of ground-based systems that are, in practical terms, further down the road toward true autonomy than the aerial systems that get most of the coverage, although these, too are poised for rapid advancement. The bottleneck with all these systems is the software – a true autonomous system that enjoys an acceptable degree of confidence from critics and the public at large will require artificial intelligence of a nuanced sophistication that is beyond what we can yet do.

There are discussions taking place in academia on what this software should be able to do, which is comforting in that the AI research isn’t happening in an intellectual vacuum. This happens a lot in high tech. There was an orgasm of criticism and commentary about cloning only after Dolly was presented and the public became aware that large mammals were now being cloned – could people be far behind? The outburst caused a rash on countries issuing bans on human cloning, but the discussion was initiated far too late in the process.

Čerenkov radiation is emitted when a charged particle (such as an electron) polarizes the molecules of the medium in which it was emitted, which then turn back rapidly to their ground state, emitting radiation in the process. The characteristic blue glow of these cooling pools is due to Čerenkov radiation.

This is a term used to describe a nuclear warhead that only needs to be assembled to be viable. In other words, one can have a finished warhead, needing only the last component to be added, and not violate NPT.

Isotopic residue after a nuclear detonation has a unique signature that can be attributed to the processing facility used to refine the fissile materials.

See, for example, G-Drive:Counterinsurgency Component/John O’Sullivan, The Malay precedent: lessons from the Brits.

See G-Drive:Counterinsurgency Component/Philippines/The Hukbalahap Insurrection/The Hukbalahap Insurrection: The Insurrection – Phase II (1950-1955).

See G-Drive: Counterinsurgency Component/David Galula, Pacification in Algeria.

An Australian colonel and counterinsurgency expert.

Hamas is the Gazan “chapter” of the Muslim Brotherhood, and al Qaeda’s post-bin Laden leader, Ayman al-Zawahiri, is a product of the Egyptian Muslim Brotherhood.

See Desktop/Benghazi/Debacle in Benghazi.

An example of this is the way in which we handled President M’barak’s departure and Saudi Arabia’s reaction. See Have You Seen Me? on, 24 November 2012.

This would include wireless internet activity, eMails, wireless cell phones, etc.

Much work has been done by DARPA at the Advanced Research Lab [Penn State University] on self-organizing sensor networks that can be air-dropped over an area of interest, and the sensors will organize themselves into a network that is fault-tolerant and establishes multiple message-passing routes to ensure connectivity. These lessons are no doubt applied to swarm self-organization.

See, for example, Patrick Lin, Drone-Ethics Briefing: What a Leading Robot Expert Told the CIA, in the Atlantic, December 15 2011; WJ Hennigan, New drone has no pilot anywhere, so who’s accountable?, in Los Angeles Times, January 26 2012; Jason Koebler, Artificial Intelligence Pioneer: We Can Build Robots With Morals, in Jewish World Review, March 26 2012; Tara McKelvey, Could We Trust Killer Robots?, in Wall Street Journal, May 18 2012.

Dolly was a domestic sheep cloned by Ian Wilmut, Keith Campbell and colleagues at the Roslin Institute and the biotechnology company PPL Therapeutics near Edinburgh [Scotland]. She was born on 5 July 1996 and she lived until the age of six, at which point she died from a progressive lung disease.

Collateral Damage is Inversely Proportional to Technological Sophistication


NOTE: This is the first section of a more detailed discussion of the issues raised in the Introduction.

THERE HAS BEEN AN UNDERCURRENT OF CRITICISM of America’s having used nuclear weapons on Japan at the close of World War II. Most of these critics are blissfully unaware of the context of the times.

There was a rush to develop nuclear explosives, as Nazi Germany and the United States each believed the other was doing the same. They were, but we didn’t know how far along Germany was. Our own Manhattan Engineering District program was a hyphenated experience, leaping ahead one day and getting stuck on some arcane problem the next. The project was to produce a technology demonstrator – could the newly discovered phenomenon of induced nuclear fission be harnessed to produce a reliable nuclear explosive – and assuming success, build two deliverable fission bombs – one for Germany and one for Japan.

Germany capitulated before the two bombs were ready. The first one, Little Boy, was a “gun”-type core using a linear technique to bring the uranium to critical mass. The second one, Fat Man, had a plutonium core and used a symmetrical implosion to bring it to critical mass. They were both sent to Tinian Island to be used on Japan.

The Pacific was under American control, with only the Japanese archipelago left in Japanese hands. Judging by the way they fought during the island campaign, it was estimated that an invasion would cost close to a million American lives and around two million Japanese lives. The decision to use the bombs was seen as a way to shorten the war and, net, save lives – both Japanese and American. And they did. We now know the utter destruction wrought by nuclear explosions – it is important to understand that we didn’t before they were actually used.

This discussion overshadows but parallels that of similar commentary about strategic bombing during World War II. Again, largely absent an understanding of the state-of-the-art in the 1940s.

The accuracy of dropping a bomb from an airplane was totally dependent on conditions in the air column between the weapons bay of the aircraft and the target on the ground. If those conditions were known, and the release timed correctly, a bombardier could expect any given bomb to strike the ground within around 500 feet of the theoretical cross-hairs. The problem, of course, is that the air column between plane and target has a dynamic signature that features cross winds, gusts and lulls, which are unpredictable. The actual CEP of a bomb dropped from 30,000 feet turns out to be around 1,200 yards – well outside the destruct radius of a 500-pound bomb (60 to 90 feet). Consequently, when a factory needed to be taken out, the only way to guarantee its destruction was to destroy the city block(s) within which the factory resided. Hundreds of bombers would be tasked to destroy a single aircraft factory, for example.

Bombsights were glorified peep sights that put cross-hairs over the target, allowing for aircraft speed, fall-time and cross winds. A B-17 flying at 160 mph at 23,000 feet and dropping a 500-pound bomb, would release a bomb 8,875 ground-feet from the target, and the bomb would fall for 38 seconds. If the speed of the airplane was off by 2 mph and the altitude off by 25 feet, the bomb would be 115 feet off-target. It was a very inexact science.

The advent of Carl Norden’s bombsight, which was actually an analog computer that flew the plane once conditions were entered, reduced the CEP to around 1,200 feet – a factor of three. This was the state of the art in the 1940s. That’s why both sides sent hundreds of bombers, each carrying scores of bombs, on raids to take out industrial sites that were a couple of city blocks in size. In order to achieve a 90% degree of confidence that a facility would be destroyed, one had to saturate the facility with enough 1,200-foot CEP rings that enough bombs would actually land on the facility to destroy it. To this end, a total of 2.7MT of high explosives were dropped during World War II, with 1.5MT dropped by the Allies on Germany alone.

The calculus was simple: if we could develop a weapon with a destruct radius equal to the CEP, the number of bombs required to yield a 90% degree of certainty could be vastly reduced, and, it follows, vastly reducing collateral damage of each raid. Assuming this could be done, deliverability becomes the primary concern. Can this weapon with a destruct radius of at least 1,200 feet be reliably delivered over a hostile target?

clip_image004Toward this end, Germany developed the Aggregal-4 (A-4), which was popularly called the Vergeltungswaffe 2 (“Retaliation” 2), or V2, the world’s first ballistic missile (shown at left). It had a range of ~200 miles, and was deployed against London, Antwerp and Liege, delivering a 2,200-pound Amatol warhead which plunged to the ground at around Mach 2. While 1.1 tons of high explosives, together with the inertia of the Mach 2 impact of a 14-ton machine and the detonation of the remnants and fumes of its liquid oxygen fuel tank, would greatly increase the destruct radius (and deliverability questions were answered by the extreme high velocity of delivery), accuracy was the limiting factor. It was nowhere near as accurate as the 1,200-foot CEP of simple gravity bombs. So still, the explosive yield/CEP dichotomy remained.

Enter the Manhattan Engineering District. In August of 1939, Leó Szilárd, a Hungarian-born physicist, became fascinated by Enrico Fermi’s sustained nuclear chain reaction experiments and calculated that, if released promptly, the energy released in Dr Fermi’s chain reaction would produce a remarkably large explosive yield. He convinced Albert Einstein to co-author a letter to President Roosevelt regarding the importance of working on this idea, as Germany’s Werner Heisenberg, one of the world’s leading nuclear physicists, was almost certainly doing so. The resulting Einstein-Szilárd letter was received and the Manhattan Engineering District was established to conduct the research and development of a weapon exploiting nuclear fission.

An incredible collection of Nobel laureates and specialized technicians were brought to bear on the problems involved, and placed under the operational command of then-Colonel Leslie Groves, the Corps of Engineers officer who oversaw the building of the Pentagon (on time and under budget!). To say that lead scientist Robert Oppenheimer, a horse enthusiast and classical pianist who read the Bagavad Gita in the original Sanskrit, had a remarkable mind is somewhat an understatement. Brilliant, temperamental and a bit of a prima donna, he was largely held in check by Groves, and led the project to fruition.

Two bomb types were designed – a rather simple “gun” type that accelerated a slug of fissile uranium (U235) into a body of fissile uranium, instantly bringing the material into “critical mass” – the concentrated amount of fissile uranium required to sustain a chain reaction of spontaneous fission. Feasibility was demonstrated by a “Dragon’s Tail” experiment whereby a tiny slug of U235 was dropped through a ring of U235, briefly releasing a burst of fission. Dragon’s Tail unfortunately resulted in the lethal radiation poisoning of the technician performing the tests.

The second design, the symmetrical implosion of a hollow sphere of U235 onto a solid core of plutonium (Pu239) promised a greater explosive yield, but was far more complex and required a test to make sure the theory would work. The test, codenamed “Trinity,” was conducted at a bombing range next to Alamogordo Army Base in New Mexico. The device, called “The Gadget” by Oppenheimer, was successfully detonated at 0530 on 16 August 1945, and released an explosive yield equivalent to ~20,000 tons of TNT (~20KT), resulting in a Trinitite crater in the desert 250 feet wide. The shock wave was felt over 100 miles away, and the mushroom cloud reached 7½ miles high. The blast was heard as far away as El Paso [TX].

One deliverable bomb was made using each method, and shipped (separately from their explosive cores) to the 509th Bomb Group, TDY on Tinian Island in the Marianas. Lone RB-29s from the 509th had been conducting pre-strike weather reconnaissance over Japan for months by this time, so the sight of a lone B-29 in Japanese skies was nothing new. At 0815 on the morning of 6 August 1945, the B-29 Enola Gay delivered Little Boy in a 1,750-foot airburst over Hiroshima, totally destroying 4.7 square miles (mi2) of the city and promptly killing 70,000 to 80,000 people. The explosive yield is estimated at 13KT.

Not having heard offers of unconditional surrender after three days, on the morning of 9 August, the B-29 Bockscar delivered the plutonium-fueled Fat Man in a 1,700-foot airburst of ~21KT, destroying 44% of the city of Nagasaki and promptly killing 35,000 people. Japan capitulated.

Two platforms, with seven crew members each, had accomplished in one day each, what 1,251 aircraft and 11,768 men did in two days of continual day and night raids over Dresden [Germany]. We finally had a weapon whose explosive yield was greater than its operational CEP, and warfare had irrevocably changed.

At this point we had a genie and bottle problem. We searched for the holy grail of aerial bombardment – one bomb per target – and found something far more powerful than strictly needed. The problems of bombing saturation were two: the explosive yield of deliverable ordinance; and the accuracy with which it can be operationally delivered. Fission weapons addressed only one of those criteria. But you can’t uninvent something, and the world’s powers were now unambiguously divided into two classes – nuclear powers and everybody else.

When confronted with a clean-paper technology – something totally new – nations tend to feel threatened, and will immediately take a two-track plan of action: countermeasures and duplicating the ability. Countermeasures are generally faster and cheaper to achieve than duplicative efforts, although both are ultimately needed in these situations.

It is during this period fighter aircraft enjoyed exponential development, going from the 480mph P-51D and turbojet powered 580mph Me-262 to the supersonic Century Series of American and MiG-17/MiG-21 series of Soviet fighters. It was now very risky to put a slow, lumbering bomber over hostile territory. Bomber development continued as well, but air superiority is in the realm of fighters, and bombers remained extremely vulnerable, especially with the advent of surface-air-missiles (SAMs). This race continued through to the American shockwave-riding Mach 2++ XB-70 Valkyrie strategic bomber and the Soviet triple-sonic MiG-25 Foxbat interceptor.

This tight spiral of airframe development was dampened by the Soviet R-7 Semyorka lift vehicle that placed Sputnik I in low Earth orbit (LEO). They, and we, supposed that a missile that could reach LEO could deliver a warhead over another continent. The intercontinental ballistic missile (ICBM) was born. And at this stage, we were back to the yield/CEP dichotomy – the size of the warhead is now limited by the throw-weight (payload weight) a given missile can carry to intercontinental distances. This is somewhat an inherent problem with nuclear weapons as the necessary shielding, guidance packages and uranium and plutonium are extremely heavy in and of themselves. And the guidance systems for ICBMs at this early stage were very rudimentary. They could hit the city they wanted, but finer-grained accuracy at those distances was unavailable.

So the ICBM race developed into a three-way contest of miniaturizing a given-yield nuclear warhead, increasing the lift capacity of missiles, and increasing the resolution of navigation and targeting systems.

It is during this phase that the collateral damage per warhead, again, began to fall as deliverable accuracy was dramatically increased through the advent of better electronics and the usual learning curve associated with the testing and use of complex systems. When GPS satellites came on-line, deliverable accuracy took an immediate quantum jump. Every significant increase in accuracy led to a noticeable reduction in explosive yield required for a given target. So, as the degree of destruction rendered to a target increased, the ancillary blast damage inflicted outside the target area decreased. But never to the degree that combat commanders favored the tactical use of nukes. The problem with them is radiation.

The use of tactical nuclear weapons renders an arbitrarily shaped area of arbitrary size uninhabitable for an indeterminate amount of time. This complicates, rather than enhances, military operations. Commanders have never looked favorably upon nuclear weapons apart from their use to break the will of a belligerent population.

What this miniaturization of warheads and increased deliverable accuracy did spawn, however, were two new developments: the ability to mount multiple, independently targeted warheads on one missile (MIRV); and targeting went from cities to sites (from counter-population to counter-force). Warhead yields dropped from the double-digit megaton range to the triple-digit kiloton range – a reduction in explosive yield of a hundred-fold. With today’s military GPS, deliverable 3-axis CEP is now less than ten feet.

This phenomenon of accuracy has filtered down to conventional weapons as well. This was widely observed on CNN during Desert Storm when an F-117 pilot put a laser-homing bomb down the air vent of a bunker. One shot, one kill. Add infrared (IR) and GPS guidance packages, and air-dropped gravity bombs are now extremely accurate. On-board computers can do the same thing with sensor-fed gun targeting – helicopter gunships, strafing fixed-wing aircraft, and attack platforms (e.g., AC-130U Spooky/Specter) can lock guns onto a designated target, and the computer will engage it with the selected weapon system. This technology has transferred over to unmanned platforms.

Needless to say, a spinoff of giving the shooter this kind of increased accuracy is a commensurate lowering of “collateral damage.” Should non-combatants be in the target zone, they will be killed along with combatants, but the ancillary killing of people outside the immediate target zone has almost been eliminated. This is a good thing. The asymmetrical response to this can be seen in watching Hamas dealing with Israel’s high tech weapons: Hamas places rocket launchers and gun emplacements next to hospitals and schools, holds high-level meetings in the homes of civilians (against their will at times), conducts operations from neighborhoods, and so forth. They are, in other words, trying to maximize civilian casualties, which can then be used for propaganda purposes.

So one of the questions that fascinates me is why does the world ignore the brutality of terrorists while condemning the honest efforts of American and Israeli forces to protect non-combatants in war zones? Is it a part of the human condition to pull for the underdog? Even if the underdog decapitates prisoners and endeavors to maximize civilian casualties?

The elimination of medieval warfare has not eliminated warfare as an extension of foreign policy, only altered the way in which we practice it. The first alternative that presented itself, the proxy war, evaporated along with the second superpower. The next alternative has been a dusting off of the colonial wars – asymmetrical wars of insurgency and, of late, the engaging of superior power by non-state actors of sophisticated knowledge and assets.

This is where we are today, and at this point, the terrorists are winning.

And that’s if nothing unexpected happened in the air column.

Fused, radioactive glass.

Remember, by summer 1945, the Japanese air force had been largely destroyed, so interception of a lone aircraft was hardly likely.

The “gun”-type uranium bomb.

Japanese High Command, being skeptical of apocryphal reports of a single bomb from a single plane destroying the city of Hiroshima, sent a team of investigators to assess the situation. They had just returned to Tokyo when reports came in that Nagasaki had suddenly gone incommunicado.

Pick a point in the air directly over the target point on the ground, and the warhead will be placed within a ten-foot horizontal circle around that point; choose an altitude to detonate, and detonation will occur within a ten-foot vertical circle around that altitude.

Transformative Warfare


NOTE: This is an introduction to a larger piece that I am writing, but thought it might interest some readers.

One of the buzzwords in today’s coverage of warfare is “collateral damage,” which actually means ancillary non-military damage inflicted during a legitimate strike. It has become a term bandied about by laymen (including the chattering and scribbling classes) to stand for “innocent civilians” killed or wounded during anti-terrorist strikes. And while the inevitable unintended human toll of combat operations is the most visible and visceral example of collateral damage, it is not the only one concerning military planners, and its misuse is skewing the public perception of modern warfare.

It may be interesting to note that we had gotten logarithmically better at waging “clean” war during the second half of the 20th century. World War II killed millions of civilians fighting what may have been the last of the Medieval wars – total war – destroying the civilization of one’s enemy. At the close of hostilities in 1945, Europe lay in ashes from the Urals to the Atlantic and the Japans were a de facto American colony. It was pure Clausewitzian warfare taken to the extreme – achieving the desired end-state by removing the political will of the enemy to continue fighting.

The Cold War threatened total war on a hitherto unimaginable scale – the two superpowers stood poised to hurl tiny suns at each other’s cities. Thermonuclear weapons of 100 megaton (MT) yields were designed, and of up to 50MT were tested. This represented the upper limit of nuclear weapons design as they had passed the point of diminishing returns, in terms of both deliverability (Tsar Bomba weighed north of 28 tons) and military utility (vastly more energy was being wasted upward than yielding downward destruction). The size-race was necessitated because, at the time, they were deliverable only by strategic bombers, which were slow enough to be vulnerable to interception. Therefore, the more punch each plane carried, the better the mission results. Also, inherent inaccuracy dictated large yields to guarantee target destruction. Circular error probability (CEP) of parachute-retarded air-dropped weapons was ~5-10 kilometers, meaning that a larger weapon was more likely to destroy the target it was aimed at. The advent of ICBMs and internal navigation systems (INS) brought CEP down to 500 meters or so, obviating the need for huge yields. Technology had reduced potential collateral damage in a nuclear strike by an order of magnitude.

Continually increasing accuracy allowed for smaller and smaller warhead yields, and targeting philosophy evolved from counter-population (aiming at cities) to counter-force (aiming at missile silos and strategic infrastructure). Improved design and manufacturing capabilities allowed us to sustain target coverage with smaller and smaller warhead yields. But this also reduced throw-weight, allowing for the MIRV’ing of lift vehicles, meaning that each ICBM could carry multiple warheads, each separately targeted. As each warhead linearly reduced the magnitude of collateral damage, the overall legitimate lethality of each missile was geometrically increased.

Ballistic missile submarines were developed, which added a mobile launch platform that could move about with relative impunity, vastly complicating the counter-force targeting task of each belligerent. All of these trends fortified the primary deterrence mechanism of the Cold War – mutually assured destruction, MAD, whereby each side was deterred from crossing the nuclear threshold because that would invite their own destruction in return. MAD wasn’t pretty, but it worked.

General war had become unavailable as an option for resolving international disputes, at least among nuclear-armed nations, and through a web of treaties, among most of the world’s nations. This raises an interesting question: wouldn’t the banishing of nuclear weapons mean a return to general war among the major powers? Their presence as a dampening agent against general war is contravened by the potential of the proliferation of nuclear weapons to otherwise non-nuclear states, either through outright sale or the passing of the technology required to build them.

Providing a belligerent state with nuclear arms would be risky as the signature left by a nuclear detonation can be traced to the country that produced the warhead. But passing along the technology and methodology of producing nuclear weapons is another matter.

The blocking of general war as legitimate superpower behavior resulted in a series of proxy wars between the US and the SU – “small wars,” as they became known. The Soviet Union (and later PRC) backed DPRK in the Korean War, while the United Nations (aka, predominately American hardware and personnel) backed ROK. Nicaragua, Guatemala, Viet Nam, all fall into this category. While it wasn’t a full test of each military’s capability against the other’s, it consistently demonstrated the superiority of American platforms and training. In fact, it was the widening of this gap – first stealth and then the threat of SDI – that forced the Soviets to face the fact that they were in a competition they could not win.

The overwhelming superiority of the major powers has led lesser states to adopt asymmetric warfare as a basic tactic when confronting them. Guerrilla tactics, up through full-blown insurgency, can be successfully waged against powers of far greater classical military capacity, exploiting mobility, faster decision cycles, hit-and-run methods, and support of the indigenous population. Guerrilla warfare, in general, and jihadis, in particular, blur the lines of the laws of war. According to the Geneva Conventions, combatants are to be uniformed (if from a sovereign state), there are other treaties that cover mercenaries and children, and spies may be summarily shot. In its commentary on the Fourth Geneva Convention, The International Red Cross stated:

Every person in enemy hands must be either a prisoner of war and, as such, be covered by the Third Convention; or a civilian covered by the Fourth Convention. Furthermore, “There is no intermediate status; nobody in enemy hands can be outside the law,” because in the opinion of the ICRC, “If civilians directly engage in hostilities, they are considered ‘unlawful’ or ‘unprivileged’ combatants or belligerents. They may be prosecuted under the domestic law of the detaining state for such action.

So terrorists are legally considered “unlawful” or “unprivileged” combatants, and may be prosecuted under the aegis of the detaining state. Exactly what we are doing.

In these asymmetrical wars, Intelligence, Surveillance and Reconnaissance (ISR) replaces the cavalry as the primary source of situational awareness for combat commanders. This activity further blurs lines of conventional thought on combat operations, laws of war and rules of engagement. The questions raised by these ambiguities are several: why does the international community simply ignore such clearly illegal acts as Hamas’ unguided rocketing of Israeli villages and farms? Where was the international outrage over al Qaeda’s public decapitation of Wall Street Journal’s Danny Pearl? Why is there public outrage over Guantanamo Bay, arguably the cushiest prison in existence?

While nuclear weapons are off the table, not so with other categories of “weapons of mass destruction” (WMD). Saddam Hussein used sarin on his own people in the Kurdish north and against Iranians in their 10-year war, and Bashar al Assad’s dad, Hafez, used chemical weapons against his fellow Syrians. So chemical weapons do not share the stigma of nukes. Biological weapons, which enjoy the longest history of WMDs, have been on hiatus since the late-18th century. This is probably because biologicals have become so much more profound than previously, and occupy a spot near that of nuclear weapons on the undesirable list. But it must be noted that chemical weapons still have currency with tyrants; all three classes of WMD have a high probability of being used by rabid Islamists.

The militarily superior power in these asymmetrical wars will inevitably apply technology to the problems caused by guerrillas and terrorists, and that we have done. Weapons development has accelerated (assault weapons, vehicle design and the entirely new field of anti- and counter-IED, and so forth), and the most influential of systems, the armed Unmanned Aerial Vehicles (UAVs). UAVs are turning out to be a bridge technology between increased ISR capabilities for the shooter and fully autonomous warfighting.

Unmanned Ground Vehicles (UGVs) and Unmanned Underwater Vehicles (UUVs) are already deployed, but like UAVs, they are in their operational infancy and not yet able to act autonomously. But they will be. This is inevitable, and fully autonomous actors with lethal capabilities will bring about a host of other questions. How, exactly, will an algorithm determine the difference between a combatant and a non-combatant? There are questions about proportionality of response and lethal or non-lethal response that defy current abilities of AI.

Autonomous warfare is the stuff of the Terminator films right now, but the ability to wage it is coming, and we had better address these, and other, questions before it gets here, not after.

Tsar Bomba, a 50MT design, was dropped by the Soviet Union over Novaya Zemlya archipelago [Arctic Circle] and detonated in a 2.6-mile ASL air-burst at 1132 on 30 October 1962. It released energy equivalent to a Richter 8.1 quake.

Pakistan’s “father of the bomb,” one Dr AQ Kahn, delivered “kits” of blueprints and instructions for a nuclear start-up to DPRK, Libya and Iran, among others. DPRK has the capability now, we talked Colonel Qaddafi out of his fledgling nuclear program and materials in 2003, and Iran is alarmingly along the road to being able to follow DPRK’s lead.

Third Geneva Convention, Article 4. See F-Drive/National Security/Treaties/Geneva Conventions/Third Geneva Convention.

Europeans are known to have catapulted smallpox-ridden blankets into AmerIndian settlements during the early history of the colonization of North America.