Centurian52

Exploitation after a breakthrough in a major operation would be its own separate scenario in Combat Mission. But an infantry company pulling back from defending a village would be under pressure from the moment they start pulling out to the few hundred meters they may need to pull back in order to break contact. That is well within the scope of a single scenario.

That's unavoidable. You will get complaints any time you implement a mechanic that makes a wargame more realistic. Combat Mission has by far the most realistic spotting mechanics of any wargame out there, and it's no coincidence that there are no shortage of complaints about the game's frustrating spotting mechanics. It turns out that reality is somewhat different from people's expectations.

I figure the "surrender" option equates to, well, surrender. As in the situation has gotten so bad that it's not just individual soldiers throwing up their hands, but the overall commander deciding to throw in the towel and march his remaining troops into captivity. You'll notice that when the enemy surrenders you aren't just automatically granted credit for all ground objectives, but also parameter objectives relating to enemy casualties, implying that the enemy force is completely destroyed. But there's also more to retreating than just giving up the battle. Retreat normally involves some attempt at pursuit on the part of the winner (if no pursuit is attempted, it's often a sign that the winner is utterly exhausted), and efforts to fend off pursuit and break contact on the part of the loser. Basically the winner is trying to convert a victory into a decisive victory, while the loser is trying prevent a defeat from becoming a decisive defeat. The fight isn't really over until after you have successfully broken contact with the enemy. It's an entire additional phase of the battle that we're largely missing out on.

You'd need to do more than just incentivize the player to take/hold their objectives with minimal losses (that's pretty much how it works right now). You'd need to find a way to incentivize the player to eventually choose force preservation over taking/holding their objectives. For one, it would need to be possible for the defender to retreat. That means exit zones would need to be present in every scenario. An attack against a defender that fights to the last man is pretty much guaranteed to suffer unusually high losses. Beyond making retreat possible, I don't know how you incentivize the player to actually do it. You might be able to program the AI to call off an attack or abandon a defense under the right circumstances. But how do you incentivize a player to willingly abandon their objectives?

Fair enough. I oversimplified things a bit by assuming a unit fought an engagement every single day and received no reinforcements while it was in the line (reinforcements do arrive a bit less reliably for both sides in Ukraine than for the Allies in WW2 though, so assuming no reinforcements was not entirely unjustified). I recall an anecdote of a US infantry division (I don't remember which one) taking 200% casualties over the course of the Battle for Normandy. Assuming they were in the front line from June 6th to August 30th, and were topped off to 100% every day, that works out to an average of about 2.3% casualties per day (3.5% if they took 300% losses over the course of the campaign). Obviously they wouldn't have suffered 2.3% casualties every day, but much fewer casualties on most days and much higher casualties on a few days. That may have been higher than normal, since I recall some WW2 general (again, I don't remember which one) estimating that a typical infantry division will lose 1% of its strength for every day it's in the line (I think I ran across this while looking up casualty estimation methods for large operations). Of course most of the individual companies and battalions in the division won't be fighting every single day. If we assume a two up one back formation, nested all the way down (two of the division's three regiments forward, two of each regiment's three battalions forward, and two of each battalion's three companies forward (I'm assuming that each formation is triangular, but I haven't double checked that for WW2 US infantry divisions before typing this)), then only 8 of the division's 27 companies would be forward on a given day of fighting. In that case a day in which the division as whole suffered 1% casualties would see each of the frontline companies suffering an average of 3.4% casualties. A day in which the division as a whole suffered 2.3% casualties would see the frontline companies suffering an average of 7.76% casualties. A while back I read about a particularly bitter struggle a British infantry battalion had over a hill in Normandy. They took the hill, but suffered 15% casualties in the process. I should emphasize again, there is no typical casualty rate in the real world. Casualties span the whole range from 0% to 100%, with lots of engagements clustering between 0% and 10% and fewer engagements with losses towards the higher end of the spectrum. A typical Combat Mission scenario in which the winner suffers 25% casualties and the loser suffers 75% casualties is not unrealistic if taken in isolation. What's unrealistic is that nearly every scenario ends with losses in the tens of percentage points (in the titles covering peer vs peer and near-peer warfare). I don't think the issue has anything to do with player tactics or weapons being modeled as over-lethal. I think it just comes down to the fact that each side (human player or AI) is determined to fight the battle to the bitter end. We push our attacks far past the point when a real attacker would stop. We defend to the very last because (unless the scenario designer has thought to put in an exit zone) retreat is literally not an option. It really just comes down to the fact that we are taking each scenario in isolation. The larger operational/strategic/political context only exists in the narrative presented by the briefing, so it can't create any real incentive to keep our casualties down or preserve combat effective formations for the long term. Even in campaigns, when we do have an incentive to preserve our forces over several joined scenarios, we are still playing to win against a defender that will not retreat, or an attacker that will push themselves to complete destruction. This is why this is a wargaming problem, not a Combat Mission problem. It's a problem for any tactical level wargame precisely because their scope is limited to the tactical level. But that's also why there isn't really an easy fix. Battlefront was able to deliver such a highly detailed and realistic tactical wargame in large part because they limited their scope to the tactical level. If the reason casualties are too high is that there are no operational/strategic incentives for keeping casualties low then the obvious solution would seem to be to expand the scope to include the operational level. But that would almost inevitably mean a less detailed representation of the tactical level. Effectively you'd be trying to make the game more realistic by making the game less realistic. If there was an easy solution, I imagine the problem wouldn't be so pervasive across all wargames.

For how to play the game, I believe @Hapless has a playlist on youtube that should have you covered: For WW2 doctrine you should be able to find anything you need at the Nafziger collection: https://nafzigercollection.com/product-category/tactical-studies-for-all-ages/ For Cold War and modern US tactics I think FM71-1 and FM71-2 are a good place to start: https://www.bits.de/NRANEU/others/amd-us-archive/FM71-1(98).pdf https://www.bits.de/NRANEU/others/amd-us-archive/FM71-2(77).pdf For Cold War Soviet tactics I'd recommend FM100-2-1 (and this has served me well for any faction operating Cold War Soviet equipment, including modern Ukraine and Russia): https://irp.fas.org/doddir/army/fm100-2-1.pdf

I feel like I'm a bit of both. I'm definitely trying to win by as wide a margin as possible, with as few casualties as possible. I also know that part of what makes a good commander in the real world is the ability to recover from mistakes and random disasters without either giving up or letting anger lead to sloppiness and more mistakes. Which is why I make a point of never reloading saves after a blunder these days (in my very early CM days, around when I first picked up the original CMSF in 2009, I reloaded saves all the time). So I definitely think of myself as a fairly competitive CM player. But the whole reason I save every turn is because I think of myself as basically recording my own war movies, which are far more realistic than anything Hollywood will ever give us, as I play. And while I never go back and play from old saves, I do go back and watch old saves all the time. The 'casualty rate too high' complaints are still alive and well. If they aren't being raised as much it's probably because there's a realization that there isn't much Battlefront can do about it in the current engine. There are individual anecdotes of entire Russian units getting wiped out in Ukraine, just as there are individual anecdotes of individual units getting wiped out in any war. But my overall impression from observing the war is that overall casualty rates, as a percentage of the unit involved in a particular engagement, are probably about on par with what you'd see in WW2. Which is to say, there is no 'typical' casualty rate. Just like in any war there is a wide spread of casualty rates across multiple engagements, with lots of engagements with fewer casualties and fewer engagements with lots of casualties. But, just like WW2, I suspect losing 10% of a unit in a single action would be considered a pretty bad day. And while total annihilations do happen, they aren't common. If you think about it, units in Ukraine spend months on the front line. If a unit lost 10% of its strength every day, it would be down to less than 50% strength after only a week, and would have lost nearly 90% of its original strength after only 3 weeks. After three months a brigade of 5,000 men would have no one left to rotate out. It's just mathematically impossible for a unit to go on fighting for months if it takes Combat Mission level casualties in every engagement. Though, the caveat that needs to be made every time this complaint is brought up is that this is really a wargaming problem, not specifically a Combat Mission problem. No tactical wargame seems to have ever managed to get this right.

Looking forward to it! For everyone else, I really think it's best if we interpret the 2024 look at the year to come as "this is what they're working on this year", rather than as "this is what will be released this year". As was pointed out a couple pages back, Battlefront has committed to quality, and that means time needs to be flexible Companies that commit to releasing their products by a promised date often end up being rather flexible on quality. In any case, I suspect it takes two or three years to develop a base game or module (I haven't payed much attention to how long it normally takes to develop Battle Packs). So it's probably reasonable for us to hope for some of the things that were announced at the beginning of 2023 (with work presumably starting sometime in 2022) to be finished in 2024 or 2025. If they announce that they're working on anything new in 2024, that doesn't mean it will be released in 2024. The better interpretation is probably that work started in 2023, and a release might be expected in 2025 or 2026. A two or three year development cycle actually isn't that long for something as complicated as a video game.

Get ready. CMBS models much more near-peer warfare than CMSF2. The jump from asymmetric warfare to near-peer warfare can be pretty shocking if you aren't braced for it. You'll find that the Russians are still a bit behind the US in most areas, but not by anywhere near as much as the Syrians. You really will need to be prepared to accept heavy casualties from time to time. If you get chewed up in your first few scenarios, don't worry. Apparently we've been focusing on COIN operations for so long that even the professionals are getting chewed up for the first few days of exercises in which the OPFOR is given near-peer capabilities.

Just chiming in to say that I've finally caught up with ISW's December 27th report (I've really fallen behind on these). That's significant because it's exactly two months after I predicted, on October 27th, that it would take the IDF 1-2 months to secure Gaza city. And I can see that ISW has not yet colored in all of Gaza city in blue on their map. So, in the spirit of maintaining accountability, my prediction is now officially wrong.

I never replay turns that don't go well for me. My feeling is that if I make a mistake in real life I'd have no choice but to live with it, so I force myself to live with any mistakes I make in-game (the point is to recreate real war after all, at least as much as the technology allows). I do save every turn, but that's mostly because I like to go back and re-watch old battles from time to time, and because I like to imagine that one day I'll go back and make Hapless-style videos of them (not that it's terribly likely that I'll ever get around to it). When I do make a costly mistake I console myself with the knowledge that even the professionals blunder from time to time. I do my best to recover from the mistake, make the most of the troops and resources I have left after the mistake, and try to understand why that particular set of actions was a mistake under those particular circumstances so I'll do better in the future. Casualties are inevitable in war. I'm playing through the WW2 titles right now. That was a high intensity peer vs peer war, so the casualties that I'm experiencing in the scenarios I'm playing these days are higher than what could reasonably be expected in CMSF2. CMSF2 is still high intensity, but much more asymmetric. So the NATO player can usually expect to take fewer casualties. But even in CMSF it's rare that I take no casualties. With some skill you may be blessed with the occasional zero-casualty mission, but that will not be the norm. I'd advise aiming for lopsided casualties, rather than zero casualties (obviously zero casualties is best if you can manage it, but it's usually not realistic). In CMFI I'm happy if I can get 2:1 (maybe 4:1 if I'm fighting the Italians), but in CMSF2 you should probably be getting at least 10:1. If 10:1 gets to be too easy for you, then try shooting for 20:1 or higher. Aim for a loss exchange rate that's high enough to be challenging, but low enough to be attainable. Keeping in mind that some scenarios will be more challenging than others.

I'm catching up on a backlog of ISW reports. It's been a while since I've seen a credible casualty estimate for IDF forces. But the December 26 report contained the following paragraph: While the IDF may not be particularly transparent about their targeting practices, I doubt they're lying about their losses. If they have suffered 161 KIA (as of December 26th), then by the rule of thumb that there are always at least 2 wounded for every 1 killed that means they've likely suffered at least 480 casualties. Of course that 2:1 rule of thumb may be outdated, as medical technology has advanced over the last century. In Ukraine there seem to be at least 3 wounded for every 1 killed. And the IDF is an advanced military that places a high emphasis on the safety of its troops, and which is fighting a smaller and lower intensity war. So the number of WIA per KIA may be much higher. Still, I'd guess their casualties probably don't exceed 1,000. That would mean they're probably suffering 8-16 casualties a day, of which about 2.6 per day are KIA. Less than I would have imagined before the ground offensive into Gaza began.

The breaks to swap the magazine aren't really that significant. Remember that, thanks to the top loaded magazine, the assistant gunner can swap the magazines very quickly (a feature not modeled in Combat Mission unfortunately). But the ease and speed with which the barrel could be changed out means that overheating still wouldn't have been much of a concern for the Bren. The MG42 is also going to have frequent enforced breaks, since it's going to run down its 50 round drum faster than the Bren will run down its 30 round magazine (and it takes longer to change out a belt than it takes to swap a magazine). Remember that when we're comparing it to the Bren we are talking about the MG42 in its LMG configuration, in which case its belt is feeding from a 50 round drum, not a 250 round box. In its HMG configuration it would make more sense to compare it to the Vickers or M1919, not the Bren. Even so, I believe you're right that overheating is a greater concern for the MG42, even in its LMG configuration. Barrel changes are quick and easy, just like the Bren. But I recall the recommendation was for a barrel change every 150 rounds, meaning that when it is firing the full 200-250 rounds per minute practical rate of fire (claimed) you are changing out the barrels more than once per minute. Which is why I believe German squads had to carry multiple spare barrels for the MG42, while British sections only carried one spare barrel for the Bren.

If people who have actually fired the MG42 or MG3 didn't notice any perceptible inaccuracy then I'm guessing that the figure of 5 MOA is probably correct. I recall from firing the M249 that 12 MOA (assuming that I am remembering the correct figure) is perceptible (though, for all the complaints about M249 inaccuracy, it didn't strike me as all that bad for a weapon meant to suppress an area target). So the MG42's accuracy must be better than 12 MOA. And I recall from firing the M16A2 that 4 MOA (at least that's what I recall being told the M16's accuracy was) is not perceptible (it always looked pin-point accurate to me). If 4 MOA isn't perceptible, then 5 MOA probably isn't either. Interestingly, that would mean that the Bren and the MG42 have roughly the same accuracy. With a figure of 4-5 MOA for the Bren based on British firing tables, and 5 MOA for the MG3 (basically an MG42) based on the word of some guy on a forum that I haven't managed to dig up again. Would be nice to find some harder evidence for the MG42. But this might do for now if no one can do better. That might explain where the myth of inaccuracy came from. Regardless of the mechanical accuracy of the weapon, if Allied soldiers notice that longer bursts frequently scatter bullets all over the place then it isn't hard to imagine that a lot of them might start to think that the MG42 wasn't very accurate.

The Bren-sniper scene from the movie about the Siege of Jadotville was the very scene addressed, and debunked, by Ian McCollum Nicholas Moran, a.k.a. The Chieftain, did a response in which he claimed that the scene itself actually might have been perfectly plausible. But his point boiled down to the fact that soldiers believe myths about their weapons all the time, so an Irish soldier might very well have believed that a Bren was a more precise weapon, and didn't at all contradict Ian McCollum's point that the Bren was not actually a more appropriate precision weapon than an actual dedicated sniper rifle. Edit: Ok, having now re-watched Ian McCollum's video I realize that he actually answered the Bren part of my question in the video. Based on British firing tables of a Bren firing 10 round bursts from a tripod out to 1200 yards it looks like its accuracy was roughly between 4 and 5 MOA. So it's pretty accurate, but not a precision weapon. The MG42 part of my question still stands.

I suppose this question could have gone into any of the WW2 western front forums, but CMBN appears to be the most active. Two of the WW2 weapons myths that I've become aware of are that the Bren was extremely accurate, to the point that it could be used as a precision rifle, and that the MG42 was very inaccurate, to the point of basically being a scatter gun. Ian McCollum of Forgotten Weapons has already done a video on the first myth. I don't know outright that the second one is a myth, but it smells fishy. The idea that the MG42 was a scattergun certainly doesn't seem very consistent with how it's modeled in Combat Mission, or any other reasonably realistic game I've encountered it in. I assume that there is probably a grain of truth to these myths. It does not strike me as remotely unlikely that the Bren really was more accurate than the MG42. But by how much? Can we actually put some numbers to the issue? The traditional units for firearm accuracy are Minutes Of Angle (MOA). 1 MOA is one 60th of a degree. These are exactly the same units as arcminutes, but for whatever reason we say MOA when we're talking about firearms, and arcminutes for every other application. After much googling I was able to find a figure of 5 MOA for the MG3, which is very nearly an MG42. 5 MOA is certainly less accurate than I would expect of a bolt action rifle of the time firing a full power cartridge, but it doesn't strike me as at all bad for a machinegun. But is the figure of 5 MOA correct for the MG42? One of the improvements of the MG3 over the MG42 might have been the accuracy, or the poster who gave that figure might have been mistaken. At the extreme end, I can't imagine that the MG42 has any worse accuracy than the M249, which I seem to recall manages about 12 MOA (I don't remember where that figure came from though). Just that little bit of scatter might even be beneficial for a weapon with such a high rate of fire. At the extreme end for the Bren, I'm sure its accuracy can't be better than 2 MOA. I've seen good marksmen achieve around 2 MOA with the SMLE and Kar98k. And if Gun Jesus says there are a host of reasons that it couldn't possibly be more accurate than an SMLE, then I'm inclined to believe him. So, can anyone out there actually put a figure to these? In MOA, just how accurate was the Bren? And just how accurate was the MG42?

These look really good. You might even being giving @IanL a run for his money

I appreciate the thought, but I'll always take realism over balance. I'm perfectly happy to take a punishment in the name of historical accuracy. So please don't dumb it down too much on our account.

I don't think warfare has a cyclic nature. Humans are infamous for seeing patterns where there are none (which is exactly why we can have so much fun staring at clouds). A lot of historians have written at length about patterns they thought they've seen in warfare, and even proposed causes for those patterns. But those patterns are likely no more real than the patterns we see in clouds. What I've heard from competent statisticians is that when wars break out actually appears to be completely random. The pattern is best explained, not by invoking cycles, but by describing a more or less constant probability that two belligerents will go to war each year (one of my favorite descriptions of this is "every moment Mars rolls the iron dice, and if they come up snake eyes he sends a pair of nations to war"). Of course individual wars have very real, non-random causes. But the scattering of wars across the centuries is completely random. There are clusters, with several wars starting in close proximity to each other, followed by calm periods with relatively few wars. But this is the normal clustering that you expect to see in any truly random process (data points are only evenly spaced apart when they are driven by a non-random process, random processes always have clusters). While it should be intuitive enough that it takes a non-random process in order to space events evenly apart, people are still often surprised by the clustering that comes with true randomness. An experiment that has been run many times in classrooms is to take two groups of students, tell one to actually flip a coin 100 times in a row and write down the results, and tell the other group to write down the results of an entirely imaginary series of 100 coin flips. Ideally this is set up in such a way that the teacher has no idea which group is which. The challenge is then for the teacher (or someone who doesn't know which group is which) to guess which group actually flipped the coin. It's very easy to tell which group actually flipped the coin when you know what to look for. The group which didn't actually flip the coin will almost never write a cluster of three or more heads or tails in a row. The group that really flipped the coin has an 80% chance of having a run of six or more heads/tails in a row (the probability of a run of seven or more is 54%). This tendency to underestimate the clustering involved in random processes is apparently the psychology behind the Gambler's fallacy. This is the fallacy someone commits when they exclaim "we've had a run of five heads, surely we're due for a tails" or, perhaps more relevant "we've had a long run of peace, so we must be due for a major war" (it's called the "Gambler's fallacy" because gamblers routinely commit it when they think they're due for some good luck after a run of bad luck). The reality is that no matter how many heads have come up in a row so far, there is still a 50/50 chance of the next flip coming up heads again. WW1 and WW2, the two bloodiest wars in human history, coming within 20 years of each other, are a good example of the clustering that comes with randomness. We weren't at the inevitable point in a cycle where it was time to have humanity's two bloodiest wars. We just had an abysmally unlucky run of die rolls in the first half of the 20th century. There are no cycles. So, will there be another major war between 2025 and 2050? Possibly. Tensions between China and the US/Taiwan seem to be pretty high right now. Those tensions may ease off without a shot being fired, as happened with NATO and the Warsaw Pact at the end of the Cold War. Or they may ignite into another major war, as happened between France/Russia and Germany/Austria-Hungary at the beginning of the 20th century. If we do get another major war, it won't be because we were due for one in the cycle, but because we had a very unlucky role of the dice.

There is a fallacy that I think it would be helpful for people to understand for this discussion. Unfortunately logicians (really all academics) prior to the 21st century had an unhealthy obsession with Latin, so it goes by the difficult to pronounce (and harder to remember) name "Tu Quoque", which apparently translates as "you also". This fallacy is committed anytime someone asserts "they broke the rules, therefore I am allowed to break the rules". If you want a more memorable name, I believe people started calling this "Whataboutism" sometime within the last decade or so. https://www.palomar.edu/users/bthompson/Tu Quoque.html The rules being broken does not invalidate the rules. Because Hamas broke the rules does not give the IDF, or anyone else, license to break the rules. If the question is "is the IDF following the rules", it is entirely irrelevant to reply with a list of all the rules Hamas has broken. We all know that Hamas has broken the rules. But how is that relevant to whether or not the IDF has broken the rules? There is no grading on a curve. It is not necessary, or even relevant, to ask which side was worse when considering whether or not one side conducted their actions legally.

We just went over how there very clearly is a distinction between guerrillas and terrorists. I did my best to lay out what guerrilla warfare is (and a guerilla fighter is just someone who engages in guerilla warfare). And @The_Capt did an excellent job of laying out what terrorism is (and a terrorist is just someone who has committed an act of terrorism). Guerrilla warfare and terrorism (and, by extension, guerrillas and terrorists) are distinct things. There is certainly plenty of room for overlap. A single action can very easily be both an act of guerrilla warfare AND an act of terrorism (it is very common for things to be more than one thing at a time (a reservist who's day job is to be an accountant would be both a soldier and an accountant)), but the features that qualify it as guerrilla warfare (it is an act of resistance, which does not attempt to control ground, against an occupying military force in the context of a war) are distinct from the features that qualify it as terrorism (it is an act of illegal violence meant to intimidate/instill terror in service of an ideological cause (political, religious, etc)). There is no need for an act to be either terrorism or guerrilla warfare and not both (it does not need to be an XOR, if you are familiar with Boolean logic or programming). To further illustrate how these are different, but can overlap, I'll try to offer a few examples. 1. An ambush on military forces, in territory that is controlled by those forces, in which the attackers attempt to inflict a few casualties and then withdraw without trying to take or hold ground would be an act of guerrilla warfare, but not an act of terrorism. 2. An attack on a populated area which is not in a war zone, in which the attackers are motivated by an ideological cause and their goal is to intimidate the population of the country in which the attack took place, would be an act of terrorism, but not an act of guerrilla warfare. 3. Extrajudicial executions of suspected collaborators in an area that is under military occupation and is an active war zone, carried out by people with an ideological commitment to fighting the occupation forces, for the purpose of terrorizing/intimidating other potential collaborators, would be both an act of guerrilla warfare AND an act of terrorism. Someone who did 1, but not 2 or 3, would be a guerrilla, but not a terrorist. Someone who did 2, but not 1 or 3, would be a terrorist, but not a guerrilla. Someone who did 3, or who did both 1 and 2, would be both a guerrilla and a terrorist. As for terrorist organizations, I don't think that it's necessary to prove that every last member of the organization is a terrorist. Only that one of the functions of the organization is to engage in terrorism.

We were definitely overoptimistic to hope for a release in 2023 given that work apparently hadn't started until late in 2022. But a release in 2024 is entirely within reason. Around 2 years from first announcement to release is not uncommon for Combat Mission base games and modules. If anything, rather than getting worked up, I was hoping to temper expectations a bit for anyone who might still be hoping for it to come out any month now (Millien's comment reminded me that there might still be a few such people). But I'll grant that I'm only making an estimate based on past precedent, and there is a fair amount of variability here. While 2 years seems to be pretty common, it wouldn't be surprising if it ended up taking them 3 years. Which would bring the release out to the third or fourth quarter of 2025. I would be surprised if it took 4 years though.

Ok, so looking back through old announcements it looks like we first got word that work on a module for CMCW had just started around early September 2022 (that was way before it was announced what the new module would cover). So just going under the assumption that it takes about 2 years to develop a module then I'd guess that we can realistically hope to get our hands on it sometime around the third or fourth quarter of 2024. If I'm right that we can hope for screenshots around 3 to 6 months before the module is released then perhaps we'll start seeing screenshots sometime between March and June of 2024.

It seems like a stretch to count targeted political assassinations as terrorism. But I think that does bring us back to my earlier concern that the term "terrorism" seems to confuse more than it clarifies these days, so we may actually be better off dropping the term altogether. Pretty much anything that has been called terrorism can be described as something more specific, and equally heinous, anyway. A bombing, a mass shooting, a political assassination, an extrajudicial execution, and so on. Whatever we end up doing with the term, whatever we end up deciding counts as terrorism, and whoever we end up deciding counts as a terrorist, I think I've derailed this thread enough by bringing up the subject in the first place. We should probably try to return to talking about the war.

We generally only admit that states can sponsor terrorism, not that they can commit terrorism. I'd be all for dropping the "goverments can commit terrorism" requirement. If we do drop that requirement, then we should be prepared to reopen uncomfortable questions around, for example, whether the firebombing of Tokyo was a terrorist attack.