This is actually something I have spent quite a bit of time looking at - in the context of financial crime rather than abuse, but the enabler/whistleblower divide is probably similar. And the solution that seems to work best so far is "make managers explicitly accountable and hold them to account". The equivalent solution for abuse would be to tell head teachers that if any of their pupils are abused by anyone at their school, the head teacher will probably lose their CRB certification (i.e. won't be allowed to work with kids) unless they can prove they were not negligent.
Yeah, the solution (or mitigant, more accurately) would need to be organisational rather than (strictly) personal, if you ask me. It's the incentives, stupid.
Heebie's take is excellent, and jay's contribution is solid as well.
I feel like there's a weird gap between mindless "the buck stops here" claims that seem to avoid looking at actual systemic problems and the fact that the people in charge are rarely held responsible. In a very different venue, you're seeing that with the Hawaii alert employee being fired when everybody can plainly see that the interface was basically designed to fail*.
Some of this is just the nature of organizations, where protecting the powerful is a natural tendency. But I feel as if the aforementioned gap has become itself a cause of the problem. That is, there's a bogus narrative that the person in charge should be fired for whatever happens below (tbh, this is mostly a phenomenon of sports fandom, but you see the same rhetoric around mundane items like snow removal: street not plowed? Replace the Mayor!**) creates increased incentives to protect the boss, who may be called upon to quit whether she's done anything amiss or not.
All of which is to say that a healthy dynamic would be to identify which organizational norms are critical, and therefore need to be laid at the boss's feet, and which are less critical, and therefore responsibility devolved.
I guess the pithy way to say it is that, right now, a company president might be fired for missing earnings targets, but not for a workplace incident that kills a dozen people or crashes global markets, and that this is precisely backwards.
*under similar circumstances, I could see draconian punishment for the line employee, but I don't for one second believe that the org in question is being run like, say, a nuclear facility or (in an ideal world) a surgical one, where perfect adherence to well-designed procedure is the norm
**and sometimes that's the right response, but you need to actually prove that, not take it as given
I feel as if the Navy, at least, has this right: captains almost always lose their commands for significant accidents, and sometimes (often?) for significant discipline infractions. But they're not demoted if the decks are insufficiently swabbed or whatever.
The Buck Is Sorted According to Size Like a Coin Sorter and Stops at the Appropriately Sized Slot.
I don't think you should have to pay for it, though.
CoinStar is free, if you get a gift certificate.
4 - there's a whole Tom Ricks book, which I basically agree with while recognizing that the limits of my knowledge are extreme so who cares if I agree or not, with the thesis that the US Army has a very serious problem of not removing bad generals from command, and that an almost hair-trigger willingness to do so is the mark of a well-run army.
OP: I really like Witt and Heebie both looking beyond individuals to their enabling systems. I suspect that you're right; a culture where easy scripts for intervention are provided will lead to more intervention.
For my first real job, I had to get a background check complete with fingerprinting. I wasn't happy because I wasn't actually in contact with vulnerable people, they just set the requirement for the whole office. But now I understand the motivation a little better plus "I can't help with that, my prints are already on file" has been a useful excuse.
Just to be clear, I am not disagreeing with the OP. That all sounds like it makes sense and Witt knows way more about her organisation that my analogy-ban-breaking self does. And pushing for formal whistleblower lines etc has definitely done some good. Rewards, or even promises of protection from retaliation, seem to be less important than simply making it known that you will listen to and act on whistleblower reports. Things like the CAA CHIRP newsletter come up a lot here: people know that CHIRP reports get attention.
captains almost always lose their commands for significant accidents
12: Apparently things are changing. The Navy captains who ran into other ships not only lost their command but are facing negligent homicide charges.
13: This dude says, IIRC convincingly, that the Navy is punishing individuals while totally failing to acknowledge systemic problems (IMO largely originating from civilian leadership, but IDK if the link says that too).
I'm sure there are systemic issues, but running into a container ship seems like a very personal failure.
It would certainly be in my top 10.
Hornblower and the Bucket List
14 sounds not very convincing. I'm sure the Navy, like every other organisation in the world, is not as well funded or staffed as it would like. But what he seems to be arguing is that unless you bring warships up to their full wartime crew levels you can't expect them to do difficult things like not run into container ships. This doesn't bode well for how they would do really difficult things like fighting, when they are undercrewed because sailors have got killed.
And furthermore it's the captain's responsibility, if he thinks his ship is unsafe, not to take it to sea, or at least to protest when ordered to do so.
19.1: I read it a while ago. IIRC he says crews aren't just too small, they're also inadequately trained because the fleet is basically too small both to train and to complete its assigned tasks.
19.2 is of course true, but I'm guessing that would be a career-ending move.
They would have more crew per ship if they didn't have so many ships.
Since quality of command has come up: this guy calculated Wins Above Replacement* for all generals throughout history based on Wikipedia. Napoleon dominates; Lee and Rommel have poor showings. Seemed suitable for this blog.
* in baseball stats, it's really more like a lifetime Win Probability Added score based on adding up each per battle WPA; it doesn't really construct a replacment general and calculate wins over that baseline, just total success over expected success, for a career.
I mean the analysis puts the "data" "science" into data science. But Napoleon really was the best general.
The true artists of war aren't in the dataset.
The comments there are full of nitpicking, of course.
22: Was anyone able to view the ratings for individual commanders? The instructions didn't work for me.
In any event, I would contend that the WAR metric undervalues commanders such as Christian DeWet, Johannes Blaskowitz and Võ Nguyên Giáp, who may have racked up more losses than wins, but nevertheless excelled at frustrating a superior adversary and inflicting casualties on the enemy in defeat -- a valuable quality in the context of a Clausewitzian understanding of warfare.
Grant shows up nicely, which is all I want out of a metric rating generals.
Since quality of command has come up: this guy calculated Wins Above Replacement* for all generals throughout history based on Wikipedia. Napoleon dominates; Lee and Rommel have poor showings. Seemed suitable for this blog.
That's fun. I appreciate that his updates responding to comments are both thoughtful and also make clear that the rankings are not intended too seriously.
In any event, I would contend that the WAR metric undervalues commanders such as Christian DeWet, Johannes Blaskowitz and Võ Nguyên Giáp, who may have racked up more losses than wins, but nevertheless excelled at frustrating a superior adversary and inflicting casualties on the enemy in defeat -- a valuable quality in the context of a Clausewitzian understanding of warfare.
Regarding the Rommel ranking, I know that his reputation* was built on his ability as a skirmisher, so the fact that the rating is based on major battles excludes that skill. The same would be true of commanders fighting against a superior force in general, who would benefit from avoiding set-piece battles.
* which probably was over-hyped, but I have some fondness for him.
In any event, I would contend that the WAR metric undervalues commanders such as Christian DeWet, Johannes Blaskowitz and Võ Nguyên Giáp, who may have racked up more losses than wins, but nevertheless excelled at frustrating a superior adversary and inflicting casualties on the enemy in defeat -- a valuable quality in the context of a Clausewitzian understanding of warfare.
Yes. I think if it measures anything it measures one skill, which is tactical command of battles. Obviously wars are won (or can be won) with much more than that.
It really does make Robert E. Lee look like a shitty general, though, probably the most overrated in history (since his reputation depends so heavily on supposedly being a great tactical battle commander) so it must be right.
I'd also note that some of the best Rommel stories* (which is not the same thing as greatest wins) come from WWI before he would have counted as a general. See, for example, this summary
Taking part in dangerous raids and reconnaissance missions throughout World War I, his men supposedly joked, "Where Rommel is, there is the front." But all of this fighting, including one 52-hour period in which his unit captured some 9,000 Italian prisoners, came with a price. In September 1914, for example, Rommel charged three French soldiers with a bayonet after running out of ammunition, only to be shot in the thigh so badly that a hole opened up as big as his fist.
Wikipedia says
The offensive, known as the Battle of Caporetto, began on 24 October 1917.[24] Rommel's battalion, consisting of three rifle companies and a machine gun unit, was part of an attempt to take enemy positions on three mountains: Kolovrat, Matajur, and Stol.[25] In two and a half days, from 25 to 27 October, Rommel and his 150 men captured 81 guns and 9,000 men (including 150 officers), at the loss of six dead and 30 wounded.
But, he as a Lieutenant so Rommell (correctly), isn't listed as a commander for the battle.
I also found this article ("How Erwin Rommel Earned Germany's Highest Honor, as a Mere Lieutenant") but it was too long and I didn't read all of it.
26.2, 30, etc. are also prefigured by the quote the post leads off with, in which Hannibal ranks Pyrrhus as the #2 general of all time.
In fairness, though, Caporetto was against the Italians and was one of the best prisoner-taking opportunities in history.
I'm not actually competent to make the argument, so this is just a besmirchment, not a well-supported attack, but I've always suspected that Rommel's reputation is largely based on the fact that it was useful to both (a) Hitler and (b) the British army to portray him as a genius. When you've got the propaganda machine working on both the winning and the losing side it's hard to overcome. The same explains Lee's reputation to some extent.
I've always suspected that Rommel's reputation is largely based on the fact that it was useful to both (a) Hitler and (b) the British army to portray him as a genius. When you've got the propaganda machine working on both the winning and the losing side it's hard to overcome.
I think that's true. I am not an expert by any means, I think Rommel was both fearless and with real strengths and savvy as a tactical leader but not overall a genius or somebody that would necessarily be well-suited to making high-level strategic decisions.
I also that he reveals some of the limitations of the ranking.
Whereas, ironically, von Manstein (who was a legit genius, and also very lucky, but very lucky matters for WPA) is not seen in popular memory as the greatest German military genius of WWII only because the fall of France isn't seen (by Anglophones) as one of the greatest, most important, most unlikely, hardest to pull off military successes in history but as the perfidious French fucking up and being cowards.
Regarding the Rommel ranking, I know that his reputation* was built on his ability as a skirmisher
Rommel's reputation is mostly based on the fact that his foe Montgomery was a world class self-promoter, who liked to pretend he had vanquished the best the Germans had to offer, and on the fact that the British press was willing to play along with the pretense, the better to bolster the morale of a population that had endured a long string of defeats. It's ludicrous to think the Oberkommando des Heeres would have deployed its top talent to a sideshow theater like Libya while the fate of the Reich hung in the balance at Stalingrad and the Rzhev salient.
...or what Halford said.
You'll see me largely agreeing in the following comment.
It really does make Robert E. Lee look like a shitty general, though, probably the most overrated in history
Both Lee and Rommel were well-served by lineal descendants who preserved their memory and carefully cultivated a particular historical narrative, one that emphasized both their tactical brilliance and their supposed opposition to the ideology they served. (To be sure, "Rommel opposed Nazism" has a higher truth content than "Lee opposed slavery".)
"Lee opposed slavery" barely has a higher truth content than "Hitler opposed Nazism."
33, 36: As alluded to in the linked post.
In particular, critics have attributed much of his reputation as a tactical genius to both German and Allied propaganda. British generals reportedly exaggerated Rommel's tactical abilities in order to minimize disapproval regarding their defeats.
I oppose your ownership of human beings but I will fight to the death for your right to do it.
42. And if not your death, then the deaths of hundreds of thousands of your countrymen.
"It's ludicrous to think the Oberkommando des Heeres would have deployed its top talent to a sideshow theater like Libya while the fate of the Reich hung in the balance at Stalingrad and the Rzhev salient."
A) Rommel arrived in Africa in February 1941. Stalingrad was November 1942. When Rommel turned up in North Africa he was taking over the main operational theatre of the war.
B) It happens; Slim was the best British general of the war and he fought in Burma.
C) Rommel was also in command in Normandy, which was not a sideshow.
Counting against Rommel,* he lobbied for reinforcements right through his time in Africa, long after the Soviet invasion. There were what, 250k captured in Tunisia? Uncool.
*Not that this metric is trying to capture this, or that it was necessarily Rommel's responsibility.
45 Far better for them to be captured in Tunisia than at Stalingrad, Kharkov, Kursk, etc.
46 is true but I'm thinking "Mitigate the conditions under which the German army is destroyed" wasn't Rommel's plan.
R also played an important role in the western campaign of 1940 - IIRC his was the first German division to cross the Meuse after he, typically, interpreted his mission to protect their flank to mean "protect the flank of the river crossing, by getting over the river first and unleashing mayhem".
Also, on 35, if anyone maximised his post-war reputation and minimised his involvement with Nazism, Manstein was Exhibit A. His enormously self-serving book "Lost Victories" basically launched the whole idea of the Wehrmacht as amazingly great, let down by Hitler, and nothing to do with the Holocaust, bought him enormous and enduring influence in the US Army, concealed what he got up to in the Crimea, and supported him in re-establishing a new German army including a lot of people from the old one.
47: I have no idea how that that got into the orders we sent him.
Officers of the deck involved in collisions are convicted at court-martial. Their commanding officers are charged to ensure that the precedent of accountability for command is maintained. Accountability does not, however, beget culpability, and lack of culpability leads to acquittal.
I'm the worst. Sending heebie a guest post and then disappearing for two days. Sorry, h-g.
enablers are made, not filtered.
That right there is what I was struggling to articulate. It wasn't really about a test but about the structural issues.
ajay: to tell head teachers that if any of their pupils are abused by anyone at their school, the head teacher will probably lose their CRB certification (i.e. won't be allowed to work with kids) unless they can prove they were not negligent.
In general I agree with the accountability argument, but in practice it's abundantly clear that creating high stakes often just leads to people covering up because "You don't want Nice Head Teacher X to get fired, do you?" I dunno how to fix this.
JRoth: right now, a company president might be fired for missing earnings targets, but not for a workplace incident that kills a dozen people or crashes global markets, and that this is precisely backwards.
100% agree.
Mooseking: a culture where easy scripts for intervention are provided will lead to more intervention
Right. I used to make my (very young) staff at a school-based youth program practice how they would call 911 in an emergency. This was meant to help them withstand the pressure from school officials to NOT call 911. It helped that I was in charge of their paychecks and I said repeatedly that we would never fire them for calling the police.
It's the sort of thing where Toyota has it right - everyone needs a big red "stop the line" button, and encouragement to press it.
Moby: I'm sure there are systemic issues, but running into a container ship seems like a very personal failure.
My exact thought.
Also, I'm very very very reluctant to resort to gender essentialism, but could the men on this thread who were intrigued by the sabermetrics discussion explain why? To me it just seems like a wildly inappropriate (in the sense of not-useful, not in the sense of socially-inappropriate) attempt to impose rankings on something that is both very important and more or less un-rankable.
I mean, the concept of an effective military commander vs. an ineffective one is not hard to understand, but even playing with the idea of calculating it like that seems to utterly miss the point. What am I missing?
. . . something that is both very important and more or less un-rankable.
I mean, the concept of an effective military commander vs. an ineffective one is not hard to understand, but even playing with the idea of calculating it like that seems to utterly miss the point. What am I missing?
In my mind, the fact that it's not even remotely close to a useful ranking (while still being clear enough that it has some meaning; it isn't like ranking the most important days of the week) is what makes it interesting -- in that one is forced to engage in the meta-discussion of "what do these rankings mean, and how does that differ from what we might think of as rankings for a general."
The original justification for the analogy ban was that once an analogy was made everybody would debate merits of of the analogy rather than the original point under discussion. This is the reverse of that; it's a conversation starter in which the only thing* of interest is the validity of the analogy.
Consider the positive and negative implications of "intuition pumps.
A popular strategy in philosophy is to construct a certain sort of thought experiment I call an intuition pump. ... Intuition pumps are cunningly designed to focus the reader's attention on "the important" features, and to deflect the reader from bogging down in hard-to-follow details. There is nothing wrong with this in principle. Indeed one of philosophy's highest callings is finding ways of helping people see the forest and not just the trees. But intuition pumps are often abused, though seldom deliberately.
Any sort of mathematical ranking is trying to do exactly what the bolded passage describes, but that can be done well or badly. Engaging with something like the ranking of generals is practice for looking at a model and trying to figure out what significance the results have.
* "only" is overstated.
53: NickS is right. The vast majority of the discussion here* was over the metric, not the results it produced.
even playing with the idea of calculating it like that seems to utterly miss the point
I think it's worth doing the exercise because it moves the discussion away from reputation and toward harder facts. Separately from commanders in particular, this kind of parlor game is also at the low end of a spectrum that goes up to massive models that, for instance, accurately predicted casualty rates in various wars in the 1990s.
*Unfogged of course not being representative of the world at large.
I mean, the concept of an effective military commander vs. an ineffective one is not hard to understand, but even playing with the idea of calculating it like that seems to utterly miss the point. What am I missing?
There are like whole gaming systems associated with those calculations. Not just D&D, but strategy games.
Honestly, it makes more sense to me that fantasy football.
Thanks. I still don't really get it, but I appreciate the additional context and explanation.
Speaking of dumb forced rankings, can I say how much I hate bank/website security questions? I always pick true answers because otherwise I have absolutely zero chance of remembering them.
But the last set I was given, 6/12 of the questions had something to do with weddings or your spouse, and I don't have any answer to "What were your wedding colors?" or "What was your wedding song?" [N.b. Frankly, I am also skeptical that 70% of married people would have answers to these questions.]
So that left me with the "favorites" questions. What is your favorite color? OK, I sort of have one. What is your favorite song? Are you kidding me? What is your favorite vacation destination? and on and on. Question after question where my answer is, "Uh, I dunno, I could pick one of a couple dozen options based on my mood" or "I have no earthly idea."
With the result that I picked "Who was your favorite childhood friend?" and answered really my only very long-time childhood friend. But the truth is he wasn't really my favorite friend; just the longest-enduring one.
So every time I log into my bank account I now have to type his name and add a mental asterisk. Sorry, Not-Really-Michael! You were a friend, just not a FAVORITE.
Sheesh.
What don't you get, specifically? You're probably right that a single definitive ranking is impossible or effectively impossible, and I share your disinterest in nitpicking rankings. But, having acknowledged quality of command (as opposed to the actual rankings) is important, analysis of the subject is surely also important, albeit more or less interesting to different people.
You're just supposed to have "favorite" things, even if you know your preferences are contextual.
But I won't remember my favorite things if they aren't actually favorites. There's just no chance. After 40+ years, I know myself at least well enough not to set up situations where I'm going to end up locked out of my own checking account because I can't remember my favorite President. (Lincoln! Wait, did I say FDR? Oh God, maybe I said Grant because I was so angry about the neo-Confederates that day....)
Mossy: I can't really explain, except to say that it feels a little like astrology to me. I can grasp the idea that there may be differences among people based on what time of year they were born -- and perhaps they are even scientifically measurable -- but astrology per se seems like such a ludicrously implausible way of determining such differences that I don't understand why anyone would waste time thinking or talking about it.
Obviously it's not an exact analogy (hence the ban!), because I'm only *slightly* certain about the differences between people based on birth season, but I'm *entirely* certain there are differences in the quality of military commanders.
62: Are you feeling this about analysis of command in general, or about doing such analysis using sabermetrics in particular? I know nothing about sabermetrics (except the Moneyball movie, which I've basically forgotten) so maybe it is ludicrously inappropriate given the far greater complexity of war.
If it's more general: astrology purports to have predictive power, and to have a theory to explain phenomena. Raw metrics like the one linked are in themselves strictly retrospective and descriptive.
TBC, I'm not trying at all to be combative. This kind of non-meeting of minds - as in, "Why the fuck would anyone even care about that?" - is something I feel fairly frequently.