Where education was tried, it turned out to be futile

Tuesday, June 26th, 2018

Gwern reviews McNamara’s Folly — which is about one particular sub-folly, The Use of Low-IQ Troops in the Vietnam War:

It’s not well-known, but one of the most consistent long-term sponsors of research into intelligence has been the US military. This is because, contrary to lay wisdom that “IQ only measures how well you do on a test” or book-learning, cognitive ability predicts performance in all occupations down to the simplest manual labor; this might seem surprising, but there are a lot of ways to screw up a simple job and cause losses outside one’s area.


Gregory’s book collates stories about what happened when the US military was forced to ignore these facts it knew perfectly well in the service of Robert McNamara & Lyndon Johnson’s “Project 100,000” idea to kill two birds with one stone by drafting recruits who were developmentally disabled, unhealthy, evil, or just too dumb to be conscripted previously: it would provide the warm bodies needed for Vietnam, and use the military to educate the least fortunate and give them a leg up as part of the Great Society’s faith in education to eliminate individual differences and refute the idea that intelligence is real.

It did not go well.

The main value of the book is providing many concrete examples of what a lack of intelligence can mean (useful for people who spend their whole lives in high-IQ bubbles and have no idea of what that means; more examples in Gottfredson’s “Why g Matters: The Complexity of Everyday Life“), the difficulty of implementing social welfare programs (McNamara’s education fantasies never materialized for lack of funds & the enlistees not being smart enough to qualify in the first place), and a forceful denunciation of the harms & cruelty committed by a willful blindness to the fact of individual differences, harms which fall on those least able to understand or withstand them. (“…He was perpetually angry and aggrieved, and he talked back to the sergeants. When they cursed him and threatened him, he would say angrily, ‘I just wanna go home! Why don’t you let me go home?’”) The phrase “banality of evil” comes repeatedly to mind in examining the ramifications of McNamara’s blank-slatism through the military system.


Gregory describes how many would be sent to remedial training, repeatedly failing the exercise requirements because they didn’t understand how to correctly execute actions; in swinging from monkey bars, they would try to swing one bar at a time, coming to a halt each time; in running an obstacle course, they would have to pause in front of each arrow and think about what an arrow meant before understanding which direction to go, costing them too much time to ever beat the deadline; they would insist on throwing grenades like a baseball directly to the target, not understanding that throwing up in a parabola would gain them the necessary distance; and in the mile run, they would sprint as fast as possible at the start and be surprised when they became utterly exhausted long before the finish line. One mutinied from the drills, under the impression that being sent to the ‘stockade’ meant ‘going home’, until it was explained to him that the word meant ‘jail’.


Where education was tried, it turned out to be futile, and those who did train them found them too slow or too dangerous to trust. A man assigned to t-shirt printing shop was unable to understand alphabetization and had to pick out each letter for printing by scanning through the box one by one; a sergeant trained two men to drive military trucks somewhat successfully but they were too dangerous drivers to be used and were transferred out; another simply forgot to get back on the helicopters after a village search forcing a second retrieval mission; another was lucky enough to be sheltered by his sergeant in mess hall duties (until a mortar hit it, killing him); one played a prank on his squad mates, tossing a defused grenade at them two times, but on the third throw forgot to disable it; another wandered away from an ambush and wandering back, was killed by his squad; while yet another almost shot his commander with a LAW rocket when startled; another did kill his commander while on guard duty when he forgot to ask for the password before shooting; another forgot to put his rifle safety on (shooting a squad mate in the foot, who died); another tripped a booby-trap while not paying attention; another was captured by the NVA and went insane, screaming endlessly and defecating on himself while being beaten… It is unsurprising that many of them would be made to ‘walk point’, or ejected somehow, in addition to the constant insults and abuse – a new recruit was told the NVA would kill them all in a few hours, went insane from fear, climbed up a flag pole, and jumped off it; and another was beaten to death in Marine basic training.

(McNamara may have had good intentions, but in the social sciences, good results follow good intentions much as the rain follows the plow; which is to say, they do mostly by accident, and we find it easier to tailor our preferences to the results than vice-versa.)

Only a few of the stories, like the recruit who was confused by having two left boots and two right boots but no complete pairs of boots, or the one who thought semen was urine, or the extremely-short man who received an honorable discharge & medical pension for contracting the terrible disability of ‘dwarfism’ in a war zone, or the draftee who tried to commit suicide “by drinking a bottle of Head & Shoulders shampoo” could be considered all that funny. Most are painful to read. (But educational, again, especially if you are in a high-IQ bubble and have a lack of empathy for what low intelligence means.) Once you’ve read some of these anecdotes, other anecdotes, like Scott Alexander’s experiences in Haiti no longer seem like such a stretch.

A commenter added some useful details:

As to the McNamara experiment, according to the RAND report on this program, they did not accept men in the bottom 10% of the IQ spectrum. The anecdotes you cite refer to men in the 11-30% range. Of course, some sub-10% men were no doubt accidentally enlisted since the recruitment was based on only 1 IQ test. I wouldn’t expect that many of these recruits were “funny looking kids.” The clearest mistake of this program was allowing guys into combat who were so dysfunctional as to be dangerous to their comrades. But, the idea that it was a priori absurd to recruit guys at the 20th percentile (about 85 IQ), is a post hoc judgment and probably not even rigorously demonstrable (it is not clear from the 250 page RAND report). Combat efficacy increases as the soldier’s IQ increases with no ceiling. The question is whether there are thresholds that matter. The only clear useful threshold is the point at which lower IQ men become too dangerous to their comrades. Another threshold might be the point at which they become too financially burdensome or too ineffective against the enemy (ie, their mortality rate is higher than the enemy’s).

In World War II, the minimum IQ permissible was even lower than it was for this experimental program. They didn’t recruit 350,000 sub-89 IQ soldiers; they recruited millions of them.


  1. Harry Jones says:

    I once had two mentally retarded step-relatives. Between the two of them they burned down two houses and molested a little girl. They weren’t evil; they quite simply didn’t know what they were doing.

    They wouldn’t have survived in a war situation. Their smarter comrades would soon have realized there’s no point in having their backs, because there’s no possibility of reciprocation, and they can’t be trusted with firearms. The training was surely a waste, but sending them to war (as infantry) can be seen as a particularly devious and murderous approach to eugenics.

    The evil, on the other hand, will thrive in combat. The non-stupid evil, that is. They have their own backs. And they get to kill lots of people and nobody minds.

    I’m told there’s a very high rate of turnover and burnout in professions that deal with the mentally disabled. I once was given a mildly retarded man to tutor. He was a very nice fellow but it was a waste of everyone’s time, and just plain depressing. I quit tutoring shortly after that.

    I don’t have the answers here. I know all too well what doesn’t work, but I’ve never seen anything ever work.

  2. Lu An Li says:

    There was a definite racial dimension to Project 100,000. Johnson wanted the streets cleaned of young men that were of military age. Too many just “hanging around” without possibility of finding suitable work and the potential for riotous behavior during a tumultuous period. Many of the 100,000 just got a less than general discharge and sent home. Those persons discharged such returning home just that much more bitter and angry with their lot and station in life. 100,000 the negative must outweigh the plus by a lot.

  3. Bob Sykes says:

    I thought the lower IQ limit for the army draftees was 85, and 90 for the Navy and Air Force enlistees. 85 is the black mean, so half of all blacks were ineligible for the draft. Only 1/6 of Whites were.

  4. Kirk says:

    The problem with this stuff is that there’s a huge spread taken in between the CAT V rejects and the CAT IV-III types they intended to take in. The program McNamara meant to implement would have taken in people who were smart enough, but didn’t necessarily “test well”. Instead, it swept up the genuinely mentally disadvantaged with those guys, and you had some sheer horror stories that took place. How they could have differentiated between the dyslexic “book-stupid, street-smart/woods-wise” types, I don’t know.

    I worked for at least one of McNamara’s Hundred Thousand when I was a young private. Had no idea at all that this long-service Sergeant Major was functionally illiterate, either–The guy had so many little scams and dodges he pulled to get other people to read things for him that it wasn’t even funny, and with his verbal memory, once he heard someone recite something for him, he had it, and could repeat it verbatim days, weeks, or months later. I don’t know what his IQ was, but his ASVAB score was low enough that I’m not sure he actually took the damn thing himself. One story I heard was that they thought he was bullshitting them down at the draft board, and when they saw his test score, they pencil-whipped it just high enough for him to pass. How the hell he got a high-school diploma back in the 1960s, I don’t know… But, he had one. Probably on the strength of his football game, though.

    He was actually an amazing leader, one of the best ones I was ever around. The functional illiteracy thing I only figured out after I’d been around him for some work, and I’d actually observed what he was up to. When I figured it out, I was around the XO, a Major, and he swore my ass to secrecy if I knew what was good for me. Sergeant Major, you see, was not only a high-school football hero, he was also a Vietnam War hero, genuine, with a rack of valor citations from two tours. The “old boys” club that the Army used to have aspects of chose to ensure he made it through to retirement, and did well. Amazing guy to watch–The stuff he used to do to get around his inability to really read was amazing… He constantly was “forgetting” his reading glasses, and having other people read stuff for him, or asking them to “read that back, so I know you understand it…”. Whenever a new regulation or FM came out, he’d have a round-robin reading of it with all the junior NCOs, and you could see him memorizing that stuff while they did it, and you would later hear him recite that new stuff chapter and verse just as he’d heard them read it–Pronunciation errors and all. Which was how I figured out he was not actually reading that stuff for himself, but memorizing it as he heard it. He demonstrated an amazing verbal memory, over the time I was around him. I wouldn’t even want to hazard a guess at the actual volume of information he had memorized and could recall–It had to be on the order of several shelves of fairly complex regulations and Field Manuals.

    In another lifetime, he’d have likely been some tribe’s loremaster.

    I’m not sure that I really buy the “conventional wisdom” on McNamara’s Hundred Thousand, or the actual utility of what we’re calling intelligence tests. The only safe thing I’d ever take from someone who did well on the ASVAB was that they did well on the ASVAB, and that might have a general correlation with how well they did on the job they were assigned or picked, but that there are a host of other factors.

    Our informant on this piece lays out a bunch of individual cases where some really dumb people did some really dumb things, in combat and in the Army. I won’t refute that, but I will happily lay out the litany of horrors I could take from my experience with people who did really, really well on the ASVAB. Anyone who made a career of the enlisted Army in the Combat Arms would likely back me up on the distressing correlation of “trouble” with the bottom end and the top end of the ASVAB scale. A certain percentage of my CAT IIIB types were just plain trouble, because sheer stupidity. Call it maybe 30% of that group–Dumb and prone to demonstrating that fact on a daily basis. On the other end, up in the rarified air of the CATI? Oh, holy ‘effing Christ… Dude, you do not know “stupid” until you’ve had it demonstrated for you by someone with an IQ in the 140 range, usually because that 140 IQ is coupled with an entirely unjustified arrogance about their own intelligence, and a degree of at least social autism that is a thing wondrous to observe in practice–So long as the asshole doesn’t report to you, or fall into your formations every morning. On this end of the scale, it’s usually a reverse of the low end percentages: CAT I guys are usually 70% trouble, 30% worth the oxygen they breathe.

    And, you need to realize I’m saying this as someone who has always tested really, really well–Even on tests for subjects I really don’t know shit about. On the ASVAB, I was a comfortable CAT IA, 94th percentile, based on a test I was taking to get out of a high school class and really didn’t give a damn about. I don’t know how well I’d have done, if I had really cared and tried at it. They wouldn’t ever let me retake the damn thing, because the recruiters were afraid I’d flub it, or something.

    And, yeah… I did some seriously, seriously stupid shit, coming up. Things my peers didn’t do, because the idea never would have occurred to them. It did to me, though, and that nearly always led to a certain amount of disaster–A flaw I later grew to recognize and dread in any subordinates of mine who “did really well on the ASVAB…”. Smart is as smart does, and to tell you the truth, were you to ask me to pick guys out based on the ASVAB? The odds of there being a single CAT I on that list are really, really low…

  5. Mostly Cajun says:

    I was in the army 1968-77, went to Korea in 1969 as a tank commander. My driver was one of McNamara’s 100,000, a ‘career private’ – the Army term for a ‘soldier’ who would get promoted a paygrade only to get busted for something.

    My particular example was a druggie – loved buying ‘reds’ on the local drug market. One morning we rolled out of the motor pool (six miles from the DMZ) on alert. My driver hadn’t had time for his last excursion into pharmaceutical recreation to clear his system. He almost ran down the battalion commander with an M48 tank.

    I pulled the dud out of the driver’s compartment, he was arrested, cashiered out of the army, and is probably a dimmocrat activist to this day.

  6. Graham says:


    Really interesting comment.

    I was struck by your suggestion that Sergeant Major would have been some tribe’s loremaster.

    The dark side of it is how much that set of capabilities, so long of such immense value, has been devalued by a literate, and especially a technological society.

    I have no model in mind for it, but I sometimes wonder if a future society, still technological if not vastly more so, will be so divided along so many lines and categories of knowledge that someone like him might not again find a niche. Still probably need literacy, but in the larger sense someone whose gifts are in memorizing, and also being able to use and relate, a lot of material on disparate subjects, should have a role in a society of specialists reliant on access to data and search. I don’t know, but I wonder if it won’t be a valuable skill after all.

    Might not even need literacy at that, if the tech is going to present more and more content in audio and visual formats to suit evolving tastes.

    The role of ‘loremaster’ will now be called “Onsite Immediate Access Knowledge Aggregator…”

  7. Kirk says:


    One of the things I think we need to pay attention to, as time goes on, is just how much effect changes in information technology and dissemination really have. Imagine, if you will, the sheer effort involved for the oral traditions of Homer to have propagated and lasted as long as they did, before anyone ever sat those immortal verses to a more permanent format. There’s a whole universe of pre-literate oral tradition we no longer have access to, and you have to wonder how much poorer we are because of it.

    Of course, on the other hand… It may be that we didn’t bother to write that stuff down once literacy came in because it was mostly crap. Either way, one does have to wonder what the reality was. Pre-literacy, the role of pure memory in human affairs had to have been tremendous; one suspects that it was especially vital for nomads who trekked over long distances on the Eurasian plain, and who would need to remember routes, water sources, and the like over vast areas and time frames. One contemplates the sheer sophistication of memory required, in order to pass on to the next generation where the various passes were, and how to find the best grazing for the animals, and you then realize just how much more advanced the nomadic herding tribes would have had to have been, over simple hunter-gatherers. Then, too, there are the Polynesian navigators, who used sophisticated symbolic models to set their routes and memories down in concrete terms–Were similar things used by the nomads of Eurasia, as they moved across the post-Ice Age steppe?

    Every notch in the ratchet of information technology changes us. Our memories are nothing compared to even those of our closer ancestors like the Victorians, who demanded that schoolboys and girls memorize vast tracts of poetry for recitation. The curriculum of yore would be considered child abuse, today. And, yet… Today’s child needs to be an information omnivore, sucking up vast amounts of data, and learning where to find answers, sifting through tons of chaff to find kernels of value.

    I wonder what will come in the future, in terms of this: Say that we manage to create implant links in the human brain, ones that work–Just what will that world look like, when you can just download a skill or knowledge base, and then make use of it. What effect on society will there be, if you can essentially download “engineer” skillsets and knowledge bases at will, vs. spending a decade acquiring them…?

    Other question is, what’s going to be the effect of those “memory implants” on the weak-willed and mentally insufficient? Would you risk overwriting the original personality, along with passing on a skill set? How much of what makes an “engineer” is actually mindset, vs. skill and memorized data?

    At some point in the future, this whole literacy thing is going to be seen as small potatoes. The issues of education and training are going to be very different, once the point is reached where you have a YouTube of human skills and techniques available to the average person. What effect will that have on the economy, and what effect will that have on culture?

    Of course, the other question is, is such a skill/knowledge summation even something that we can distill to transmit between ourselves? Certainly, there’s a lot more to being a Rudolf Nureyev than just the bare knowledge of ballet, but… How much more is there? Could you download the majority of Nureyev’s knowledge, and then practice the skills enough in your own body and mind, to make his genius yours? Will the people able to do those things in the future be the real heirs of Homer, along the longitudinal history of human endeavor?

  8. Bomag says:

    “In another lifetime, he’d have likely been some tribe’s loremaster.”

    Reminds me of some indigenous people I’ve met who, while otherwise seemingly dull witted, had a prodigious ability to remember names and clan lines.

    Makes me think that survival in olden times consisted of who to flatter and who to stay away from.

    The written word and the printing press have markedly reduced the value of such mental skills coupled with the revealed importance of symbolic logic in today’s world.

  9. Graham says:


    Some of the possible futures you intuit fill me with horror. Then again, most futures on offer these days do that so I’ve achieved a sort of panic-equilibrium.

    On your idea of downloaded knowledge- there is an Asimov story on this. he has a couple on the theme of future education but this one “Profession” sticks with me. https://www.abelard.org/asimov.php

    Of course, the main point is Asimov’s faith in the Social Sciences to guide man’s future through the agency of enlightened Social Scientists and their supple minds.

    But that aside, in this future most people get their knowledge from being “taped”- basically, all the current knowledge of a specific field is downloaded at 18 or so and then off you go. You cannot be taped again or the knowledge ever updated, so your career has built-in obsolescence even more than today. But Earth’s society and export economy are built on this. It’s very old-fashioned, yet eerily forward looking even now.

  10. Kirk says:


    I’m not really seeing it in the sense that Asimov did. He dwelt in a mental world that was static, and not network-enabled, which drastically limited his imagination. His horizons were stunted by his worldview and knowledge.

    The way I see it going is more like “Oh… Need to build that deck this weekend… Better find a good carpenter engram on the web tonight, and download that “engineer” package of skills to design the thing, so I know what materials to pick up…”.

    That may not be an actual possibility, but it’s an interesting gedankenspiel. Imagine being the guy out on a remote asteroid with a broken drill, not knowing a damn thing about repairing it, and then simply uploading the datasets of repair information and perhaps even what I’d term an “engram”, to borrow and misuse a term, which would encompass the entire “repair guy” mentality that a really accomplished repairman has. Upon completion of the repair, wipe the memories and engram, and move on to the next phase, downloading navigation data and the “asteroid prospector” engram…

    Necessarily, you’d have to have a very flexible, very “personality certain” set of mental attributes, able to swap out the various mental toolsets from task to task, much as your desktop PC does when you shift focus from wordprocessing to spreadsheets. What changes this sort of thing would mandate in human mental architecture and thinking…? Probably at least as great as the shift between the old oral traditions and writing, but probably a lot more than the one between writing and mass printing.

    At some point, education is going to be more about teaching people the “how” of dealing with this sort of thing, rather than imparting knowledge and practical skills. Strong, confident, and stable personalities that won’t be overwhelmed by the downloaded “mentality engrams” and the knowledge packs will be necessary, and there will be a bunch of therapeutic uses for these packages in dealing with the mentally ill and terminally ineducable. Johnny wants to be a waste of air…? No problem; overwrite his lack of a work ethic with a therapeutic engram that gives him one…

    The possibilities for misuse and abuse are endless, as well as the disadvantages. I could see a totally different set of “success models” for human behavior coming in, along with it all. A successful human, under these conditions, would have to have a very strong will, a powerful personality, and not be susceptible to being overwhelmed by the experience of pulling in a set of artificial mental tools like that “engineer mindset” I was talking about.

    Of course, it might not be a bad thing for the weak-willed and easily influenced, to be “overwhelmed” by a superior mental toolkit they downloaded… At least, they’d be productive members of society. Or, maybe not–The correlation between Western-educated Muslims who were trained to be engineers, and then who took up terrorism might well be a constructive example of why it might be a good idea to lock down that “engineer engram”, considering how much damage the sort of kid who does tagging and graffiti might cause, were they to suddenly have access to a mental toolset that heretofore required decades of diligent work to acquire.

    If this stuff proves possible, ’tis gonna be interesting watching it all roll out.

  11. Kirk says:


    Love the screen handle–You using that in the sense of waste compaction, or construction…?

    Anyway, what you’re saying goes to what I’ve been trying to get across to other people, for years: Intelligence, as it really is, isn’t measured by how well you do on a written test. Actual intelligence is, in my belief, a measure of how well you are adapted to your environment and milieu.

    That tribesman who has thirty generations of genealogy, relationships, and feuds memorized? He’s way better adapted to living in a jungle nightmare environment of feud and internecine slaughter than you are, but he probably can’t read and/or do well on a standardized Western IQ test. But, by his lights and experiences, you’re the idiot, because you don’t know that great-uncle Igban started a feud with that sept living over the ridge, and will kill you on sight because of it. Likely, you’d have to write down what he remembers simply from having heard it once from Igban’s daughter when he was a kid…

    Intelligence is a wil-’o-the-wisp that remains indefinable. What’s “intelligent” and utterly necessary in Environment “A” may be unnecessary and positively deleterious in Environment “B”, while utterly irrelevant but annoying in “C”. To think that being able to do well on a written test taken in a classroom somewhere in the West is at all a measure of anything important outside that classroom…? That’s the arrogance and foolishness of it all. You may be really smart along one axis of human behavior and knowledge, while remaining a complete moron along another–And, to puff yourself up and pronounce your innate superiority because you did really, really well on a written test measuring what is really a very small subset of human skill and behavior…? Rank foolishness.

    I’d love to know how well someone like Homer, from the ancient oral traditions, would do, stacked up against some of the members of Mensa. Were you to give him an IQ test, orally, would he do well? Or, would the test fail, being built in a totally different mental world than the one he lived in, and fail to effectively measure his genius?

    Likely, Homer would consider his interlocuters complete idiots, having to write everything down while interviewing him, and he would be able to retain, process, and remember the majority of what they told him simply via his skills and mental tools as an oral historian and storyteller. From his standpoint, the moderns would be the mentally crippled morons, unable to remember and recite as well as he could, on a routine basis.

    We don’t know as much as we think we do, about intelligence, its value, or its measurement. What we think of when we say “IQ” is more accurately termed not as “raw intelligence” but more as “raw ability to do well on a written test”. That’s a thing that is of value in our world, but would be utterly useless to evaluating someone for how well they’d likely survive on a Eurasian steppe during the end of the last Ice Age…

Leave a Reply