The file was sitting on a desk in the Directorate of Military Personnel in Moscow when a senior general picked it up for what he assumed would be a routine review. It was 1974. The general, a decorated veteran of the Great Patriotic War, who had survived Stalingrad, who had watched men die by the tens of thousands, and who had signed the orders that sent them there, had read his share of personnel documents.

He knew what soldiers look like on paper. He had read enough of these files to have long since stopped seeing the individual and begun seeing only the category. Infantry, artillery, sapper, special assignment. He opened the file, he read the first page, then he put it down and reportedly said something that would be repeated with variations for the next two decades in the quarters of Soviet military administration.

He said, “These men are not human.” He did not mean it as a compliment. The file contained the intake assessment for a candidate applying to the 45th Guards Spettn’s regiment. What troubled the general was not the candidate’s combat record, which was exceptional, nor his loyalty scores, which were impeccable, nor even the classified psychological evaluation, which was in the language of Soviet military bureaucracy strongly positive.

What troubled him was the physical assessment. Not because the numbers were bad, because the numbers were impossible. This was the paradox at the center of Soviet special forces development from the moment the GRU first began systematically selecting and training what it called diversionary reconnaissance forces in the years following World War II.

The standards that defined an acceptable Spettznaz candidate were so far in excess of what the Soviet military medical establishment considered the documented [music] ceiling of human physical performance that the selection officers responsible for evaluating candidates were for the better part of three decades genuinely uncertain whether the documents they were signing constituted accurate assessments or elaborate fraud.

The men who passed the selection process did not look like what the Soviet military system had been designed to process. [music] They looked, as one medical officer famously noted in an afteraction report that eventually reached the desk of the GRU deputy director, like something the system had not considered.

to understand how this happened. How an institution as methodically bureaucratic as the Soviet military found itself in possession of human beings who could not be adequately described by its own documentation [music] standards. You have to go back to the specific moment when the Soviet high command decided to rebuild [music] its special forces capability from the ground up and the specific decision [music] that would over the following quarter century produce the men who made general officers uncomfortable. The key word there is selection. The year was 1950. The decision was made not in Moscow, but at a military research institute on the outskirts of Lennenrad, where a committee of military scientists, physiologists, and combat veterans were tasked with answering a question the Korean War had made suddenly uncomfortably urgent. If you needed to put a small group of men deep inside enemy territory, 20 kilometers behind the line, perhaps 50, perhaps

200, to gather intelligence, destroy infrastructure, assassinate command figures, and then extract themselves without support. What would those men need to be able to do, and how would you find them? The committee worked for 14 months. The document they produced ran to several hundred pages and contained among its conclusions a physical performance standard that the committee’s own chief medical officer formally noted in the margin was and he used the technical phrase physiologically implausible for unugmented human subjects. The standard required candidates to sustain a combat load of 35 kgs, weapons, ammunition, demolition equipment, radio, water, food across terrain conditions ranging from dense forest to mountain slope to urban rubble [music] for a minimum of 72 hours of continuous operation before the onset of combat performance degradation. It

required the ability to engage in close quarters combat of maximum intensity at the end of that 72-hour period. It required cold water immersion tolerance, altitude tolerance, caloric deprivation tolerance, and sleep deprivation performance floors that individually sat at the outer edge of documented human human capability.

Together, they sat somewhere beyond it. The committee did not reduce the standards. They reasoned with a kind of institutional logic that is either brutal or admirably cleareyed depending on your perspective that the operational requirement was real. The enemy was real and the cost of sending inadequate men into those conditions was not a training failure but a strategic catastrophe.

If the standard was physiologically implausible, the answer was not to lower the standard [music] but to find or create the men who could meet it. What followed was one of the most intensive and systematically brutal selection and training programs in the history of organized military forces. Not because the GRU was indifferent to human cost, though there is evidence that it was, but because the program’s architects had concluded [music] that the men they needed could not be made from ordinary material. They had to be found. The selection process that emerged by the mid1 1950s began not at the military level but at the youth level. physical education instructors across the Soviet Union, particularly in Siberia, in the Caucasuses, in the Central Asian republics, in the regions where the population had been hardened by climate and poverty and the lingering trauma of the war were given informal

guidelines, never formally codified about the kinds of young men they should flag to local military recruitment committees. The criteria were not technical. They were observational. You were looking according to a training manual recovered from a GRU archive that was partially declassified in the 1990s for the young man who when the exercise becomes impossible does not stop.

You were not looking for the fastest or the strongest or the most aggressive. You were looking for the one who had in whatever form it takes in a given individual decided that stopping is not a category that applies to him. Those young men and the numbers were never large.

perhaps a few hundred per year who survived the initial screening entered a pre-selection pipeline that lasted anywhere from 18 months to three years depending on the candidates’s age and the specific units requirements. The pipeline was not designed to build capabilities. It was designed to identify the limits of a candidate’s existing capabilities [music] and then push systematically past them.

Candidates who discovered that they had limits were in the euphemistic language of the program’s internal documents [music] returned to conventional service. Candidates who appeared not to discover their limits or who discovered them and then proceeded anyway were advanced. By the time a candidate reached formal Spettznaz selection, the process that would have produced the file that caused the general to make his remark about inhuman soldiers, he had already been filtered through a system designed to eliminate everyone who was merely exceptional. What remained were the people for whom exceptional was a flaw. The physical standards at formal selection were not administered as an as a test in any [music] conventional sense. There was no structured examination. There was a [music] continuous operational environment maintained for a minimum of 14 days and sometimes as long as 30 days in which [music] candidates were required to

function under conditions caloric restriction, sleep deprivation, thermal stress, heavy physical load, and psychological pressure that any competent sports [music] physiologist would have identified as medically dangerous. Candidates were not formally told they could quit. There was simply no mechanism provided for quitting.

The assumption baked into the design of the selection environment was that the men who were supposed to pass it would not raise the question. What the selection system produced in its decades of operation from the 1950s through to the Soviet collapse was a documented and recurrent anomaly in the military personnel records of the units that used these men.

Medical officers assigned to Spettznaz units filed reports that when reviewed in bulk have a consistent and peculiar character. They note repeatedly in the clinical language of military medicine that the men they were treating presented with injuries that should have ended operational effectiveness but were treated in the the field and continued operating.

that men completed missions in physical conditions such as documented frostbite, documented fractures, and documented blood loss that should have produced incapacitation. That the recovery timelines for Spettzna’s personnel following operations that would have hospitalized conventional soldiers for weeks are routinely measured in days.

The general who made his remark about inhuman soldiers was responding to a specific file. But the sentiment had been building in the Soviet military administrative apparatus for years. [music] The problem was not only that the men were frightening, though they were. The problem was that the Soviet military had built an extremely precise and technically sophisticated bureaucratic apparatus for understanding and categorizing human beings and their performance.

And these men kept breaking the categories. This problem came to a specific and documented head in Afghanistan. The Soviet intervention in Afghanistan [music] began formally in December 1979 and would continue for nearly a decade for the GRU Spettznaz units. Afghanistan provided an operational environment that was in the most clinical sense ideal for demonstrating the capabilities the selection and training system had spent 30 years constructing.

The terrain was brutally demanding. The enemy was skilled, motivated, and intimately familiar with a landscape that Soviet conventional forces found nearly incomprehensible. The logistical infrastructure was thin. The operational distances involved in getting reliable intelligence on Mujahedin supply routes, command nodes, and weapons caches required exactly the kind of extended, unsupported, deep penetration operations the program had been built for.

The 40th Army deployed to Afghanistan in 1979 with Spettznaz elements that fell under the operational authority of GRU. They were used initially in a relatively conventional way as elite light infantry, better trained and better equipped than the motorized rifle units that made up the bulk of Soviet forces, but employed in roles that did not fully exploit what what they had been built to do.

This changed after the first 18 months of the war made it clear that the conventional approaches were not producing the desired results against an enemy that operated in small groups, knew the terrain, and had no fixed infrastructure to destroy. By 1981, Spettzna’s units were being deployed on operations that a conventional Soviet military planner reviewing the mission parameters would have concluded were not executable.

penetrations of 50 to 100 kilometers into territory controlled by Mujahedin forces on [music] foot carrying full combat loads with radio silence protocols that meant no extraction assistance could be requested for specified periods measured in days. The expectation from the operational planners was [music] that the men conducting these missions would be discovered and would have to fight.

The further expectation which would have struck any student of military history as remarkable was that this was acceptable. A Spzn team of 8 to 12 men discovered deep in enemy territory was expected to be able to fight its way out. There is an account documented in declassified GRU after action materials that have been partially accessible to researchers since the mid 1990s of an incident in Kunar province in the spring of 1983 that illustrates the kind of operational reality that was producing the generalized UNES among conventional military officers that had been building since the 1950s. [music] A Spetszn’s reconnaissance team of nine men had been inserted by helicopter approximately 80 kilometers from the nearest Soviet position of any size. Tasked with locating and documenting a supply route being used to move weapons from [music] Pakistan into the Kunar

Valley. The Spettznaz team moved into [music] position to observe. The insertion was undetected. The team moved to its observation position, conducted its uh its reconnaissance over a period of 4 days and was preparing to move to its extraction point when it made contact with the Mujahedin [music] force that was in the afteraction assessment estimated at somewhere between 60 and 90 [music] fighters.

That contact ended the quiet of those four days. The team leader, whose name in the declassified documents is given only by his patronymic, made what the afteraction report describes as an immediate assessment and acted accordingly. [music] The team did not attempt to break contact and run. Breaking contact and running with 80 km between them and support and an enemy force that knew the terrain had a lower probability of success than what the team leader chose.

Instead, he chose to attack. What followed was a 6-hour engagement. The afteraction report describes it in the flat technical language of Soviet military documentation which has the effect not unintentional given the audience these reports were written for of making something extraordinary sound almost ordinary.

The s iix hur fight is rendered in dry professional terms. The team the report notes conducted a series of coordinated fire and movement sequences. [music] It notes that the team leader used the terrain effectively. It notes that casualties were sustained. The report’s summary assessment concludes with a sentence that has been quoted in several subsequent analyses of Soviet Spettznaz operations in Afghanistan.

[music] The team completed the mission, extracted on schedule, and delivered the required intelligence product. Three team members required medical attention upon extraction. Six were fully operational within 72 hours. The conventional military officers who read that report and documents like it had two responses.

The first was professional admiration which they expressed in the guarded qualified language that Soviet military culture required for anything that might sound like excessive enthusiasm. The second was something closer to institutional vertigo, a sense that these men and the system that had produced them were operating according to rules that the larger institution had not written and did not fully understand.

It was this vertigo that produced the documented observation which appears in various forms in various official and semi-official Soviet military documents from the 1970s through the late 1980s that Spettznaz soldiers were somehow not entirely continuous with normal human beings. The observation was not made admiringly.

It was made the way a systems engineer makes a note about a component that is performing outside its specified parameters with a kind of professional unease that comes from the suspicion that you do not fully understand what you are working with. What the Soviet military was confronting without the conceptual vocabulary to describe it clearly was the consequence of having spent 30 years systematically selecting for a specific characteristic that the military system had never formally defined.

But the selection officers had always been able to recognize a constitutional incapacity for accepting limitation as binding. The men who passed through the Spettznaz selection system were not simply physically superior, though they were that. They were men who had been identified from early in their development as people for whom the normal mechanisms by which human beings recognize and respond to their own limits had either been reduced to background noise or had never operated at full strength in the first place. The selection system had not created this quality. It had found it and then spent years subjecting the people who had it to conditions designed to verify that it was genuine rather than situational. The general who read the file in 1974 and said these men are not human was wrong in the narrow sense that the men described in the file were unambiguously human subject to all the vulnerabilities and mortalities that implies. But he was identifying something real. What the

Soviet military had built with three decades of systematic effort was a method for finding the specific human beings whose relationship to their own physical and psychological limits was fundamentally different from the statistical norm and then training those human beings until that difference was both maximized and reliable.

The result was soldiers who when evaluated by a military bureaucratic system calibrated for normal human performance produced numbers that looked like errors. They were not errors. They were the point. The Soviet military had spent three decades trying to answer the question its Leningrad committee had posed in 1950.

And the answer turned out to be that you did not build such men. You found them by constructing a system so demanding that the men who could not be found by it would remove themselves until what remained were the men who could not be made to stop. The general’s discomfort then was not really with the men at all.

It was with the recognition that his institution had succeeded and that success when you actually looked at what it had produced was not entirely comfortable to contemplate.