Old School Strength (III) – Ad Optimam Valetudinem Fingendam

After tracing Old School Strength to the training of the Roman legions, we can finally compare it with today’s elite military training programs. (Around 5.500 words, estimated reading time 25-30 minutes.)

Throughout the history of warfare, military fitness has had but one purpose: see to it that soldiers can carry gear to the theater of operations, use it to the greatest effectiveness against the enemy, and bring it back so that it can be re-used.

Preferably the same soldiers, since adequate training and proper indoctrination cost time and resources. And for all the technical innovations that millennia have brought in the subtle art of slaughtering your neighbors for territorial gains warfare, few principles have remained as absolutely constant as that one.

This makes possible a comparison between today’s infantry and the muli Mariani while eschewing the proverbial apples-to-oranges issue. Which is not one in the first place, since apples and oranges are both fruits that can be compared along a variety of dimensions (this is not a metaphor).

Similarly, compared along the appropriate dimensions (again, not a metaphor), the US Army Rangers, Green Berets and Navy Seals, the French Foreign Legionnaires and the UK Royal Marines Commandos, and other modern elite infantry units, have more in common with Roman legionnaires than with Private Snafu.

Private Snafu, by Chuck Jones – Public Domain

Now, I reckon that this last statement may seem a bit surprising. But bear with me and surprise will abate soon enough.

Infantry Then and Now

Today’s regular troops can be transported so close to a theater of operations that it makes the fitness requirement to actually get there almost negligible.

Consequently, regular infantry troops focus primarily on combat-specific readiness rather than on general fitness. An example of this would be the United States Marine Corps motto that “Every Marine is a rifleman” and the frequency and variety of the USMC marksmanship training compared to the content and frequency of the USMC physical tests (I’ll come back to them later).

Elite infantry units are different, though, and are closer to the Roman Legions in that respect. As I said earlier, a direct comparison between Ancient and Modern infantry is possible if carried along the right dimensions.

I also said that that talk was not a metaphor. So, let’s get to that first.

The timeless dimensions of military fitness

My initial description of the purpose of military fitness and combat readiness training was the following:

see to it that soldiers can carry their gear to the theater of operations,
use it to the greatest effectiveness against the enemy,
and bring it back so that it can be re-used

This description includes a set of implied parameters that unfold as follows:

  1. Carry the gear to the theater of operations has two implicit parameters: the weight of the gear (x_1) and the distance soldiers have to carry it before reaching the theater of operations (x_2).
  2. Use it to the greatest effectiveness could be broken down in as many parameters as there are physical and mental qualities involved in using the specific gear. For now, let’s only assume that the list of these parameters is finite, which gives (y_1,…,y_n).
  3. Bring it [the gear] back so that it can be re-used has, like (1) above, two implicit parameters: the weight of the gear after use (z_1) and the distance soldiers have to carry it to get back from the theater of operations (z_2).

Now for some general observations about the parameters. First, parameters x_i and z_i are insensitive to historical time differential. Obviously, measurement units have varied with historical periods (Roman vs.Imperial vs. metric) but what is measured has not. Weight remains weight, distance remains distance. There are some fine prints, but they are for nitpickers only.

In the above, I intentionally neglected two possible complications: the first is that step (1) and (3) may involve some skirmishes, and then requires the same qualities as (2); and the second is that (2) may involve covering substantial distances. The first complication can be addressed by military exercises combining load-carrying with weapon-training (or practicing them on the same days). The second can be addressed by building a margin of safety into the load-carrying standards. Thus, neither of these complications is of much consequence, I don’t feel bad for neglecting them, and if anyone bitches nitpicks about them, I’ll just repeat verbatim what I just wrote.

Second, the list (y_1,…,y_n) is only marginally sensitive to historical time differential. Of course, it is affected by technological evolution: marksmanship with a pilum or a sling requires different qualities than marksmanship with a rifle. However, even very different tasks, such as being able to hamstring a Numidian warrior with a gladius after one full day of fighting at the Battle of the Muthul, or taking a shot at a German soldier after dodging artillery shells all day long at any day of the Battle of the Somme, require similar qualities (stamina, eye-hand coordination, and ability to function under stress). And I’ve not even mentioned bayonets.

In what follows, I will make one simplifying assumption, namely: that there is a list (y_1,…, y_n) where n >1, such that: (1) any y_i in the list is a physical or mental quality commonly required from infantry troops from all historical times; and: (2) any quality not in that list depends on historical differences in gear (e.g. shields, guns, etc.). In practice, the details do not matter much: you can think of (y_1,…,y_n) as encompassing the physical and mental fitness requirements for load-bearing marches and hand-to-hand combat. Also, this makes n finite and relatively small and decreases the overall dimensions of the vector space under consideration, which is always nice.

And now for the geeky stuff: (x_1, x_2, y_1,…,y_n, z_1,z_2) can be seen as a vector in a (4+n)-dimension space where each of the (4+n) represents a dimension of infantry fitness. A particular vector of values will represent a type of units under the assumption that the values are those ‘typical’ for this unit, where ‘typical’ means “what the unit is trained for” (alternatively: substitute “typical” with ideal and “is” with should be, if you have no problem with metaphysical possibilities).

Consequently, all infantry units of all historical times are comparable because their vectors are. Of course, my simplifying assumption leaves out a few dimensions, but merely prevents apples-to-oranges comparisons that cannot be carried by want of appropriate dimensions, such as “who would be the best marksman between a Roman veles and a U.S. Marine?”.

I’ll leave to your judgment as to how much is lost.

Timeless principles

We can now make three observations about the values of the parameters across historical times. I could actually make more than that. But those three will do.

The first observation is that the relation between x_1 and z_1 is the same for ancient and modern troops. Modern troops carry food rations and ammunition belts that are depleted as the mission progresses. Roman legionnaires carried food rations and pila and auxiliaries, food rations and lead shots (for slings). So, in general, x_1 > z_1, but under special circumstances it can happen that x_1 = z_1 (short patrol where no foodstuff is consumed, absence of combat) or even that z_1 > x_1 (salvage operation, post-combat supply drop). Accordingly, troops should be trained to carry at least x_1 and at most max(x_1, z_1) over some distance d which depends on x_2 and z_2 (see below).

Of course, there are some historical differences: modern dehydrated rations weigh less than Roman flour rations, modern body armor may weigh more than a lorica segmentata and a full ammunition belt, more than two pila. But historical differences in gear are by-and-large inconsequential. As we will see in the next sections, there is a load l such that optimal training is pretty much the same as soon as max(x_1,z_1) > l, and the load carried by soldiers both ancient and modern typically exceeds l.

Second, the relation between x_2 and z_2 follows a historical trend characterized by a reversal. Specifically, x_2 > z_2 was very common in ancient warfare, whereas the relation x_2 < z_2 has become commonplace in modern warfare. An example of the first is the Battle of Bibracte: C. Julius Caesar had left his camp in the morning for a supply run of about 28 km (x_2, around 5 hours of load-bearing march at the regulatory 5.5km/h pace), was forced to make a stand before reaching Bibracte, fought the whole afternoon and well into the night, and then camped for three days to rest (z_2=0). At the other end of the historical spectrum we find seaborne forces on D-Day carrying their gear inland after taking the beaches demonstrating x_2 < z_2.

British Royal Marine Commandos demonstrating a modern instance of x_2 < z_2 after the assault on Sword Beach on 6th of June 1944  (Public Domain)

And yet, this historical trend has a limited impact on fitness requirements. What matters is the longest distance the troops must be prepared to cover under load. Specifically, troops should be trained to cover at least max(x_2, z_2) while carrying a load which depends on x_1 and z_1.

The third remark is that defining general-purpose infantry fitness in terms of xs and zs immediately yields a ‘floor’ for general-purpose infantry fitness goal. Indeed, troops must be trained for: (1) carrying x_1 for z_1; and: (2) carrying x_2 for z_2. Therefore, they would be guaranteed to meet the minimum demand for their doctrine of employment (as characterized by their vector) if they are trained for a maximum minimorum: carrying the heaviest of their pre- and post-combat load for the longest of their pre- and post-combat traveling distance. More pedantically precisely, troops should be trained to carry at least X for Z where (X,Z) = (max(x_1,x_2), max(z_1,z_2)). Again, there are some fine prints for nitpickers.

The above definition of minimal general-purpose infantry fitness is ‘timeless’ by stipulation. It is unaffected by qualities that are left out of (y_1,…,y_n), and so is the general principle parasitic on it. Whether the definition is empirically adequate a good one depends on how much overlap there is between general-purpose and combat-specific infantry fitness. I’ve already implicitly taken care of the case where (y_1,…,y_n) imposes a higher demand on loaded-carrying abilities in the previous for-geeks-only aside. We can make it explicit by defining (X,Z) = [max(x_1, x_2)+i, max(z_1, z_2)+j] where i and j are safety margins based on some parameter(s) in the (y_1,…,y_n) list. Now, notice that there is also some (‘timeless’) overlap between the general-purpose and combat-specific fitness. For instance, physical strength and endurance are useful, on the one hand, for carrying a sarcina or a backpack, and on the other hand, for hitting someone with a gladius or a bayonet. Thus, the ‘timelessness-by-stipulation’ is of not much consequence, I don’t feel bad for making that stipulation, and if anyone nitpicks about it, I’ll just repeat verbatim what I just wrote.

A more tangled issue is whether the troops should be trained for a minimum maximorum, or ‘lowest ceiling’. The latter would be characterized by (X,Z) = (m(x_1+x_2), m(z_1, z_2)) where m is essentially the number of days you want your troops to be able to operate without proper rest. Gruelling training like the U.S. Army Ranger School indicates that doing so actually puts the trainees at a risk of burnout. Accordingly, the favored solutions invariably seems be a maximum minimorum with some safety margin added. Again, this issue is a complicated one, and I actually feel bad for not exploring it further, but it exceeds the scope of this post. And if anyone nitpicks about it, I’ll just repeat verbatim what I just wrote

Forging elite military fitness

Armed with the above analysis we can now meaningfully ask the two following questions: (1) where does the Roman Legion stand in comparison to modern infantry units? and: (2) how good was the training of Roman Legions compared to modern infantry training?

Even better: we can answer them. For the first, we need to specify the Roman Legion vector with enough precision to compare it to the vectors of modern infantry units. For the second, we can extrapolate the effect of Roman training relative to the physical demand of their doctrine of employment, with just as much precision as we need to answer the first question (which is not much).

Interestingly, the main grounds for comparison with modern infantry hinges on the existence of a cut-off value for max(x_1, z_1) of ⅓ of one’s bodyweight. Roman legionnaires carried loads that routinely exceeded that cut-off and so do modern infantry personnel. Therefore:

  • Looking at which units carry loads for distances comparable to those the muli Mariani were trained for yields an answer to the first question.
  • Comparing the best training for those loads and distances to the Roman Legion training (based on our current knowledge of exercise science for the former, and historical sources for the latter) yields an answer to the second question.

Galea, képi blanc, and coveted green beret

It can be safely assumed that military training, by and large, abides by the maximum minimorum principle laid down in the previous section. Indeed, virtually all modern infantry personnel are trained to carry loads comparable to those carried by Roman legionnaires. However, few modern infantry units are trained to carry loads over distances comparable to those covered by Roman legionnaires. But some are.

Two examples that come to mind are the French Foreign Legion (FFL) and the U.K. Royal Marine Commandos. Both require the completion of marches in full gear in basic training and these marches have become iconic to these branches of service.

  • The Foreign Legion basic training includes a 2-day 50 km Marche du Képi Blanc (25 km/day in full kit) which earns the Foreign Legionnaires their white kepi.
  • The final test of the Royal Marine Commandos’ initial training is the dreaded 30-mile Dartmoor yomp with a 32 lbs kit (14kg) which must be completed in at most 8 hours (implying a minimum speed of 6.25 km/h) in order to receive the Coveted Green Beret.

Due to organizational constraints, durability is paramount in the FFL. Foreign Legionnaires are only taught about 400 French words through basic training, specialize early thereafter, and do not cross-train specialties. This leaves them generally unable to take over the role of a fallen comrade within the same platoon or to understand complex orders from their officers, who are always French nationals detached from other branches.

For all these reasons, keeping Legionnaires operational is a priority and the Foreign Legion emphasizes physical fitness over other qualities. As the saying goes “un Légionnaire ne pense pas” (a Legionnaire does not think). But a Legionnaire walks, carries loads, and ends up capable of covering longer distance at a faster pace and under greater loads than in basic training. The only exception is (perhaps) the FFL Music Band, which is an operational unit but whose training is probably closer to the maximum minimorum than other FFL units.

While the FFL is a conventional warfare branch (albeit an elite one), the Royal Marine Commandos are a special operation force (SOF). As such, they are trained to carry way more weight than 32lbs/14kg on greater distances than 30 miles in order to approach the enemy and engage them in the absence of conventional support when conditions dictate. When the Brits decided to take back Port Stanley during the Falklands War (1982) they had to drop the Royal Marines at the other end of the East Falkland Island to prevent their transport vessels from being shot down by Argentinian Exocets. The Battle of Mount Longdon and the Battle of Mount Harriet (11/12 June 1982) both involved night approach yomps (part of which was across minefields) under 30km/h winds and heavy rain.

I have not found a precise figure for the distances covered and I’m too lazy to try and figure it out through Google Earth. But I found this source indicating that the commandos’ battlefield kit was of about 70lbs (30kg); and that they were carrying well in excess of that during the night approach yomp. Then, they engaged the Argentinian forces without even taking a pee tea break.  The episode would become a national sensation thanks to the picture of Royal Marine Peter Robinson carrying the Union Jack on his backpack during the 45 Royal Marine Commando approach yomp towards Port Stanley. Since then, the picture has become a statue, The Yomper, welcoming visitors of the Royal Marine Museum in Portsmouth.

Based on the above, the Roman Legion compares to elite troopers from conventional and special operation force (the French Foreign Legion and the U.K. Royal Marines, respectively). I did not mention the U.S. military because the Cold War led them to de-emphasize physical fitness (why bother with it when the most demanding task of the upcoming war would have been to press buttons?). The post-Cold-War U.S. embroilment in long-duration, low-intensity counterinsurgency wars changed the perspective, but the situation has only notably improved in special operation capable ground forces. Even the U.S. Marines have a fitness problem (see below).

I don’t know much about the training routines of contemporary units but I will assume that modern military training is well-suited to their needs as characterized by their doctrine of employment (that, again, can be expressed mathematically as a vector). And as we’ll see in the next section, how to best train infantry personnel for extended marches under heavy load (i.e. over ⅓ of their bodyweight) is actually known.

So, the above assumption amounts to assuming that the FFL and the Royal Marines Commandos are trained in the known best way. The U.S. Marines are not, it’s a known issue by the U.S. military’s own reckoning and it has been increasingly addressed in the last 10 years (but maybe not optimally). I’d be tempted to skip the subject but for the sake of my North American readers (assuming there are any) and for nitpickers (again), here’s a USMC aside.

The USMC physical tests mirror changes in U.S. military doctrine. Nowadays, the United States Marine Corps has two benchmark tests, the USMC Physical Fitness Test (PFT) and the USMC Combat Fitness Test (CFT). The former is a stripped-down version of a test introduced in the late 1940s. The latter was introduced in 2008, reinstated some of the trials of the 1940s test, and added a few twists. Both were initially pass/fail tests but a scoring system was introduced in 2017 in order to incentivize Marines to train for them all year long, although each of them can only be attempted annually once.

How did the PFT become so bad? The PFT includes a 3-mile run and a sit-up test. Before 2017, it included a pull-up test for male personnel and a pull-up hold for female personnel. It was initially decided that female Marines would have to complete at least 3 pull-ups before the hang test from 2014 on, but the measure was delayed due to failure rates. Eventually, all Marines (male and female) were eventually given the option to substitute pull-ups with push-ups when the scoring system was introduced (in 2017) although maxing out the push-ups only yields half the points of maxing out the pull-ups.

The current PFT test is a toned-down version of a previous, more demanding test that included push-ups and pull-ups both indoor and outdoor, as well as sprints (outdoor) and shuttle runs (indoor, a staple of today’s NFL combine that requires a rather systematic training). The harder test was used in the 1940s, at the time where the USMC (and more generally, the US Army) also drilled marches in full gear. Overall, today’s PFT is not only a rather poor test of combat readiness but also a rather poor test of overall physical fitness. Not to mention that it includes a movement that is potentially harmful (max sit-ups).

According to Cpt. Andrew J. Thompson’s MBA thesis Physical Fitness in the United States Marine Corps: History, Current Practices and Implications for Mission Accomplishment and Human Performance (2005), this situation is essentially the consequence of the evolution of operational doctrines during the Cold War and a priority given to nuclear deterrence. The fittest personnel where thus the submariners and bomber crews, and the infantry was left pushing pencils and fattening, by want of real operational need. As I mentioned above, the increase of long-duration, low-intensity conflicts from the 1990s led to a re-evaluation of this doctrine. But the army was slow to change.

What did the USMC do? Thompson’s memoir was rather critical of the PFT and of the USMC approach to fitness that he knew from the inside. His recommendation was, unfortunately, moronic suboptimal. Thompson suggested that CrossFit® would be a viable alternative to the then-current fitness regimen of the USMC. Fortunately, the Corps has not followed that recommendation. A few years after Thompson’s work, in 2008, the USMC introduced the CFT in order to test combat-specific fitness. The new test includes a half-mile run in boots and fatigue, crawling, loaded carries (ammunition boxes and a fellow Marine) and submaximal lifts for reps (lifting an ammo box overhead). Training for the CFT specifically is more conducive to actual combat readiness than training random CrossFit® WODs.

Now, like I said, Thompson’s memoir is an MBA thesis. I don’t know much about MBA standards, but this one reads a bit like an infomercial for CrossFit®. I don’t think he’s bullshitting (in Harry Frankfurt’s sense) but he is clearly biased, and the source of his bias is right out in the open (see Appendix E, p. 87). Thompson trained at CrossFit® Headquarters while writing his memoir and was impressed by the performances he saw there. Unfortunately, his argument is little more than a generalization from personal gym experience, aka ‘broscience’. Thompson’s lack of attention to the scientific evidence on load-bearing marches (see below) further explains why he saw CrossFit® as a viable option.

Elite fitness: NATO, rucking, and the Roman way.

Due to their importance for military fitness, and given the general availability of military funding for research, there is no shortage of high-quality scientific evidence on load-bearing marches. A good chunk of it was gathered at the behest of NATO, so it’s kinda strange that the conclusions did not end up informing the revision of the training of U.S. Marines (cf. aside above). Let’s chalk that up on American exceptionalism.

The scientific studies on load-bearing marches are where the cut-off value of ⅓ bodyweight comes from. Still, I have not yet said much about what it is a cut-off for. It’s damn time I do. But I’ll flash a diagram first. All you need to understand about its relevance to the topic at hand is this: “Ruck” is the American English word for “Yomp”.

Due credit where credit is due, this diagram comes from Mike Prevost’s Ruck Training Programs – Part I which I use without permission (but I take it to be fair use). In fact, I highly recommend that you read Mike Prevost’s 2-Part series (Part 1 covers the science and Part 2 how to apply it). In fact, if Prevost had written a book on the topic, it would be in my top 5 fitness reading list. I hope he does write one someday. I’d advertise it. Actually, I’d even buy copies and hand them over. But for now, his 2-part series will do.

Enough evidence has already been collected to show that Greco-Roman nutrition […] was certainly significantly better than in the societies of the European ancien régime or the modem Third World which are most often offered as a model for likely living conditions in Classical antiquity.
G. Kron Anthropometry, Physical Anthropology, and the Reconstruction of Ancient Health, Nutrition, and Living Standards, 2005, p. 69.

We can extrapolate a fair bit about the Roman legionnaires’ anthropometry based on the recent re-evaluation of our appreciation of nutrition and health standards in Ancient Greece and Rome. I’ll examine the topics of Old School health and nutrition later in this series, but for now, I’ll point to a few resources and just go for generalities and mean values, with some additional details for those who are sticklers for them.

Long story short, the average legionnaire at the time of Marian Reforms must have been around 171cm tall. That’s slightly above the Marian reforms minimum requirement of 5’10” Roman feet (165cm). This average height was later given as an ideal by Vegetius, at a time where the regular legions were depleted and recruits were rushing to join the auxiliaries. The ‘ideal’ may have been just jingoism from his part, as it was probably still the average height of Roman Italians.

Do we really know how tall were the Roman Legionnaires? A landmark study reviewing the available anthropological data is Geoffrey Kron’s 2005 article in Historia: Zeitschrift für Alte Geschichte article (title and link in the quote, also in my info dump: Anthropology/History⇒Roman Legions). Kron found that the average male body buried in Italy between 500 BC and 500 AD was 168 and notes that military recruits would have been about 3cm taller (hence the 171cm figure). Kron’s figures are consistent with earlier findings, for instance, the data from Herculaneum and Pompeii, two obvious sources of well-preserved bodies. This data had however not been considered until then representative of the population-at-large. Herculanum and Pompei where indeed home to the privileged, and for all we knew, slaves to the rich might have eaten better scraps than the poor Romans from the asshole of Italy.

Kron argues convincingly that data from burials (including accidental ones in Herculanum and Pompeii) can be treated as randomly drawn samples. When we do so, the figure of 168 cm can be extrapolated as the Roman Italian male average. I’ll leave the details of Kron’s argumentation and assume he’s correct. The next step of Kron’s argument is a comparison with more recent historical data. For instance, in order to fill the rank of his armies in Italy, Napoleon had to lower the minimum height of recruits to about 150cm. More generally, most modern European nations for which data is available achieved the average height of Roman Italians well into the 20th century. Ditto for the minimum height for enrollment in the Marian legions. Exceptions are Nordic countries, who were already that tall at the turn of the 20th century, and European immigrants in the North American continent, who reached a comparable height sometime in the late 1600s (cf table below, pulled from Kron and completed by guesstimates from yours truly).

Why does height matters so much? Height can be estimated from archaeological findings both direct (human remains) and indirect (height of furniture, door frames, etc.). Since it’s well correlated with health (which is in turn well correlated with diet quality) it’s as good as it gets when we come to health markers. Moreover, when cross-referenced with what can be determined of the caloric intake of Roman legionnaires (from surviving regulation texts and fossil garbage heaps) and their training regimen, average height allows for a good estimate of how much Roman legionnaire weighed. And therefore, of what percentage of their body weight they carried.

As for the diet, the most detailed source I found is an old review study from Emile Fornaris and Marc Aubert on Roman military nutrition published (in French) in 1998 in Histoire des Sciences Médicales and republished by the French website Theatrum Belli (Le Légionnaire Romain, cet athlète méconnu) which estimates the caloric intake of the Roman legionnaires as 3.500 kcal from wheat bread and porridge alone, plus anywhere between 500-1500 kCal of additional calories from oil, wine, forage (mostly nuts and fruits), meat (both preserve and hunted for, adding to the daily physical activity) for a grand total daily intake of 4.000-5.000 kcal calories.

Fornaris and Aubert estimate the energy expenditure of Roman Legionnaires through comparison with the training regimen of modern athletes and elite military forces and find that: (1) Roman legionnaires had a workload comparable to top-level Olympic athletes in high-volume training cycles and elite military during the most demanding phases of their training; and: (2) that the diet was well-suited to the level of physical activity (perhaps a little low on the protein side by modern sport nutrition standard). Hence the title of their study, which translates as The Roman Legionnaire, this unacknowledged athlete.

Summing up, we have anthropological and archaeological evidence about height, estimated caloric expenditure, and estimated caloric intake for the Roman Legions after the Marian reforms. We also have textual and archaeological evidence about their training regimen and overall level of physical activity (respectively: Vegetius; and roads, bridges, and forts, and the like). And all we have supports the conclusion that the training regimen of the Roman legionnaires was optimally suited for maintaining top yomping performance and sustain their physical activities when they were not on the move (building roads, bridges, and the likes, with heavy stones, logs, and the likes)

All of which requires an abundance of strength-endurance and power-endurance work, in addition to just walking, hustling and sometimes running under load, rather than merely jogging or running (again, see Mike Prevost’s series). Again, this is what the daily and weekly routines of Roman legionnaires provided, as shown by the infographic that I flashed in conclusion of Part II.

Now, based on my own magic sauce knowledge of sport nutrition and exercise science, I’d guesstimate the weight of a Roman legionnaire somewhere around 75-85 kg, which seems to be what most specialists and semi-specialists in the field agree upon (according to the figures my research assistant pulled from a history forum on StackExchange.com, in the absence of direct scholarly work, so obviously more research needed). And a very, very lean 75-85 kg, to boot.

Which, for 171cm on average, translates as this: Roman legionnaires approached the fitness level of top CrossFit® athletes while being a bit shorter. I’ll explain the 75-85 kg figure and the CrossFit® comparison in a follow-up to this post, but for now, suffice to say that top-tier CrossFit® athletes don’t ‘train CrossFit®’ (that’s where Thompson got it wrong). And although they most likely juice like crazy use performance-enhancing drugs wisely, that’s a red herring. They probably cannot do as well as they currently do without them due to the breadth of events they have to be good at, and the time they have to train. But if they were as focused as the Roman legionnaires were, they could.

The short version is this: (1) Roman legionnaires’ bi-weekly (Scipio in New Carthage, cf. Part II) or daily (Metellus in Numidia, same) load-bearing marches elicited more (anaerobic) strength adaptations than (aerobic) endurance adaptations while still contributing to maintaining an aerobic base; (2) training at the post was comparable to high-intensity interval training (HIIT) that inspired the whole CrossFit® and is known to indirectly improve aerobic fitness when combined to even a moderate dose of low-intensity steady-state cardio (see that post); (3) castrametation and ancillary engineering work (=lifting heavy but sub-max for reps) provided additional strength stimulus in the hypertrophy-inducing range. The diet did the rest.

Bottom line, the Roman legionnaires’ training was similar to that of top CrossFit® in both design and functional demands. Therefore, the muli Mariani may very well been the ‘Fittest on Earth’ of their time.

Built like a legionnaire? 175cm/88kg Rich Froning (center), From Chris Shugart’s In defense of CrossFit, 14/04/2014 (uncredited picture, used without permission, fair use)

Conclusion: Forging Elite Fitness Ad Optimam Valetudinem Fingendam

Anthropological and archaeological evidence on the one hand and exercises science and nutrition on the other suggest that Roman legionnaires were comparable to today top-tier generalist strength- and power-endurance athletes and elite modern military personnel.

Comparisons between Roman legionnaires and modern elite troops and all-around athletes is warranted by the comparison of the fitness demand of their respective jobs (top CrossFit® is a full-time job). As long as it is carried along the right dimensions, the comparison is the right type of apples-to-oranges.

I outlined explicitly the methodology to carry the comparison as far as military fitness is concerned, as an exercise of Analytic Fitness™. I also hinted at the comparison between training methodologies. There’s still a lot to say about that last one, though.

I bet you’d like to know how to look like train like a legionnaire. And that will be a topic for another day.

In the meantime, vale!


Leave a Reply