Movement Training for Humans
Close this search box.
Close this search box.
CrossFit, Movement Culture & The Rise of the Generalist Mover
Jump To

CrossFit, Movement Culture & The Rise of the Generalist Mover

“We do what you do almost as well as you do, you can’t do what we do at all, and what neither of us do, we do better than you.”

This is a wonderfully charismatic boast.

The person or group who can claim this has a great claim to being a true generalist… but is it even possible to attain in the first place? And if it is, then how?

I think it is to a significant degree, but to do so requires the right approach, and while I don’t think CrossFit is the answer, it was a compelling start.

In 2002 CrossFit layed out a bold vision for what fitness meant.

To be truly fit was to balance capacity in:

This definition aimed to be “the broadest and most general fitness possible.”

When I encountered CrossFit in 2005, the most compelling part of that definition for me was the second element which CrossFit illustrated with this analogy:

“Imagine a hopper containing an infinite set of physical tasks. The athlete who performed best across the entire set would be the fittest.”

Put another way, the fittest athlete is the one who performs best across the entire set of potential movement problems.

I believe this concept nicely aligns with the work done by the great neurophysiologist Nikolai Bernstein on dexterity, which he defined as:

“the ability to find a motor solution for any external situation, that is, to adequately solve any emerging motor problem…
– correctly (i.e., adequately and accurately)
– quickly (with respect to both decision making and achieving a correct result)
– rationally (i.e., expediently and economically), and
– resourcefully (i.e., quick-wittedly and initiatively).”

Ido Portal then helped ignite a second revolution (the Movement Culture) in the wake of CrossFit by recognizing what CrossFit didn’t: that the 10 athletic attributes and 3 metabolic engines are in fact best seen as subcomponents of competence in athletic tasks.

Fitness is subsidiary to the broader concept of dexterity when it comes to functional ability.

The general fitness of CrossFit was the primary precursor to the idea of the generalist mover of Movement Culture. That is true both in the general cultural zeitgeist of the fitness world and also in the areas of fitness culture Ido and his early students were involved with.

Ido’s model was an improvement on the CrossFit model by prioritizing movement or motor dexterity, but it is still incomplete.

To me, Ido’s model seems to be an attempt to solve the infinite hopper problem but without any clear starting point in movement.

Movement is an infinite space of potential training modalities. CrossFit offered a clear prescription of basing your training on olympic lifting, middle distance running and gymnastics.

Movement Culture’s model has never been as clear, but ground-based locomotion derived first from the floreio of Capoeira, then incorporating elements from Russian Martial Arts and contemporary dance, along with handbalancing, gymnastics strength work and various joint integrity protocols was his initial base.

This is a relatively arbitrary walk through different movement modalities which lacks a clear organizing structure (Note: Ido has spoken recently about an update to his model that is more congruent which I became aware of after starting this article).

Being a Generalist and the Problem of Relevance.

The problem the generalist faces is that one can not train everything, so how does one choose what to train?

The problem with both the CrossFit and Movement Culture definitions of fitness/movement is that they ignore the problem of relevance.

We can’t just say the best athlete is simply the best averaged over every possible movement task because the fact is that some movements matter more than others.

Consider a classic CrossFit WOD like Fran: 21-15-9 Pull ups and 135 lb thrusters.

This is a brutally intense workout, but whether you fail to finish it, set a pr, or set a world record, at the end you can simply go home. Well, unless you have given yourself Rhabdo.

Consider now escaping a burning building or fighting off a mugger, these movements are far more relevant. They have a direct and immediate implication to your survival.

We don’t just want to be the best at the set of all potential movement tasks, we want to be the best at the most important movement tasks!

An optimal approach to general movement education needs to make the most relevant movement capacities its foundation.

The Core of the Generalist Movement Practice

So what are those movements?

To understand them, I believe we need to consider human movement through the frames of evolutionary theory, motor learning theory, and play research.

Before we dig deeply into the science though, I want to offer a simple observation: Some movement skills transfer better than others.

For example, in my experience, parkour athletes who have an extensive history of training in nature tend to adapt quickly to training in urban environments, whereas athletes who primarily practice in urban environments tend to have more difficulty adapting their movement to natural environments.

One of the most accomplished students I have worked with in parkour was Justin Sweeney.

Just a couple months into training with me, Justin and I climbed some trees together and he shocked me by jumping between trees 30 feet off the ground.

It turned out that Justin had spent a huge amount of his childhood climbing trees in Hawaii.

Justin and I would go on to train in the tree trees together almost every week for about a year, and I was always impressed at his abilities.

Justin eventually went on to focus primarily on urban training, and quickly became one of the best speed competitors in the world, podiuming or winning many large competitions.

What was particularly interesting was his ability to sequence complex climbing challenges on sight that very few other athletes could solve, let alone do so rapidly.

Conversely, over the years I have had many world class parkour athletes join me to train in the trees and I have been shocked at how often their abilities are substantially limited in the new environment.

This is just anecdotal but I think it points to something important:

There are asymmetrical relationships between movement practices, and if we want to create the most robust movement capacity, we need to leverage those relationships.

Identifying the Central Hubs of Movement Practice

CrossFit prioritized gymnastics in its early conceptualization with the observation that gymnastics offered the greatest transfer benefit to other sports of any specific discipline.

CrossFit was proposing that gymnastics was what is referred to in the scientific literature as a Donor Sport, i.e., the skills and qualities developed within it have high carry over to other disciplines.

In fact, CrossFits claim was that gymnastics was the ultimate donor sport.

Gymnastics itself offers us a very clear example of asymmetrical transfer between movement disciplines.

In Men’s gymnastics there are six disciplines based on six specific apparatus: The floor exercise, vault, high bar, parallel bars, rings and pommel horse.

It is traditionally understood in gymnastics that floor exercise is the most transferable of these six disciplines.

A quick look at past Olympic and world champions supports this conclusion.

Comparing the past twenty world all-around champions, the single most common discipline that an all-around champion wins is indeed the floor, followed by parallel and high bars, vault, rings and finally pommel horse.


-Floor Exercise
-Parallel Bars
-High Bar
-Pommel Horse

Total Champions


Medals on Apparatus


Now this is an informal look at a small sample so it’s hardly definitive; feel free to dig deeper, but as a former gymnastic coach this is very much what I would expect to see.

If you understand the theory and mechanics of gymnastics, the primacy of floor exercise seems unavoidable, as does the relatively low ranking of rings and pommel horse.

Many skills in gymnastics are possibly only on the floor initially, or are built up from progressions that start on the floor.

For instance the high bar, parallel bar, and rings all include the back giant swing, the first progression of which is a back roll, which is followed by a back extension roll.

The fundamental mechanics of the skill have to be built in part on the floor exercise.

It is also on the floor that handstands, handstand pirouettes, rolls, cartwheels, handsprings, and flips are most easily developed, and these are fundamental to every other apparatus.

The pommel horse, on the other hand, requires swinging the body around the hands in a support position. These are unique body mechanics that only have analogues in a small number of floor and parallel bar skills, which makes the pommel horse the most specialized and least transferable skill in gymnastics

The floor exercise is the foundation of gymnastics because it donates specific skills and general abilities to every other discipline, and does so asymmetrically. For instance floor offers more transfer to parallel bars than vice versa; indeed, developing high-level parallel bar skill is almost impossible without a solid foundation on the floor.

In contrast, pommel horse is the least foundational apparatus because it donates the fewest skills to any other apparatus.

The ideal foundations of any movement practice needs to identify those hubs of skill with many near positive transfers.

We can conceptualize gymnastics as in this figure.

The lines represent relationships, arrows represent where those relationships are asymmetrical, and the size of the nodes represent their overall importance in development of the most successful athlete.

I believe this is similar to how movement can be more generally conceptualized— with nested asymmetrical nodes.

The task of the generalist is to find and prioritize the biggest central nodes within the network.

Let’s consider another example of asymmetrical transfer.

Bouldering vs Calisthenics

Consider twin brothers of equivalent athletic history. They can both climb a V1 boulder problem and do five pull ups in a row.

One decides to focus on developing technical skill in bouldering, so he trains by climbing hard for an hour a day, three days a week.

The other decides he needs to get strong for bouldering, so to condition himself he does calisthenics, also for one hour a day, three days a week, focusing primarily on the pull up.

I don’t know of any proper scientific experiments like this, but based on my experience, the boulderer will be much more similar in pull up strength than the calisthenics athlete will be in climbing ability.

I wouldn’t be surprised to see the boulderer improve by 2-3 grades and add around 10-15 pull ups to their max. While the calisthenics brother might add 15-20 pull ups but find this only allows them to improve by a single grade in climbing.


This is because, while both modalities are a powerful stimulus for upper body pulling strength, bouldering is a more complex activity that challenges far more aspects of coordination, mobility and strength.

During a bouldering session, the lats, biceps, brachialis, grip and core will be substantially challenged, often nearly maximally, especially if dyno’s and campusing are explored. The demands on these muscles will substantially overlap with pull up training.

In addition to pulling strength, bouldering will develop leg strength, mobility, hand eye coordination, finger strength and fully body positional awareness.

Simply training pull ups will leave many of these adaptations on the table.

It’s easy to see that the ceiling for overall improvement in general athleticism is going to be much higher for the boulderer.

I have done a similar experiment on myself. 

In 2006, I spent three months climbing three different trees as fast as possible with no direct pull up training, and during that time my max pull ups went from 8 to 15.

After an unfortunate shoulder injury in 2009, my max pull ups regressed back to only 8. Three months of dedicated pull up training alongside my parkour training simply returned me to 15 pull ups.

I have seen this repeatedly in parkour as well. For instance, I know of many parkour athletes who have deadlifted 1.5-2x bodyweight on their first day of weight training simply from the strength they acquired from parkour training.

However, the benefits of focusing on bouldering over simple pull up training extends beyond the one to one comparison.

If we go back to the CrossFit boast, the final part was about performing better at tasks that neither of us do.

In this context, which brother’s training will have given him the best set of potential transfers?

Which of the two brothers will perform better in parkour or grappling after their year of focused training?

Again, I would predict it will be the boulderer because bouldering builds a diverse set of motor and perceptual skills that can transfer well to parkour and grappling in a way that simply doing pull ups does not.

There are limits to this, of course, I would expect no difference between the brothers in swimming ability or cycling, but bouldering offers a set of near transfers that are much broader than simple calisthenics.

In this way CrossFit fundamentally missed the boat because they prioritized training abstracted, simple exercises over complex, concrete movement tasks.

In fact they would later update their definition of fitness to “work capacity across broad time and modal domains.”

Despite the new claim to these “broad modal domains”, CrossFit simultaneously abandoned almost all the tumbling, mobility, and hand balancing elements of gymnastics in its program, and dropped its flirtation with parkour.

So if you want to live up to the CrossFit boast, CrossFit will not get you there!

To actually get there, you want to build your movement practice around the skills and disciplines that are most relevant and offer the largest set of positive transfer potentials.

Evolution, Play Research and the 4 Fundamentals of Movement

So what are those practices?

To answer that, let’s pick up the thread of evolutionary biology, and play research; for as Margarette Streicher said.

“Physical education could well be seen as a form of applied biology.”

I have argued in the past that we need to root our conceptualization of fitness in evolutionary biology. The metabolic systems and athletic attributes by which we have defined fitness can not only be seen as subsidiary to movement, but movement itself is an expression of our evolved nature.

We can move in specific ways because we evolved those capacities to allow us to thrive in the specific niches we have occupied throughout our evolution

My hypothesis is that the movements we evolved with over the longest periods of time are those that determined our physiology.

The best expression of our capacity for movement will need to reflect our evolved function.

This is revealed in the way we play.

What is revealed by play

Every species plays in ways specific to its evolutionary niche; kittens especially love to stalk and pounce, and dogs love to chase and tug because those are patterns specific to the hunting strategies of their species.

Human play similarly reveals the fundamentals of the human movement adaptation.

There are four primary forms of movement play that have been described by Play researchers; these are Locomotor-Exploratory Play, Object-Oriented Play, Rough and Tumble Play and Dance.

Intriguingly, if we rewind the clock back to the early 20th century, we discover another movement for movement generalism that presaged our current one. There we discover thinkers like George Hebert and his Natural Method and the Natural approach to Gymnastics of Margarete Streicher.

Hebert famously organized his training around 10 fundamental movement tasks: Walking, Running, Jumping, Climbing, Moving on all Fours, Swimming, Balancing, Lifting, Carrying, and Self Defense.

If we take the 10 elements of George Hebert, we can divide them into three primary categories which directly correspond to the first three elements of movement play.

Locomotion (Locomotor-Exploratory Play)
-Moving on all fours
-Also Rolling, Acrobatics, Sliding, etc

Manipulation (Object-Oriented Play)
-Also Throwing, Catching, Striking, Chopping, Cutting, Weaving etc

Combat (Rough and Tumble Play)
-Self Defense

Margarete Streicher who I quoted above divided movement again into Locomotion and Manipulation with combative movements seen as a component of manipulation.

So from play, to evolutionary reasoning, to early generalist movement thinkers, we find repeated patterns of these three fundamental movement capacities (four, once we include dance, which by itself is a deep and interesting topic that deserves it’s own article)


A movement generalist needs to understand that it is impossible to train everything, so we need to bias towards the most relevant movements, these are the movements that give us the greatest benefit for our time spent training.

We need to look for asymmetrical transfer relationships, and practices that donate to many other practices. These practices are most likely to be congruent with our specific evolved movement nature & to reflect our inherent patterns of movement play.

Play research, evolutionary theory, and the work of pioneering general movement thinkers all point us to exploratory locomotor play, object oriented games, arts and crafts, interactive and combative movements, and traditional games as the best foundations for that general movement practice.

We will dig deeper into all these relationships in future articles. For now, if you want a deeper dive into evolutionary thinking and the specifics on how we approach the elements of general movement practice, hit the link below to check out our new Return to Movement five day online workshop.

Make The Leap! Become an EMP Insider

Get our best tips and strategies in your inbox to keep you moving plus get notified about special offers and events sent only to our subscribers.