Internal Training: The Quest for Systemic Efficiency
Updated: 6 days ago
The case for internal training as a cost-reduction and substrate-reorganisation system is made elsewhere in this series. This article examines what lies behind that case: the structural and neurological mechanisms through which internal training reduces the biological cost of movement, and how the form and the interoceptive feedback loop make those mechanisms refine one another across decades of practice.
1. The Problem of Internal Friction
When you move, some of the force you generate goes toward the intended task. The rest is lost to internal friction: unconscious guarding, fascial restrictions that resist movement, misaligned joints that leak force, and muscles working against one another for no useful purpose.
Metabolic friction. Unnecessary muscular activity burns oxygen and fuel without contributing to the task. Guarding that has nothing to do with the movement still has to be paid for.
Mechanical friction. Force that should travel through the body is lost through misalignment, joint instability or breaks in the load-bearing structure, escaping as heat, joint compression or erratic micro-movements rather than useful work.
Recovery friction. Every unit of unnecessary work leaves metabolic byproducts to clear, so part of each recovery period is spent dealing with costs that efficient movement would not have incurred.
Coordination friction. The nervous system becomes cluttered managing competing motor commands, which reduces the precision and timing of everything it does and adds to central fatigue.
These four forms of friction compound one another. Guarding generates metabolic cost that has to be recovered from. Misalignment forces compensatory bracing, which adds more guarding. A cluttered nervous system controls movement less precisely, which produces more misalignment. The total cost is greater than any single source would suggest.
Accumulated Biomechanical Debt amplifies all of this, imposing a baseline cost before any useful work is done. But even without substantial debt, these sources of friction are present in most bodies and most movement. They are not separate problems. Guarding held long enough becomes structural as the fascia remodels around it, and a restricted or misaligned structure, in turn, gives the nervous system reasons to guard. The next two sections approach the same friction from each side: section 2 from the structure, section 3 from the nervous system. The feedback loop described later explains why the effects keep compounding.
2. Structural Efficiency: From Segmented to Integrated Load Distribution
Most bodies function as a collection of parts working semi-independently. The shoulder does its job, the hip does its job, the trunk does its job, but they are not truly integrated. Load is managed locally, force is generated locally, and each joint is held stable by muscular guarding around it.
That segmented organisation has a predictable cost. When load concentrates in particular joints rather than spreading across the whole structure, those joints need constant guarding to stay stable. The knee, the lower back and the shoulder become bottlenecks where force and stress accumulate, and effort that could go toward the task goes toward holding the structure together.
Internal training reorganises this. Through the development of Peng, the trained quality of elastic integration that runs through the whole body, the practitioner learns to distribute load across the entire connective tissue network rather than concentrating it locally. Stability comes increasingly from the organisation of the whole structure, and less from guarding at each joint.
That stability is not passive. Peng involves slight, continuous activation. Muscles can only pull, but in a tensegrity structure, cables that only pull are what hold the struts apart, keeping the whole structure expanded. Peng works in a similar way: muscles around the shoulder blades, back and arms keep up a continuous tension that holds the frame open against the bones, and over a long practice session that work is genuinely tiring. What appears to change with training is where the work goes. The defensive, uneven guarding of the untrained body gives way to effort that is organised, balanced and distributed across the whole structure.
The practical consequence is that less of the muscular work is spent guarding, and more of it serves the structure. When load distributes globally rather than concentrating locally, the joints that were bottlenecks become throughways. Force generated in the legs travels through the trunk because the connected load paths are structurally functional, without each segment needing to be consciously coordinated. The body moves as one system rather than a collection of coordinated parts. This is what practitioners mean when they talk about moving as one piece: a precise account of a mechanical reorganisation, from high-cost, locally guarded stability to lower-cost, integrated structural support.
That integration develops slowly, because it requires genuine structural reorganisation: changes in how the fascial network is loaded, how the joints are habitually positioned, and how the nervous system distributes motor commands across the whole body rather than addressing each segment separately. It cannot be rushed or faked. But once established, it is far more durable than stability built on guarding.
3. Neurological Efficiency: The Art of Song
Structural integration addresses how load is distributed. Song addresses how much of the nervous system's output actually serves the task.
Most movement carries motor noise: the nervous system's side of the friction described in section 1, muscular activity that contributes nothing to the task but is paid for as though it did. It takes several forms:
Unconscious guarding. The most pervasive form: habitual holding that most bodies carry into every activity, such as raised shoulders, a braced lower back or a held breath. It is rarely chosen and usually invisible to the person doing it, and it distorts the structure it is meant to protect.
Anticipatory bracing. The system tenses before movement begins, paying in advance for a demand that may never arrive at the predicted intensity or from the predicted direction, and reducing precision before a single useful action has been taken.
Unnecessary co-contraction. Opposing muscles fire together locally and unevenly, creating internal resistance the prime movers must overcome before any external work is produced. Unlike the organised opposing tension of Peng, it is defensive rather than structural.
Phasic muscles doing postural work. Large, superficial muscles take on sustained stabilising work better suited to the deeper postural muscles, and do it at higher metabolic cost and with faster fatigue.
These patterns are not random. They typically reflect the nervous system's protective response to perceived instability or uncertainty: a safety tax imposed by a brain that does not fully trust the structure it inhabits. If a joint does not feel stable, the brain locks it. If a movement feels uncertain, the system braces before engaging. The protection is real, and it is scaled to how trustworthy the structure feels, which is why it can fall as the structure becomes more integrated.
Song is the progressive removal of this noise. It is subtractive: it takes away guarding, bracing and unnecessary activity, and Peng organises what remains. Removing that noise depends on three capacities developing together: interoceptive resolution fine enough to detect unnecessary tension; the capacity to release that tension even when the body is under load; and structural confidence, the brain's learned trust that the integrated structure will hold without guarding.
As these capacities develop, the changes in movement quality are clear. Movement becomes better coordinated because the nervous system is no longer managing competing motor commands. Timing becomes more precise, because less tension distorts the initiation of action. Force becomes cleaner and more directed, because it is no longer fighting internal resistance. And recovery between efforts is faster, because less unnecessary work was done in the first place.
Song and structural integration are two aspects of the same process. As the structure becomes more trustworthy, the brain lowers the safety tax. As the safety tax falls, the structure can be felt more clearly, which makes further refinement possible. Each step makes the next one available.
4. The Form as Forcing Function
Structural integration and Song are principles. The Laojia Yi Lu form of Chen-style Taijiquan is the mechanism through which they are trained, and it is built to make inefficiency progressively harder to sustain.
The form is often understood as a catalogue of techniques or a moving meditation. While it is both of these, its primary function is a continuous audit of structural and neurological efficiency: a movement sequence designed so that departures from integrated movement become detectable as breakdown. It works as a constraint in the sense used in motor learning, setting the conditions within which the body method has to be found. Its design logic is specific:
From settled postures into movement. The form alternates between postures in which the body settles and the movement that connects them. The principles are easiest to find in the settled postures, where the structure is quasi-static; the movement then asks the practitioner to carry them out of stillness without losing them.
Continuously changing load paths. Complex transitions, performed while maintaining Peng and Song, reveal whether the principles are genuinely integrated or hold only along familiar lines of load.
Spiralling force paths. Spirals resist linear, muscular solutions: produce force linearly and the spiral breaks, so guarding and unnecessary co-contraction become hard to sustain within correct execution.
Shifts between slow and fast. Release of fajin acts as a diagnostic moment, where structural coherence either holds or fragments, exposing any segment that has dropped out of the integrated chain.
Whole-body coordination as a condition of correct execution. If any segment drops out of the chain, the movement fragments in ways that become apparent to the practitioner and to an informed observer.
The form is not automatically self-correcteing. A practitioner who moves mindlessly can perpetuate their habitual patterns of bracing and inefficiency indefinitely. But a practitioner who attends carefully to internal feedback will find that the form continually surfaces the next layer of inefficiency to be addressed.
This is why the same form can be practised for decades without becoming redundant. It functions as a diagnostic tool whose resolution increases with the practitioner's sensitivity, always one level ahead of where they currently are.
5. The Interoceptive Feedback Loop
Structural integration, Song and the form that trains them do not develop independently. They develop together, driven by a continuous, self-reinforcing feedback loop that is one of the most distinctive features of internal training.
The loop works like this. Interoceptive sensitivity detects unnecessary tension, structural misalignment, or Peng being lost. The practitioner brings their awareness to address the issue. The movement becomes smoother, less effortful and more coherent. With less interference, the interoceptive signal becomes clearer, and subtler inefficiencies that were previously below the threshold of perception become detectable.
Each cycle makes the next more sensitive, because the body plays two roles at once. It is the output, the quality of the movement being refined, and it is the input, the sensing apparatus doing the refining. As unnecessary tension falls away, the movement improves, and the same tissues become better instruments for feeling what remains. Each gain in efficiency sharpens the perception that makes the next gain possible, so the two develop together in a spiral.
This is why progress in internal training feels different in kind from progress in conventional skill training. Skill training improves by refining movement against its results, guided mainly by proprioception: where the body is, what it did, whether it worked. In internal training, improvement is largely a function of perceiving more: detecting subtler layers of tension and inefficiency from the inside, through interoception, and removing them.
It is also why refinement can continue across decades. The largest gains probably come relatively early in practice, and age eventually sets its own limits, but interoceptive resolution appears to remain refinable for a very long time: practitioners with forty or fifty years of practice describe still finding new layers of inefficiency in movements they have performed tens of thousands of times. The loop keeps spiralling inward, always finding a new level of refinement.
6. How the Mechanisms Combine
Structural integration and Song are two sides of one process. Structural integration, expressed through Peng, replaces local guarding with organised, whole-body support; Song removes the neurological noise layered on top of it. The form is the method through which both are trained, continually surfacing what remains to be refined, and the interoceptive feedback loop drives that refinement, compounding its effects across years and decades of practice.
The result is a system that becomes progressively more efficient. In practice, the work goes increasingly into the structure rather than into guarding; outside practice, the reduced guarding persists as a baseline that expresses itself in every context the body meets. Underlying it is a shift toward stronger autonomic regulation: less of the defensive arousal that most bodies carry by default, and a nervous system that spends less of its resources simply maintaining the status quo. The same reorganisation also makes the body progressively more resistant to injury.
These efficiency gains are also unusually durable. Training builds two kinds of thing: capacity, such as muscle mass and aerobic fitness, and organisation, the coordination and skill through which capacity is used. Capacity fades within weeks to months without continued input and declines with age regardless; organisation lasts far longer, which is why someone who learned to swim as a child can still swim decades later.
The efficiency internal training builds is largely organisation: interoceptive awareness developed over years of practice, and fascia remodelled over a similar timescale. Both are hard-won and slow to revert, so the gains are resilient and long-lasting, holding up better than capacity as the body ages.
Written with Claude Opus 5.5
Comments