Donald Epstein

The claim

Donald Epstein described a rhythm the spine produces for itself

Donald M. Epstein developed Network Spinal Analysis and named the movement at its center the network wave. The wave is a visible undulation that travels the length of the spine while a person lies still and is touched lightly at a few defined places. Epstein first demonstrated it clinically in 1987. For eleven years it was a clinical observation and nothing more. Then engineers measured it. In 2016 Simon A. Senzon, Epstein and Daniel Lemberger published The Network Spinal Wave as a Central Pattern Generator in the Journal of Alternative and Complementary Medicine, volume 22, issue 7, pages 544 to 556 (Senzon 2016). Their conclusion was narrow and large at once. The network wave carries the mathematical properties of a central pattern generator, and it may be the first one found in the human spine that has nothing to do with walking.

Ask what that commits you to. A central pattern generator is a circuit that produces rhythmic output without rhythmic input. Locomotion has one. Breathing has one. Chewing has one. If the spine carries another, then the spine is not only a relay for signals arriving from the brain and the periphery. It writes. It sets a timing of its own and holds it. That single reframe is why Epstein belongs in a history of tone. Tone is the living state of the nervous system, expressed as tissue tension at every scale. A structure that generates its own rhythm is a structure whose tension is authored from inside rather than only imposed from outside.

1987

The wave was seen for years before anyone could measure it

The network wave entered the record as an observation, not as a hypothesis. Epstein describes it as a visible undulation and rocking of spinal segments elicited by gentle contacts applied in a defined sequence at specific areas. The word elicited is doing real work there. The practitioner does not push the spine through the motion. The contact is light, brief, and placed away from the region that moves. What follows travels. It begins at one end of the column, runs the other way, and in many people it eventually collides with a wave returning. Nobody teaches a patient the movement. It arrives, or it has not arrived yet, and in the protocol described in the 2016 paper it commonly takes several weeks to several months to develop in a given person.

Why does that delay matter? Because it rules out the simplest explanations. A reflex arrives immediately. A learned movement arrives when the person decides to make it. Something that takes months to appear and then appears reliably is a capacity being built, not a response being triggered. Ask what builds over months inside a nervous system. Not muscle length. Not joint position. What builds is regulation, meaning the ability of a system to hold a pattern and to return to it after disturbance. Epstein made a clinical claim that the spine can learn a rhythm. The research program he opened was an attempt to make that claim testable by people with no stake in the answer.

Breig

The starting point was a neurosurgeon writing about tension in the cord

Epstein did not begin with waves. He began with tension. The 2016 paper states that the paradigm of the Swedish neurosurgeon Alf Breig has been fundamental in Epstein teaching since 1985, and it cites Adverse Mechanical Tension in the Central Nervous System, published by Almqvist and Wiksell in 1978 (Breig 1978). Breig argued something mechanical and unromantic. The spinal cord and its coverings are not slack. They are tethered and they are loaded. Move the spine and you change the tension inside the cord itself. Breig was a functional neurosurgeon writing for neurosurgeons. He was not writing about chiropractic and took no part in it. That is worth stating plainly, because the link between the two is often reported as though he endorsed a technique he never encountered.

The anatomy he described gives the tone reading its literal spine. The 2016 paper points to the dural attachments at the foramen magnum, through the arch of the atlas, with varying attachments from the second to the seventh cervical vertebra, and again at the coccyx by way of the filum terminale, and it argues that these create an extra feedback loop. Follow that. If the cord is anchored top and bottom and slung through a moving column, then tension anywhere along the column is tension in the cord. Adverse mechanical cord tension becomes a target rather than a metaphor. The paper describes it as assessed by examining the stabilizing spinal subsystems, which it names as muscular, bony, and neural.

Two waves

The respiratory wave and the somatopsychic wave name two different jobs

Network Spinal describes two waves, and they are not interchangeable. The respiratory wave is the entrainment of breathing motion with spinal motion. In the table of clinical outcomes published in the 2016 paper it sits at Level One, alongside reduction in the parameters of adverse spinal cord tension, release of tension from the spinal stability subsystems, enhancement of spinal and neural integrity, and an increase in basic somatic awareness. The somatopsychic wave is the older name for what the 2016 paper simply calls the network wave. It belongs to Level Two and above, where the paper describes the development and refinement of the wave as the entrainment of two vertebral oscillators, temporally and spatially, through what Epstein called the spinal gateway.

Read the two names as a sequence and the logic becomes plain. First you couple breath to spine. Then you couple spine to spine. Breathing is the rhythm every nervous system already owns, and it is the only autonomic oscillation a person can take hold of at will. Entrain a guarded segment to that rhythm and you have borrowed a working oscillator to teach a silent one. Ask what happens once two oscillators along the same column begin keeping time with each other. You get a system with an internal reference. That is the difference between a spine that is moved and a spine that moves itself, and the whole protocol is built around that difference.

Measurement

Engineers put electrodes on the wave in the fall of 1998

The wave stopped being anecdote when it was handed to people outside the profession. The 2016 paper records that empirical research began in the fall of 1998 and the spring of 1999 with Edmond Jonckheere and his team in electrical engineering at the University of Southern California. They used surface electromyography, treating the paraspinal muscles as a window into the central nervous system, and they built an instrument for the job. The device was published in 2005 as ChiroSensor, an array of non-invasive sEMG electrodes in Studies in Health Technology and Informatics, volume 111, pages 234 to 236 (Jonckheere 2005). Its stated purpose was blunt. It existed to give objective confirmation of the physiological reality as perceived by the practitioner.

What did the signal turn out to be? Not simple. The research applied nonlinear canonical correlation analysis, implemented through alternating conditional expectation, to test the burst activity for nonlinearity. The verdict was that the wave was better predicted by nonlinear models than by linear ones. Ask what that rules out. A linear system responds proportionally, so a bigger input gives a bigger output of the same shape. A nonlinear one does not. It has thresholds, jumps, and regimes. The paper also reports that the signal at higher levels of care shows a distinctly higher level of organization than at lower levels. Organization increased with exposure, and that is a claim you can put a number on.

2004

The mathematics arrived at a central pattern generator

By 2004 the sEMG signal of the network wave was best described using the mathematical modeling for a central pattern generator. The 2016 paper is direct about the significance. This was the first time a CPG was observed in the spine apart from locomotion or respiration, and the authors describe the network wave as the first known segmental oscillatory CPG in humans unrelated to locomotion or arm movement. Central pattern generators are not exotic objects. Neuroscience has described them in lamprey swimming, in mammalian stepping, in breathing, and in chewing. What was new here was the location and the purpose. A rhythm generator distributed along the human spine whose output is not a behavior aimed at the outside world.

Then what is it for? The 2016 paper does not answer that cleanly, and the honest version of this page says so. What the mathematics establishes is the class of thing being observed. A CPG has a signature: rhythmic output that persists without rhythmic drive, entrainment across a range of frequencies, and stable phase relationships between distant elements. The network wave was found to carry that signature. Once you accept the classification you inherit a question. If the spine hosts an oscillator that is not producing locomotion, then its output is regulation rather than movement. Regulation of what? Tension along the column and through the cord. Which is the working definition of tone.

Coherence

Coherence at a distance is the finding that matters most for tone

Coherence is the part of this research that survives every argument about technique. In signal terms, coherence between two recordings means their phase relationship is stable. They are not merely both active. They are keeping time with one another. The 2016 paper reports coherence between the sacral and the cervical spine, across the length of the column, and it reports the boundary that gives the finding teeth. Coherence was demonstrated only with healthy spines. It was not demonstrated in persons with spinal cord injury. The authors go one step further and propose that the detection methods for this coherence may represent a new measure for central nervous system health.

That last sentence is the hinge. A measure is not a treatment. If phase coherence along the spine tracks the integrity of the cord, then you have a non-invasive read on how well a nervous system holds itself together, independent of what anyone does about it. The clinical case behind the boundary is reported in the same paper. A quadriplegic patient experienced the wave below the level of injury with some recovery of sensory and motor function, and the wave traveled through the injury but did not settle into a stationary wave as it does in a healthy person. One case proves nothing about outcomes. It says something precise about the signal, which is a different and more useful kind of claim.

Solitons and gateways

Waves that collide without dispersing describe a spine that holds structure

The wave behaves oddly in a way that is technically informative. The 2016 paper notes that network waves move from the neck down to the sacrum and from the sacrum up to the neck, collide, and do not disperse, which suggests they propagate as solitons. A soliton is a solitary wave that keeps its shape. Two of them can pass through each other and emerge unchanged. Ordinary waves in a lossy medium blur and die. The paper further describes the spinal soliton waves bifurcating into specific regionally located oscillations, and at advanced care it describes a standing wave established between the front and the back of the body together with the development of a third oscillator in the thoracic region.

In 2000 Epstein introduced the spinal gateway to name where the practitioner enters this system. He postulated that gateway regions are generally located in areas of spinal cord attachment, and that they are regions of probability within the intersegmental connective tissues rather than fixed anatomical points. Read that carefully, because it is unusual language for a manual method. A region of probability is not a landmark you palpate. It is a place where the system is most likely to accept an input. This site reads that as a general principle rather than a technique specific one. Every effective input to a nervous system lands where the tissue is currently willing to negotiate, and that is true of a needle, a breath, a word, and a hand.

Stages

The protocol is staged, and each stage names a different quality of tone

Network Spinal is organized into levels, and the 2016 paper publishes what each level is meant to produce. Level One lists awareness and entrainment of respiratory motion with spinal motion, reduction of the parameters of adverse spinal cord tension, release of tension from the spinal stability subsystems, enhancement of spinal and neural integrity, reorganization of spinal structures, development of the ability of the body to self regulate tension, and an increase in basic somatic awareness. Level Two lists stretching of areas of spinal tension, resolution of dominant spinal defense patterns, and refinement of the network wave through the spinal gateway. Advanced care lists the standing wave, the absence of defensive physiology, the third oscillator, and a shift in awareness and consciousness.

Notice what that sequence describes. It starts with tension you can measure, moves to patterns you can name, and ends with organization you can only infer from behavior. It is a ladder of tone, written by a clinician tracking what changed and in what order. Epstein wrote the experiential half of it for patients as well. The 12 Stages of Healing: A Network Approach to Wholeness was published by Amber-Allen in 1994 (Epstein 1994) and runs 225 pages. Healing Myths, Healing Magic followed from the same publisher in 2000 (Epstein 2000) at 216 pages. Epstein also developed Somato Respiratory Integration, a set of breathing and attention exercises paired with those stages, which carries the work between visits.

The outcome data

The wellness research is survey research, and it still says something

Where is the patient evidence? It exists, and its shape should be stated accurately. In 2004 Tonya Schuster, Marnie Dobson, Maritza Jauregui and Robert Blanks published a pair of papers in the Journal of Alternative and Complementary Medicine. Wellness Lifestyles I ran at volume 10, issue 2, pages 349 to 356 (Schuster 2004) and built the theoretical model. Wellness Lifestyles II, at pages 357 to 367 (Schuster and Dobson 2004), tested it against 2,596 patients drawn from 156 offices of practitioners belonging to the Association for Network Chiropractic, with an estimated response rate of 69 percent. Using structural equation modeling, the authors reported a direct path from care to wellness of 0.43 and a further path from care to positive lifestyle change of 0.65.

Now be precise about what that is. It is self report, collected retrospectively, from people already receiving the care. It is not a randomized trial and it does not establish that the wave caused anything. What it does establish is a pattern that deserves an explanation. Across thousands of respondents in more than a hundred and fifty practices, the largest modeled route from care to wellbeing ran through changed behavior rather than around it. An earlier retrospective survey by Blanks, Schuster and Dobson appeared in the Journal of Vertebral Subluxation Research in 1997 (Blanks 1997) and set that program in motion. Anyone reading this literature should hold the numbers and their limits in the same hand.

Epstein and the model

Epstein gave the model the meninges and a fourth subsystem

The Unified Model of Tone credits Epstein with two specific additions to the anatomy of spinal stability. Manohar Panjabi described the spine in 1992 as three interacting subsystems: the passive one of vertebrae, discs and ligaments, the active one of muscles and tendons, and neural control (Panjabi 1992). Epstein extends that frame. He proposes the meninges as a transducing component of the neural control subsystem, which makes dural tension state a regulatory variable in its own right rather than a passive consequence of position.

That one move changes what a spine is. Read Panjabi straight and the dura is packaging, the sleeve the cord arrives in. Read Epstein and the dura is an instrument. The cord is anchored top and bottom and slung through a column that moves, so tension anywhere along that column is registered by the covering itself, and the covering reports it inward. Breig measured that tension. Epstein made it something a practitioner reads and works with, which is the whole premise of the Network Spinal protocol.

The second refinement is a fourth subsystem, the emotional one. It is hypothesized to occupy portions of the same anatomical and physiological space as the other three. It becomes most responsive when those three can no longer dissipate the tension placed upon them. The model does not leave that hypothesis floating. It names the closest established neuroanatomical parallel: the emotional motor system described by Gerard Holstege in 1992 (Holstege 1992). That system projects from prefrontal cortex and caudal brainstem into the spinal cord. It influences sympathetic and parasympathetic tone, establishes specific emotional behaviors, and triggers rhythmical spinal reflexes. Emotion reaches the spine by a known route.

Each subsystem holds and dissipates energy through its own medium. Coordinated, they hold the far from equilibrium responsiveness that life runs on. Uncoordinated, energy that should be moving through the system is stored instead, and the four begin pulling in incompatible directions. Here the model takes its own step. Epstein names four subsystems. The wider literature names biotensegrity, soft tissue continuity, autonomic regulation, and the body’s retention of what has happened to it. They are one multilayered architecture described from different vantage points. Each subsystem is tone observed at a different stratum. The refinements are Epstein’s. The identification of the four with one variable is the model’s, and it is claimed as the model’s.

The ledger

What is measured, what is modeled, and what this page adds

Separate the layers and the whole story stays honest. Measured: a repeatable spinal undulation, recorded on surface electromyography from 1998 onward, reproduced a decade later with different participants, software and equipment, and showing phase coherence from sacrum to neck in healthy spines and not in injured ones. Modeled: the classification of that signal as a central pattern generator, worked out in the engineering literature by Jonckheere and colleagues in Biomedical Signal Processing and Control in 2010 (Jonckheere 2010) and extended there by Martin del Campo and Jonckheere in 2017 (Martin del Campo 2017), then presented for clinicians in the 2016 narrative review, along with the soliton description of how the waves travel and collide. Interpreted: everything concerning meaning, stages, and consciousness. Epstein set out his framework for that layer with Senzon and Lemberger in 2009 (Epstein 2009) under the name Reorganizational Healing, built on the Four Seasons of Wellbeing, the Triad of Change, and the Five Energetic Intelligences.

Here is what this page contributes to the story of tone. Epstein is the point in this history where spinal tension stopped being only something done to the body and became a rhythm the body generates and can be trained to hold. That is a different order of claim from the cortical measurements of Heidi Haavik that precede it here, and it is testable in a way most claims about tone are not. Where this site reads the network wave as evidence that tone is an oscillatory property of the whole nervous system rather than a local tissue state, that reading goes further than the 2016 paper does. The paper classifies a signal. The extension is ours, and we name it as ours.

What the record shows

What Epstein built, in seven dated findings

• 1987. Donald Epstein first demonstrated the network wave clinically, a visible undulation and rocking of spinal segments elicited by gentle contacts applied in a defined sequence at specific areas of the spine.
• 1992. Manohar Panjabi set out the stabilizing system of the spine in the Journal of Spinal Disorders, volume 5, as three interacting subsystems: passive, active, and neural control (Panjabi 1992). Epstein’s refinements are built onto that frame.
• 1992. Gerard Holstege described the emotional motor system in the European Journal of Morphology, volume 30, pages 67 to 79 (Holstege 1992). It projects from prefrontal cortex and caudal brainstem into the spinal cord, influences sympathetic and parasympathetic tone, and triggers rhythmical spinal reflexes.
• Fall 1998. Empirical recording of the wave began with Edmond Jonckheere and his electrical engineering team at the University of Southern California, reading the paraspinal muscles by surface electromyography (Jonckheere 2005).
• 2004. The recorded signal was best described by the mathematics of a central pattern generator, the first such oscillator reported in the human spine unrelated to locomotion (Jonckheere 2010).
• 2009. Epstein, Senzon and Lemberger published Reorganizational Healing in the Journal of Alternative and Complementary Medicine, volume 15, pages 475 to 487 (Epstein 2009), the framework for the interpretive layer of the work.
• 2016. The same journal, volume 22, pages 544 to 556 (Senzon 2016), reported phase coherence from the sacral to the cervical spine in healthy spines and not in persons with spinal cord injury. The authors proposed that measure as an index of central nervous system health.