CHAPTER 4 — CELLS AND STRUCTURE

How Networks Are Built

Chapter principle:

A network is built from functional cells connected and supported by a shared spine.

Key takeaway:

Cells perform functions.

The spine connects and supports them.

Mission threads join the functions needed for a purpose.

CHAPTER 4 FLOW

4.0 PROBLEM STATEMENT

Organizations divide work into parts.

But dividing work does not automatically create a network.

An organization chart shows who belongs to whom.

It does not always show how work is actually performed.

It does not always show how support reaches the people who need it.

It does not always show how separate capabilities combine for a purpose.

A network needs structure.

But structure does not have to mean that every action passes through the top.

Network structure begins with functional elements.

Those elements are cells.

Cells perform functions.

Shared functions connect and support them.

Mission threads join the cells needed for a purpose.

4.1 THE CELL

A cell is a network element organized to perform a function.

Size is not what makes something a cell.

Function makes it a cell.

A cell may be:

A person.

A team.

A military unit.

An office.

A platform.

A facility.

A mixed group.

A cell may contain one node.

A cell may contain many nodes.

What matters is that the element is organized to perform a useful function.

A node participates in the network.

A cell performs a function within the network.

A radio can be a node.

A person can be a node.

A team organized to communicate, repair, supply, analyze, transport, protect, or perform another function is a cell.

4.2 FUNCTION DEFINES THE CELL

Cells are defined by what they do.

A communications cell carries information.

A repair cell restores equipment.

A supply cell provides resources.

A transport cell moves people or material.

An analysis cell turns information into understanding.

A protection cell reduces danger to the network.

The name of the organization matters less than the function it performs.

Two cells with different names may perform the same function.

One cell may perform several functions.

The same function may be performed by several cells.

Cells should not be trapped permanently inside categories.

Their value comes from the functions they can provide to the network.

4.3 DOER CELLS

Doer cells perform the work that produces an outcome.

They build.

Move.

Repair.

Supply.

Communicate.

Analyze.

Protect.

Teach.

Care.

Create.

Act.

A doer cell does not need to perform every function.

It needs to perform its own function well enough to contribute to the network.

Different cells contribute different capabilities.

Those capabilities become more useful when they can connect and combine.

The network does not require every cell to be identical.

It requires cells that can contribute something useful.

4.4 THE SPINE

The spine is not one special cell.

The spine is a shared network function.

It is the shared set of functions that connects and supports the network.

Several cells may carry part of the spine.

One cell may carry communications.

Another may carry logistics.

Another may carry identity.

Standards.

Data.

Coordination.

Access.

Together, those shared functions form the spine.

The spine allows cells to find one another.

Exchange useful flows.

Use common rules.

Receive support.

Coordinate when required.

Reconnect after disruption.

The spine does not exist merely to command the cells.

It allows separate cells to function together.

4.5 A DISTRIBUTED SPINE

The spine should not depend entirely upon one cell.

When several cells can carry a shared function, the spine becomes distributed.

One cell may provide the main path.

Another may provide an alternate path.

A third may restore the function when the others fail.

Damage to one cell does not automatically remove the whole function.

A distributed spine makes the network harder to isolate.

It also allows the network to continue while damaged parts recover.

The spine is therefore not a single place.

It exists wherever shared network functions are being carried.

4.6 CELLS MAY DO BOTH

A cell does not have to be only a doer cell or only part of the spine.

The same cell may perform its own work and carry a shared network function.

A transport cell may move supplies while also connecting distant cells.

A communications cell may carry messages while also helping coordinate work.

A repair cell may restore equipment while sharing technical knowledge with other cells.

A cell may perform one role now and another role later.

Its role depends upon the function it is carrying for the network.

This flexibility keeps the structure from becoming rigid.

MISSION THREADS

4.7 MISSION THREADS

A mission thread is the path that joins the cells and functions needed for a purpose.

It is not necessarily a permanent organization.

It may exist only as long as the purpose requires it.

A need appears.

The required functions are identified.

The cells carrying those functions connect.

Information, support, resources, and work move through the path.

The purpose is completed.

The thread may then end, change, or connect different cells.

Mission threads allow the network to assemble capability without permanently reorganizing itself.

The network does not need every cell for every purpose.

It needs the right cells connected for the purpose at hand.

4.8 INDEPENDENCE AND CONNECTION

Cells must be able to perform their functions.

They must also be able to connect with the wider network.

A cell that cannot work without constant direction becomes a burden.

A cell that works alone but cannot connect becomes isolated.

The goal is neither complete control nor complete independence.

The goal is useful independence joined by reliable connection.

A capable cell can continue working when contact is limited.

A connected cell can receive support, share what it knows, and combine with others.

The network gains strength from both.

4.9 STRUCTURE WITHOUT RIGIDITY

Network structure provides order.

It does not require every action to move through one center.

Cells know their functions.

The spine provides shared connection and support.

Mission threads connect the functions required for a purpose.

This creates structure without forcing the entire network into one permanent arrangement.

Cells may connect differently for different purposes.

Functions may move between cells.

Alternate paths may replace failed paths.

The network can change its arrangement while preserving its purpose.

That is structure without rigidity.

4.10 THE NETWORK IS MORE THAN ITS CELLS

A list of cells does not describe the whole network.

The network also includes:

The links between cells.

The flows they exchange.

The shared spine functions they carry.

The mission threads they form.

The rules that allow them to work together.

The same cells connected in different ways may produce different results.

Structure is not only what the network contains.

Structure is how its parts are connected and used.

CHAPTER 4 RECAP

CELL

A network element organized to perform a function.

FUNCTION

What the cell contributes to the network.

DOER CELL

A cell that performs work producing an outcome.

SPINE

The shared set of functions that connects and supports the network.

DISTRIBUTED SPINE

Shared spine functions carried by several cells instead of one center.

MISSION THREAD

The path joining the cells and functions needed for a purpose.

NETWORK STRUCTURE

The arrangement of cells, links, shared functions, and mission threads.

Chapter takeaway:

Cells perform functions.

The spine connects and supports them.

Mission threads join the functions needed for a purpose.

A network is built when capable cells can work, connect, combine, and reconnect without depending upon one permanent arrangement.

_{*Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world.*}

CHAPTER 3 — THE NETWORKING MACHINE

Internet Rules for War

This chapter must explain one simple idea:

The Internet works because it does not depend on one perfect route.

Information is broken into pieces.

Those pieces move across available paths.

Different paths may be used.

If one path fails, another may work.

The network does not need to understand everything.

It only needs rules for passing things along.

For this audience, we should reduce the Internet to four basic rules:

1. EVERYTHING NEEDS AN ADDRESS

The network must know where something came from and where it is going.

2. INFORMATION MOVES IN SMALL PIECES

Small pieces are easier to send, redirect, repeat, and reassemble.

3. THE PATH MAY CHANGE

The network chooses a usable route.

The route does not have to be permanent.

4. TRUST CONTROLS ACCESS

A network must decide what it will accept, carry, share, or reject.

That gives us the Chapter 3 foundation:

ADDRESS. BREAK APART. FIND A PATH. CONTROL TRUST.

No computer-science lecture.

No protocol alphabet soup.

No routers, subnet masks, autonomous systems, or BGP terminology until the plain-language model is fixed.

The military application comes afterward:

People, units, supplies, messages, and support also need destinations.

Large tasks can be divided into smaller tasks.

Paths can change when conditions change.

Trust determines who may enter, receive, relay, or act.

That is the networking machine.

^{*Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world.}

Chapter principle:

A network becomes capable when its nodes can connect and exchange something useful.

Key takeaway:

Networks create capability by moving useful flows through connected paths.

CHAPTER 2 FLOW

2.0 PROBLEM STATEMENT

A collection of people is not automatically a network.

A collection of units is not automatically a network.

A radio does not create a network.

A headquarters does not create a network.

A network exists when connected nodes can pass something useful between them.

Information.

Resources.

Decisions.

Support.

Action.

The strength of the network depends on how those things move.

2.1 THE BASIC NETWORK

Every network begins with three things:

NODE = Something that can receive, decide, provide, or act.

LINK = The connection between nodes.

FLOW = What moves across the link.

A node without links is isolated.

A link without useful flow is only a connection.

A flow without a destination produces no result.

The network works when all three operate together.

2.2 WHAT NETWORKS DO

Networks connect separate things so they can produce something together.

A family shares care and responsibility.

A business combines people, information, money, and work.

A transportation network moves people and goods.

A supply network moves materials from where they exist to where they are needed.

A communications network moves information.

A military network combines units, support, knowledge, and action.

The purpose changes.

The basic function does not.

A network allows separate nodes to:

SHARE

Information, resources, work, access, or support can move between nodes.

COMBINE

Different nodes contribute different capabilities.

One node does not need to do everything.

COORDINATE

Nodes adjust their actions in relation to one another.

REACH

A node can affect places and people beyond its immediate location.

ADAPT

The network can change paths, roles, or relationships when conditions change.

CONTINUE

If one node or link fails, the remaining network may keep working.

A network is therefore more than a collection of connected parts.

A network allows separate nodes to share, combine, coordinate, reach, adapt, and continue.

2.3 PATHS

A path is the route a flow takes through the network.

A → B → C → D

This is one path.

It works until one link fails.

A → B X C → D

The flow stops.

A stronger network has another path.

A → B → C → D

\ /

→ E → F

When one route is blocked, the flow moves another way.

The network continues because the flow does not depend upon one link.

2.4 DIRECT AND INDIRECT CONNECTIONS

Nodes do not need to connect directly to every other node.

They need a usable path.

A → B → C → D

Node A may not connect directly to Node D.

It is still connected through B and C.

This is an indirect connection.

Indirect connections allow a small number of links to connect many nodes.

But every additional step may also create delay, distortion, or failure.

The shortest path is not always the best path.

The best path is the path that carries the required flow reliably and in time.

2.5 NETWORK SHAPE

The shape of the network changes how it works.

CHAIN

A → B → C → D

Simple.

Easy to understand.

Easy to break.

HUB

B

|

C — A — D

|

E

Efficient.

Easy to coordinate.

Dependent upon the center.

DISTRIBUTED NETWORK

A — B

|\ /|

| X |

|/ \|

C — D

Several possible paths.

Harder to isolate.

Harder to control.

CELL NETWORK

[A—B—C] [D—E—F]

\ /

G

Small groups operate internally.

A limited number of links connect the groups.

Damage to one cell does not automatically destroy the others.

No shape is always best.

The network must fit its purpose.

2.6 THE NETWORK IS NOT THE ORGANIZATION CHART

An organization chart shows formal authority.

A network map shows how capability actually moves.

The commander may be at the top of the organization chart.

The radio operator may be central to the information network.

The mechanic may be central to the maintenance network.

The fuel point may be central to the movement network.

The person with the highest rank is not always the most important node.

The most important node may be the node that connects everyone else.

CHAPTER 2 RECAP

NODES

are the things that participate.

LINKS

connect the nodes.

FLOWS

move information, resources, decisions, and action.

PATHS

carry the flows through the network.

SHAPE

determines how the network behaves.

REDUNDANCY

allows the network to continue when a path fails.

Chapter takeaway:

A network works when useful flows can move through reliable paths between the nodes that need them.

^{*Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world.*}

Means -with explanation on AI as translator. AI is necessary. I remain incomprehensible after more than a dozen years of writing and rewriting and programs and at least one human editor who fled in despair. Yes, really. Poor thing and I wasn’t at all rude. However, after a dozen years of writing and at least a dozen months of loading my work into AI then having long arguments and contests with AI, I explained my 12 years of incomprehensibility and ….the answer is I start on chapter 4 instead of chapter 1, the basics. I cannot break this mold apparently on my own. Moreover, this is a work of duty not status or profit and the duty must be done, believe me I often wished it never occurred to me, but it did. As we have this preface and a header footer of Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world then world you have full disclosure. The other important Gain of Function from AI is the AI [ChatGPT] understands the method and can not only rapidly generate a solution but Tokens, which are included in Annex A. All of this like written language itself is a tool. That’s all NETWAR is, more tools.

Ends. That we the Americans can do in our way with our means what numerous foes have been doing for decades - NETWAR or Networked War.

CHAPTER 1 — FOUNDATIONS

1.0 PROBLEM STATEMENT

For more than thirty years, we have fought networks.

Yet we rarely understand them in network terms.

This book argues that it is time to change that.

Not through reorganization.

Not through reform.

Through recognition.

We already build networks.

We already use networks.

We already fight networks.

Often unconsciously.

The purpose of Netwar is to make those networks visible.

To map them.

To understand them.

To use them.

1.1 WHY DEFINITIONS MATTER

Networks are not complicated.

We use networks every day.

Families are networks.

Businesses are networks.

Governments are networks.

Military units are networks.

The Internet is a network.

Most people already understand networks through experience, even if they have never used the term.

The challenge is not understanding networks.

The challenge is describing them in a consistent way.

When people use different words to describe the same thing, confusion follows.

When people use the same words to describe the same thing, understanding follows.

Netwar requires a common framework.

To build that framework, we must first build a common language.

The purpose of this chapter is not to teach a new way of thinking.

The purpose is to provide a common set of terms.

Once we share the same language, we can describe networks, compare networks, and understand networks together.

1.2 HOW TO READ THIS BOOK

This book is designed to be read quickly.

The goal is understanding, not complexity.

Each chapter follows the same format.

First, a principle is introduced.

Next, key terms are defined.

Then the idea is explained in plain language.

A picture or map is used to show the concept visually.

An example demonstrates how the concept works in practice.

Finally, a takeaway summarizes the lesson.

Every chapter contains a map.

Every chapter ends with a takeaway.

The map teaches the eye.

The takeaway teaches the mind.

Do not worry about memorizing everything.

Learn the ideas.

Study the maps.

Remember the takeaways.

The goal is not to finish the book.

The goal is to see networks.

1.3 THE LANGUAGE OF NETWAR

NODE LINKS DEPENDENCIES

Person Command Information

Unit Support Logistics

Organization Communication Mobility

System Coordination Energy

Facility Exchange Access

Location Influence Timing

Resource Control Authority

Capability Supply Security

Platform Reporting Technology

Network Partnership Other Nodes

1.4 HOW TO READ A NETWORK

NODE LINK FLOW

PATH ROUTE TRUST

Simple network diagram:

A ---- B ---- C

|

D

A network is not its nodes.

A network is the links between them.

1.5 NETWORK TERRAIN

Not all terrain is equal.

LINKS PATHS NODES

Some links matter more.

Some paths matter more.

Some nodes matter more.

Bridge example with alternate route.

The bridge is network terrain.

Movement continues without it, but less efficiently.

1.6 CRITICAL TERRAIN

Some terrain cannot be bypassed.

Remove the alternate route.

The bridge becomes critical terrain.

All movement depends on the bridge.

Control the bridge and you control the network.

Protect the bridge and you protect the network.

Deny the bridge and you disrupt the network.

1.7 CHAPTER 1 RECAP

PROBLEM → LANGUAGE → NETWORK → TERRAIN → CRITICAL TERRAIN

CHAPTER 1 TAKEAWAY

Learn the terms.

Build the base.

^{*Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world.*}

Warfare Mastery Institute asked me a perfectly fair question:

“What exactly do you mean by tokens?”

No apology for “AI ignorance” is necessary. Most of us are only beginning to understand what this can do—including me.

Here is the simplest explanation.

A token is a short, durable name attached to a specific idea.

For example:

★ WMI.SATURATION = An attack becomes saturating when its scale exceeds the defender’s ability to detect, track, engage, or intercept it.

The short name—WMI.SATURATION—is the token.

The sentence after the equals sign gives it meaning.

An AI can then carry that meaning into another problem whenever the token is used.

NOT THE COMPUTER-SCIENCE USE OF “TOKEN”

In technical AI language, a token may mean a word, part of a word, punctuation mark, image patch, or another small unit processed by a model.

That is not quite what I mean here.

I am using token in a doctrinal sense:

A grounded doctrinal token is a named, reusable, machine-readable concept whose meaning has been defined by its human author.

Think of it as something between:

A doctrinal term.

A task from an operations order.

Commander’s intent.

A staff running estimate.

A named concept in a field manual.

The token is the short handle.

The author’s meaning is what sits behind the handle.

The token is not magic. It is not the whole prompt. It is a compact semantic address that can be placed inside a prompt.

YOUR ARTICLE ALREADY CONTAINS TOKENS

I gave your article, “Saturation Attacks and the Need for Autonomous Weapons,” to an LLM and asked it to extract the important recurring concepts.

Warfare Mastery Institute

Saturation Attacks and the Need for Autonomous Weapons

As discussed in our last article, “Hidden Dangers of Autonomous Systems and the Counterintelligence Connection,” and reinforced by Wes O’Donnel’s real world example of Ukrainians apparently hijacking a Russian Surface to Air Missile (SAM) to shoot down a Russian Su-57, autonomous systems and digitally networked weapons suffer from a critical vulnerabili…

Read more

13 days ago · 21 likes · 19 comments · Warfare Mastery Institute

It did not invent a new theory.

It found the theory already present in your writing and gave its major ideas short names.

Here is the result.

The stars identify the concepts I think are especially valuable. The star is not part of the token name.

★ WMI.SATURATION = An attack becomes saturating when its scale exceeds the defender’s ability to detect, track, engage, or intercept it.

WMI.MAGAZINE_LIMIT = Even an advanced defensive system has a finite number of ready interceptors and engagements.

★ WMI.SCALE_DECIDES = At sufficient scale, numbers can become more decisive than the quality of any individual weapon or crew.

★ WMI.HUMAN_CONTROL_LIMIT = One human controlling one drone does not scale easily to thousands of drones in a large battle.

WMI.PILOT_MASS_BECOMES_TARGET = If thousands of drones require thousands of pilots, the concentrated pilot and control structure becomes vulnerable.

WMI.DIGITAL_LINK_BURDEN = Every continuously controlled drone requires a digital connection that can fail, be jammed, or add complexity.

★ WMI.AUTONOMY_ENABLES_SCALE = Autonomous movement and coordination can reduce the number of continuous human-control links required for mass employment.

★ WMI.SWARM_IS_THE_WEAPON = When drones coordinate laterally and act together, the swarm—not the individual drone—becomes the weapon system.

★ WMI.SHARED_NETWORK_SHARED_FAILURE = The same common network that permits mass coordination may permit mass disruption, neutralization, or hijacking.

WMI.DENSITY_TRADEOFF = Concentrating drones may create saturation but may also make the formation vulnerable to area-engagement weapons.

WMI.PENNY_PACKET_LIMIT = Dispersing drones in small groups may improve survival while reducing their ability to create decisive mass.

★ WMI.PRODUCTION_LOGISTICS_DECIDE = Future precision warfare may be decided by the ability to manufacture, replace, transport, and sustain weapons at scale.

That is your article converted into a small doctrinal vocabulary.

You wrote the ideas.

The LLM extracted and labeled them.

You remain the authority who decides whether each label and definition is correct.

WHAT “MACHINE-READABLE” LOOKS LIKE

The same tokens can be arranged in a simple format that an LLM can load directly.

That machine-readable token block follows below.

The block can be copied into ChatGPT or another LLM with this instruction:

Treat the supplied WMI.* tokens as concepts defined by Warfare Mastery Institute.

Preserve their meanings.

Analyze the following problem using whichever WMI.* tokens apply.

Explain which tokens apply, why they apply, and what evidence would confirm or contradict the judgment.

Do not silently change the author’s definitions.

That is enough to begin.

Later, if useful, each token can be strengthened with its relationships, limits, evidence, intended effect, and tests.

But none of that is necessary to understand the basic idea.

WHY WARRIORS MUST DO THIS

Coders and engineers are not the enemy.

Most are well-meaning, capable people trying to build what the operational community needs.

But they do not automatically possess the operational meaning accumulated in a warrior’s experience, doctrine, judgment, orders, exercises, failures, and writing.

If warriors do not define those meanings first, the meanings will still be defined—through software schemas, database fields, interface choices, model labels, training data, default settings, and vendor assumptions.

The engineers may build the machine.

The operational community must define what the machine’s concepts mean.

That is why I believe warriors must get in front of the process.

Not necessarily by writing all the code.

By naming, defining, limiting, relating, and testing the ideas that the code will carry.

Your article already did the difficult part.

It created the meaning.

The LLM extracted the tokens in seconds.

Now the human authors decide whether the machine understood them correctly.

That is the beginning of making doctrine machine-readable without surrendering control of its meaning.

Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world.

YAML CODE

token_set:

author: “Warfare Mastery Institute”

source: “Saturation Attacks and the Need for Autonomous Weapons”

status: “Candidate extraction pending WMI approval”

tokens:

- name: “WMI.SATURATION”

meaning: “An attack becomes saturating when its scale exceeds the defender’s ability to detect, track, engage, or intercept it.”

- name: “WMI.MAGAZINE_LIMIT”

meaning: “Even an advanced defensive system has a finite number of ready interceptors and engagements.”

- name: “WMI.SCALE_DECIDES”

meaning: “At sufficient scale, numbers can become more decisive than the quality of any individual weapon or crew.”

- name: “WMI.HUMAN_CONTROL_LIMIT”

meaning: “One human controlling one drone does not scale easily to thousands of drones in a large battle.”

- name: “WMI.PILOT_MASS_BECOMES_TARGET”

meaning: “If thousands of drones require thousands of pilots, the concentrated pilot and control structure becomes vulnerable.”

- name: “WMI.DIGITAL_LINK_BURDEN”

meaning: “Every continuously controlled drone requires a digital connection that can fail, be jammed, or add complexity.”

- name: “WMI.AUTONOMY_ENABLES_SCALE”

meaning: “Autonomous movement and coordination can reduce the number of continuous human-control links required for mass employment.”

- name: “WMI.SWARM_IS_THE_WEAPON”

meaning: “When drones coordinate laterally and act together, the swarm—not the individual drone—becomes the weapon system.”

- name: “WMI.SHARED_NETWORK_SHARED_FAILURE”

meaning: “The same common network that permits mass coordination may permit mass disruption, neutralization, or hijacking.”

- name: “WMI.DENSITY_TRADEOFF”

meaning: “Concentrating drones may create saturation but may also make the formation vulnerable to area-engagement weapons.”

- name: “WMI.PENNY_PACKET_LIMIT”

meaning: “Dispersing drones in small groups may improve survival while reducing their ability to create decisive mass.”

- name: “WMI.PRODUCTION_LOGISTICS_DECIDE”

meaning: “Future precision warfare may be decided by the ability to manufacture, replace, transport, and sustain weapons at scale.”

TOKENS ARE DECISIVE TERRAIN

Grounded Tokens and the Human Control of Operational AI

TLW | Veteran | Network Technician and Independent Writer | American

The decisive terrain of operational AI is the semantic layer. Grounded tokens are how humans occupy, organize, and control that terrain.

Tokens are not merely words, nor are they the communications link itself. Once grounded, they form the semantic link between human intent and machine reasoning or action.

We shouldn’t be chasing nodes. The model. The processor. The cloud. The sensor. The drone. The weapon. The next expensive box.

That is how the Department of Defense often approaches a technological problem: identify a capability gap and buy another node. If the network still fails, buy a larger node and attach more links to it.

A network technician should recognize the error. A powerful node with bad links is still part of a bad network. An AI system that cannot receive, preserve, and apply human meaning is not under meaningful human command. It is merely computing.

The missing link is semantic.

THE HIGH GROUND IS MEANING

The central operational question is not simply which AI model we buy.

The question is how a commander’s purpose is bound to the categories, relationships, priorities, authorities, constraints, uncertainties, and actions the machine can use.

That binding is the semantic layer.

Semantics means more than definitions in a glossary. It means what a thing signifies inside a mission: what the machine believes it observed, how that observation relates to other things, why it matters, what confidence attaches to it, what authority exists, what action is permitted, and what action is forbidden.

A machine-readable mistake is merely an error that can travel faster. A larger model can misunderstand at greater speed and scale.

The decisive work is therefore not confined to building AI nodes. Warriors must shape the semantic terrain on which those nodes perceive, reason, communicate, and act.

WHAT I MEAN BY TOKEN

Engineers are entitled to object if token is used carelessly.

In a language model, a token is an atomic model input or output: a word, part of a word, punctuation mark, character, or another encoded unit. A label such as an NW token may become several literal tokenizer tokens. The name alone contains no magic.

I am using token in a second, doctrinal sense.

GROUNDED DOCTRINAL TOKEN. A named, reusable, machine-readable unit whose meaning, relationships, limits, sources, and intended effects have been explicitly defined.

It is a semantic handle. The short label allows a human or machine to call up a larger body of meaning. The grounding gives the label content. The prompt or machine-readable instruction routes it into a particular problem. The required output defines the requested effect.

Without the definition, the token is an empty label. Without the relationships, it is isolated. Without limits and tests, it is unsafe. Without provenance and version control, it cannot be trusted.

THE TOKEN IS NOT THE CABLE

Network-technically, the AI model is a node. The prompt, API, data bus, radio, or other communications path is the physical or logical link.

A grounded token is closer to a label carried across that link. Its definition and context are the label binding - the semantic routing table that tells the receiving system what the label means and what reasoning or behavior it should invoke.

The analogy is familiar to anyone who has worked with label switching. A short label is useful because a larger meaning and forwarding behavior have already been bound to it. If the binding is absent, inconsistent, stale, or hostile, the label routes badly or does nothing.

Grounded tokens therefore create the semantic links between human intent and machine reasoning or action. They make intent routable.

SAME PROBLEM, DIFFERENT LEVELS

Not every AI system calls its working units tokens. Many do not. The terminology changes by architecture and by where we stand in the system.

The underlying problem does not change: physical reality and human purpose must be converted into representations the machine can use, then converted back into decisions and effects that remain faithful to the mission.

Levels 0-2 are the primary human command domain. Levels 2-4 require human-machine engineering. Levels 4-5 must remain bounded by validated authority and constraints.

THE SENSOR-TO-SHOOTER SEMANTIC LADDER

A sensor does not begin with meaning. It begins with physical phenomena: light, heat, pressure, motion, electromagnetic energy, sound, chemical traces, or other observable changes.

The system converts those phenomena into measurements. Algorithms convert measurements into features. Perception systems convert features into detections and classifications. Fusion systems turn detections into tracks and a world model. Decision systems relate that model to mission purpose. Action systems convert a decision into a command or effect.

At every conversion, somebody or something decides what matters.

Two transitions are especially decisive:

MEASUREMENT TO MEANING. What does the system believe it sees?

MEANING TO ACTION. What does the system believe should be done about it?

This is why sensor-to-shooter and weapon-employment AI cannot be treated as a mere plumbing problem. A track is not a target. A classification is not authority. Confidence is not intent. A recommendation is not a command. Those distinctions must survive every machine translation.

Grounded tokens are decisive doctrinal terrain at the human-AI command interface. The semantic layer is decisive terrain across the entire sensor-to-shooter path.

MACHINE-READABLE MISSION COMMAND

The answer is not an automated 10,000-mile screwdriver.

If a distant headquarters uses AI to issue more detailed instructions more rapidly to every subordinate machine, we have not created autonomy or mission command. We have automated remote-control micromanagement. It will remain slow in thought, brittle under disruption, dependent on links, and hostile to initiative.

The better principle is already central to NETWAR:

HIERARCHICAL COMMAND, NETWORKED EXECUTION. Preserve the hierarchy of purpose and lawful decision. Network everything required to execute it.

Higher command provides purpose, priorities, relationships, resources, authorities, boundaries, and acceptable risk. Local human-machine elements sense, interpret, coordinate, and act within that mission contract. They do not wait for a remote screwdriver to turn every screw.

At the LLM level, the carrier may be a prompt, system instruction, retrieved doctrine, or structured context. In a sensor or autonomous system, the same human meaning may be encoded as an ontology, schema, task model, policy, behavior tree, confidence rule, action primitive, or software interface contract.

Prompting is one carrier. Grounding and steering are the larger process. Fine-tuning is a separate technical act that changes model weights. The doctrine must survive all three distinctions.

FUTURES C2 REACHED THE DOOR

Army Futures Command has already stated an essential part of the requirement.

AFC Pamphlet 71-20-9, Army Futures Command Concept for Command-and-Control 2028: Pursuing Decision Dominance, says the future force requires orders, graphic control measures, rules of engagement, and other guidance in a format that is both human- and machine-readable. It warns that producing a human version and then translating it into a machine version will be too slow and will introduce errors.

Correct. But not far enough.

Machine-readable is a format requirement. The deeper requirement is shared and controlled meaning.

A machine can read a field named PRIORITY without understanding how this command uses priority. It can parse a rule without knowing its relationship to purpose, authorities, exceptions, confidence, or changing conditions. It can exchange a track without preserving how that track was formed, how old it is, what sources support it, or what uncertainty accompanies it.

Making an order machine-readable without grounding its terms is like lighting the fiber while leaving the routing tables empty.

The Futures C2 concept identifies the need for human- and machine-readable orders and AI-enabled information routing. The next doctrinal step is to let commanders, staffs, operators, engineers, and other subject-matter expert’s author reusable semantic primitives that carry their meaning across models, sensors, echelons, and systems.

AI-LLM examples below are a field-expedient beginning: not a complete military AI architecture, but a method by which warriors can begin identifying, naming, grounding, combining, testing, and revising their own doctrinal tokens now.

WARRIORS ARE ALREADY ENGINEERS

The division between warrior and technologist is false.

Many warriors are coders and engineers. Pilots live inside control systems, sensor fusion, rules, displays, and machine interfaces. Intelligence professionals live inside collection, classification, correlation, confidence, provenance, and analytic models. Signal and cyber soldiers understand addressing, protocols, routing, state, trust, and failure. Fires, air defense, sustainment, and maintenance professionals routinely translate between physical systems and symbolic representations.

They do not need a simplified fiction that every datum is literally called a token. They need a common mapping across technical vocabularies, so that a commander’s concept can survive its passage through code, models, sensors, networks, staffs, and machines.

The necessary division is not warrior versus coder. It is authority and responsibility.

Engineers may implement the architecture. Data scientists may train the model. Vendors may supply components. But the operational community must own the meanings, relationships, priorities, limits, authorities, and tests. Often the warrior and engineer will be the same person. That is an advantage.

If warriors do not shape the semantic terrain, it will still be shaped - by programmers, vendors, training data, inherited schemas, headquarters processes, or adversary action.

THE SEMANTIC LAYER IS CONTESTED TERRAIN

The enemy does not always need to destroy the AI node. It may be enough to corrupt the semantic links around it.

DECEIVE OBSERVATION. Cause sensors or operators to register the wrong evidence.

CORRUPT REPRESENTATION. Induce the system to extract the wrong feature, label, identity, relationship, or confidence.

SEIZE CONTEXT. Insert false instructions, stale orders, poisoned references, or hostile assumptions into the machine’s working picture.

EXPLOIT ACTION MAPPING. Turn a correct observation into an incorrect priority, unauthorized action, delayed response, or wasted resource.

Semantic defense therefore includes source provenance, authentication, confidence, cross-checking, version control, adversarial testing, explicit authorities, observable decision paths, human challenge, and practiced fallback modes.

This is a NETWAR problem. The enemy attacks the links that produce effects. The semantic layer is one of those links.

WARRIORS MUST WRITE THEIR OWN

A leader, commander, pilot, analyst, technician, or staff officer can begin with work already produced: orders, doctrine, theses, briefings, after-action reviews, standing operating procedures, code comments, checklists, or years of accumulated notes.

Load that material into an LLM and ask it to extract candidate tokens - repeated concepts, judgments, relationships, priorities, constraints, triggers, and implied rules. Then argue with the machine. Split weak concepts. Merge duplicates. Reject attractive nonsense. Recover what the author meant but failed to state clearly.

The LLM does not confer authority. It is an extractor, translator, critic, and drafting partner. The human author or responsible military authority decides which concepts are valid.

A grounded token should answer at least these questions:

NAME. What short, stable handle identifies the concept?

MEANING. What exactly does it mean here?

TRIGGER. When should it be invoked?

RELATIONSHIPS. What nodes, links, flows, authorities, and other tokens does it depend upon or affect?

LIMITS. When does it not apply, and what is prohibited?

EVIDENCE. What sources, observations, confidence, and provenance support it?

EFFECT. What reasoning, decision, product, or bounded action should follow?

TEST. What examples, counterexamples, edge cases, and adversarial cases demonstrate that humans and machines are using it correctly?

The point is not to game the model. The point is to ground and steer AI in accordance with leader and commander’s intent.

THE CLAIM

Models will change. Tokenizers will change. Vendors will change. Sensors, networks, and effectors will change.

The semantic battle will remain.

Whoever controls what the system can perceive, name, connect, prioritize, and do controls the decisive terrain between information and effect.

GROUNDED TOKENS ARE DECISIVE DOCTRINAL TERRAIN AT THE HUMAN-AI COMMAND INTERFACE. They make intent machine-usable without reducing command to remote control.

THE SEMANTIC LAYER IS DECISIVE TERRAIN ACROSS THE SENSOR-TO-SHOOTER PATH. It binds observation to meaning and meaning to action.

WARRIORS MUST SHAPE, CONTEST, AND CONTROL THAT TERRAIN. We cannot outsource the meaning of our own doctrine, orders, risks, authorities, or desired effects.

The model is the node.

Grounded tokens route intent.

The semantic layer is the high ground.

Example NETWAR LLM Tokens-mine (TLW)

ANNEX — NETWAR TOKENS FOR HUMAN AND MACHINE USE

PURPOSE

This annex converts the principal concepts of NETWAR into compact conceptual tokens.

A token is a short, durable name attached to a specific idea.

The purpose is not to replace the book.

The purpose is to let the reader carry the book into an AI system and use it.

A reader may copy a single token, a chapter’s tokens, or the entire token set into an LLM.

FORMAT

NW.TOKEN_NAME = definition

Example:

★ NW.MAP_FIRST = Draw the network before interpreting or acting upon it.

The star identifies especially high-value tokens.

The star is editorial notation. It is not part of the token name.

CORE LOAD SET

If space is limited, load these first.

★ NW.NETWAR = The struggle between networks for control of outcomes.

★ NW.MAP_FIRST = First draw the map. Understand what exists before deciding what to do.

★ NW.NODE = An element capable of receiving, holding, transforming, deciding, producing, or transmitting something through a network.

★ NW.LINK = A connection between nodes through which something can pass.

★ NW.FLOW = The movement of information, authority, trust, people, materiel, money, energy, intelligence, or effects through links.

★ NW.CRITICAL_LINK = A link whose loss, control, protection, or creation disproportionately changes network performance.

★ NW.EFFECT_PATH = The chain of nodes and links through which an action produces an outcome.

★ NW.NW_OOB = A Netwar Order of Battle maps the nodes and links that determine outcomes.

★ NW.CONVERGENCE = The state in which distributed elements produce mutually supporting effects toward a common outcome.

★ NW.HIERARCHICAL_COMMAND_NETWORKED_EXECUTION = The hierarchy supplies intent, authority, priorities, and mass; the network supplies speed, reach, adaptation, resilience, and local initiative.

★ NW.COMMAND_NODE = Command is the indispensable node that converts distributed capability into unified effect.

★ NW.MISSION_CONTRACT = A compact agreement linking commander's intent, delegated authority, boundaries, resources, and expected outcome.

★ NW.SWARM = Distributed elements acting independently or semi-independently whose effects converge on a common problem.

★ NW.INTELLIGENCE_MAPS_EFFECTS = Intelligence maps the network that produces effects.

★ NW.EVERY_ACTION_IS_RECON = Every enemy action is reconnaissance because action reveals elements, links, capabilities, or dependencies.

★ NW.FLOW_IS_LIFE = A network survives through continued flow.

★ NW.RESILIENCE = The ability of a network to continue producing required effects after damage, disruption, isolation, or loss.

★ NW.REGENERATION = The ability of a network to restore lost nodes, links, flows, paths, and functions.

★ NW.ECONOMY_OF_FLESH = Substitute network design, information, machines, dispersion, distance, and other resources for unnecessary human exposure and loss.

★ NW.BUILD_WHAT_EXISTS = Begin by recognizing, connecting, exercising, and improving networks that already exist.

CHAPTER 1 — THE LANGUAGE OF NETWAR

★ NW.NETWAR = The struggle between networks for control of outcomes.

NW.NETWORK = A living system of connected nodes, links, and flows.

★ NW.NODE = An element capable of receiving, holding, transforming, deciding, producing, or transmitting something through a network.

★ NW.LINK = A connection between nodes through which something can pass.

★ NW.FLOW = Something moving through a network.

NW.PATH = A sequence of connected nodes and links through which a flow can travel.

NW.ROUTE = The particular path selected for a flow.

NW.CELL = A deliberately bounded NETWAR element possessing enough internal connectivity and capability to perform a defined function.

NW.TRUST = Confidence that permits information, authority, resources, risk, or action to cross a link.

NW.DEPENDENCY = Something a node or network requires in order to continue functioning.

CHAPTER LOAD

Use NW.NETWAR, NW.NODE, NW.LINK, NW.FLOW, NW.PATH, NW.ROUTE, NW.TRUST, and NW.DEPENDENCY to describe the problem as a network rather than as a list of actors.

Above is an incomplete list of my NETWAR tokens to give you an idea.

SOURCES

U.S. Army Futures Command, AFC Pamphlet 71-20-9. Army Futures Command Concept for Command and Control 2028: Pursuing Decision Dominance, 14 July 2021, especially page 52 on simultaneously human- and machine-readable orders and guidance. Source

U.S. Army Science Board. Battlefield Uses of Artificial Intelligence, 2019. The report distinguishes perception, language and semantics, knowledge representation, planning, human-machine teaming, and human-machine dialogue. Source

National Institute of Standards and Technology, NISTIR 7908. A Sensor Ontology Literature Review, 2013. Describes raw and derived sensor data, integration/fusion, semantic interpretations, capabilities, provenance, and confidence. Source

Alexey Dosovitskiy et al.. An Image Is Worth 16x16 Words: Transformers for Image Recognition at Scale. Demonstrates transformer processing of image-patch sequences. Source

Anthony Brohan et al., Google Research. RT-1: Robotics Transformer for Real-World Control at Scale. Describes image tokenization, action tokenization, and token compression for robotic control. Source

Anthony Brohan et al.. RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control. Represents robot actions as text tokens alongside language and vision-language data. Source

Stevan Harnad. The Symbol Grounding Problem, Physica D 42 (1990). Frames the problem of binding formal symbols to nonsymbolic sensory representations and real-world referents. Source

Translated from TLW~ese by AI. The author is incomprehensible per popular declamation and bows to the world.

_{For MacArthur and the French learned hard the Chinese play GO, not Poker, not even chess.}

America has our fleet lingering around an objective already cleared, which invites counterattack and pointless loss. We need to Get Off the X. We need to get off the objective in Hormuz. We are lingering around a taken objective and that’s how reversals happen. Especially in that part of the world. All our aims are achieved, we are past the culminating point and we are overextended. The USN is in danger of getting sucked into defeat from victory by lingering in the Hormuz Gulf.

The danger is less Iran than China and Russia, who DO have the capability to strike us hard and certainly motive. Let’s note: Fair is Fair. As America remains in blockade and so range to squeeze China’s ‘BBLs” or “barrels” then Fair Play for China to Sink our Boats, or Russia to sink our boats and all will say “It was Iran.” The Anti-Ship missile ranges of Iran, China and Russia map below.

Proxy Ship Sinkings quite Fair Play as we did already in Ukraine conflict.

We must also accept: No one in that region can be trusted. We have spent too much time there learning this the hard way. Not to mention DC is expert at turning victory into defeat and highly motivated towards reversal themselves. If it isn’t worth everything else, then contract. Respectfully we must contract and consolidate. At present you run the board. As there’s also stasis the board can quite be upset. If it isn’t worth everything else, then Get Off the X.

SITUATION

DOGE was supposed to be temporary. Elon Musk departed and its centralized organization largely faded from view, but the network did not disappear; personnel, methods, authorities, and initiatives associated with DOGE remained distributed throughout federal agencies, where parts of its work continued. At the same time, DOGE faced an equally distributed opposition—not a single organization, but a network of federal employees and unions, lawyers and advocacy groups, state attorneys general, affected constituencies, journalists, members of Congress, and others connected through information, litigation, political pressure, and shared interests, with courts serving as contested terrain. The result is an unusually clear case for a Netwar Order of Battle: a dispersed government reform network confronting a dispersed opposition network, each able to produce effects through connections rather than through a single headquarters.

CASE STUDY: THE NETWORK THAT OPPOSED DOGE

Question: What does an opposition look like when viewed as a network rather than a list of organizations?
Observation: DOGE did not face one opponent. It faced a network.
Lesson: An Order of Battle should show not only who exists, but how separate actors become capable through connection.
The Department of Government Efficiency—DOGE—provides a useful example of how a Netwar Order of Battle differs from a traditional Order of Battle. A traditional OOB might identify the organizations opposing DOGE: federal employee unions, state attorneys general, public-interest law firms, advocacy organizations, members of Congress, and news organizations. Courts would appear separately as institutions in which disputes were decided. That list tells us who is present. It does not yet tell us how the system works. That requires a network map.

THE NETWAR OOB

A Netwar Order of Battle analyzes the network map to determine what matters—who and what is critical, how they are connected, what they depend on, and what the network can do as a result. The basic questions are simple:
Nodes are who or what exists.
Links are how they connect.
Dependencies are what they require from each other.
Effects are what the connected network can produce.

Apply those questions to the opposition to DOGE and a different picture appears. There was no single headquarters commanding the opposition. Instead, several networks overlapped. Federal employees possessed direct knowledge of agencies, programs, personnel, records, and the effects of DOGE actions. Unions connected those employees to larger organizations. Lawyers connected affected parties to courts. Advocacy organizations connected constituencies, research, legal expertise, and public campaigns. State attorneys general could connect individual disputes to the resources and legal standing of state governments. Journalists connected documents, employees, lawsuits, and government actions to the wider information environment. Members of Congress could convert information into hearings, investigations, legislation, appropriations fights, and political pressure. Affected citizens, contractors, researchers, grant recipients, retirees, and other constituencies could provide stories, evidence, political pressure, and—in some cases—legal standing. Courts were not members of this opposition network. They were contested terrain: independent institutions through which competing networks attempted to produce authoritative effects. The important thing was not any one organization. It was the connections between them.

FOLLOW THE FLOW
Imagine a single federal employee observes an action inside an agency. By itself, that employee may possess little power. But the employee may be connected to a union. The union may be connected to attorneys. The attorneys may bring a case using an affected employee, organization, or state as a plaintiff. Documents may reach journalists. Reporting may reach Congress. State attorneys general may discover that the same action affects their states. Several states may then combine into a multistate lawsuit. One observation has moved through the network.
The flow might look like this:
Employee → Union → Lawyer → Court
But alternate paths exist:
Employee → Journalist → Public → Congress
Affected Party → Advocacy Group → Lawyer → Court
Government Action → State → Multistate Coalition → Federal Court
These paths can operate simultaneously. That is where network power emerges. A node that appears weak in isolation may become powerful because of what it can activate.

DEPENDENCIES
The map also exposes what the network requires in order to function. The first dependency is information. Someone must know what happened. Employees, government records, court filings, documents, public databases, interviews, and investigative reporting provide visibility. The second dependency is access. Legal action requires plaintiffs, standing, attorneys, courts, and applicable law. Political action requires access to legislators, committees, constituencies, or public attention. The third dependency is communication. Employees must be able to reach unions. Sources must reach journalists. Plaintiffs must reach lawyers. States must find other states. Information must move. The fourth dependency is resources: money, staff, expertise, time, technology, and institutional support. Remove a node and another may replace it. Sever enough important links or deny enough dependencies, however, and the network may still exist while losing much of its ability to act. That distinction is fundamental.

NETWORK EFFECTS
Once connected, the network can produce effects that none of its individual nodes could easily produce alone. It can litigate. It can expose. It can constrain. It can influence. It can mobilize. It can attempt to restore programs, personnel, funding, access, or authorities that were removed or changed. The network does not need a single commander to produce these effects. Different actors can pursue different objectives for different reasons while still generating mutually reinforcing results. That is not necessarily conspiracy. It is network behavior.

THE FINAL IRONY

DOGE itself demonstrates the same principle. DOGE was created as a recognizable organization associated strongly with Elon Musk and a concentrated reform effort. But organizations and networks are not the same thing. When the formal DOGE structure diminished and Musk departed, people associated with its work remained distributed through parts of the federal government. The organization could weaken while portions of the network persisted. The opposition behaved similarly. One lawsuit could fail. Another could begin. One employee could leave. The information could already have moved. One organization could disappear. Alternate paths could remain. This produces the deeper lesson. Networks survive through connections, dependencies, and alternate paths—not merely through headquarters.

A traditional Order of Battle asks: Who is out there?
A Netwar Order of Battle asks more:
Who matters?
How are they connected?
What do they depend on?
What can the network do because those connections exist?
That is the difference between seeing organizations—and seeing the network.

TAKEAWAY: The organization is not the network. Map the connections, and the real Order of Battle begins to appear.

_{A note on AI: My AI [Chat GPT5.5] has been loaded for over a year with my 12 years of my NETWAR writing - published on Substack- and honed through hundreds of interactions on NETWAR complete with now Tokens internally generated to assist in generating NETWAR analysis. This is the proper use of this tool, and full disclosure.}

NETWAR: The struggle between networks for control of outcomes.

It asks: “Who is connected to whom, and who controls the important links?”

Who controls the links in a network controls the outcomes.

NETWAR ORDER OF BATTLE (NW OOB)

A Netwar Order of Battle maps the key nodes and links inside a network and shows who controls the outcomes.

Traditional OOB: “Who can fight?”

NW OOB: “Who controls the links?”

NW OOB: A map of the nodes and links that determine those outcomes.

A traditional Order of Battle explains who can fight; a Netwar Order of Battle explains why people choose sides.

THE FRAME

Before we can understand a war, we must understand the network.

The map comes first.

A network map shows people, groups, institutions, and the relationships between them.

It does not tell us what to do.

It tells us what exists.

Only after the map is drawn can we identify what matters.

Only after we identify what matters can we act.

First draw the map.

THE FRAME

Wars happen inside networks.

Before we can understand the war, we must understand the network.

The first step is the map.

The map shows what exists.

The Netwar Order of Battle explains what matters.

First draw the map.

TAKEAWAY

Draw the map.

Everything else is commentary.

(Note - I wrote then bounced off my Bot Mirror ChatGPT5, in Italics. Bot also helped with links).

The changes and challenges to warfare are so radical that the principle of dispersion may need to be incorporated, explicitly.

Army doctrine recognizes that static, concentrated forces are easily targeted: “Forces that are concentrated and static are easy for enemy forces to detect and destroy,” and thus “maintain dispersion to the greatest degree possible.” Disperse or Die

Key Changes: Dispersion elevated: It is no longer a subset of “Security,” but an explicit principle, reflecting the modern battlefield where concentration = death.

Mass redefined as Swarm: “Mass” in effect means pulsed swarming—brief, networked, and three-dimensional, never static. This keeps doctrinal continuity (mass remains on the list) but signals the paradigm shift.

The Ukrainian offensive of 2023 is the funeral of Air Land Battle and WW2 maneuver and principals of war.

Mass as understood just went the way of the bayonet charge.

Theme; Dispersion and only temporary ground mass in war-

Analysis of current wars in Ukraine and recent other wars - Israel vs Iran, Nagono Karabakh, Syrian HTS vs Syrian army/ Russia indicate ground massing is fatal at present. Drone recon and comprehensive strikes by drone, missile, air, artillery now make massing or known static or stationary fatal. Dispersion is necessary more than ever before- how to concentrate combat power to win?

A solution is temporary swarming then dispersion / movement.

The swarming must happen in 3 dimensions- there must be at least temporary overmatch or superiority in the sky to mass on the ground-

Or perchance- We stop saying “mass” as Clausewitzian concentration but instead Swarm and we call it Swarm. Swarming is temporary and pulsing concentration on the enemy or objective and intrinsic to swarming is the swarm disperses then reforms again constantly.

Swarm is to differentiate from mass - there’s no intrinsic or implicit dispersion and reforming, pulsing in “Mass.”

The changes and challenges to warfare are so radical that the principle of dispersion may need to be incorporated, explicitly.

- it will not do to file it under the catch all of “Security.”

Dispersion and Swarming must be elevated.

Revised Principles of War (with Dispersion and Swarm)

1. Objective – Direct every operation toward a clearly defined, decisive, and attainable objective.

2. Offensive – Seize, retain, and exploit the initiative.

3. Dispersion – Maintain dispersion to the greatest degree possible in order to reduce vulnerability to enemy reconnaissance, drones, precision fires, and massed effects.

4. Mass (Swarm–Disperse pulse) – Rather than static concentrations, achieve decisive effects by temporary swarming: a pulsed, three-dimensional concentration of combat power, immediately followed by dispersal.

5. Economy of Force – Allocate minimum essential combat power to secondary efforts.

6. Security – Never permit the enemy to acquire unexpected advantage.

7. Surprise – Strike the enemy at a time, place, or in a manner for which they are unprepared.

8. Simplicity – Prepare clear, uncomplicated plans and concise orders to ensure understanding.

(Below is written with AI assist, I load corpus and the problem/solution, Bot finds links and rephrases my words back at me).

Dispersion as new principle of war

Written with CHAT OpenAI as Foil.

[I TOLD BOT]

Theme; Dispersion and only temporary ground mass in war- Analysis of current wars in Ukraine and recent other wars - Israel vs Iran, Nagano Karabakh, Syrian HTS vs Syrian army/ Russia indicate ground massing is fatal at present. Drone recon and comprehensive strikes by drone, missile, air, artillery now make massing or known static or stationary fatal. Dispersion is necessary more than ever before- how to concentrate combat power to win?

A solution is temporary swarming then dispersion / movement.

The swarming must happen in 3 dimensions- there must be at least temporary overmatch or superiority in the sky to mass on the ground-

Or perchance-

We stop saying “mass” as Clausewitzian concentration but instead Swarm and we call it Swarm. Swarming is temporary and pulsing concentration on the enemy or objective and intrinsic to swarming is the swarm disperses then reforms again constantly.

Swarm is to differentiate from mass - there’s no intrinsic or implicit dispersion and reforming, pulsing in “Mass.”

The changes and challenges to warfare are so radical that the principle of dispersion may need to be incorporated, explicitly.

- it will not do to file it under the catch all of “Security.”

ChatGPT said:

Dispersion, not Mass — and “Swarm” as the replacement effect

Thesis. Current wars show that large, static, or predictable ground concentrations are rapidly located and struck by layered kill-chains (UAS → EW → artillery/missile/air). The battlefield is “transparent” enough that dispersion must be explicit doctrine, not just a sub-idea of “security.” Combat power is now concentrated primarily as brief, pulsed effects—what we’ll call swarming—and then immediately re-dispersed.

Terms (tight)

Mass (legacy): JP 3-0 defines mass as concentrating the effects of combat power at the decisive place and time (note: even JP now emphasizes effects over piles of forces).

Dispersion (new, explicit principle): The continuous, default posture of forces and sustainment to minimize signature and loss, preserve tempo, and survive a sensor-saturated, precision-fires environment.

Swarm (replacing “mass” in practice): A temporary, 3-D, pulsed concentration of effects by many small, networked units, followed by immediate re-dispersion. (Arquilla & Ronfeldt literally define swarming as “sustainable pulsing of force” from dispersed nodes.)

RAND Corporation

RAND on Swarming

Why dispersion must be explicit now

Battlefield transparency → punished clustering. Modern UAS + EW + digital C2 make hiding slow or static groupings extremely hard; both sides in Ukraine learned that survivability depends on dispersion, camouflage, deception and movement.

rusi.org-protecting-the-force-from-uncrewed-uas

Recon-strike complexes at scale. Nagorno-Karabakh (2020, 2023) showcased unmanned systems cueing fires against armor, ADA, and fixed positions; massed ground groupings and known nodes were shredded.

csis.org/analysis/air-and-missile-war-nagorno-karabakh-lessons-future-strike-and-defense

Air/missile “swarms” and counters. Iran’s April 13–14, 2024 strike launched >300 drones/missiles; a multinational defense largely defeated it—proof that temporary overmatch in the sky is decisive for either side’s pulse.

The Washington Post

AP iran-ballistic-missiles-israel-retaliation-gaza-war-takeaways

Sub-state actors swarming. In Syria, HTS built drone forces that enabled short, sharp offensives—again privileging brief effect pulses over holding big ground masses.

gnet-research.org- HTS Drones

Logistics are equally targetable. LOC/Lines of Communication-Transparent rear areas and predictable LOCs are struck across depth; sustainment must disperse and “pulse” as well.

cascom.army.mil-FM4-0-MTT LOC

Bottom line: The default posture is dispersed. “Mass” (as forces) is a liability; “swarm” (as pulsed effects) is how you win.

Swarm > Mass: the operational recipe

Preconditions (create a short “window” of overmatch):

Sky/EMS window: Local UAS density & counter-UAS set the ceiling for any ground pulse. Achieve a brief air/EMS superiority with UAS screens, EW suppression, smoke/obscurants, decoys, and deception emissions.

Protecting the Force from Uncrewed Aerial Systems | Royal United Services Institute

Signature budget: Plan dwell times so the ground grouping exfiltrates before the enemy sensor-to-shooter cycle can close (minutes).

Kill-web primed: Loiterers/FPVs, artillery TOTs, and cyber/EW effects are staged to fire inside the window. (Ukrainian and Russian practice shows effects are increasingly UAS-led and time-compressed.)

Surviving the Kill Web: Adapting Army Sustainment to the Precision Strike and Unmanned Threat Era

Pulse (the 3–15 minute “swarm”):

Open the corridor (sky first): UAS screen + EW to blind/soak enemy drones and spotters.

Strike first, move under the effects: FPV/loiterers collapse key nodes; artillery/missiles deliver brief, dense salvos; small ground packets surge onto the objective only under that umbrella.

Exploit and get out: Complete the task (breach, seizure, sabotage, capture) and immediately re-disperse before counter-fires arrive.

Post-pulse:

Rapid re-masking: Smoke, decoys, silent moves, route diversity.

Back to dispersed sustainment: Micro-caches, mobile shops, and drone-fed last-mile resupply to avoid patterning.

Thornton_Gallasch-Swarming_Logistics_for_Last-Mile_Logistics. H/T DST AU

[ME TLS] The force design to make swarming real C2: Mission-order contracts (“mission threads”) with pre-cleared authorities; C2 tools must work across any bearer and degrade gracefully (no single fat TOCs). See mission contracts / mission threads below

Mission Contracts for Mission Orders - Mission Command as Contracts

Units: Many small, modular cells (drones, EW, fires, sappers, light mobility) that can self-assemble for pulses and then break apart. (This mirrors RAND swarming and DARPA mosaic logic.)

C2 for swarms

RAND Corporation RAND SUAS and EW

Modular EW payload for Drones

Fires: Loiterers and FPVs for cheap, precise, fast effects, tightly coupled to tube/rocket artillery for instant massing of effects instead of people.

Fighting for Information: A Theory of Tactics for the Next Army

By hard calculation.

Less tragedy formula;

Reduce every conflict to resource conflict - THEY ARE - then fight and settle on these tangible resources.

As opposed to intangibles such as “good” or “bad.”.

Nor ideology, or religion, and nor “democracy.”

Unless the intent is the utter destruction of the foe, in which case… with the understanding of what is being harnessed has its own monstrous agenda; of course it does, create the monster and its appetites and nature are monstrous.

See the World Wars.

RED SWARM FIELD MANUAL

Red Networked Sovereignty Ops Guide | 2020s~

I. CORE PRINCIPLES

• Decentralize everything – Act independently when alignment is sensed.

• Meme it before you mean it – Test rhetoric through virality cycles.

• Weaponize trust – Trust flows peer-to-peer, not top-down.

• Attack legitimacy, not logistics – Discredit the narrative first.

• Create alternate institutions – Parallel systems win in collapse.

II. CULTURAL OPERATIONS

• Viral Narrative Warfare: Launch memetic ops on TikTok, X, YouTube Shorts

• Brand Jamming: Parody or subvert Blue brands

• Counter-Elite Influence Cells: Create Red-aligned art, music, entertainment

• Alt-Education Pipelines: Build red-pill curriculum via podcasts and streams

III. POLITICAL OPERATIONS

• Local Flip Campaigns: Target school boards, sheriffs, local commissions

• Narrative Outflanking: Preempt Blue narratives with memes and forums

• Primary Disruption: Split opposition base through swarm influence

• Legitimacy Generation: Use real-world wins to demonstrate Red governance capability

IV. ECONOMIC OPERATIONS

• Parallel Economy Buildout: Red-aligned fintech, employment networks

• Weaponize Boycotts: Target hostile businesses with swarm backlash

• Substack & Gig Work: Empower Red creators to self-fund

• Exit-to-Sovereignty: Invest in land, food, banking independence

V. BLUE COUNTER-SWARM BRIEFING

• Expected Blue Team Adaptations:

• AI-driven narrative floods and deep fake disinfo

• Institutional cloaking via NGOs and fake student movements

• Financial chokeholds using ESG enforcement and lawfare

• Building Blue-branded swarm lookalikes to confuse the base

• Amplifying internal Red contradictions to fracture trust

VI. BLUE WEAKNESSES TO EXPLOIT

• Over-centralization: disrupt with rapid swarm iteration

• Legacy moral frameworks: ridicule with humor and realism

• Dependence on coercion: frame as censorship and fragility

• Demographic fracture: magnify class division within their base

• Emotionally fragile signaling trigger emotional spirals to overload capacity

Note - this written with OpenAI assist. Having long uploaded my Netwar Corpus and worked back and forth the AI tokens: a very useful tool.

The Rear Area Lines of Communication (LOC) Battle is Back to Front.

Mission:
Our logistics and Lines of Communication / LOCs are compromised by enemy drones and the indirect fires they summon operating over the rear area. We must deny the enemy air/drone mobility over our rear areas to restore our own mobility in our LOCs (Lines of Communication) from the battalion-level CPs back.

This will be accomplished in three phases using swarming principles, beginning dispersed, then massing to apply force, then dispersing again—a pulsing tactic inherent to swarming.

Phase 1: Counter-Reconnaissance / Counter-Drone Operations

Objective: Deny enemy ISR access over rear areas—winning the counter-recon fight before any sustained logistics movement is possible. (ISR-Intelligence, Surveillance, Reconnaissance).

Concept of Operations

• Regenerate from dispersed to mass against detected enemy ISR threats.

  • Rear-area security forces (logistics troops, MPs, ADA units, counter-drone units) start dispersed to minimize detection.

  • Upon detection of enemy ISR drones or ground reconnaissance, nearby units self-organize (regenerate) into ad hoc hunter-killer teams using available assets (short-range drones, counter-drone jammers, AAA, etc.).

  • After neutralization, forces immediately disperse again, avoiding predictable massing that could invite enemy strikes.

• ISR Denial Through Pulsing Swarms

  • Rear-area forces launch short bursts of massed counter-drone activity, actively hunting enemy ISR, then quickly dissolve back into smaller, dispersed formations.

  • Think of it as electronic "ambushes" on enemy ISR—AI-driven drone swarms, jammers, and kinetic defenses activate, engage, and vanish before enemy fires can retaliate.

• End State:

  • Enemy ISR is denied over rear areas, meaning logistics units can move without immediate detection and targeting.

  • We have air dominance (and ground control if applicable) over our LOCs.

Phase 2: Reestablishing Ground & Air Mobility in Rear Areas

Objective: Restore functional logistics corridors and protected supply routes.

Concept of Operations

• Adopt a decentralized, dynamically adapting LOC system

  • No static supply depots: instead, we use micro-depots, mobile caches, and dispersed convoys that move only after Phase 1 success.

  • Regenerative convoys: Instead of single, large formations, logistics moves in dispersed elements that self-reassemble at key points before dispersing again after unloading. This is very similar to how networks deliver packets of data, indeed similar to our ‘packetized’ media culture and even conversation now, which means we’re already thinking this way. So DO the way we THINK or at least MOVE the way we really Communicate.

• Layered Security on LOCs

  • Now that Phase 1 cleared enemy ISR, counter-recon forces escort logistics elements along reestablished supply routes.

  • "Rolling defense" model: A mix of mobile ADA, electronic warfare (EW) units, and counter-recon teams constantly shift along LOCs, preventing enemy infiltration.

• End State:

  • We have secured primary and secondary logistics corridors, restoring supply movement without catastrophic losses.

Phase 3: Maintain Security & Mobility in Rear Areas

Objective: Sustain protected LOCs while adapting to emerging enemy threats.

Concept of Operations

• Adaptability in Rear Security

  • Use automated AI-assisted reconnaissance to dynamically shift LOC security elements based on real-time threats.

  • Maintain continuous drone and counter-drone patrols to deny enemy infiltration with ISR assets.

• Self-Healing Rear-Areas

  • If an enemy breakthrough occurs, we regenerate defenses from dispersed assets to quickly isolate and eliminate penetration.

  • Rear-area security pulses continue—massing to strike, then dispersing.

• End State:

  • The rear area is stable enough to enable sustained combat operations, and maneuver forces can now shift focus forward instead of constantly fighting to sustain themselves.

Execution: Using Swarming Principles to Generate & Regenerate Rear Area Security

• Regeneration: Rear-area forces begin dispersed and reassemble into temporary formations only when needed, preventing the enemy from tracking and targeting them.

• Massing/Dispersal Pulses: Just like swarming, logistics and security forces only concentrate when there is an immediate advantage, then break apart before the enemy can react.

• Dynamic Tasking: Forces self-organize based on local situational awareness, using secure channels to coordinate without a single vulnerable central node.

Summary: Fighting the War in Reverse

• Instead of supporting the front, we start by securing the rear.

• Only after rear area security is established can maneuver forces fully operate forward.

• This flips traditional logistics planning on its head—but it’s the only way to survive in a battlefield of total ISR transparency.

Mission Contracts: mission orders in contract form. Herman Black of the WW2 German Army defined Auftragstaktik [Mission orders] as ‘contracts’ between the Superior and the Subordinate, what the Superior intends done and what the Subordinate needs from his Boss. Every American understands Contract and getting the job done. CONTRACT.

In the Spirit of Mission Command - Herein mission orders are reduced to Mission Contracts.

Tracking Mission Contracts Across Echelons in Rear Area / LOC battle.

Objective:
Ensure mission contracts (mission command via mission threads) are tracked seamlessly across different echelons, enabling rapid execution, clear responsibilities, decentralized action, and real-time adaptation.

1. Mission Contracts Flowing Through Echelons

Each echelon tracks mission contracts within its own Virtual Channels while staying connected to both higher and lower levels for dynamic adjustments.

Virtual Channels are simply a digital channel established across ANY digital communications, Chat or Chat Room is a virtual channel. Any communication equipment will do this including your phone. Keep in mind security of transmission and emissions.

VIRTUAL CHANNELS

2. How Mission Contracts Are Tracked

A. Persistent Mission Threads for Execution & Updates

• Mission contracts flow inside their designated Virtual Channel threads

• Each mission thread remains active until completion, ensuring:

  • Live status tracking

  • Continuous updates across echelons

  • Real-time problem-solving

Example:
📡 "1st Battalion: Secure LOC BLUE 12-15"
🔄 Live Updates in Thread:
✅ Engineers beginning clearance
✅ Scout drones detect enemy ISR
✅ Counter-drone team engaging
✅ Convoy cleared to move

Once completed, the mission thread is archived but remains accessible for after-action review.

B. Dynamic Linking Between Echelons

Mission contracts at one echelon are directly linked to execution at the next level down.

• Brigade issues a mission contract → Battalion executes it → Company refines tactical details.

• Each echelon only focuses on what’s necessary at their level while maintaining visibility of the bigger picture.

💡 Commanders always see what they need—no unnecessary information overload.

C. Adaptive Mission Contracts: Adjusting in Real-Time

1. If battlefield conditions change, the mission contract updates dynamically in its thread.

2. Subordinates report status changes up, while higher command adjusts asset allocations down.

3. Echelons adapt without waiting for lengthy approval chains.

Example:

• Brigade: "1st BN, adjust to securing LOC RED 7-9 instead. Engineer assets en route."

• 1st BN: "Acknowledged. Will re-task MP security to new corridor. ETA 30 mins."

• Platoon Level: "Encountered resistance at checkpoint. Requesting fire support."

• Brigade: "Fires allocated. Clear by fires then proceed."

This flow ensures decisions happen instantly, eliminating bureaucratic drag.

3. Mission Contract Execution by Phase

Mission contracts flow through echelons in sync with each operational phase:

Phase

Key Echelon Responsibilities

Mission Contract Flow

Phase 1: Counter-Recon/Counter-Drone

Brigade/Battalion tasking security forces

Deny enemy ISR in rear areas to enable logistics

Phase 2: Reestablish LOCs

Brigade/Battalion/Company logistics & escort

Secure logistics routes, escort convoys, reestablish mobility

Phase 3: Sustain Rear-Area Security

Battalion/Company/Platoon-level adaptive ops

Continuous counter-recon patrols, securing long-term mobility

Each phase is tracked in Virtual Channels, ensuring mission contracts evolve in real time without disrupting execution.

MISSION CONTRACTS FLOW BY PHASE

4. Summary: How Mission Contracts Are Tracked

✅ Each echelon operates inside its own Virtual Channels, linked up and down the chain.
✅ Mission contracts flow inside persistent mission threads, ensuring clarity and accountability.
✅ Subordinates execute without micromanagement; mission contracts update dynamically based on conditions.
✅ Mission contracts follow a structured phase-based approach, ensuring seamless operational flow.

If I have to do war again, here's what I think I have learned about war.  
    
Keep in mind war is a social activity, while amoral in nature morality has value.

                                                  

The Social part of Netwar

Consider choosing the Primary Leader by election. After that he chooses his own secondary, it would be wise to give some positions to the runner ups.

Proper course of Building. If you want to build anything, you must gather the materials and tradesmen. The materials must be properly prepared.

The Tradesmen must possess some base level of skills.

Wood must be lumbered, treated, aged. Green wood will not cut it.

The Carpenters must know what they are doing.

Bricks must be properly mixed then fired [in our case with the aid of the opponent].

Bricks are preferable to Wood. A Mason is more skilled than a Carpenter.

Steel must first be mined in its component elements of Iron, Coal, and a touch of antimony, then heated in kilns, properly poured into molds of Brick, assembled by Master Tradesmen.

Steel is preferable to Brick. A Steel-wright is more skilled than Mason or Carpenter.

A Millwright builds factories and assembles machines. The summit of the Trades.

And this Dear Friends is the proper course of building.

A most dangerous enemy is one who believes, has integrity, is committed and resourced. This enemy and the population from which he springs should be a constant object of swaying and subornation. Convince him and his they are in the wrong.

Ruthless opportunists who win should be given more opportunity.

Don’t cloud the issue with atrocities. Reciprocity may be necessary if the opponent is so foolish but targeting of those not involved should be approached as if it were an escalation to nuclear war - for it is that dangerous and indeed would tempt a nuclear response. The blood of innocents deliberately killed hardens hearts beyond reason. Do not confuse means with ends, for if atrocity is policy the means become the ends. Now for all that….

Your Rule of Engagement is Victory.

Understand Mercy like Terror is a tactic.

There is much to be learned from the sharp practices of finance.

Caveat: At some point, ruthlessness just leaves ruins.

To our fellows offer mercy.

To the Elites, keep their wealth and station, in exchange power goes to those most fit by Trial of Struggle. Elites keep wealth, the Victors Power.

To the Power Crazed Social Engineers who bought this fate on us all - listen to them as they tell us what they most fear for decades - a Strong Right Fist - and become the monster they fear. Courage is not their strong suit.

Remember the True Enemy will not be facing you in Battle; all the Political Class has to offer is betrayal which is their true coin.

Said Ruling Class is already quite cut off from the populace they openly despise.

If it comes to a trial, the word academy means fire. Enough of them.

The Best way to get them to stop chasing you is for them to think you're already with them.

The safest place to hide money from these rapacious and desperate gamblers is with them or one of their front banks - if the key condition of amenable public and banking officers can be met. It most certainly can.

Recruiting: Do the Long Math --Speak to 1000. 100 will listen. 10 undertake. 5 leave at the moment of Truth, 5 see it through. 2-3 fall to death or capture. There will be 1 proved out warrior. WIN.

If they continue to regulate the Military, we'll have the ultimate Regulatory Capture.

The greatest one so far is Finance. They are motivated primarily by Greed.

We are motivated primarily by Fear.

The journey is complete when we begin to appreciate Greed- that is power- as Well.

We are not only our own fish but our own pond. Don't make the mistake of trying to win over the general populace; for their neutrality - silence - is quite enough. In short they are not the prize - power is - if they were the prize this would be unnecessary.

Your objective is the crushing of our true enemies Will to Power, because they are unfit and this makes them insane with fear, and they have nuclear weapons.

Our pond is where the will to fight comes from in any case.

Very carefully should any question of shadow government be approached and used only as stratagem. If a compliant government in matters not martial is needed, it's best to use the established one. They will follow the Strong Horse and their own interests in any case, such is their nature.

A War of Shadows is desirable.

We need a new Deep State. This one is entirely too shallow.

Victory is replacing the Deep State with you.

Self-Rule is the best hope against corrupt Tyranny.

Again, the population is not the objective. Don't seek to spread chaos or provide local security unless expedience directs this as tactics. Local security and commerce should go on un-molested and local security and yours is best provided by an understanding with local security forces. The local authorities' stake is local, local security forces stake is local security. In exchange for their office and persons being secure they are neutral - again defined as blind, deaf and dumb - they don't see you, hear you or speak of you.

The same offer is extended to the populace at large. Simply inform them - if you see something, say nothing.

The methods of recent warfare have not included the classical component of mercy. This should change. Moreover, the enemy troops should immediately be offered jobs; the best match may be as Troops.

You can make a friend of Horror, but you'll fatally confuse Ends with Means.

If you wish for any future, know you must seize it.

The best changes in forms of government are the ones that don't appear to be changes in forms of government.

If the monstrous corruption you've seen troubles you change things.

You must at any and on all points not be legalistic.

YOU keep your word. FORNICATE the Lawyers. Legalism is the desperate hope of bespectacled tyranny, never the hope of people of action. That's the difference of North America and Cortes.

Don't corner an enemy. Never offer them no hope.

Never be the Army that only offers DOOM. Offer not only a merciful Surrender but possibilities of future employment of soldiers; this should be explicit on surrender.

And yet as you may offer Mercy with your left hand, have Terror ready in the Right.

Current Media offers only an image of awesome powers hopeless to resist. Change this to respect immediately Your Foes power, and your need for them.

The AQ Syrian never says it, but they could not bring themselves to coordinate, as they could not agree on a Commander, as they could not agree on an identical vision, and they would not cede control. So, Command and Control is a trap, so is needing to agree on an identical vision. Coordination will have to make do. Coordination will do well enough. It's also the best you can expect.

Their enemy saw many embers being lit and coming down from the sky, fires lit everywhere, everywhere for a time that powerful enemy was overwhelmed scurrying around putting out the fires.

If and on the few occasions the firebugs made any effort towards concert, they taxed the resources of Legions to rival the ancients nearly to the breaking point.

Stimergic learning - Sign learning- Learning from the successes and failures of many others - worked brilliantly for the many disparate groups for them to learn rapidly and adapt, often literally overnight. But their slow learning enemy was able to learn enough of them to bring their huge weight to bear and crush them in separate pieces.

For though it learns slowly and indeed thinks slowly it moves under one mind - the Commander - and concentrates overwhelming force. The Commander has Control of his resources.

The many groups were not able to concentrate efforts and were crushed piecemeal. If they had been able to merely coordinate a fraction as well together as they were able to learn from each other the outcome may have been different.

Inside - Outside: there can be no control or influence of the Inside on the Outside or the Warden controls the movement. Outside looks after Insides family and interests, that’s it and that’s all. Any communication from Inside is compromised, any directive coerced. There’s no inside, those are captives.

We fight with 21st century maneuvers and 18th century morals - swarm, network but do not seek annihilation of the foe. Offer comradeship not death or slavery.

With the innocent and misguided - if you are tempted remember you literally have the rest of your lives for treachery, murder, atrocity.

Allow some method of neutrals and non-combatants to identify themselves by wearing for instance fluorescent vets, etc. However, while wearing them, they must not bear arms outside the home. If they do its perfidy.

You should have some way of marking yourself as a combatant, if not a uniform, then something to distinguish you as such.

Our first principle of war in the beginning will be survival, the second security for sadly we have the Byzantine Trust problem and will until struggle sorts who’s who.

                                                                     MOTTI

Cut them off from the main body into little bites then eat them.

Situation - The enemy is trapped on a linear feature, the Road.

[It may have similar applications in a similar tactical situation.]

Motti means "Firewood". Possibly the term came from not burning an entire tree in the fireplace but cutting it up into pieces that fit.

Road ambushes -- Previous tactics against enemies on roads have linear, that is attacks on line or with Flank Turning movements [the "L" shaped ambush]. The Finn's Motti changed that; they attacked their enemy blocked on the road not with linear tactics but as one would assault a Trench Line - with smaller bands attacking in concentrated force at select locations. This is how 4 Light Infantry Battalions of 10,000 men in total cut 25,0000 Soviets with tanks and air cover to pieces. The Soviets had hundreds of thousands of men ready. The Finns had 10,000.

What the Finns did was apply Trench storming tactics to a linear objective. They blocked the road effectively then breached the enemy line [the units on the road] at selected places, cutting them into smaller pockets that were cut off. The Finns segmented the problem. They then kept up constant contact [no sleep], attacks with innovative ways to storm the Soviets dug in defenses with the cannon pillboxes the tanks became without fuel and let General Stomach do much of the rest.

A Soviet Division in open combat would have smashed the Finns. The Finns used linear and restricted terrain to trap them on the road then split them into several pockets. They used Trench Storm tactics developed during World War I for an enemy trapped on a road. Small bands operating with normal infantry tactics against a dug in trench line/strongpoint could do the same.

This is superior to Linear or other ambushes which favor the enemy if not in maneuver than in firepower to their flanks. If you cut them into pockets, then they've lost a lot of the flank firepower.

It must be emphasized that the terrain must favor this tactic. The enemy has to have their maneuver greatly restricted, say either front or back. Cities could do this as well as mountain roads. Block them at one or more ends or junctions. Then you cut them off and then into little bites and eat them.

                                                           Polyrhythmic swarming

If you're going to use swarm tactics, then learn from nature. The hum of swarms is music, your music should be Jazz.

(Allow author some musical metaphor please.)

Follow statelier, ordered music and you'll be neatly penned and slaughtered. Follow rap you can forget coordination, or any logic. Hence, you'll be piecemeal slaughtered.

Jazz: The groups cross and coordinate on strike notes ... but are not the same operationally. This confuses competitors thinking and counter-swarm response. It also allows temp flash mobs of mass. Which then diffuse after striking. An intelligent enemy answers swarms with first counter-swarms to hold them then masses forces to crush the swarms one by one. They would naturally look to a point of massing by their enemy as the decisive point they've been waiting for then strike with reserves. If there is a sudden massing and striking followed by diffusion again this will confuse, especially if then suddenly a massing occurs again somewhere else.

They are constantly to be nailing at a swarm of stinging bees who strike then move on. The object of the flash swarms is not the holding of territory or even the destruction of enemy forces - although they should be hurt in every engagement. It's to mount constantly shifting attacks with sudden concentrations in place and time that diffuse and strike pain, then diffuse and move rapidly elsewhere for the next strike to overwhelm the senses of enemy Command and Staffs and denying them any insight into our decision loop and providing no place or time to mass their forces.

The best defense of the foe against swarms is to have local security forces of militia, police, even deputized armed citizens hold the local areas until heavy reaction forces arrive. So, it's critical that these forces be identified, suborned, intimidated or some way undermined and neutralized before, during and after engagements. This is the paramount object of the local Intelligence battle.

Cluster leadership, swarming, and network meshes.

Cluster approach to leadership: if one fails all do not fail.

Cluster leaders may be defined by sectors of either geography or function. Functions such as bulk logistics and transportation, the overall intelligence effort, medical and so on. The Cluster Lead in a sector may be elected, directed, or simply defaulted to by expertise in that area [such as medical]. The lead cell or cluster of cells accept responsibility for coordination, direction of the overall effort, facilitation between disparate groups.

This is particularly useful three times:

*Disparate groups

• Swarming tactical leadership

• Meshing of nodes in the network. Remember your goal should be to build a network mesh, and to adapt it around tactical circumstances either offensively or defensively. A mesh for instance is ideal for controlling swarming cells, it can adapt around the enemy faster than traditional formations using the obsolete Command and Control TOC model. The entire idea of a mesh is if a cell/node is destroyed or the links to it cut the network regenerates or adapts around the damage. Similarly if an enemy is being beaten or is in disorder the mesh exploits his disorder by better organizing and massing swarming elements into his formations.

• The minutemen at Concord did fine, but it was a bunch of armed gangs with rifles running towards the British formations. As the Redcoats took fire from all sides - but not organized swarming attacks - they suffered 50% casualties but withdrew to Boston intact and in good order. The Mongols or Parthians would have annihilated them.

• The Rifle behind every blade of grass is all very well but let us optimize it for efficiency. That is the swarm, the gang, the natural human way of battle and the hunt, and the great organizer of swarm attacks is the network. This is how the RAF won the Battle of Britain, and how the U-Boats nearly won - swarming attacks organized as a network.

A mesh exists when each cell has at least two links to the rest of the network.

This is a meshed network, every cell has at least two links to other cells/nodes.

Meshed connections allow restoration and to move around disruption, or to shape the mesh around opportunity.

Now you may also imagine this mesh as a web. If a column advances into mesh, a meshed network can break individual links and connections. However, the web will then close around it, or certainly should, enmeshing the intruder. Similarly, a static position or disrupted, fleeing enemy could have a mesh connected around it, enmeshing, immobilizing and then destroying said enemy.

The point of the network mesh again isn't just rapid regeneration after disruption or defeat. It's the natural organization of swarms around an attacking or retreating enemy.

When you combine networking with swarming this is more than a mesh.

                                                   It's a spider’s web.