Modelling Command and Control Neville Stanton

pdf download

https://ebookgate.com/product/modelling-command-and-control-
neville-stanton/

Get Instant Ebook Downloads – Browse at https://ebookgate.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Digitising Command and Control Neville Stanton

https://ebookgate.com/product/digitising-command-and-control-neville-
stanton/

ebookgate.com

Freedom from Command and Control Rethinking Management for

Lean Service 1st Edition Seddon

https://ebookgate.com/product/freedom-from-command-and-control-
rethinking-management-for-lean-service-1st-edition-seddon/

ebookgate.com

Modelling Control Systems Using IEC 61499 2nd ed. 2014

Edition Alois Zoitl

https://ebookgate.com/product/modelling-control-systems-using-
iec-61499-2nd-ed-2014-edition-alois-zoitl/

ebookgate.com

True Digital Control Statistical Modelling and Non Minimal

State Space Design 1st Edition C. James Taylor

https://ebookgate.com/product/true-digital-control-statistical-
modelling-and-non-minimal-state-space-design-1st-edition-c-james-
taylor/
ebookgate.com
Aircraft Modelling Modelling Masterclass Brett Green

https://ebookgate.com/product/aircraft-modelling-modelling-
masterclass-brett-green/

ebookgate.com

Two English Republican Tracts Henry Neville

https://ebookgate.com/product/two-english-republican-tracts-henry-
neville/

ebookgate.com

Fix Freeze Feast 2nd Edition Kati Neville

https://ebookgate.com/product/fix-freeze-feast-2nd-edition-kati-
neville/

ebookgate.com

Practical Plant Failure Analysis Neville W. Sachs

https://ebookgate.com/product/practical-plant-failure-analysis-
neville-w-sachs/

ebookgate.com

True Stories of Alien Abduction 1st Edition Stanton T.

Friedman

https://ebookgate.com/product/true-stories-of-alien-abduction-1st-
edition-stanton-t-friedman/

ebookgate.com
MODELLING COMMAND AND CONTROL
This page intentionally left blank
Modelling Command and Control
Event Analysis of Systemic Teamwork

NEVILLE A. STANTON
CHRIS BABER
DON HARRIS
Human Factors Integration Defence Technology Centre
© Neville A. Stanton, Chris Baber and Don Harris 2008

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording
or otherwise without the prior permission of the publisher.

Neville A. Stanton, Chris Baber and Don Harris have asserted their moral right under the
Copyright, Designs and Patents Act, 1988, to be identified as the authors of this work.

Published by
Ashgate Publishing Limited Ashgate Publishing Company
Gower House Suite 420
Croft Road 101 Cherry Street
Aldershot Burlington, VT 05401-4405
Hampshire GU11 3HR USA
England

Ashgate website: http://www.ashgate.com

British Library Cataloguing in Publication Data

Stanton, Neville, 1960-
Modelling command and control : event analysis of systematic teamwork. - (Human
factors in defence)
1. Command and control systems - Simulation methods
2. Command of troops - Simulation methods
I. Title II. Baber, Christopher, 1964- III. Harris, Don, 1961-
355.3’3041

Library of Congress Cataloging-in-Publication Data

Stanton, Neville, 1960-
Modelling command and control : event analysis of systemic teamwork / by Neville A.
Stanton, Chris Baber and Don Harris.
p. cm. -- (Human factors in defence)
Includes bibliographical references and index.
ISBN 978-0-7546-7027-8
1. Command and control systems. 2. Command and control systems--Data processing. I.
Baber, Christopher, 1964- II. Harris, Don, 1961- III. Title.

UB212.S72 2008
355.3’3041011--dc22
2007030959

ISBN 978-07546-7027-8
 Contents

List of Figures vii

List of Tables xi
Preface xv
Acknowledgements xix
Senior Author Biographies xxi
Contributing Authors xxiii

Chapter 1 Overview 1
Chapter 2 Modelling Command and Control 1
Chapter 3 Event Analysis of Systemic Teamwork 2
Chapter 4 Case Study at HMS Dryad 3
Chapter 5 Case Study in RAF Boeing E3D Sentry 4
Chapter 6 Case Study in Battle-Group HQ 4
Chapter 7 Development of a Generic Process Model of
Command and Control 5

Chapter 2 Modelling Command and Control 7

Purpose and Scope 7
Models 7
Specification 10
Structural Models of Command and Control 13
Network Models 19
Dynamic Models 25
Agent Models 32
Socio-technical Models 36
Example: Contextual Control Model 43
Summary of Socio-technical Models 45
Summary of Modelling Review 45

Chapter 3 Event Analysis of Systemic Team-work 49

EAST Review 49
Methods Review 51
Summary 117

Chapter 4 Case Study at HMS Dryad 119

Introduction 121
Observations 124
Conclusions 154
vi Modelling Command and Control
Chapter 5 Case Study in RAF Boeing E3D Sentry 157
Introduction 157
Observations 159
Propositional Networks 172
Conclusions 177

Chapter 6 Case Study in Battle Group HQ 181

Introduction 181
Findings 192
Conclusions 219

Chapter 7 Development of a Generic Process Model of Command and

Control 221
Three Domains for Command and Control 221
Common Features of the Domains and Application of
Command and Control 225
Taxonomies of Command and Control Activities 225
Construction of the Model 231
Network Enabled Capability 233
Model Validity 234
Summary 235
Conclusions 235

Bibliography 239
Index 245
 List of Figures

Figure 2.1 SAS-050 Space of C2 12

Figure 2.2 An adapted version of Lawson’s model of the command and
control process 15
Figure 2.3 The OODA loop 16
Figure 2.4 Example of closed loop feedback control system 17
Figure 2.5 Network Centric Operations Conceptual Framework 18
Figure 2.6 The N-squared (N2) chart from Hitchens (2000): A simplistic
generic model of command and control 20
Figure 2.7 Pyramid structure metrics 21
Figure 2.8 Prototypical social networks 22
Figure 2.9 Propositional network from the air traffic control work domain 24
Figure 2.10 Centralised architecture without information sharing 27
Figure 2.11 Split architecture without information sharing 27
Figure 2.12 Distributed architecture without information sharing 29
Figure 2.13 Negotiated architecture without information sharing 29
Figure 2.14 Distributed with information sharing 31
Figure 2.15 Performance of different network structures 31
Figure 2.16 Architecture of a semiotic agent 35
Figure 2.17 Simulated agents produce emergent flocking behaviour within
a computer simulation 35
Figure 2.18 Kotter’s model of organisational dynamics (1978) 37
Figure 2.19 Smalley’s functional command and control model 38
Figure 2.20 Definition of the abstraction decomposition space
(Rasmussen, 1986) 40
Figure 2.21 Rasmussen’s decision-ladder applied to a command and
control domain 41
Figure 2.22 Example output of strategies analysis. Shaded regions refer to
a specific role or actor; the numbered circles refer to task steps 42
Figure 2.23 Hollnagel’s contextual control model 44
Figure 2.24 NEC benefits chain 47

Figure 3.1 Social network diagram 60

Figure 3.2 CUD extract 61
Figure 3.3 OSD glossary 61
Figure 3.4 OSD extract 62
Figure 3.5 Propositional network for objects referred to in CDM tables 68
Figure 3.6 Propositional network for CDM phase one 68
Figure 3.7 Propositional network for CDM phase two 69
viii Modelling Command and Control
Figure 3.8 Propositional network for CDM phase three 69
Figure 3.9 Propositional network for CDM phase four 70
Figure 3.10 Hazardous chemical spillage event flowchart 76
Figure 3.11 CUD template 95
Figure 3.12 Comms usage diagram for an energy distribution task 95
Figure 3.13 Return to service social network diagram 98
Figure 3.14 Example OSD template 104
Figure 3.15 Example propositional network 114

Figure 4.1 Illustration of workstations onboard a Type-23 frigate 121

Figure 4.2 Seating lay out of Type-23 frigate operations room 122
Figure 4.3 Combined task model for the air, subsurface and surface
threat 125
Figure 4.4 Social network encompassing all three scenarios 132
Figure 4.5 OSD air threat scenario 133
Figure 4.6 OSD subsurface threat scenario 135
Figure 4.7 OSD surface threat mission 137
Figure 4.8 Propositional network for the air threat task 139
Figure 4.9 Plan resources and strategy 140
Figure 4.10 Identify and classify targets 140
Figure 4.11 Assess threat and locate targets 141
Figure 4.12 Control external resources 141
Figure 4.13 Engage targets 142
Figure 4.14 Posture platform and attack 142
Figure 4.15 Re-allocate assets and weapons to new targets 143
Figure 4.16 Shared knowledge objects of the air threat scenario 143
Figure 4.17 Propositional network for the sub-surface threat task 144
Figure 4.18 Plan resources and strategy 144
Figure 4.19 Identify and classify targets 145
Figure 4.20 Assess threat and locate targets 145
Figure 4.21 Control external resources 146
Figure 4.22 Engage targets 146
Figure 4.23 Posture platform and attack 147
Figure 4.24 Re-allocate assets and weapons to new targets 147
Figure 4.25 Shared knowledge objects of the air threat scenario 148
Figure 4.26 Propositional network for the surface threat task 148
Figure 4.27 Plan resources and strategy 149
Figure 4.28 Identify and classify targets 149
Figure 4.29 Assess threat and locate targets 150
Figure 4.30 Control external resources 150
Figure 4.31 Engage targets 151
Figure 4.32 Posture platform and attack 151
Figure 4.33 Re-allocate assets and weapons to new targets 152
Figure 4.34 Shared knowledge objects of the air threat scenario 152

Figure 5.1 Typical seating plan onboard an RAF E3D 158

 List of Figures ix
Figure 5.2 Mission crew structure 159
Figure 5.3 Illustration of workstations on board the E3D 161
Figure 5.4 Task model of E3D mission 161
Figure 5.5 Social network diagram constructed from the standard
operating procedures 163
Figure 5.6 FA (1) social network diagram 168
Figure 5.7 FA (2) social network diagram 168
Figure 5.8 OCA social network diagram 169
Figure 5.9 SO4 social network diagram 169
Figure 5.10 SC social network diagram 170
Figure 5.11 FA(1) operational sequence diagram 171
Figure 5.12 Propositional network for the E3D mission 172
Figure 5.13 Assume station (TD) 173
Figure 5.14 Manage airborne battle (FA, WC, TD) 173
Figure 5.15 Coordinate crew 174
Figure 5.16 Manage communications (SO, LM, TD, CO, CT, SC, RT) 174
Figure 5.17 Ensure operational procedures are followed (TD) 175
Figure 5.18 Manage self defence (TD) 175
Figure 5.19 Manage operations/comms security (TD) 176
Figure 5.20 Control surveillance/operate ESM equipment (SC, ESM) 176
Figure 5.21 Overlap of knowledge objects 177

Figure 6.1 Simplified command hierarchy 181

Figure 6.2 IPB system and products 183
Figure 6.3 Illustration of simplified BAE overlay showing water features
and restricted terrain 184
Figure 6.4 Illustration of BAE overlay showing possible avenues of
approach to engage enemy force 184
Figure 6.5 Illustration of a situation overlay; icons show formation of
enemy forces, arrows show possible routes/courses of action,
dotted lines indicate time phases from the enemy’s current
position 185
Figure 6.6 Illustration of the commander’s effects schematic; the battle
winning idea expressed in terms of effects on the enemy 186
Figure 6.7 Illustration of a draft decision support overlay showing some
key features of the BAE in relation to target and named areas
of interest (TAI and NAI respectively) and decision points (stars) 187
Figure 6.8 Illustration of a completed synchronisation matrix showing
the time line (time based decision points) upon which enemy
and own courses of action are mapped along with the specific
activities of the various force elements 189
Figure 6.9 Example of operational graphics 190
Figure 6.10 Relationship between the combat estimate and Wargaming 191
Figure 6.11 Task network of observed military command and control
activities derived from HTA 193
Figure 6.12 Photograph of the command tent 197
x Modelling Command and Control
Figure 6.13 Social network diagram illustrating a systems level view of
the combat estimate scenario 200
Figure 6.14 ‘Briefing’: illustration of social activity and communication.
The one way flow of information from commanding officer to
subordinate staff is evident as is the close coupling between
nodes 201
Figure 6.15 ‘Reviewing’: illustration of social activity and
communication. The collaborative, 2-way nature of
communication is evident as are the ad-hoc ‘open’ links to
planning materials (shown by dotted links) 201
Figure 6.16 ‘Semi-Autonomous working’: illustration of social activity
and communication. The agents are primarily linked to the
planning materials that are being worked upon with ‘ad-hoc’
open channels to other Headquarters staff (shown by the
dotted line) 202
Figure 6.17 Illustration of social network where the communication
links between nodes have been annotated with the media or
modality that facilitates it. The sub-units and higher formation
are geographically remote from the Battlegroup Headquarters 204
Figure 6.18 Key to enhanced OSD symbology 206
Figure 6.19 OSD representation of the ‘prepare to make a plan’ phase 207
Figure 6.20 OSD representation of the phase dealing with question 1 of
the Combat Estimate 207
Figure 6.21 OSD representation of the phase dealing with questions 2 and
3 of the Combat Estimate 208
Figure 6.22 OSD representation of the phase dealing with questions 4 and
5 of the Combat Estimate 208
Figure 6.23 OSD representation of the phase dealing with questions 6 and
7 of the Combat Estimate 209
Figure 6.24 OSD representation of the concluding phases of the Combat
Estimate and execution of the plan 210
Figure 6.25 Overview of systems level knowledge for the CAST scenario
represented via propositional network 211
Figure 6.26 Knowledge objects associated with question 1 of the Combat
Estimate technique 212
Figure 6.27 Knowledge objects associated with questions 2 and 3 of the
Combat Estimate technique 213
Figure 6.28 Knowledge objects associated with question 4 of the Combat
Estimate technique 214
Figure 6.29 Knowledge objects associated with question 5 of the Combat
Estimate technique 215
Figure 6.30 Knowledge objects associated with questions 6 and 7 of the
Combat Estimate technique 216
Figure 6.31 Knowledge object activities with putting the plan into effect 217

Figure 7.1 Generic process model of command and control 232

 List of Tables

Table 2.1 Sample of mathematical metrics for key constructs

in command and control 14
Table 2.2 Propositional network metrics to detect emergent
property of SA in relation to key knowledge objects 23
Table 2.3 Relating SAS-050 variables to Scud Hunt Coefficients 28
Table 2.4 Mapping Dekker’s SAS-050 models 30
Table 2.5 Worked example of an abstraction decomposition
space from a militaristic work domain 40
Table 2.6 Summary of modelling perspectives and broad emergent
properties 46
Table 2.7 Informal classification of model types/typologies
with defined NEC benefit criteria 48

Table 3.1 EAST analyses 50

Table 3.2 Summary of EAST methods review 52
Table 3.3 Agents involved in switching scenario 58
Table 3.4 Switching scenario CDA results 58
Table 3.5 Extract of CDA analysis 59
Table 3.6 Agent association matrix 60
Table 3.7 SNA results 60
Table 3.8 Operational loading results 62
Table 3.9 CDM probes 63
Table 3.10 CDM Phase 1: First issue of instructions 64
Table 3.11 CDM Phase 2: Deal with switching requests 65
Table 3.12 CDM Phase 3: Perform Isolation 66
Table 3.13 CDM Phase 4: Report back to NOC 67
Table 3.14 Observation transcript extract 73
Table 3.15 Example HTA plans 81
Table 3.16 CDA teamwork taxonomy 87
Table 3.17 CUD summary table 89
Table 3.18 Agents involved in the return to service scenario 100
Table 3.19 Agent association matrix 100
Table 3.20 Agent centrality (B-L Centrality) 100
Table 3.21 Agent sociometric status 101

Table 4.1 Glossary of abbreviations 120

Table 4.2 The main agents involved in the mission 123
Table 4.3 Air threat scenario CDA results 126
xii Modelling Command and Control
Table 4.4 Air threat scenario CDA results in HTA stages 126
Table 4.5 PWO subsurface scenario CDA results 127
Table 4.6 Subsurface threat scenario CDA results in HTA stages 127
Table 4.7 PWO surface scenario CDA results 129
Table 4.8 Surface threat scenario CDA results in HTA stages 129
Table 4.9 Extract of CDA analysis 130
Table 4.10 Operational loading for the air threat scenario 134
Table 4.11 Operational loading for the subsurface threat scenario 134
Table 4.12 Operational loading for the surface threat scenario 136
Table 4.13 Analysis of core knowledge objects against the
seven phases of operation 154

Table 5.1 E3D mission scenario CDA results 162

Table 5.2 E3D mission CDA results in HTA stages 162
Table 5.3 Extract of CDA analysis where 1 = low, 2 = medium
and 3 = high 164
Table 5.4 List of agents involved in the E3D operations 166
Table 5.5 Matrix showing association between agents on the E3D 166
Table 5.6 Comparison for positional centrality (degree dimension) 167

Table 6.1 Illustration of a partially completed DSO matrix.

The matrix lists the order in which TAIs are to be
dealt with and the associated resources to be used 188
Table 6.2 Illustration of a simplified decision support matrix 189
Table 6.3 Coordination demand dimensions 194
Table 6.4 CAST scenario CDA results 195
Table 6.5 CAST scenario CDA results in HTA stages 196
Table 6.6 Advantages and disadvantages of existing comms media 198
Table 6.7 Network metrics illustrating centrality (key agents
in the scenario) and density (network connectivity) for
the social network as a whole 203
Table 6.8 Network metrics illustrating centrality (key agents in
the scenario) and density (network connectivity) for
the activity stereotypes of Briefing, Reviewing
and Semi-Autonomous Working 203
Table 6.9 Technology/facilitation/modality matrix. Shading
shows the match between technology and modality 205
Table 6.10 Summary of key knowledge objects active within
each scenario 218

Table 7.1 Taxonomy of command and control activities 226

Table 7.2 The ‘receive’ activities taxonomy 227
Table 7.3 The ‘planning’ activities taxonomy 228
Table 7.4 The ‘rehearsal’ activities taxonomy 229
Table 7.5 The ‘communicate’ activities taxonomy 229
Table 7.6 The ‘request’ activities taxonomy 230
 List of Tables xiii
Table 7.7 The ‘monitor’ activities taxonomy 230
Table 7.8 The ‘review’ activities taxonomy (The taxonomies
are related to comprehensive HTAs of each individual
scenario in question. These can be found separately in the
individual EAST analysis reports that deal with each
live scenario.) 231
This page intentionally left blank
 Preface

Readers of this book might be interested to know something of the genesis of ideas
and projects that led to its conception. The basic idea to develop a new approach to
the analysis and representation of command and control began early in 2002, when
we decided to form a consortium to bid for the Ministry of Defence research contract
for a Human Factors Integration Defence Technology Centre (HFI DTC). At that
time there was a paradigm shift in defence research spending, with the desire to set
up consortia based on collaboration between academia and industry. We formed a
consortium led by Aerosystems International, with Birmingham, Brunel and Cranfield
Universities together with Lockheed Martin, MBDA and SEA. In the competitive
tendering process, we had to develop a set of research ideas that would define the
work of the consortium for three years, from 2003 to 2006. One of the strands of
work focused on Command and Control. Part of this work was to be focused on the
analysis of military Command and Control using a new approach (which was to be
developed). Because part of the remit of the HFI DTC was to examine spin-in to the
defence domain from the civil sector (and vice versa), we looked at Command and
Control in the emergency services and other civil applications. However, this book
focuses almost exclusively on research in the military domain.
As history testifies, we were successful in winning the initial HFI DTC contract
and have also won a follow on contract, which takes our programme of research
from April 2003 to March 2009. From those humble beginnings in 2002, we have
developed a new approach to the description and analysis of Command and Control
called Event Analysis of Systemic Team-work (EAST) as well as a sister method
called Workload, Error, Situation awareness, Time and Team-work (WESTT).
This book presents the EAST method and applications of the method in the Navy,
Air Force and Army Command and Control. EAST brings together a collection of
Human Factors methods in an integrated manner, rather than developing completely
new methods from scratch. The novelty of the approach is in the way in which the
various methods have been integrated and the network models that the application
of the methods produce as outputs. These models enable the analyst to consider
the Command and Control system under scrutiny in terms of the task, social and
knowledge networks, as well as the inter-relations between these networks.
At the time of writing the HFI DTC comprises:
xvi Modelling Command and Control

Aerosystems The University of Brunel

International Birmingham University
Dr David Morris Dr Chris Baber Professor Neville A. Stanton
Dr Karen Lane Professor Bob Stone Paul S. Salmon
Stephen Brackley Dr Huw Gibson Dr Guy H. Walker
Linda Wells Dr Rob Houghton Dr Dan Jenkins
Kevin Bessell Richard McMaster Amardeep Ajula
Nic Gibbs Dr James Cross Kirsten Revell
Kelly Maddock-Davies Robert Guest

Cranfield Lockheed Martin UK Systems Engineering

University and Assessment Ltd
Dr Don Harris Mick Fuchs Pam Newman
Dr John Huddlestone Lucy Mitchell Dr Anne Bruseberg
Dr Geoff Hone Mark Linsell Dr Iya Solodilova-Whiteley
Jacob Mulenga Ben Leonard Mel Lowe
Ian Whitworth Rebecca Stewart Ben Dawson
Andy Farmilo Jonathan Smalley
Antoinette Caird-Daley Dr Anne Bruseberg
Jon Pike
Louise Forbes

DSTL MBDA Missile Systems

Geoff Barrett Dr Carol Mason
Bruce Callander Grant Hudson
Jen Clemitson Roy Dymott
Colin Corbridge
Roland Edwards
Alan Ellis
Jim Squire
Debbie Webb

This book may be approached in different ways by different readers. For an

overview of the entire book, read chapter one. For those wishing to get to grips
with the modelling literature, read chapter two for a review and chapter seven for
our own Command and Control model. Those readers that want to apply the EAST
method to their domain, we recommend reading chapter three plus one of the case
studies in chapters four, five or six. For those readers who want to understand some
aspects of Command and Control in the military domain, read chapters four, five and
 Preface xvii
six. We have tried to write the book so that each of the chapters can stand alone, but
inevitably there is some inheritance from previous chapters, which may require the
reader to dip in and out to make the most of each chapter. The index section should
help guide the reader.
The research work and spirit of the HFI DTC is such that collaboration between
all of the organisations was integral to our approach, combining the best mix of
academic and industrial knowledge and skills. We have learnt a lot about the military
and about collaborating in this new approach to research funding in the Ministry of
Defence. The willingness to participate and the can-do attitude of our Armed Forces
is inspiring. We are very grateful, in particular, to the personnel at HMS Dryad in
Portsmouth, CASTT at the Land Warfare Centre in Warminster and the E3D Sentry
crews at RAF Waddington (8 and 23 Squadron) who have made this research
possible. We hope that you, the reader, will find this book both useful and interesting
in your own research endeavours. We have certainly enjoyed all of the experiences
that have led to us bringing the fascinating area of Human Factors research onto the
printed pages in front of you.

Neville A. Stanton
Chris Baber
Don Harris
This page intentionally left blank
 Acknowledgements

The Human Factors Integration Defence Technology Centre is a consortium of

defence companies and UK Universities working in cooperation on a series of defence
related projects. The consortium is led by Aerosystems International and comprises
Birmingham University, Brunel University, Cranfield University, Lockheed Martin,
MBDA and SEA.
We are grateful to DSTL who have managed the work of the consortium, in
particular to Geoff Barrett, Bruce Callander, Jen Clemitson, Colin Corbridge, Roland
Edwards, Alan Ellis, Jim Squire and Debbie Webb.
This work from the Human Factors Integration Defence Technology Centre
was part-funded by the Human Sciences Domain of the UK Ministry of Defence
Scientific Research Programme.
This page intentionally left blank
 Senior Author Biographies

Professor Neville A. Stanton

HFI DTC, BIT Lab, School of Engineering and Design
Brunel University, Uxbridge, Middlesex
UB8 3PH
UK
neville.stanton@brunel.ac.uk
http://www.brunel.ac.uk/about/acad/sed/sedres/dm/erg/

Professor Stanton holds a Chair in Human-Centred Design and has published over
75 international academic journal papers and 10 books on human-centred design.
He was a Visiting Fellow of the Department of Design and Environmental Analysis
at Cornell University in 1998. In 1998 he was awarded the Institution of Electrical
Engineers Divisional Premium Award for a co-authored paper on Engineering
Psychology and System Safety. The Ergonomics Society awarded him the prestigious
Otto Edholm medal in 2001 for his contribution to basic and applied ergonomics
research. Professor Stanton is on the editorial boards of Ergonomics, Theoretical
Issues in Ergonomics Science and the International Journal of Human Computer
Interaction. Professor Stanton is a Chartered Occupational Psychologist registered
with The British Psychological Society, a Fellow of The Ergonomics Society. He
has a BSc in Occupational Psychology from Hull University, an MPhil in Applied
Psychology from Aston University, and a PhD in Human Factors, also from Aston.

Dr Chris Baber
HFI DTC, Electronic, Electrical and Computer Engineering
The University of Birmingham, Birmingham
B15 2TT
UK
c.baber@bham.ac.uk
http://www.eee.bham.ac.uk/baberc

Dr Baber is a Reader in Interactive Systems Design, a member of the Human Interface

Technology Group, Educational Technology Group and the Pervasive Computing
Groups within the School of Electronic and Electrical Engineering, and an Associate
Member of the Sensory Motor Neuroscience Group in the School of Psychology. He
joined the University in 1990 as a lecturer on the MSc Work Design and Ergonomics
course, before moving to his new post in 1999. In 2002, he was promoted to Senior
Lecturer, and to Reader in 2004. Dr Baber has been involved in ergonomics research
since 1987. His main area of interest is in the ways in which people make sense of
xxii Modelling Command and Control
and make use of ‘everyday technology’ and how ‘everyday’ skills can be brought to
bear in people’s interaction with computers. At present, he is working on two funded
projects: Wearable Computers for Crime Scene Investigation (funded by EPSRC
2004–2007) and the Defence Technology Centre for Human Factors Integration
(2003–2009). Work on the latter project covers further development of wearable
and mobile computers, design and development of a performance-modelling tool,
and research into distributed operations. His early research was into speech-based
interaction with computers, and he was particularly interested in issues of error
correction and dialogue design.

Dr Don Harris
HFI DTC, Department of Human Factors
School of Engineering, Cranfield University
Cranfield, Bedford
MK43 0AL
UK
d.harris@cranfield.ac.uk
http://www.cranfield.ac.uk/soe/hf/biog_d_harris.htm

Don Harris is a Reader and Course Director for the MSc in Human Factors in Health
and Safety at Work and MSc Ergonomics and Safety at Work. He is Director of
Flight Deck Design and Aviation Safety Group and of the Defence Human Factors
Group. His principal teaching and research interests lie in the design and evaluation
of flight deck control and display systems, accident investigation and analysis, system
safety and flight simulation and training. Don is a Fellow of both the Ergonomics
Society and the Higher Education Academy. He is also a chartered Psychologist. He
is Chairman of the International Conference series on Engineering Psychology and
Cognitive Ergonomics. He sits on the editorial boards of the International Journal
of Applied Aviation Studies and Cognition, Technology and Work and is Co-Editor in
Chief of the Journal Human Factors and Aerospace Safety (published by Ashgate).
In 2006 Don was a member of a team that received the Royal Aeronautical Society
Bronze Award for advances in Aerospace and was invited to address the Chinese
Academy of Sciences in Beijing.
 Contributing Authors

Dan Jenkins, Paul S. Salmon, Guy H. Walker and Mark Young

HFI DTC, BIT Lab, School of Engineering and Design
Brunel University, Uxbridge, Middlesex, UB8 3PH UK

Rob Houghton and Richard McMaster

HFI DTC, Electronic, Electrical and Computer Engineering
The University of Birmingham, Birmingham, B15 2TT UK

Alison Kay
HFI DTC, Department of Human Factors
School of Engineering, Cranfield University
Cranfield, Bedford, MK43 0AL UK

Linda Wells
HFI DTC, Aerosystems International Ltd
Alvington, Yeovil, Somerset, BA22 8UZ UK

Roy Dymott
HFI DTC, MBDA Missile Systems
Golf Course Lane, Filton, Bristol, BS34 7QW UK

Mark Linsell and Rebecca Stewart

HFI DTC, Lockheed Martin UKIS
Building 7000, Langstone Technology Park
Langstone, Havant, Hampshire, PO9 1SW UK

Geoff Hoyle, Mel Lowe and Kerry Tatlock

SEA House, Building 660, The Gardens
Bristol Business Park, Coldharbour Lane, Bristol, BS16 1EJ UK
This page intentionally left blank
 Chapter 1

Overview

The Defence Technology Centre for Human Factors Integration (DTC HFI) is a
research consortium comprised of academic institutions and defence companies, part
funded by the Human Sciences Domain of the UK Ministry of Defence Scientific
Research Programme. Human Factors Integration is about:

… providing a balanced development of both the technical and human aspects of equipment
provision. It provides a process that ensures the application of scientific knowledge about
human characteristics through the specification, design and evaluation of systems.’ (MoD,
2000, p.6)

The aim of the book is to show the application of the HFI-DTC’s Event Analysis
for Systemic Teamwork (EAST) methodology to military Command and Control
applications. The chapters in the book report on a series of investigations from a
Human Factors perspective. The book begins with an overview of Command and
Control models. Then the EAST methodology is explained. This is followed by three
case studies in different domains: Sea, Air and Land. The final chapter draws the
material from the case studies, and other research, to present a generic activities
model of Command and Control.

Chapter 2 Modelling Command and Control

With contributions from Rob Houghton, Paul S. Salmon and Guy H. Walker

Since its inception, just after the Second World War, Human Factors research has
paid special attention to the issues surrounding human control of systems. Command
and Control environments continue to represent a challenging domain for Human
Factors research. We take a broad view of Command and Control research, to
include C2 (Command and Control), C3 (Command, Control and Communication),
and C4 (Command, Control, Communication and Computers) as well as Human
Supervisory Control paradigms. This book will present case studies in diverse
military applications (for example, land, sea and air) of command and control. While
the domains of application are highly diverse, many of the challenges they face share
interesting similarities.
2 Modelling Command and Control
Chapter 3 Event Analysis of Systemic Teamwork

With contributions from Dan Jenkins, Paul S. Salmon and Guy H. Walker

The EAST methodology is particularly suited to the analysis of team-based or

collaborative activity, such as that seen in C4i environments (command, control,
communication, computers and intelligence). The method was originally developed
for this purpose and applications so far have proved extremely successful, highlighting
its suitability for such applications. The EAST methodology is an exhaustive set of
data collection, analysis and representation methods. A number of different analyses
are conducted and various perspectives on the scenario(s) under analysis are offered.
In its present form, the EAST methodology offers the following analyses of a
particular C4i scenario:

• A step-by-step (goals, sub-goals, operations and plans) description of the

activity in question.
• A definition of roles within the scenario.
• An analysis of the agent network structure involved (for example network
type and density).
• A rating of co-ordination between agents for each team-based task step and an
overall co-ordination rating.
• An analysis of the current technology used during communications between
agents and also recommendations for novel communications technology.
• A description of the task in terms of the flow of information, communications
between agents, the activity conducted by each agent involved and a timeline
of activity.
• An analysis of agent centrality, sociometric status and betweenness within the
network involved in the scenario.
• A definition of the key agents involved in the scenario.
• A cognitive task analysis of operator decision making during the scenario.
• A definition of the knowledge objects (information, artefacts etc) required and
the knowledge objects used during the scenario.
• A definition of shared knowledge or shared situation awareness during the
scenario.

The EAST methodology is relatively simple to use. The method requires an initial
understanding of HF and experience in the application of HF methods. However,
from analyst reports it can be concluded that the EAST methodology is relatively
simple to apply. The method’s ease of use is heightened when compared to the
exhaustive output that is generated from an EAST analysis. The EAST methodology
is generic and can be applied in any domain in which collaborative activity takes
place. As an overall conclusion, it is felt by the authors that the EAST methodology
has been successful in its applications thus far and the methodology is perfectly
suited to the analysis of Command and Control activity. Each EAST application
has produced valid and useful results that are currently forming the basis for the
development of a Command and Control model.
 Overview 3
Chapter 4 Case Study at HMS Dryad

With contributions from Roy Dymott, Rob Houghton, Geoff Hoyle, Mark Linsell,
Richard McMaster, Paul S. Salmon, Rebecca Stewart, Guy H. Walker and Mark
Young

This study was conducted in order to analyse C4i in the Royal Navy domain.
Researchers were given access to one of the Navy’s training establishments – the
Maritime Warfare School – HMS Dryad in Southwick, Hampshire. Observations
were made during Command Team Training (CTT). This programme involved
training the Command Team of a warship in the skills that would be necessary for
them to defend their ship in a multi-threat environment.
The aim of this study was to apply the EAST methods to assess the communication
and command on board the Type 23 frigate. The study uses the methodology to
explore a complex communication system between sixteen team members where
effective communication, decision making and coordination are essential to task
success, for this highly distributed communication network. Three scenarios (air
threat, subsurface threat and surface threat) were observed and analysed. The
scenarios were different, and due to the complexity of the task only individual crew
members could only be observed at any time as opposed to the scenario being viewed
as a whole. However an overall idea of communications could be ascertained from
these individual observations.
Data are presented in the form of Social Network Analysis, Coordination Demand
Analysis and Propositional Network Analysis. Shared situation awareness is also
considered in this case study using the propositional networks. Knowledge objects
are identified for the whole of the mission as well as for the phases of the mission.
This gives an indication of where there is sharing of information. The networks
also suggest where knowledge is built on as the phases of the mission progress. The
sharing of information could be as a result of direct communications between agents
or through the use of the radio networks.
The individual scenarios show that the social networks are not particularly well
distributed. However when the scenarios are amalgamated into one scenario, as may
happen in a real life threat, it becomes a much denser network with better levels of
participation. The network then becomes a split communications network with the
Anti-Air Warfare Officer (AAWO) and Principal Warfare Officer (PWO) being the
central nodes. Overall the analyses indicate that the Type 23 crew use a distributed
network. Each crew member is connected (communication links) to other crew
members and hence there are several channels with which communication can travel
or information be shared.
4 Modelling Command and Control
Chapter 5 Case Study in RAF Boeing E3D Sentry

With contributions from Alison Kay, Mel Lowe, Paul S. Salmon, Rebecca Stewart,
Kerry Tatlock and Linda Wells

This study was conducted on board a Boeing ED3 AWACS (Airborne Warning
and Control System) aircraft in the Royal Air Force. The crew’s role is as airborne
surveillance, command and control, and weapons control and operations. They are
essentially an aerial patrol that is able to provide information on enemy movement
as well as providing control and the direction of friendly defensive and offensive
airborne operations. The study used the EAST methodology to explore a complex
communication system between eighteen team members and external agencies,
where effective communication, decision making and coordination are essential to
task success for this highly distributed communication network.
Five different missions were flown and observed. The missions were different,
and due to the complexity of the task only individual crew members could be
observed as opposed to the scenario being viewed as a whole. However an overall
idea of communications could be ascertained from these individual observations.
Data are presented in the form of Social Network Analysis, Coordination Demand
Analysis and Propositional Network Analysis. Shared situation awareness is also
considered in this case study using the propositional networks. Knowledge objects
are identified for the whole of the mission as well as for the phases of the mission.
This gives an indication of where there is sharing of information. The networks
also suggest where knowledge is built on as the phases of the mission progress.
The sharing of information could be as a result of direct communications between
agents or through the use of the radio networks or radar screens. Overall the analyses
indicate that the E3 crew use a fairly distributed network. Each crew member is
connected (communication links) to other crew members and hence there are several
channels within which communication can travel or information can be shared.

Chapter 6 Case Study in Battle-Group HQ

With contributions from Dan Jenkins, Richard McMaster, Rebecca Stewart, Guy H.
Walker and Linda Wells

These studies were conducted at the Command And Staff Training centre in
Warminster. Military command and control relies heavily on tasks that require
interaction with other team members, and where this is manifest, team working is
principally concerned with the communication of information and development of
Situation Awareness (SA). A relatively simple, yet robust, technological infrastructure
underpins team tasks. It is heavily reliant on a combination of verbal communications
and/or the translation of various planning ‘products’ into an integrated, collective,
4D spatial and temporal ‘image’ of the battle-space. It appears to be in this domain,
based on the Communications Usage Diagram (CUD) method, that Network
Enabled Capability (NEC) technology has much to offer. The assumption is that if
 Overview 5
the state of SA can be more rapidly and accurately acquired (and there seems little
doubt that new technology offers this potential), then decision superiority can be
achieved more quickly. If SA can also be shared in optimal ways throughout the
system (which again, new technology appears to provide for), then unity of effort
can also be achieved.
The Hierarchical Task Analysis (HTA) specifies how the configuration of people
and technology changes in a task and context dependant manner. Three activity
stereotypes are defined: semi-autonomous working, briefing and reviewing. The
social network configures (and re-configures) itself numerous times during the
enactment of military command and control (and the Combat Estimate specifically).
As the network is re-configured, its constraints in terms of communications, density
and centrality change. The design of NEC paradigms, therefore, is revealed to
be more than just a consideration of technology in isolation. The specification of
technology may be appropriate for one configuration, but inappropriate for another.
The combination of HTA and SNA appears to provide one route into addressing this
issue.
The knowledge base that underpins effective SA at the systems level changes in
response to task phase, but also arises as a property of the constraining features of
the configuration of people and technology. Systems level SA, at this summary level
of analysis, appears to support, and be congruent with, task goals (as specified by
the HTA).
In summary, the emergent properties associated with military (and indeed any)
command and control scenario relate to the interplay between task, social and
propositional networks.

Chapter 7 Development of a Generic Process Model of Command and

Control

With contributions from Rob Houghton, Dan Jenkins, Richard McMaster, Paul S.
Salmon, Rebecca Stewart, Guy H. Walker and Mark Young

Despite the differences in the domains, the command and control applications share
many common features. First, they are typified by the presence of a central, remote,
control room. Data from the field are sent to displays and/or paper records about the
events as they unfold over time. Second, there is (currently) considerable reliance on
the transmission of verbal messages between the field and the central control room.
These messages are used to transmit report and command instructions. Third, a good
deal of the planning activities occurs in the central control room, and these are then
transmitted to the field. There are collaborative discussions between the central
control room and agents in the field on changes to the plan in light of particular
circumstances found in-situ. Finally, the activities tend to be a mixture of proactive
command instructions and reactive control measures. It is hypothesised that one of
the determinants of the success or failure of a command and control system will be
the degree to which both the remote control centre and agents in-the field can achieve
shared situational understanding about factors such as: reports of events in the field,
6 Modelling Command and Control
command intent, plans, risks, resource capability, and instructions. This places a
heavy reliance on the effectiveness of the communications and media between the
various parties in the command and control system.
Analysis of the task analyses from these three domains led to the development
of a taxonomy of command and control activities. The resultant data from the
observational studies and task analyses were subject to content analysis, in order
to pick out clusters of activities. These clusters were subjected to thematic analysis
consistent with a ‘grounded theory’ approach to data-driven research. It was possible
to allocate most of the tasks in the task analysis to one of these categories. To this
extent, the building of a generic model of command and control was driven by the
data from the observations and task analyses.
From development of a series of taxonomies (called receive, planning, rehearsal,
communicate, request, monitor and review), and an analysis of the previous command
and control models, it was possible to develop a generic process model. Construction
of the model was driven by the data collected through observation from the different
domains, and the subsequent thematic analysis and taxonomic development. In the
tradition of ‘grounded theory’ the generic command and control model was as a
result of our observations, rather than an attempt to impose any preconceived ideas
of command and control. This may account for many of the differences between the
current model developed in the course of the current research and those models that
have come before it.
It is proposed that the command and control activities are triggered by events,
such as the receipt of orders of information, which provide a mission and a description
of the current situation of events in the field. The gap between the mission and the
current situation lead the command system to determine the effects that will narrow
that gap. This in turn requires the analysis of resources and constraints in the given
situations. From these activities, plans are developed, evaluated and selected. The
chosen plans are then rehearsed before being communicated to agents in the field.
As the plan is enacted, feedback from the field is sought to check that events are
unfolding as expected. Changes to the mission or the events in the field may require
the plan to be updated or revised. When the mission has achieved the required effects,
the current set of command and control activities may come to an end.
The model distinguishes between ‘command’ activities and ‘control’ activities.
Command comprises proactive, mission-driven, planning and co-ordination
activities. Control comprises reactive, event-driven, monitoring and communication
activities. The former implies the transfer of mission intent whereas the latter implies
reaction to specific situations.
 Chapter 2

Modelling Command and Control

With contributions from Rob Houghton, Paul S. Salmon
and Guy H. Walker

Purpose and Scope

In order to situate the generic process model within its wider research context a review
of the command and control literature has been conducted. Of particular concern to this
review are the questions of what would define a useful ‘model’ of command and control.
The literature provided several approaches to the challenge of modelling command
and control, and the review addressed a number of broad questions, including:

• Model metrics: How are aspects of command and control measured or

expressed either quantitatively or qualitatively?
• Measures of modelling outcome: How does the model define good or bad
command and control system performance?
• Degrees of model reconfigurability: Is the model tied to a particular type of
activity or situation? Is it flexible enough to be reconfigured for use across a
wide range of settings and contexts?
• Construct validity and reliability: Is the theoretical basis of the model sound?
• Extent, nature and degree of dependency upon constraints and assumptions:
Are the assumptions the model is based on reasonable? Are the formal
constraints within which the model falls appropriate or are they overly
restrictive or too poorly specified?

These concerns cover the wider aim of extracting broad modelling trends from the
prevailing literature, examples of paradigms and approaches, and the derivation of
typologies and categories of outcome measure. The aim of the review, therefore, is
not to provide an exhaustive account of every permutation of C4i related models
but to illustrate the range and scope of relevant approaches and paradigms. Prior to
presenting this review, the chapter will address generic issues associated with the
problem of creating and using models.

Models

As Pew and Mavor (1998), in their report on the activity of the US National Research
Council’s Panel on Modeling Human Behavior and Command Decision Making:
Representations for Military Simulations, the word ‘model’ can cover a host of
definitions, from physical mock-ups to analytical descriptions. In Pew and Mavor
8 Modelling Command and Control
(1998) the term ‘model’ was taken to imply that ‘...human or organizational behavior
can be represented by computational formulas, programs or simulations’ (p.11). In
this book, the notion of ‘model’ is much lighter, that is, our concern is to develop a
useful description that can be applied, as a framework, to understand the operation of
command and control systems. Thus, the intention is not to generate algorithms that
can predict how a command and control system should operate, so much as to develop
a framework in which to describe how such systems do operate. There are numerous
modelling challenges underlying command and control. The difficulty contained by
these challenges can perhaps be summarised by the fact that the ‘real world is made
from open, interacting systems, behaving chaotically’ (Hitchins, 2000), and in the
case of human actors, non-linearly. Complex systems like command and control
also possess various real-time properties that cannot be considered ‘designed’ as
such, they sometimes merely ‘happen’ (Hitchens, 2000). Therefore the notion of a
commander representing something akin to the conductor of an orchestra is in some
cases entirely false (Hitchens, 2000). Also, unlike neat linear systems the possibility
exists (increasingly so with NEC) for there to be no clear boundaries between certain
system elements, as well as no beginning and no end, given that goals are more or
less externally adaptive. For the purposes of this book, a descriptive model serves
four primary purposes: abstracting reality, simplifying complexity, considering
constraint and synthesising results.

Abstracting reality

A model is an abstraction of reality (Wainwright and Mulligan, 2004), or ‘a

representation that mirrors, duplicates, imitates or in some way illustrates a pattern
of relationships observed in data or in nature. …’ (Reber, 1995, p.465). A model
is also a kind of theory, ‘a characterisation of a process and, as such, its value
and usefulness derive from the predictions one can make from it and its role in
guiding and developing theory and research’ (Reber, 1995, p.465). The purpose of a
model is, therefore, to explain attendant facts, to characterise them, to represent the
relationships between them, in a way that represents some form of direct analogue
to the phenomena under analysis but in the most parsimonious way possible or that
is appropriate.

Simplifying complexity

A model aims to simplify complexity. Complexity is related to the amount of information

needed to describe the phenomena under analysis. The closer that the phenomena under
analysis approaches complete randomness, the more data are needed until it ‘cannot
be described in shorter terms than by representing the [phenomenon] itself’ (Bar Yam,
1997). However, ‘Something is complex if it contains a great deal of information that has
a high utility, while something that contains a lot of useless or meaningless information
is simply complicated’ (Bar Yam, 1997 cited in Grand, 2000, p.140). The primary
purpose of the current generic model of C4i is to reduce complexity and particularly
when partnered with additional modelling techniques, to offer outcome metrics that
can detect and describe non-random emergent properties. Emergent properties exist
 Modelling Command and Control 9
where the ‘characteristics of the whole are developed (emerge) from the interactions
of their components in a non-apparent way’ (Bar Yam, 1997). Previous analyses of live
C4i scenarios using the EAST methodology have detected the presence of emergent
properties related to task, social and knowledge networks and in so doing demonstrate
that information contained within these scenarios may indeed be complex, but is far
from random. The optimal model of command and control can be defined as one that
is sensitive enough to detect cogent emergent properties, whilst containing merely
‘sufficient’ complexity to explain (and predict) these ‘widely observed properties
and behaviours in terms of more fundamental, or deeper, concepts’ (Wainwright
and Mulligan, 2004; Builder, Bankes and Nordin, 1999). These apparently simple
requirements are heavily tempered by the fact that:

the world being modelled has an inviolable nature; it cannot be exhaustively described.
We can model the world but we can always go back to find new perspectives for
describing what we are modelling, usually involving new perspectives on what constitutes
information (data), new languages for modelling, and new perspectives on the purpose for
constructing models. (Clancey, 1993, p.41)

Considering constraint

A crucial aspect of any modelling enterprise is the quality of the constraints and
assumptions within the model. For a model to have scientific viability it must feature
some variety of formal constraint. In simple terms, constraints are the limits on what
a model can and cannot do and should (for the most part) correspond with the real
world. For example, if we were modelling some aspects of human manual labour,
this would naturally dictate a constraint that within the model any one individual
has at maximum two arms and two legs. A model may be otherwise impressive in
any other aspect but if its constraints are incorrect or inappropriate then the rest of
the model is invalid. A model may even appear to give the ‘right’ answer if treated
as a ‘black box’ model where we concern ourselves only with the input and the
output, but if it requires the representation of impossibilities (for example, three
armed radio operators, speed of light reaction times and so on) to generate those
results it is questionable whether the model is of genuine theoretical value: a good
model ‘plays by the rules’ so to speak. Whilst it might be obvious how many limbs to
ascribe to an individual, less concrete variables are more difficult to accurately and
uncontroversially constrain; things like attentional capacity, situation awareness and
teamwork. Whilst these are essential constructs in the conceptualisation of command
and control, they have long suffered from problems in their precise characterisation
(and in some cases possibly always will given that mental qualities like attention
and awareness are possibly products of reification). Furthermore, where models are
themselves particularly complex or abstract by nature of their operating principles
(for example, a statistical Hidden Markov-Chain Model) there can be additional
challenges in translating realistic constraints into parameters (for the author) but also
assessing the constraints (for the reader); not all models are so transparent that one
might easily relate, say, a set of arrows on a graph or a set of numbers in a vast matrix
with an inappropriate ‘take’ on the real world.
10 Modelling Command and Control
Synthesising results

Chapters 4, 5 and 6 contain a series of studies conducted in a wide range of domains.

A generic model allows the results from these studies to be combined into a coherent
account, and the model itself provides a motivation for the analysis, that is, if the
results do not bear out the relationships implied by the model, then it needs to be
modified. The model would also provide a means for comparing and contrasting
different approaches to command and control across the different domains, and also
allows analysts to tease out those factors that seem to be contextual and those that are
more generic aspects of performing command and control activity.

Specification

Defining command and control

In this book, we follow NATO (1988) in separating the concept of ‘command’ from
that of ‘control’. For NATO (1988), ‘command’ is ‘…the authority vested in an
individual…for the direction, coordination and control of military forces’. This
implies that an individual will be given the role of Commander, that this individual
will (through this role) be imbued with sufficient authority to exercise command,
and (by implication) this command will involve defining the goal (intent, effect) that
Forces under the individual’s command will achieve.
Builder, Bankes and Nordin (1999) follow NATO (1988) in their definition of
command and control:

Command and control: The exercise of authority and direction by a properly designated
[individual] over assigned [resources] in the accomplishment of a [common goal].
Command and control functions are performed through an arrangement of personnel,
equipment, communications, facilities, and procedures which are employed by a
[designated individual] in planning, directing, coordinating, and controlling [resources]
in the accomplishment of the [common goal]. (p.11)

Combining command (authority) with control (the means to assert this authority)
leads to ‘unity of effort in the accomplishment of a [common goal]’ (Jones, 1993,
p.2). Despite the militaristic undertones, the notion of command and control is
itself generic. The separation of activity associated with command from that of
control has a number of useful benefits. For example, in military parlance, ‘Mission
Command’ involves passing a Commander’s Intent down a chain of command in
such a way as to allow lower levels to elaborate and develop the command in the
light of contextual demands. Thus, a Commander might have an Intent to ‘secure
a route from town A to town B’. This might be interpreted by Officers in the field
as (a) patrol the road between the towns; (b) secure a bridge along the route; (c)
repel any attacks on the bridge; and (d) disrupt any attempts to damage the road.
Each of these activities could be passed to specific Units, under the command of an
Officer who will then define these activities according to contextual demands, for
example, activity (a) could be performed according to different time-intervals, using
 Modelling Command and Control 11
different resources, with greater or lesser shows of strength etc. In a similar manner,
the notion of Sectorisation in the Fire Service requires Officers to divide activities in
major incidents between different teams, for example, managing water supply, using
breathing apparatus. Each ‘Sector’ exercises different skills under the leadership of
different Officers. Taken together, these examples point to the exercise of Command
as the broad definition of a goal and the provision of appropriate resources to achieve
that goal, whereas Control is more concerned with the management of these resources
in the ongoing pursuit of that goal in the light of changing contextual demands.

Generic properties of command and control

Given the definitions above the (relatively) invariant properties of command and
control scenarios can be distilled down to the following three features:

• A common overall goal (this may, however, be comprised of different but

interacting sub-goals).
– Corollary: systems of command and control are goal-oriented systems.
• Individuals and teams acting individually or in unison.
– Corollary: there is the need to coordinate activity.
• Teams and sometimes individuals are often dispersed geographically.
– Corollary: there is the need to communicate and share ‘views’ on the
situation

Beyond the descriptive level, command and control by definition is a collection of

functional parts that together form a functioning whole. Command and control is a
mixture of people and technology, typically dispersed geographically.
In their consideration of command and control, Alberts and Hayes (2006) consider
the SAS-050 ‘cube’ space of possible command and control structures. This assumes
three broad Independent Variables,

• Distribution of information – this could be sent from one person to another, or

could be broadcast to all members of a network;
• Patterns of interaction – this could take the form of a top-down, hierarchical
command structure or could take a more open, or ‘distributed’ form of
management;
• Allocation of decision rights – this could have Intent originating from a single
source, for example, a Commander, or arising from some form of ‘democratic’
decision making.

Figure 2.1 shows how these three Independent Variables can be mapped onto a cube.
It is possible to place different forms of command and control structure in this cube.
Thus, for example, a ‘traditional’ hierarchical command structure could be located in
the bottom left-hand corner, whereas a ‘power to the edge’ structure could be located
in the top right-hand corner. Other forms of command and control structures could
be located throughout the cube.
12 Modelling Command and Control

Distribution of Edge
information organisation

Fully distributed

Patterns
of
interaction

Fully hierarchial

Allocation
Classic C2
of
decision rights

Figure 2.1 SAS-050 Space of C2

Network Enabled Capability

Network Enabled Capability (NEC) is a way of ‘doing’ command and control.

It is a term used to describe what is at present a nascent paradigm within British
military command and control, but also an extremely cogent modelling issue. It has
been described thus: ‘NEC is about the coherent integration of sensors, decision
makers, weapons systems and support capabilities to achieve the desired effect’
(MoD, 2005a). Diverse command, reconnaissance, weapons, support and decision
making assets will be interlinked by way of a ‘network of networks’, which will
deliver and transfer data and communications in a flexible manner. In essence this
is an information-based approach to command and control; aside from their obvious
operational capabilities, assets also take the role of active producers and consumers
of information. Expected benefits include:

• Increased provision of timely information allowing rapid response (ideally to

the point where the opposing force’s own decision loop is undercut).
• Improved interoperability (across domains, services, agencies, nations etc.).
• Increased tempo of operations.
• The development of more effective command and management structures
(likely to be marked by a reduction in or flattening of existing hierarchies and
increasing freedoms given to commanders lower in the chain of command).
 Modelling Command and Control 13
• Shared situation awareness across actors (that is, ‘singing from the same hymn
sheet’) that will in turn allow the emergence of the so-called synchronisation
effect. In other words self-coordinating group actions will be enabled by
improved information about what colleagues intend to do, are doing and have
done, and the operational picture they are responding to.

Success in achieving these aims appears to be contingent on a set of central

presuppositions about the nature of technology and human cognition: that shared
information can be actively shared through networking; that shared information
could become in the mind of operators shared knowledge (that is, it is meaningful
and supports action); shared knowledge in turn leads to shared situation awareness;
that shared situation awareness enables synchronisation and that ultimately
synchronisation results in operational effect.
In essence NEC represents the application to command and control of principles
and technologies of the so-called information age. In particular, the effect of the
Internet and related technologies within the commercial sphere to increase the
responsiveness and flexibility of businesses appears to have been a primary
inspiration (Kaufman, 2004).

Structural Models of Command and Control

The dominant perspective – the cybernetic paradigm

C4i is variously defined as a form of ‘management infrastructure […] for any […] large
or complex dynamic resource system’ (Harris and White, 1987). It is immediately
apparent based on this review that the dominant modelling perspective, the so-called
‘cybernetic paradigm’ (Builder et al., 1999) accords the structural aspect of command
and control particular prominence. Dockery and Woodcock (1993) are unequivocal
in stating that, ‘Since [command and control] […] concerns itself with providing
structure to combat, its description should be in terms of structure’ (p.64).
Under this structural/cybernetic paradigm a command and control scenario is
divided into linked functional parts that exchange and modify signals that can be
specified according to mathematical formulae (Builder et al., 1999). Or as similarly
expressed from a systems dynamics point of view, ‘the mathematical modelling of
an assemblage of components so as to arrive at a set of equations which represent the
dynamic behaviour of the system and which can be solved to determine the response
to various sorts of stimuli’ (Doebelin, 1972, p.4). An example of these ‘equations’,
and their relation to the dynamic behaviour and structure of command and control
‘systems’ are reproduced in Table 2.1 from Dockery and Woodcock (1993) merely
as an illustration of the outcome measures that can be derived.
14 Modelling Command and Control
Table 2.1 Sample of mathematical metrics for key constructs in command
and control (from Dockery and Woodcock, 1993, p.66)

Mathematical Tool Modelling Requirement

Catastrophe Theory To capture in both a time independent and time
dependant manner the global non-linear responses
of combat in terms of a few control variables,
therefore essentially presenting the commander’s
command and control perspective on the battle.
Category Theory To define measures of effectiveness and
to embed the whole command and control
modelling process in a larger context.
Cellular Automata To capture the self-structuring of combat
based on minimal nearest neighbour rule sets,
essentially to present the small unit command
and control perspective of combat.
Chaos Theory To capture the observed chaotic dynamical
nature of combat and to identify attractors and
to present the combat evolution perspective.
Fuzzy/Partial/Stochastic To describe attrition based combat processes.
Differential Equations
Entropy Computation To capture the effects of casualties
upon the structure of the fighting.
Fractals To capture deployment hierarchies.
Fuzzy Sets To capture the imprecision in all phases of the
military command process from data to orders.
Relativistic Information Theory To capture the idea that organisations move
relative to each other in a metaphorical sense
of internal organisational efficiency.
Langrangian Formulation To define classical, albeit heuristic, expressions
such as combat momentum (tempo).
Petri Nets To express the transactions that occur in a
command and control system between its elements.
System Dynamics To capture feedback and feed-forward
aspects of command and control
interacting with combat processes.
Games Theory ‘Ways in which strategic interactions among
rational players produce outcomes with
respect to the preferences […] of those
players, none of which might have been
intended by any of them’ (Ross, 2005).
 Modelling Command and Control 15
Lawson’s model of command and control

Under Lawson’s model (1981), command and control can be viewed as an

information processing chain with data flowing between the environment, one’s own
forces and the command centre. The model in Figure 2.2 epitomises this perspective.
The model is rooted in the idea that there is some desired state that the command
centre seeks to achieve. Data are extracted from the environment and processed. The
understanding of these data are then compared with the desired state. If there is any
discrepancy between the desired state and the current state, the command centre has
to make decisions about how to bring about the desired state. These decisions are
turned into a set of actions, which are then communicated to their own forces. The
data extraction cycle then begins afresh.
Lawson’s model owes much to the ideas of control theory. The comparison of
actual and desired states implies a feedback process and some form of regulation.
Central to his model, therefore, would be the ‘compare’ function. The feedback
involves control of ‘own forces’ to affect a change to the environment. The notional
‘actual’ and ‘desired’ states imply phenomena that can be described in terms of
quantitative, discrete data; in other words it is not easy to see how the model would
cope if the actual state was highly uncertain. Nor is it easy to see what would happen
if the changes to the environment led to consequences which lay outside the limits
defined by the discrete state. The model does indicate the central issue that command
can be thought of as working towards some specified effect or intent but suffers,
however, from its apparent reliance on a deterministic sequence of activities in
response to discrete events.

Sense COMMAND
CENTRE
ENVIRONMENT

Process

Own Forces
Compare Desired State

Decide

Act

Figure 2.2 An adapted version of Lawson’s model of the command and

control process
16 Modelling Command and Control
The OODA Loop

One of the most commonly used structural models of Command and Control
is referred to as the OODA Loop. This provides a tactical-level perspective on
Command and Control as a process involving Observation (of events in the field),
Orientation (of resources to deal with those events), Decision (of how best to manage
those resources in order to deal with those events, under some set of constraints), and
Action (the use of resources to deal with the events). Figure 2.3 shows an example
of the OODA Loop, together with some of the factors that could have a bearing on
performance.

The Headquarters Effectiveness Assessment Tool

Another example derived from the ‘cybernetic approach’ is the Headquarters

Effectiveness Assessment Tool (HEAT). HEAT’s raison d’etre is to provide an objective
measure of headquarters effectiveness based on the premise that effective headquarters
approach command and control activities in quantitatively different ways than
ineffective headquarters (Choisser and Shaw, 1993). The HEAT method, like Lawson’s
model above, is based on (mathematical) normative systems and optimal control
theories, whereby the ‘objective of the [commander] is to determine control activities
that will induce the evolution of the system towards an acceptable goal’ (Choisser
and Shaw, 1993; p.48). The command and control system, in common with Lawson’s
model, is viewed as a variation on a deterministic closed loop system, expressed in
very simple terms in Figure 2.4. The idea expressed in the HEAT methodology/theory
is that rational command and control systems will possess some of the anticipatory and
self-optimising properties of a normative system. In a practical sense HEAT is used
within war-game simulations and various mathematically derived outcome measures
are derived from the underlying control (and decision making) model.

Observation Orientation Decision Action

Implicit Implicit
Guidance Cultural Guidance
& Control Traditions & Control
Outside
Information
Genetic Unfolding
Heritage Analyses/ Decision Interaction
Unfolding Action
Observations Synthesis (Hypothesis) with
Circumstances (Test)
Environment
New
Information
Unfolding Previous
Environmental Experience
Interaction

Feedback

Feedback

Figure 2.3 The OODA loop

 Modelling Command and Control 17

Random
Disturbances

Observer Events System

Actions

Observations Delay Actuator

Commands

Estimator Perceptions Controller

Figure 2.4 Example of closed loop feedback control system (from Choisser
and Shaw, 1993)

The Network Centric Operations Conceptual Framework

The Network Centric Operations Conceptual Framework (NCO CF) comprises four
main elements. Information sources, which could be sensors, people, or any other
originator of intelligence; value-added services, which could involve the integration
of information from several sources; the Command and Control function; and
Effectors, which are the means by which actions are performed. Figure 2.5 shows
a schematic of the NCO CF process model. Notice how the intention is to contrast
the activity relating to Individuals (on the left-hand side) with that of Groups (on the
right-hand side), with a central role for ‘Quality of Interactions’. This figure can be
read as an elaboration of the previous structural models, in that its focus is on higher-
level ‘quality’ issues and their interactions.

Issues in structural modelling

The models that arise from this dominant perspective all tend to approximate in
varying degrees towards control theory models and as Builder et al. (1999) put it,
could just as easily apply to a thermostat as it could to a C4i system. Although
practitioners of the cybernetic paradigm would argue that much of the complexity
lies ‘behind’ such simple systems (within the mathematical metrics expressed in
Table 2.1, for example), the fact, at bottom, still remains. Overall, the specific
outcome metrics from these models are relatively hard to divine, but examples
include ‘catastrophe manifolds’, ‘butterfly landscapes’ (a form of five dimensional
graph), ‘control coefficients’, ‘time dependant behaviour’, etc. These mathematical
metrics provide measures related to overarching ‘mission based’ constructs such as
survivability and effectiveness. Changes in these overarching constructs are based
18 Modelling Command and Control

Figure 2.5 Network Centric Operations Conceptual Framework (adapted

from Alberts and Hayes, 2006)

on changes to (and the mathematical interplay of) independent variables such as

force strength, fire power, decision aids, etc. (Dockery and Woodcock, 1993).
The attractiveness of, yet inherent danger associated with, this modelling
perspective is perhaps hinted at by Doebelin (1972), who states, ‘From our intuitive
ideas about the cause-and-effect nature of the physical world, it is clear that if we
precisely define a model of a physical system and subject it to specific known inputs, the
outputs are completely determined’ (p.4). The problem is that the robust relationship
between cause and effect does not necessarily hold true for systems involving
human cognition as it might do for systems involving merely physical entities with
known properties and known input-output characteristics (Rasmussen, Pejtersen
and Goodstein, 1994). Simply put, cybernetic models ‘inadequately represent the
complex and idiosyncratic activities of humans in [command and control]’ (Builder
et al., 1999). ‘When the [structure] is put to work, the human elements change their
characteristics; they adapt to the functional characteristics of the working system, and
they modify system characteristics to serve their particular needs and preferences’
(Rasmussen et al., 1994). ‘It is the un-specifiable messiness of the neural system
– becoming organised in new ways at the time of the interaction itself – which gives
human behaviour its robust, always adaptive character’ (Clancey, 1993, p.41). In
modelling terms two strategies within the cybernetic paradigm can be implied as
 Modelling Command and Control 19
routes taken to attempt to overcome this inconvenience. Firstly, the human is simply
subsumed into a complete physical system, where any non-linearitys are catered
for using (ever more) complex mathematical techniques (for example, Dockery and
Woodcock, 1993). Secondly, the complex role of the human may be recognised, but
is still ultimately reduced and overlain on to an underlying control theoretic model
(Choisser and Shaw, 1993; Edmonds and Moss, 2005; Levine, 2005).

Summary of the cybernetic paradigm

In summary, the issue of model constraints appears to be felt particularly acutely and
while cybernetic models might provide a robust basis for understanding control, they
appear to be restricted in their ability to model command. This is simply because
the models tend to rely on the assumption that the role of C2 is to react to changing
events in the world. Thus, the models are ‘event-driven’ and reactive rather than
proactive or anticipatory.

Network Models

Introduction

If taken as a form of doctrine the structural/cybernetic perspective yields several

serious limitations for modelling the multi-faceted nature of command and control.
If, however, the notion of structure is assumed to be a component of a system (as
opposed to a complete characterisation), then structural aspects of command and
control can be modelled in several alternate and useful ways.
From the perspective of Organisational Theory an organisation can be defined
as ‘a collection of interacting and interdependent individuals who work toward
common goals […]’ (Duncan, 1981). Here the focus is directed onto ‘individuals’,
and their links and interrelations. The straightforward organisational chart is a simple
example of this and ‘shows the relationship between specific jobs or roles within
[an] organisation’ (Arnold, Cooper and Robertson, 1995, p.2). Put more explicitly,
an organisational chart represents the links between roles, where the hierarchical
organisation of roles is reflective of command (and the cybernetic paradigm), and
the links that exist between roles reflective of control (and the communications
between individuals).

Example: Hitchen’s N-squared chart

This notion of roles and links is related to a simplistic descriptive model of military
command and control presented by Hitchens (2000) in Figure 2.6.
In Hitchens’ N-squared (N2) chart, information intersects pairs of roles (shown
in capitals on the diagonal). The example given is militaristic, where the content
of information flow downwards from the commander is represented in the upper
quadrant and information flowing upwards to the commander in the lower quadrant.
The commander, therefore, interchanges ‘decision’ information to the operations
20 Modelling Command and Control

Figure 2.6 The N-squared (N2) chart from Hitchens (2000): A simplistic
generic model of command and control

role. Similarly, the logistics role interchanges information on ‘constraints’ to the

operations role. And so on. In basic terms this simplistic model expresses the links
between roles and the broad topic of information that the links are facilitating.

Example: Hierarchical Task Analysis

Hierarchical Task Analysis (HTA) can also be viewed, perhaps unconventionally

as a form of network model. Task analysis is the activity of collecting, analysing
and interpreting data on system performance (Annett and Stanton, 2000; Diaper and
Stanton, 2004). It is one of the central underpinning analysis methods within the DTC
HFI’s EAST method alluded to earlier. According to Stanton (2004), task techniques
can be broadly divided into five basic types: hierarchical lists (for example, HTA and
GOMS), narrative descriptions (for example, the Crit and Cognitive Archaeology),
flow diagrams (for example, TAFEI and Trigger Analysis), hierarchical diagrams
(for example, HTA and CCT), and tables (for example, Task-Centred Walkthough,
Interacting Cognitive Subsystems-Cognitive Task Analysis: ICS-CTA, HTA, Sub-
goal template: SGT, and Task Analysis For Error Identification: TAFEI). Some
methods have multiple representations, such as HTA, which can be viewed as a
hierarchical text list, a hierarchical diagram or in tabular format.
HTA is often referred to as a means to ‘model’ an interaction. The emergent
properties are related to functional groupings of tasks within a hierarchy (which is
not readily apparent in most cases), and the reductions in complexity that arise from
structuring tasks in this way. HTA permits multiple modelling perspectives; at the
bottom layer the entire task is represented in full, whereas higher levels within the
hierarchy represent progressively more parsimonious descriptions. Either may be
more or less appropriate to the type of analysis required. HTA is a more rigorous
means of describing a system in terms of goals and sub-goals (as opposed to roles
and sub-ordinates) and departs from an organisational chart in terms of its more
rigorous and highly defined internal logic used to create a nested hierarchy, and also
in the provision of rules and feedback that define the deterministic (cause and effect)
enactment of tasks contingent upon specific external conditions being met.
 Another Random Document on
Scribd Without Any Related Topics
“Who? Elam? Oh, yes! You can trust him anywhere.”
“I mean he wouldn’t let harm come to Bob without making a fuss
about it.”
“No, sir,” said I, rather astonished at the proposition. “I don’t see
that any harm can come to him out here.”
“Well, I don’t know,” said Mr. Davenport, with a heavy sigh, which
told how heavily the matter bore on his mind, “I don’t know.”
Not to dwell too long on incidents that are not connected with this
story, I will simply say that we were presented to two of the
cowboys that night at supper time as the fellows Mr. Davenport had
employed to help him drive his cattle north, our duties to begin on
the day the march commenced. I took a great notion to the two men
—tall, rawboned, and rough, and the simple and earnest manner in
which they agreed with their employer on all questions concerning
the conduct of the farmers, in keeping his cattle out on the barren
prairie where there was neither water nor grass to be had, made me
think that their hearts were in the matter.
During the next week I noticed that Bob and Elam went off
somewhere immediately after breakfast and did not get back before
night. That was all right to me, but I wanted to make sure that Elam
knew what he was doing, so one day when I got a chance to speak
to him in private I said:
“What do you and Bob do when you are gone all day?”
“Sho!” said Elam, with a laugh. “He just makes me lay under the
trees and tell him stories.”
“You are sure no harm comes to him?”
“Harm? What is going to harm him out here?”
“I don’t know and his father doesn’t know; but if you are wise you
will keep your eyes open.”
“Harm!” repeated Elam. “Well; I should like to see somebody harm
him. He’s got a good heart, that boy has. Be they going to shoot
him?”
“I don’t know what they are going to do, I tell you. If his father ever
tells me I will tell you.”
During all this time Mr. Davenport kept Tom and me close to himself.
It was a companionship that was entirely new to him in that country,
and he wanted to make the most of it. Before I had been acquainted
with him twenty-four hours I could see that he was different from
most men who made stock raising a business, that for years he had
been out there where he had nobody to talk to, and I was sure he
had some secret to tell us. One day it all came out, as I knew it
would, if we let the matter alone and did not trouble him with it. It
was a hot day during the first of August and we were sitting there
on the porch, trying to raise a little breeze by fanning ourselves with
our hats. It was after dinner, and the Mexican cook had gone
somewhere to sleep and we were there alone.
“I haven’t always been what you see me now,” said Mr. Davenport,
settling back in his chair as if he had resolved upon his course. “I
have a secret which I want to tell Bob, but I don’t know how to go
about it. It isn’t anything of which I am ashamed,—many men have
done the same before me,—but somehow I have let it go so long
that it has become a task to me. I want to ask your advice about it.
You are comparative strangers to me, but somehow I have taken to
you and want to trust you. I haven’t had anyone around me to
whom I was willing to confide it, and now I know that I am not long
for this world I want to see Bob have his rights.”
With these words the invalid began his story. It was short, but we
could both see how great an effort it cost him.
Mr. Davenport was an old “forty-niner.” He spent a few successful
years in the gold mines and then returned to the States, and
established himself as a wholesale merchant in St. Louis, his native
city, and soon became known as one of its most enterprising
business men. The only relatives he had in the world, except his son
Bob,—who was not his son in reality,—were an unmarried uncle,
who went to Texas and became a ranchman, and a half brother, who
was not a relative to be proud of. Too lazy to work, this half brother,
whose name was Clifford Henderson, gained a precarious living by
his wits. He gambled when he could raise a stake, and borrowed of
his brother when he couldn’t. He was more familiar with the police
court than he was with the interior of a church, and when his
generous brother’s patience was all exhausted and he positively
refused to pay any more of his debts, he left that brother’s presence
with a threat of vengeance on his lips.
“I will get even with you for this,” said he. “Bob is not your son, and
I will see that you don’t adopt him, either. Whenever I see a notice
of your death,—and you can’t live forever,—I will hunt that boy up
and make him know what it is to be in want, as I am at this
moment.”
The fact that Bob was not his son ought not to have weighed so
heavily with the invalid as it did, but still he could not bear to
enlighten him. He was the son of a friend in the gold mines, who,
dying there, left Bob alone, and Mr. Davenport took him up. He
christened him Davenport, and the boy always answered to his
name. There never had been any doubt in his mind that Bob would
some day come in for all his money, until this Clifford Henderson
began his threatenings; and even after that Mr. Davenport did not
wake up and attend to things as he ought.
In process of time Mr. Davenport’s unmarried uncle died, and in his
will he made him executor and heir to all the property he had
accumulated in Texas. In the hope that a change in the climate
might prove beneficial to his health, as well as to leave that
miserable Clifford Henderson and all his threatenings behind, Mr.
Davenport moved to Texas and took possession of his legacy,
bringing Bob with him. In fact, the two did not act like father and
son, but like two brothers who could not bear to be separated. All
they found when they reached Texas was a rather dilapidated old
house, which was very plainly furnished, and presided over by a
half-breed Mexican cook, who was so cross and surly that one could
hardly get a civil word out of him. The rest of the help—there were
four of them in all—were cowboys. They spent the most of their lives
on the open prairie, looking out for the safety of Mr. Davenport’s
cattle.
“I have got everything——”
Mr. Davenport suddenly paused and put back into his coat the large
pocket book which he had been in the act of showing to us. Then he
got upon his feet and carefully closed the door leading into the
cabin, and walked cautiously to one end of the porch and looked
around the house, then to the other end, but came back without
seeing anybody.
“One has to be careful,” said he, in explanation. “I am as afraid of
my help as of anything else.”
“Of your help!” I exclaimed. “If there is anybody here that you are
afraid of, why don’t you discharge him?”
“Because I want to see what he is here for,” said the invalid. “He
works for nothing at all, but yet he always seems to have plenty of
money. You know ’Rastus Johnson?”
Yes, we did know him, and he was one of the few people about the
ranch to whom I had taken a violent dislike. He was just the man to
excite the contempt of a Texan, because he couldn’t ride; but when
he came to Mr. Davenport’s ranch six months ago, and told a pitiful
story about the luck that had befallen him in the mines, he was
given odd jobs to do about the ranch for his board. There were two
things that struck Mr. Davenport as peculiar, or we might say three,
and tempted by something, he knew not what, he kept the man
around the house as much as possible and watched his movements.
One was the care he took of his six-shooters. He had a splendid pair,
and when engaged in no other occupation, he was always rubbing
them up until they shone like silver. The other was his story about
the mines. He did not know that Mr. Davenport was an old forty-
niner, and he thought he could say what he pleased to him and he
would believe it. The nearest mines that Mr. Davenport knew
anything of were those located about Denver, the very place we had
come from; and the idea that anyone could walk a thousand miles,
right through a country settled up by cattlemen and farmers, and be
as poor as he was when he struck Mr. Davenport’s ranch, was
ridiculous. But Mr. Davenport kept this to himself. He had Clifford
Henderson in mind, and he resolved if ’Rastus attempted anything
out of the way he would expose him on the spot.
As ’Rastus grew more and more at home about the ranch, other
qualities developed themselves. He took to “snooping” around the
house to see what he could find there, and once, when Mr.
Davenport entered the ranch suddenly, he was certain that he saw
’Rastus engaged in trying to pick the lock of his desk; but ’Rastus
began tumbling up his bed, and turned upon his employer with such
a hearty good-morning that the invalid was inclined to believe he
was mistaken.
“Yes,” said I, in response to Mr. Davenport’s question; “I believe we
know something about ’Rastus. Some of the cowboys have told us a
good deal about him. Is he the one you are afraid of?”
“I’ve got the whole thing right here,” said Mr. Davenport, seating
himself in his chair and drawing a big fat pocket-book from his inside
pocket. “It contains my will, and also instructions in regard to what I
want Bob to do with the rest of our herd in case any escape the
effects of the drought. It also contains a full history of the manner in
which he came to me, and hints regarding those threats of
Henderson—whom I sincerely trust he may never see again. In
short, nothing that I could think of has been omitted.”
“You don’t think that Henderson would follow you down here, do
you?” said Tom.
“My dear boy, you don’t know anything about that man if you think
he wouldn’t follow me to Europe,” said Mr. Davenport sadly. “If he is
alive, Bob will hear from him; and that he is still alive I am forced to
believe from the actions of this man Johnson. I don’t expect to come
back here, and I want you two boys to swear to what I have told
you. You will, won’t you?”
Of course Tom and I agreed to it, and then we wondered what sort
of a man Clifford Henderson could be to scare his half brother so
badly as that.
 CHAPTER III.
’RASTUS JOHNSON.

H aving no wish to pry into Mr. Davenport’s affairs any further

than he was willing to reveal them to us, we did not question
the invalid, although there were some points in his story that I
should have liked to have cleared up. He seemed to know that
’Rastus Johnson was employed by Clifford Henderson, and I wanted
to know what reason he had for thinking so; but he was sadly used
up by his talking, and settled back in his chair in a state of complete
exhaustion. It was this state that troubled me. I began to think that
when his time came to go he would go suddenly.
Presently Bob came up accompanied by Elam. I strolled off to find
’Rastus Johnson. You see I was as much interested in that pocket-
book Mr. Davenport carried in his coat as I was in anything else.
’Rastus Johnson must have known that he carried it there, and if
anything should happen while the invalid was alone the pocket-book
would be found missing; and without a will where would Bob be?
Henderson could claim his property as next of kin, and Bob would be
left out in the cold. I knew that Tom understood all this as well as I
did. At any rate I would speak to him about it the very first chance I
had, and arrange it with him so as to keep Mr. Davenport under
guard the whole time.
It did not take me long to find ’Rastus Johnson. The ranch stood on
the edge of a little grove, and there, under one of the trees, I found
the man of whom I was in search. His hat was pulled over his eyes,
as if he were fast asleep, and the belt containing his revolvers lay
near him on the ground. Evidently they had just received an extra
rubbing. He started up as he heard my footsteps and pulled the hat
off his face.
“Oh, it’s you, is it?” said he, with a long-drawn yawn. “How-dy. What
does the old man have to say to you? He says more to you than he
has to me, and I’ve been on this ranch for three months.”
“Yes, he has had a good deal to say to Tom and me. He has been
telling us about the threats of Clifford Henderson. Seen anything of
him lately?” I asked, as if I didn’t care whether or not he answered
my question.
I asked this abruptly, as I meant to do, and the answer I got set all
my doubts at rest. The man was in the employ of Henderson—that
was a fact; and while he used his own time in getting his wits about
him, I busied myself in giving him a good looking over. He was a
giant in strength and stature, long haired and full bearded, and
when he sat up and looked at me, I knew I was looking into the
eyes of a desperado of the worst sort. His clothes were not in
keeping with the story of poverty he had told when he first came to
Mr. Davenport’s ranch. They were whole and clean, and his high-top
boots looked as though they had just come from the hands of the
maker. There was something about the man that made me think he
was wanted somewhere else—that there was a rope in keeping for
him, if the parties who held it only knew where to find him. He
looked at me for fully a minute without speaking, then rested his
elbows on his knees and looked down at the ground.
“I don’t know the man,” said he, and he spoke so that anybody
could have told that he was angry.
“There is no need of getting huffy about it,” said I carelessly. “Where
is he now?”
“I tell you I am not acquainted with the man,” said he. “Henderson!
I never heard the name before.”
“No offence, I hope; but I thought from the way you acted that you
were in his employ. Be honest now, and tell me when you have seen
him lately.”
“How have I acted?” enquired the man.
“Oh, snooping around the ranch and trying to find out things that
are not intended for you to know,” I answered carelessly. “You know
you have been doing that ever since you have been here, and Mr.
Davenport is sorry that he ever consented to let you remain.”
“Did he tell you what I have done?”
“There is but one thing he could put his finger upon, and that was
when you tried to pick the lock of his desk.”
“I never——” began Johnson.
“If you had got into it you wouldn’t have made anything by it. The
man’s papers are safe.”
“I know he carries them on his person, and he’s got a little revolver
handy, bless the luck. There now, I have let the cat out of the bag!
There’s no one around who can hear what we say, is there? Sit
down.”
I tell you things were going a great deal further than I meant to
have them. I had come out there on purpose to induce Johnson to
drop a hint whether or not he was in Clifford Henderson’s employ,
but I had succeeded almost too well. It looked as though the man
was going to take me into his confidence. It was a dangerous piece
of business, too, for I knew if I did anything out of the way, I would
be the mark for the bullets in one of Johnson’s shining revolvers.
“I don’t see why I should sit down,” I replied.
“Sit down a minute; I want to talk to you. You have had bad luck
with your cattle,” said the man, as I picked out a comfortable place
to seat myself. “You once possessed a large drove, but they were
taken away from you at one pop.”
“That’s so,” I said. “If I could find the men who did it, I wouldn’t ask
the law to take any stock in them. I would take it into my own
hands.”
“Well, I don’t know anything about that,” said the man. “I wasn’t
there, although, to tell you the truth, I have been in at the bouncing
of more than one herd of cattle that was all ready to drive to
market.”
“What got you in this business, anyway?” I asked suddenly.
“What business?”
“Oh, you know as well as I do. A man of your education can make a
living a great deal easier than you do.”
“Look a-here, young fellow, I did not agree to make a confidant of
you in everything. Perhaps I will do that after a while. What I want
to get at now is this: Are you willing to work with me to have this
property go where it belongs?”
“Where does it belong?”
“You mentioned the name of the man not two minutes ago—Clifford
Henderson.”
“Aha! You do know that man, don’t you?”
“Yes; and now you know my secret, for I have got a secret as well
as the old man,” said Johnson; and as he spoke he reached out and
pulled his six shooters within easy handling distance, turning the
butt of one up, so that he could catch it at a moment’s warning.
Now, I suppose some of my readers will think I was in no danger
about that time, but I knew I was. My life hung upon the words I
uttered during the next few minutes. If I had refused I would never
have known what hurt me. Johnson would have shot me down and
then reported to Mr. Davenport that I had insulted him; and as there
was no one present to overhear our conversation, that would have
been the last of it. Law was not as potent then as it is in Texas in
our day, and Johnson’s unsupported word would have been taken,
there being no evidence to the contrary. I tell you I was in
something of a fix.
“How does it come that Henderson has so much interest in this
property?” I enquired.
“Why, Bob is no relative of Davenport’s at all. He picked him up in
the gold mines,—where his father died and left him,—named him
Davenport, and the boy has been brought up to believe that he has
an interest in all his stocks and bonds. I wish I had known a little
more about that when I came here. I told the old man some funny
stories about my being in the gold mines,” he added, with a laugh.
“And Henderson doesn’t want him to have it. It seems to me that it
would be the part of policy for Henderson to come here and live with
Mr. Davenport.”
“Oh, that wouldn’t do at all!” exclaimed the man hastily. “He used to
live with him in St. Louis, but they had an awful row when they
separated, and he is afraid the old man will go to work to adopt the
boy. I tell you he don’t want him to do that!”
“It seems very strange that Mr. Davenport hasn’t adopted him before
this time.”
“I lay it to his illness as much as anything. Like all persons who are
troubled with an incurable disease, he thinks something will happen
to take him off the minute he adopts Bob, and I tell you it’s a lucky
thing for us. Well, what do you say?”
“I don’t propose to go into this thing until I know how much there is
to be made out of it,” I answered, as if I had half a mind to go into
it. “How much are you going to get?”
“I am not going to take my pay in half-starved cattle, I tell you,” said
Johnson emphatically. “The old man has a few thousand dollars in
bonds in some bank or another,—I don’t know which one it is,—and
when I get that pocket-book in my hands I shall get some of those
bonds. I won’t let it go without it. He ought to give you as much as
he gives me.”
“How much are you going to get?” I said again.
“Twenty thousand dollars; and what I want more than anything else
is that pocket-book. He has got his will in there, and I must have
that before anything is done. Now, if you can steal that pocket-book
and give it to me, I’ll see that you are well paid for your trouble. If
Henderson gives you five thousand dollars it would go a long way
toward straightening up your cattle business.”
“Well, I want some time to think about it. It is a pretty dangerous
piece of work.”
“Take your own time. We shall not go off until next week. You won’t
say anything to Bob or the old man about it?”
“Never a word,” I replied, hoping that he wouldn’t ask me to keep
still where Tom and Elam were concerned. I couldn’t possibly get
along without taking them into my confidence, for although it was
new business to them, I felt the want of a little good advice.
“Because if you do—if I see you riding off alone with either one of
those fellows I shall know what you are up to, and then good-by to
all your chance of getting any money.”
“You need have no fear,” said I, getting upon my feet. “I shall not
say a word to either one of them.”
I walked slowly toward the ranch, feeling as if I had signed my own
death warrant. There was no bluster about Johnson, he wasn’t that
sort; but I knew that I not only would lose all chances of getting any
money by going off riding with Mr. Davenport or Bob, but I would
lose my chance of life. I would be shot down at once the first time I
was caught alone, and, with all my practice at revolver drawing,—
that is, pulling it at a moment’s warning,—I would not stand any
show at all. These Texans are a little bit quicker than cats when it
comes to drawing anything.
“Of all the impudence and scandalous things that I ever heard of,
that ’Rastus Johnson is the beat,” I soliloquized as I walked toward
the house, wondering what I should do when I got there. “A man
comes out to steal a will from another man and pitches upon me, an
entire stranger, because I have had ill luck with my cattle. Of course
I have no intention of doing anything of the kind, but if something
should happen to get this fellow into serious trouble—— By gracious!
if this man was lynched he could take me with him.”
When I reached the ranch and mounted the steps that led to the
porch I found Tom and Elam sitting there alone. Mr. Davenport had
talked himself into a state of complete exhaustion and had gone in
to take a nap, taking Bob with him as guard. In order to secure the
quietness he wanted they had closed the door after them. I felt that
now was my only chance. I saw by the look of surprise on Elam’s
face that Tom had been hurriedly whispering to him what Mr.
Davenport had told us.
“Where have you been?” enquired Tom. “We have been waiting half
an hour for you.”
“Is it a fact that this Johnson has been working for Clifford
Henderson?” exclaimed Elam. “If I was in Davenport’s place I would
drive him off the ranch.”
“Sh—! Don’t talk so loud,” I admonished him. “I’ve been gone half
an hour, and during that time I have heard some things that will
astonish you. I have learned that Johnson is in Henderson’s employ,
and that he wants me to act as his accomplice.”
I uttered these words in a whisper, thinking of the listening ones
there might be on the other side of that door, and when I got
through I tiptoed first to one end of the porch and then to the other
to keep a lookout for Johnson. I was afraid of the “snooping”
qualities that the fellow had developed, and if he had suddenly come
around the corner of the house and caught me in the act of
whispering to my friends I would not have been at all surprised at it.
Tom and Elam were both amazed at what I had told them, and
looked at one another with a blank expression on their faces.
“Tom, he wants me to steal that pocket-book Mr. Davenport showed
us to-day,” I continued. “He says the will is in there and he can’t do
anything without it. He says the property rightfully belongs to
Henderson.”
“If I were in your place I would go right straight to Mr. Davenport
with it,” said Tom, speaking in a whisper this time.
“And be shot for your trouble,” chimed in Elam, waking up to the
emergencies of the case.
“That’s the idea, exactly,” I went on. “He would shoot me down as
soon as he would look at me, and then report to Mr. Davenport that
I had insulted him; then what could anybody do about it? You
fellows would have to shoot him, and that would end the matter. I
promised I wouldn’t say anything to Bob or his father about it, but I
had a mental reservation in my mind when it came to you. Now I
want to know what I shall do about it.”
“Tell us the whole thing, and then perhaps we can pass judgment
upon it,” whispered Tom. “I don’t know that I understand you.”
With that I began, and gave the boys a full history of my short
interview with Johnson. It didn’t take long, for I did not hold a very
long conversation with ’Rastus; and when I came to tell how readily
he had included me in his plans I saw Elam wink and nod his head in
a very peculiar manner. Then I knew that I had hit the nail squarely
on the head when I made up my mind what ’Rastus would do to me
if things didn’t work as he thought they ought to. I tiptoed to the
end of the porch to see if I could discover any signs of him, and then
I came back.
“You see he knows that I have had bad luck with my cattle, and he
takes it for granted that I am down on everybody who has been
fortunate with theirs,” I said, in conclusion. “He thinks I want to steal
enough to make up for my lost herd.”
“The idea is ridiculous,” said Tom. “How in the world does he
suppose Mr. Davenport had anything to do with your loss?”
“That aint neither here nor there,” said Elam. “That feller has stolen
more than one herd of cattle, an’ I’ll bet on it. I shouldn’t wonder if
he was one of them desperate fellows—what do you call them——”
“Desperadoes,” suggested Tom.
“I know he is,” said I. “And he is a man of education. He doesn’t talk
as the Texans do at all, and I told him that a person of his learning
could make a living easier than he did.”
“What did he say to that?”
“He said he didn’t agree to make a confidant of me in everything. He
might do it after a while. He acknowledged that he had been in at
the stealing of more than one herd that was all ready to be driven to
market. Now, fellows, what shall I do about it?”
This was too much for Tom, who settled back in his chair and looked
at Elam. Our backwoods friend arose to the emergency, and I
considered his advice as good as any that could be given.
“You can’t do nothing about it,” he said, after rubbing his chin
thoughtfully for a few minutes. “Let him go his way, an’ you go
yours.”
“Yes; and then see what will happen to me if I don’t do as he says.
Suppose he thinks I have had time to steal that pocket-book? If I
don’t give it over to him, then what?”
“Tell him that Mr. Davenport keeps a guard over it all the while,” said
Elam, “an’ that you can get no chance. Heavings an’ ’arth! I only
wish I was in your boots.”
“I wish to goodness you were,” said I. “What would you do?”
“I’d let him go his way, an’ I’d go mine. That’s all I should do.”
“I guess that’s the best I could do under the circumstances,” said I,
after thinking the matter over. “By the way, I think it is about time
you two went out on your ride. I am of the opinion that it will be
safer so. Leave me here alone, so that when Johnson comes up——
I do not believe his name is Johnson; do you?”
“’Tain’t nary one of his names, that name aint,” said Elam
emphatically. “His name is Coyote Bill.”
“How do you know?” Tom and I managed to ask in concert.
“I aint never seen the man; I aint done nothing but hear about him
since I have been here, but I know he is Coyote Bill,” replied Elam
doggedly. “At any rate that’s the way I should act if I was him.”
Coyote Bill was emphatically a name for us to be afraid of. We had
done little else than listen to the stories of his exploits since we had
been in Texas. He didn’t do anything very bad, but he would steal a
herd of cattle,—it didn’t make much difference how many men there
were to guard them,—run them off to a little oasis there was in the
Staked Plains, and slaughter them for their hides and tallow; and
when the story of the theft had been forgotten, two of his men
would carry the proceeds of their hunt to some place and sell them.
He never killed men unless they resisted, and then he shot them
down without ceremony. Many a time have we sat on the porch after
dark when the cowboys were there, listening to the stories about
him, and if this man was Coyote Bill he must have been highly
amused at some things that were said about him. We were both
inclined to doubt the story of his identity. No one had ever seen
Coyote Bill, and how could Elam tell what he looked like?
“Elam, you are certainly mistaken,” said I; and the more I thought of
his story the less credit I put in it. “If you had seen Coyote Bill I
should be tempted to believe you; but you know you have never met
him.”
“And then just think what he has done?” added Tom. “He comes up
here and agrees with Carlos, a man whom he had never seen
before, to go in cahoots with him. The idea is ridiculous. And how
did Clifford Henderson fall in with him?”
“I don’t know anything about that,” returned Elam, as if his mind
was fully made up. “I’ll tell you what I’ll do: I’ll bet that Carlos
dassent call him Coyote Bill to his face!”
“You may safely bet that, for I aint going to do it,” said I, looking
around the corner of the house. “Here he comes, boys. You had
better get on your horses and make tracks away from here.”
The boys lost no time in getting off the porch and to their horses,
which they had left standing close by with their bridles down, so that
they would not stray away. They swung themselves into their
saddles with all haste, and I sat down to await the coming of Coyote
Bill, if that was his real name, and to think over what I had heard.
 CHAPTER IV.
ELAM’S POOR MARKSMANSHIP.

“C oyote Bill!” I kept repeating to myself. That name had

probably been given to him by the Texans on account of his
being so sneaking and sly—so sly that none of the men he had
robbed had ever been able to see him. What his other name was I
didn’t know. While I was turning the matter over in my mind Bill
came around the corner. I confess he did not look like so dangerous
a fellow, and if I had met him on the prairie and been in want, I
should have gone to him without any expectation of being refused.
He looked surprised to see me sitting there alone.
“Where are they?” he asked, in a whisper.
“Whom do you mean?” I enquired, being determined, if I could, to
answer no questions except those he had on his mind. How did I
know whom he referred to when he spoke of “they,” and wanted to
know where they were?
“I mean the old man and Bob, and all the rest of them,” he added. “I
thought they were here with you.”
“Tom and Elam have gone off riding,—there they go,—and Mr.
Davenport and Bob have gone into the ranch to have a nap. I can’t
steal the pocket-book now, even if I wanted to, for Bob is keeping
guard over it. It is true he don’t know what there is in it, but he is
keeping watch of his father all the same.”
“Look here, Carlos,” said Bill, coming up close to the porch, “do you
ever have charge of the old man in that way?”
“In what way?”
“Well, I haven’t been able to do any business in almost a year, and I
am getting heartily tired of it.”
“What business do you mean?”
“Aw! Go on, now. You know what I mean. I can’t steal cattle that are
half starved, for I wouldn’t make anything out of them if I did. I am
getting impatient, and my boss is getting impatient, too.”
“Well?” said I, when he paused.
“I want you to see if you can’t secure possession of that pocket-book
by to-morrow night,” said Bill, in a quiet way that had a volume of
meaning in it. “You see, it isn’t the will that Henderson cares for. The
cattle are pretty well gone up, and there won’t be a third of them
left when we get to Trinity. What he cares most about is the bonds.
If he can get them in his hands he will be all right.”
“Why, Coyote Bill——” I began.
I stopped suddenly, with a long-drawn gasp, for I had done the very
thing I was willing to bet Elam I would not do. Bill started and
looked at me closely, and one hand moved to the butt of his
revolver. My heart was in my mouth. Coyote Bill’s face was a study,
and I was sure my slip of the tongue had hit him in a vital spot.
Understand me, I didn’t speak his name knowing what I was doing,
but because I couldn’t help myself. The idea that I was to steal that
pocket-book at twenty-four hours’ notice was more than I could
stand, and I blurted out the first words that came into my mind. I
never had had much practice in studying out the different emotions
that flit across a person’s mind, but I was sure that in Coyote Bill’s
expression both rage and mirth struggled for the mastery—rage,
that I had suddenly found out his name since I had left him; and
mirth, because I, an unarmed boy, should stand there and call him
something which he didn’t like too well anyway. So I resolved to put
a bold face on the matter.
“See here, Bill——” was the way I began the conversation.
“Who told you that was my name?” he asked.
“Why, Bill, I have done nothing but hear about you and your doings
since I have been here,” I answered. “You certainly do not pretend
to say you are not what I represented you to be?”
“Well, that’s neither here nor there,” said he, taking his hand away
from his pistol. “You are a brave lad; I will say that much for you,
and you ought to be one of us. What’s the reason you can’t steal the
pocket-book by to-morrow night?”
I drew a long breath of relief. The worst of the danger was passed,
but the recollection of what might be done to me after a while made
me shudder. I had half a mind to slip away that very night, but I
knew that Elam would scorn such a proposition. He meant to stay
and see the thing out. I tell you I wished he stood in my boots, more
than once.
“Because Bob is keeping guard over it,” I said. “He don’t know what
there is in it, I tell you; but he has been made to understand that
there is something in it that concerns himself, and so he is keeping
an eye on it.”
“Does he know that he is in danger of losing it?”
“Yes, he does; but he don’t know where the trouble is coming from.”
“Well, you have got hold of my name, and I wish you hadn’t done
it,” said Bill, looking down at the ground and kicking a chip away
with his foot. “Be careful that you don’t use it where anybody else
can hear it. Perhaps I can find some other way to get it. Do you
sleep very sound?”
I don’t know what reply I made to this question, for it showed me
that Bill was about to attempt something after we had retired to
rest. I made up my mind that he would try it too, but whether or not
he would succeed in getting by Elam was a different story altogether.
I made it up on the spur of the moment to take Elam into my
confidence. He was a fellow who could remain awake for three or
four nights, and in the morning he would be as fresh and rosy as
though he had enjoyed a good night’s sleep.
“You want to sleep pretty soundly to-night, whatever you may do on
other occasions,” said Bill, in a very decided manner. “I shan’t be
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com