Tomorrow’s commander must be able to sense the answer to this question if network-centric warfare is to be realized.
Network-centric warfare is the philosophical overcoat for the collection of systems and organizations that promises to enable greater sensitivity, totality, and accuracy in gathering information and a reduction in decision-making time. Descriptions of this idea derive primarily from the technical systems that empower the network-centric concept, but the importance of labor in the network has been understated. The most significant lesson from the information revolution in civilian enterprises has been the importance in the network of the people and their understanding of the information that is generated, tabulated, and displayed. Current literature indicates this lesson has not yet been learned by the U.S. military.
Sixteen years ago, the Buchsbaum Panel established the goal of removing uncertainty from the information used by the U.S. military. Admiral William A. Owens's "system of systems" is acknowledged to be the practical vision of Buchsbaum's goal—the detection, classification, and tracking of the militarily relevant objects moving on land, through the air or in space, and on the surface of the ocean within a battle space some 200 miles on a side. But regardless of the technical elegance of information-gathering devices and displays, differences among the people using the network create ambiguity and from that, uncertainty. The fog of war will continue to exist regardless of the mechanical perfection of the machinery of command, control, communications, computers, intelligence, surveillance, and reconnaissance (C4ISR).
Understanding of Technology Is Imperfect
The global positioning system (GPS) is the most widespread and dramatic example of a radio + spacecraft + computer system that is revolutionizing military command and control. GPS has changed drastically command and control at the tactical, operational, and, as the Afghanistan campaign demonstrated, strategic levels. Not many years ago, soldiers stumbled through dark forests with only a vague idea of their exact location or direction of movement. Sailors' worries about where they were after days without stars or a sun line or even a horizon are legendary. Now, not only can a unit and even an individual know their locations, but in many cases so can all their seniors. Gone are the clumsy gridlocks that did not move at a rate commensurate with the speed of advance and the dated reference points often lost in a maze of order revisions. In their place are accurate locating data at headquarters as far up the chain of command as there may be interest.
But GPS carries a virus that unduly buttresses belief in the accuracy of other sensors, references, and databases. Because the positions of one's own forces now can be portrayed accurately on cathode ray tube, acetate overlay, or Plexiglas status board, other data similarly portrayed often are endowed with the same veracity. In particular, cathode ray tube displays impart an unwarranted precision to the data displayed. Knowledgeable users may appreciate the divergence between truth and fiction in sensor input and remote databases, but those less knowledgeable may not grasp the multitude of variables behind the icon on the screen and are at the mercy of information systems that "create illusions that complex problems are understood when they are not."
The Navy is more comfortable with network-centric warfare concepts than other services because of its intimate use of radar for 60 years. But this experience with radar adds to unrealistic expectations in a wider network that includes distant sensors. Operators raised using onboard radar (most Navy and Air Force officers) tend to assume all sensors can achieve that standard. Most operators are ill informed about the quality of the data developed by sensors other than their own. In particular, warriors are especially ignorant about the off-board "national" sensors that supply much of the data to theater C4ISR systems; at the same time, the operators of those systems often are accused of not meeting the needs of those in the field. Compounding the problem, not enough officers possess the knowledge or experience to frame the questions to the off-board sensors or to the network.
As anyone who has witnessed a junior seaman making the noon report to a ship's captain can attest, to properly evaluate a report, recipients must understand the capabilities and limitations of the reporter. Remote sensors are no different from human scouts; all have traits that can cause their reports to vary in accuracy, timing, and insight. Understanding the technical apparatus that senses, correlates, and displays data is key to the functioning of a military network. Without this understanding, automated systems may contribute to the disease they are supposed to cure.
The software that collects, fuses, interprets, and displays data is as important as the sensors. Understanding the rules under which these various systems are programmed is essential to comprehending the results. Software behaves the way a programmer has dictated and is not subject to the constraints of natural laws. Without careful supervision and adjustments by competent managers, for example, the Navy Tactical Data System can multiply contacts ad infinitum. Furthermore, a computer's embedded belief system allows only finite inputs. Presented with information that does not fit the pattern, the machine will answer the question within the programmed framework, never recognizing that it is stymied. The resulting misinformation can infect the entire network. Finally, computers ignore factors that cannot be quantified, programmed in advance, or concisely displayed.
It is true that rapidly making routine decisions with near-perfect information is something computers do better than humans. The problem with this notion is that in battle, little input is perfect. The temptation to let the computers run unfettered by human oversight may be overwhelming because the information to be processed is extensive and fast reaction seems important. Yet, in real situations, good command and control depends not so much on having superior technology as on understanding the technology's capabilities and limits.
Intuition Clarifies Ambiguity
After generation of information in the network, action to use that information requires several steps: fusing databases and remote sensors with the current tactical situation, integrating this information within the command(s), generating an appropriate response or action, and directing forces. These steps must be repeated through each level until final action occurs. Then the cycle starts again with battle damage assessment. Opportunities for uncertainty lie in each step. As data staggers down through the system of systems, uncertainty cascades. Even with a thorough understanding of the parts and pieces, the validity of the resulting data is subject to concealment or manipulation by enemy or friend. At every level, operators question whether the display is telling them the truth or whether the input has been manipulated to show something someone would like them to believe. In a system as complex as the wartime network, this hesitancy can be a major obstacle that is increased by operators' ignorance of system components and diversity of experience. Persons tend to see what is expected rather than what is happening. As the expectations of participants become more diverse, ambiguity multiplies and uncertainty jumps into the openings.
Ambiguity is much more difficult for machines to handle than for humans: no mechanistic system can replace intuition. The more information is gathered, the more ambiguity will exist, and the more important people become. Ambiguity does not necessarily translate to uncertainty or vice versa, but it takes practice to judge the difference. Negative information is real information only when the evaluator understands the nature of the situation at hand. There is no substitute for native intelligence, but intellect alone is not enough to grasp the essentials of a fast-moving military problem. To convert that into useful knowledge, intuition must be fortified by experience in the realm being evaluated.
The network-centric concept's key arguments center around the display, where all relevant information is available to all levels of command, thereby shortening the time for action and reaction. The resulting understanding comes not just from the data displayed but also from the experience and intuition of the observers. The largest potential pitfall to the network's effectiveness arises at this man-machine interface. All relevant information never will be present and new information will be processed only to the extent necessary to determine where it fits into the structure of people's previous knowledge.
Common Experience Offers Advantages
As operations become more joint, they will involve more persons of dissimilar education, training, and experience. The expansion of information-age sensors and processors changes but does not remove the uncertainty stemming from differences in field experience, service culture, and education. As the network's operators become a more heterogeneous group, the certainty with which sensed data is treated becomes more problematical and the human interactions around uncertainty more complex.
The more two people know about a common subject, the less need be said between them to communicate new information or create shared appreciations. This common understanding becomes more valuable the more difficult a subject is to understand, the more rapidly it changes, or the more new information is developed. Conversely, the greater the disparity in the backgrounds of the individuals, the more difficult will be resolution of ambiguous data and the greater the general uncertainty within the organization as to the quality of information and the appropriateness of resulting plans or actions.
Well-developed intuition contains knowledge of the range of uncertainty in a given circumstance, allowing commanders to function in spite of a lack of information. Machines have no ability to analyze, fuse, or display such non-data. Intuition can fill the gaps around sensed information to the extent that experience has generated the necessary background. Danger arises when the commander thinks he knows something he does not or lacks the knowledge to question assistants trained to value truth over authority.
Persons who can draw inferences from small amounts of data always have been of value. They will become even more valuable with increases in the amounts of data, the physical distance of the commander from the sensor and scene of action, and time pressure. The nature of military command is to make decisions in stressful situations without having all the information necessary to determine their outcomes. The system of systems will not change this. To operate in this specialty in the future, officers will need what Napoleon described as "a gift, called a coup d'oeil militaire [the ability to take in the military situation at a glance], which great generals have received from nature." Whereas Napoleon was speaking about an intuition for sensing what lay over the hill, tomorrow's commanders will need to develop a sense of what lies behind the screen.
Every naval officer will have to become a fluent user of information technology. In the words of retired Navy Vice Admiral Arthur Cebrowski, former president of the Naval War College, "If we are hoping to fight first for information superiority ... then surely our leaders must focus on this, not leave it to junior staff officers or technologists." Facility in quickly and accurately assessing information will become a central strategic skill. To master the operational art in the future, officers must be able to think in abstractions, fed by technologies of the most complex kind.
These considerations about people are important in constructing any network, but especially so when the network's function involves life-threatening situations. The importance of common understanding among the commanders cannot be overemphasized. This commonality is the lubricant that allows the network to operate efficiently. Rather than seeking to use many different forces in a mission, using as few different specialties as possible makes best use of the network's attributes. Joint operations should be the last choice, not the first.
The fog of war will not go away—it will appear in new and different forms. Understanding the technical aspects of the profession has been fundamental to successful prosecution of the mission since the introduction of the square-rigged ship. New dimensions have replaced study of royals and spankers, dimensions that must not be ignored or relegated to support functionaries, especially as the focus of operational art shifts to higher and higher levels of command.
Admiral Holland has been a frequent contributor to Proceedings since 1975 and was named 2003 Proceedings author of the year. He served most of his 32 years of active service in submarines.