As my dissertation is reaching it’s conclusion, I am working in the direction of making extra of its analysis accessible by way of the online. This submit continues the collection of interviews that I carried out as a part of that research on Agent-Based mostly Modeling and intelligence analysis. Within the interests of full-disclosure, Joe was my boss for more than three years (2001-2004) once I worked for him as a analysis associate within the Middle for Know-how and National Security Coverage at National Protection College. Underneath’s Joe’s and Desmond Saunders-Newton’s mentoring, I used to be introduced to Agent-Based mostly Modeling and complexity principle. I owe Joe a particular debt professionally and intellectually and am proud to have benefited from his mentoring. This interview was carried out on September 6, 2012.
Background: Joseph Eash, III (JE) had a distinguished career in the US Air Pressure and Pentagon, the place he led the event of superior technologies for a lot of many years. He served at the Nationwide Reconnaissance Workplace, for which he acquired the Pioneer Award, served as the Deputy Undersecretary of Defense for Advanced Know-how, and retired from authorities in 2005 after serving as a Chief Scientist for Computational Social Science Modeling on the Middle for Know-how and National Security Coverage at National Protection College.
Dialogue: I requested JE several questions on his time in authorities, where he served in intelligence, know-how, and policy positions. In consequence, he had a singular set of experiences upon to draw upon when contemplating the development and integration of latest analytic technologies and methods into the intelligence group.
JE began by noting that we’ll never good info or intelligence, but that the ranges of what can occur might be understood by understanding technical issues and the processes that folks comply with to develop artifacts or execute plans. What we do is spend time creating templates or patterns of how actions are performed, and then sample from the world in an effort to see if we will fill within the templates.
JE noted that his characterization of intelligence launched several challenges. First, the problem of templates was largely a theoretical one. Templates are primarily characterizations of how one thing is perhaps completed, whether one seeks to develop a nuclear weapon or run a political marketing campaign. In a sense, they may be thought-about algorithms that constitute a set of steps to be followed as a way to achieve a specific aim. Assortment then samples from the world and makes an attempt to match observations towards templates with a view to understand what is occurring and what may happen. Which means intelligence requires both imaginative, theoretical work to develop libraries of templates, and also an understanding of collection methods and limitations in an effort to precisely perceive and talk what info is gathered and what shouldn’t be. Importantly, JE noted that on this context a single piece of data or intelligence has very little which means – what matters is the way it relates with different info so as to fill in a template.
JE famous that in lots of instances we look for things with no template, or an unknown template. This leads to problems relating to what to do with collected info. We stretch it, exaggerate it, do a bunch of things to make sense out of it, however absent a common template reflecting a strong understanding of individuals, processes, and so on. we are more likely to misunderstand what we collect.
JE continued by noting the importance of cognitive elements in the analytic process. He discussed the transition from movie to digital pictures inside the intelligence group. He noted that one of the curious elements of study was how the brain affected the perception of pictures. When photographs have been obscured, analysts typically confirmed larger confidence in their capacity to determine its contents than when it was clearer. Furthermore, once an analyst dedicated to specific identification or interpretation, it was very troublesome to again them off their conclusion, even because the image was made sharper, the brain acquired caught and hooked up to a specific interpretation.
JE believed that the cognitive challenges of deciphering imagery have been representative of a larger range of analytic challenges. He famous that in many instances, the obtainable info is usually of low high quality, akin to the obscured image, but analysts must make judgments from them, and this process can lock errors into the analytic course of. He also famous how essential good, top quality knowledge is early within the analytic process. Because analysts have been extra accurate when first shown a high quality picture than a low high quality image, however typically resisted revising assessments from low quality photographs after being show a high quality image, an asymmetry existed the place extra info was needed to vary an observers thoughts than to form it in the first place.
JE argued that one of many major issues in intelligence and coverage is that analysts and decision-makers are making assessments and insurance policies with mental fashions or templates of unknown quality, and that the knowledge going into them can also be of poor or unknown high quality. Consequently, everyone is weak to seeing what they want to see and until challenged by top quality info, their perceptual biases are more likely to carry the evaluation. He continued by noting that in many instances, the underlying theoretical work required to develop templates has not been executed, and that the knowledge we will gather can’t justify the conclusions we try to succeed in.
JE compared the problem of study to stealth in radar world. He famous that accumulating towards a stealthy car was a troublesome drawback – assortment techniques solely get glimpses of the goal, and these glimpses might be quite troublesome to combine into a standard picture, e.g. have been they of the same car of or a number of automobiles? Within the abstract, the problem was the same – how one can get probably the most out of low-resolution info? Higher assortment was one risk, which meant accumulating with a better sample price or decision. A second answer was to develop new templates that might permit evaluation to proceed with much less info.
I adopted up by asking JE concerning the differences between intelligence issues within the technical domain and the development of latest technologies compares with social science problems that turned more and more essential in the course of the latter years of his profession. JE famous that the social domain prolonged the vary of issues and period of time to be concerned with, but in addition the alternatives to develop an understanding of the international system. He noted that in a simple case, e.g. figuring out if an assault was occurring, one might look into the technical domain to take a look at the horizon utilizing optical, infrared, acoustic, and other sensors. Nevertheless, extending into the social domain opened the problem as much as provide new indicators for warning, lots of which might be prior to the technical ones, e.g. the execution of organizational procedures for assault preparation or social indicators akin to communications between relations.
JE argued that the social dimension offered new dimensions across which alerts might be correlated to realize a more complete and correct understanding of present occasions and potential developments. His turn in the direction of the social sciences was then largely about finding new methods to process other technical assortment techniques by cuing off of social findings, e.g. if it was discovered a rustic’s financial system was specific sensitive to a specific commodity worth, then monetary markets might provide indicators of impending political and social instability. JE argued that by extending evaluation beyond technical indicators and templates by incorporating social info, custom-made regional, national, and subnational fashions could possibly be developed to offer custom-made warning for specific instances and enhance the standard of warning out there to shoppers. Importantly, JE noted that whereas these templates would draw upon social science and social indicators, every can be totally different and tied to specific instances and circumstances.
JE concluded by observing that one of many difficulties in integrating social science and indicators into the intelligence system was the entrenched institutional pursuits of costly, technical assortment methods which are the thing of finances and corporate priorities. Thus, investments within the social sciences and models faced troublesome inner competitors when paired towards costlier, conventional assortment capabilities, regardless that the 2 essentially enhanced the capabilities and effectiveness of the other.
My subsequent query to JE was about his time as a shopper of intelligence slightly than a producer. I was notably all for his perspective on what made evaluation efficient or helpful from the attitude of policy-makers. JE answered that he all the time in the direction of analysts that employed a transparent and clear course of for locating, explaining, and eliminating various explanations of events. Each time analysts did not contemplate, explore, or present various rationalization he famous that he turned cautious and fearful that analysts have been engaged in advocacy for specific views relatively than analysis. He also paid nice attention to the range and use of sources, noting that checks towards overseas denial and deception and efforts to cross-validate info have been essential features of intelligence and establishing his confidence within the high quality of their merchandise. Every time collection and analysis relied on a single source, he turned cautious and all the time needed a option to hint again analytic judgments to the underlying supply materials.
Lastly, I requested JE about the future of intelligence and what he thought can be most essential going ahead into the longer term. He noted that our growing understanding of cognition and the mind was key to enhancing evaluation and understanding the bounds, strengths, and failures of human perception and using info. He believes that ongoing work in brain scanning and imaging as a means of understanding how beliefs develop and change will make necessary contributions to enhancing analysis. Finally, JE famous that understanding cognition will essentially need to incorporate a greater understanding of emotion and its position in analysis and decision-making. Indeed, JE believed that most of the troublesome intelligence issues are questions of understanding ‘why’ individuals behave as they do (individually or collectively), and that successfully addressing such considerations requires an understanding of their emotional states, wants, and predisposition in addition to more conventional rational issues.