GEOG 885
Advanced Analytic Methods for the GEOINT Professional

Is Geospatial Intelligence Analysis an Art or Science?


It should be no surprise that there are competing views of geospatial intelligence analysis. One school of thought is that intuition, experience, and subjective judgment predominate. Analysis here is an art, and non-quantitative methods predominate. Another school of thought is that quantitative data and analysis are most relevant. Analysis here is a science, and quantitative methods predominate. This controversy somewhat mirrors a long-standing debate in the intelligence community: if good analysis depends largely on subjective, intuitive judgment (an art) or systematic analytic methods (a science). Understanding this question is important to developing an effective approach to geospatial intelligence creation. To help understand these points of view, I will define the terms using the Merriam-Webster Collegiate Dictionary, tenth edition, as:

  • Art - the conscious use of skill and creative imagination in the production of aesthetic objects.
  • Science - knowledge or a system of knowledge covering general truths or the operation of general laws, especially as obtained and tested through scientific method.

Interestingly, there are those that consider integrative geospatial data tools, such as those found in GIS, as primarily aids to intuition and experience-based analysis and not the application of quantitative analytic methods. This seems contrary to the technical capabilities GIS brings to the geospatial intelligence. It is sufficient to say that there is no bright dividing line between art and science, and a pure scientific approach to geospatial analysis is undesirable. The dissatisfaction with the push toward a science perspective in GIS has been seen as a step backwards by some. In their thinking, GIS’s models and analysis methods are not rich enough in geographical concepts and understanding.

Geospatial intelligence is geospatial analysis, and geospatial analysis, at its core, is geography. Geography is both the conscious use of creative imagination in the representations of the earth and the science of developing general truths about the earth. For something to be automatable, it must be modeled and the facts (inputs) quantified. Since a model is a simplified abstract view of the complex reality, the model represents a limited set of rules which allows an analyst to work out an answer if they have certain information. Quantifiability of the information is important because unquantifiable inputs cannot be tested, and thus unquantifiable results can neither be duplicated nor contradicted. However, we know that reliable models and data are not available for all analyses.

The table below illustrates the broad types of geospatial analyses. The lower right panel of the matrix identifies the ideal of analysis as a Scientific Process (upper right quadrant) in which there is good knowledge of the data and models surrounding an output. In the model, analysts understand the problem that confronts them and can take into account the key factors that bear on the problem. The notion of fixed-in-advance standard procedures typically plays an important role in such geospatial analysis.

However, many of the analytic tasks in geospatial intelligence fall outside of the scientific quadrant. Consider the Puzzle Solving Process (lower left) quadrant in which there is agreement on models, but disagreement on data. The notion of "hunting" for the data to solve the problem plays an important role in such analysis.

Analysis as an Opinion Process (upper left quadrant) is the opposite. In this analytic environment, there is agreement on data, but disagreement on model. Analysis is characterized by analysts involved in a struggle for influence, and decisions emerge from that struggle. This kind of analysis necessitates bargaining, accommodation, and consensus, as well as controversy. The bottom line is that conclusions are most often the result of bargaining between diverse and strongly held beliefs.

Intelligence analysis as a Heuristic Process (lower left quadrant) is the most contentious, with disagreement on data and models. Under these conditions, science and technology tools have significantly less direct relevance. Here, conclusions depend on parameters that change over the period the analysis is being made. As a consequence, the analytic process is experience-based . In the end, this is the framing of questions. They can only be framed, not solved, and thus the logic of argument and analysis is as important as the evidence.

 Types of geospatial analyses (Opinion Process, Scientific Process, Heuristic Process, and Puzzle Solving Process)  as described in text.
Figure 3: Types of Geospatial Analyses
Source: T. Bacastow, D. Bellefiore, D. Bridges, S. Harter, The Pennsylvania State University, 2010.

Is geospatial intelligence an art or science? Analytic problems can fall into any of the four quadrants ---- you, the analyst, need to understand the problem solving environment and the nature of the problem solving process. The term “sensemaking” is used as a term to describe the analysis process. Sensemaking is defined is the deliberate effort to understand events using explanatory structure that defines entities by describing their relationship to other entities. Data elicit and help to construct the frame; the frame defines, connects, and filters the data.