A unified theory for intelligence analysis

ABSTRACT

Causation has traditionally been an under-theorized topic. Until Hendrickson’s work, very little effort had been devoted to creating a compelling theory of causation in intelligence analysis. In line with the recent attempts to integrate intelligence theory with philosophy, this article is intended to contribute to the philosophy of intelligence by defining a dedicated account of causation for it. The Unified Theory for Intelligence Analysis, as this account of causation is named, is intended to integrate into a single account Betts’ Normal and Exceptional Theories as well as Hendrickson’s target challenges. It is then proved that a pluralistic account of causation that combines both counterfactual and probabilistic accounts of causation is the most successful option. Finally, it is shown that Bayesian tools are the natural manifestation of this Unified Theory, and that Subjective Logic can help refute criticism against Bayesianism in intelligence analysis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1. Phythian, “Intelligence Analysis and Social Science Methods: Exploring the Potential for and Possible Limits of Mutual Learning,” 601.

2. Marrin, Improving Intelligence Analysis: Bridging the Gap between Scholarship and Practice.

3. Kaupi, “Counterterrorism Analysis 101,” 47.

4. Phythian, “Intelligence Analysis and Social Science Methods: Exploring the Potential for and Possible Limits of Mutual Learning,” p. 605.

5. Betts, Enemies of Intelligence: Knowledge and Power in American National Security, p. 57.

6. Laplace’s demon is a hypothetical being that has perfect knowledge of the position and momentum of every particle in the universe. If such a being existed, it could use this knowledge to predict the future with certainty. Laplace and his contemporaries believed that the universe was deterministic, meaning that every event was caused by a previous event and could be predicted with perfect knowledge. However, later research showed that the universe is not deterministic at the quantum level, meaning that there is an element of randomness in nature. This led to the crisis of classical determinism and the development of probabilistic quantum mechanics, which is a more accurate description of the world.

7. Minkel, “If the Universe Were a Computer”.

8. Phythian, ‘Intelligence Analysis and Social Science Methods: Exploring the Potential for and Possible Limits of Mutual Learning’. P. 603.

9. Hendrickson, Reasoning for Intelligence Analysts: A Multidimensional Approach of Traits, Techniques, and Targets.

10. Heuer, “The Evolution of Structured Analytic Techniques,” p. 4.

11. Hendrickson, Reasoning for Intelligence Analysts: A Multidimensional Approach of Traits, Techniques, and Targets.

12. According to Hendrickson, there is an epistemic continuum from data collection to strategic advice, and it can be divided into four separate stages that can be added up depending on the kind of intelligence to be produced. In particular, he divides that continuum into: what is happening? Why is this happening? When and where might this change? And how can the client respond to it? In turn, each of Hendrickson’s four problems becomes the protagonist in one of these stages. Moreover, this classification is similar to Edward Waltz’s epistemic domains: prescriptive, descriptive, exploratory, and predictive-evaluative.

13. Godson and Wirtz, “Strategic Denial and Deception,” 426.

14. Betts, Enemies of Intelligence: Knowledge and Power in American National Security, p. 60.

15. Betts, “Warning Dilemmas: Normal Theory vs. Exceptional Theory,” 829.

16. Betts, Enemies of Intelligence: Knowledge and Power in American National Security, 62

17. Ross, Prospects for Crisis Prediction: A South Pacific Case Study., p. 31

18. Hendrickson, Reasoning for Intelligence Analysts: A Multidimensional Approach of Traits, Techniques, and Targets

19. Hart and Simon, “Thinking Straight and Talking Straight: Problems of Intelligence Analysis,“ 41.

20. Waltz, Quantitative Intelligence Analysis: Applied Analytic Models, Simulations, and Games.

21. Gaspard and Pili, “Integrating Intelligence Theory with Philosophy: Introduction to the Special Issue,” 763.

22. Ben-Haim, “Positivism and Its Limitations for Strategic Intelligence: A Non-Constructivist Info-Gap Critique,” 912.

23. Knightian uncertainty, coined by economist Frank Knight, is a form of uncertainty that arises when faced with new or unfamiliar situations where probabilities cannot be assigned. Shackle-Popper indeterminism is the inherent unpredictability of human behavior and complex systems due to the numerous and interconnected variables at play.

24. Danks, “The Psychology of Causal Perception and Reasoning,” 458.

25. Loewer, “Determinism and Chance,” 612.

26. Psillos, “Regularity Theories,” 132–133

27. Berofsky and Mackie, “The Cement of the Universe: A Study of Causation,” 86.

28. Lebow, Forbidden Fruit: Counterfactuals and International Relations, 259–286

29. Nolan, “Why Historians (and Everyone Else) Should Care about Counterfactuals,” 333.

30. Abel and Ofer, “Subjective Causality and Counterfactuals in the Social Sciences: Toward an Ethnographic Causality?,” 15.

31. Lebow, “What’s so Different about a Counterfactual?,” 557.

32. Menzies and Beebee, “Counterfactual Theories of Causation”.

33. Lebow, “What’s so Different about a Counterfactual?,” 554–555

34. Weber, “Counterfactuals, Past and Future,” 278.

35. Lebow, Forbidden Fruit: Counterfactuals and International Relations, p. 30

36. Lebow, “What’s so Different about a Counterfactual?,” 566.

37. Kiser and Levi, Using Counterfactuals in Historical Analysis: Theories of Revolution., p. 189

38. Kiser and Levi, Using Counterfactuals in Historical Analysis: Theories of Revolution., p. 192

39. Lebow, “What’s so Different about a Counterfactual?,” 556.

40. Lebow, “What’s so Different about a Counterfactual?,” 574.

41. Harbecke, “Counterfactual Theories of Causation and the Problem of Large Causes,” 1652.

42. Papineau, “Can We Reduce Causal Direction to Probabilities?,” 239–240

43. Pearl, “Probabilities Of Causation: Three Counterfactual Interpretations And Their Identification,” 94.

44. Hájek, “Probabilities of Counterfactuals and Counterfactual Probabilities,” 237.

45. Longworth, “Causation, Pluralism and Responsibility,” 54.

46. The Second Law of Thermodynamics is responsible for the future being uncertain. It basically describes why processes in the world are irreversible and defines why the Arrow of Time always goes forward and why the past is defined, but the future is not. In fact, this law is one of the most solid claims Science has ever made. In Einstein’s opinion, ‘the second law of thermodynamics is the only physical theory of universal content concerning which I am convinced that, within the framework of the applicability of its basic concepts, it will never be overthrown’.

47. de Finetti, “Foresight: Its Logical Laws, Its Subjective Sources,” 139.

48. Marrin and Torres, “Improving How to Think in Intelligence Analysis and Medicine,” 649.

49. Zlotnick, ‘Bayes’ Theorem for Intelligence Analysis”.

50. Marchio, “Overcoming the Inertia of ‘Old Ways of Producing Intelligence’ – the IC’s Development and Use of New Analytic Methods in the 1970s,” 990.

51. Sloman, Causal Models: How People Think about the World and Its Alternatives., pp. 116–131

52. Zlotnick, “A Theorem for Prediction,” 5.

53. Martin and Popper, “Objective Knowledge: An Evolutionary Approach,” 276–277

54. Lewis, “Postscripts to ‘Causation’,” 50.

55. Jøsang, “A Logic for Uncertain Probabilities”; Jøsang, “Subjective Evidential Reasoning,” p. 282.

56. Although Subjective Logic is a powerful tool, it requires considerable knowledge on Statistics and Logic. An introduction to Subjective’s Logic can be found at Subjective Logic: A Formalism for Reasoning Under Uncertainty. Subjective Logic is a solid and well-developed theory that permits complex operators in causal networks to be evaluated to determine nested causal chains, for example. Some simple calculators are available online at: https://folk.universitetetioslo.no/josang/sl/Op.html

57. M. Isaksen and McNaught, “Towards a Better Framework for Estimative Intelligence – Addressing Quality through a Systematic Approach to Uncertainty Handling”.

Additional information

Notes on contributors

Carles Ortola

Carles Ortola Associate Professor of Strategic Intelligence at Universitat de Barcelona and PhD Candidate at UNED (Universidad Nacional de Eduación a Distancia) in Spain.