Date: Tue, 24 May 1994 18:31:31 -0600 (MDT) From: CIOFFI CLAUDIO Subject: LORANOW article "The Long-Range Analysis of War (LORANOW) Project: A Brief Description," by Claudio Cioffi-Revilla, Director, Long-Range Analysis of War (LORANOW) Project, Department of Political Science, Campus Box 333, University of Colorado, Boulder, CO 803090333, U.S.A. Tels. (303) 4925445, 4927871, Fax (303) 4920978, E-mail: cioffi@colorado.edu May 1994 When and where did warfare originate? What role did it play in the rise and fall of polities in the global system? Did warfare originate in one area and then diffuse to others? Was warfare spontaneously "invented" in separate unrelated areas? Can knowledge about the remote origins of warfare in antiquity shed new light on contemporary and future war? In 1988 I established the Long-Range Analysis of War (LORANOW) Project at the University of Colorado at Boulder (see "The Long-Range Analysis of War," Journal of Interdisciplinary History, Vol. 21, no. 4, Spring 1991, pp. 603-29). The goal is to advance our scientific understanding of the origins and evolution of war and politics--starting from antiquity to the present age. The LORANOW project is interdisciplinary because others besides political scientists also study war. LORANOW has been funded by the University of Colorado, the Apple Computer Foundation, the Conflict Resolution Consortium, and the William and Flora Hewlett Foundation. Earlier theoretical research and mathematical modeling was funded by the National Science Foundation. The project has an interdisciplinary board of external peers, formed in 1992, during the 25th Silver Anniversary of the Peace Science Society (International). The project also draws on expert advice from panels of anthropologists, paleontologists, military historians, archaeologists, sociologists, area specialists, and others with established expertise in the appropriate areas. The LORANOW project covers both theory and measurement. In the past six years many students have conducted hands-on research in this largely unexplored territory. Although other scholars will one day use the LORANOW dataset for other purposes--such as testing their own theories or hypotheses, as with the Correlates of War (COW) Project--the primary purpose of our data collection effort is to test the project's own theory of war and political dynamics. The main purpose of the theory is to explain the long-range origins and evolution of war as recorded in the dataset, particularly as warfare interacts with politics (see "The Long-Range Analysis of War Project: A Progress Report," Peace Science Society International, Ann Arbor, Michigan, November 1992). Data collection has focused on reliable cross-societal information on war, using sources from anthropology, archaeology, ethnohistory, and other fields besides political science and history. We also collect data from "dirt archaeology," such as wars depicted in lithic monuments. The LORANOW dataset is probably the first to be based at least in part on hard paleographic and archaeological evidence--something which I did not envision when the project began, but which has proven both essential and exciting. Empirically, we mark progress in two directions. First, we have completed the first pilot datasets on the three areas of the ancient world were war first originated as a reliably documented historical phenomenon in human evolution: the Near East (from ca. 3000 BC), China (2700 BC) and Mesoamerica (900 BC). Second, we are also making some progress on the first Paleolithic conflicts on which there is hard evidence. By our operational definition, war began occurring at least 10,000 years ago, toward the end of the last Ice Age, or Late Paleolithic and Early Neolithic periods. However, the first reliably recorded wars are currently dated to ca. 2700 BC (e.g., the 2725 BC Sumer-Aratta War in Mesopotamia), or about 5 millennia ago. Before the LORANOW project began it was assumed that the first historically recorded war was the famous 1469 BC Egyptian War in Palestine, at Meggido. However, since 1988 we have recorded numerous earlier qualifying wars in ancient Mesopotamia and China, although not in Mesoamerica. Our preliminary results now push back the first wars by more than one millennium than was previously known. Considering this long-range record, our current political science knowledge about war is mostly based on theory and data of the recent medium range--i.e., less than 3% of the total timeline which could be analyzed if reliable systematic data were available. The data collection effort is divided into five surveys, each representing an original dataset: (1) Mesopotamia, (2) China, (3) Greece and Rome, (4) Japan and (5) Mesoamerica. These pilot surveys contain today over 1,000 wars, a number already several times larger than the COW data for both international and civil wars. Modern datasets have recorded an average war onset rate of approximately 0.5 war onsets/year, depending on the type of war events being measured, or an average of one onset every two years. Our theoretical expectation was that onset rates would be higher in antiquity, sometimes by as much as an order of magnitude during intensely bellicose periods. However, this general expectation turns out to be wrong. Data take a long time to collect, but theories take even longer to conceptualize, formalize, test and revise. Thus far our theoretical efforts have concentrated on two core puzzles. The first concerns the origin(s) of war and political evolution, the second has to do with the causal forces of war. After years of addressing the many questions contained within these puzzles, our tentative conclusion is somewhat different from what I had expected earlier (I think I was brought up in the tradition of cultural diffusionism), but also more exciting. We explain the origins and evolution of warfare in global human history based on a "trident model"--a model used to specify various hypotheses on the origins and evolution of warfare. >>>From our findings thus far it appears that warfare was invented on three separate occasions (around 4000 BC, 2000 BC, and 900 BC), by three very different groups of belligerents that had no contact with each other (Mesopotamians, Chinese and Amerindians, respectively). We have found no evidence of contact that would indicate that warfare as a cultural pattern was diffused, transmitted or learned across these areas. Warfare therefore came about as a spontaneous development, not as a diffused or learned social pattern that was original to a single group (as Margaret Mead and other "diffusionist" social scientists first believed). The second major theoretical puzzle concerns our efforts to explain observed patterns of warfare as measured by the data being collected. My theory builds on earlier efforts by Wright, Richardson, Sorokin and others, using a causal, probabilistic, dynamic, and long-range approach. Dimensions of war such as time of onset T, magnitude M, or duration D are the dependent random variables whose behavior is explained by the form of distributions and the causal factors that give form to such distributions. Different probability functions define different forms of observed war uncertaintybecause war will have different patterns of uncertainty across war variables (T, M, or D), across epochs, and across belligerents. By contrast, most of the earlier literature has treated randomness as if it were all of one form (Poissons) and stationary. Warfare is governed by multiple nonlinear forceseach of which is politically founded and mathematically specified. What are these specific causal war forces? How are they characterized in substantive political terms? How can they be modeled mathematically to derive results and measure their empirical existence? I call my theory a "tri-force" theory of war because it is based on three universal political forces, each with a clear substantive interpretation that is well-known to theorists and belligerents alike. (Philosophically, my theory's long-range cross-polity perspective also attempts to go beyond the narrow realist-idealist debate that endures in our discipline, a debate absent in non-Western thinking on the causes of war.) I distinguish between (a) background, slowly-changing environmental conditions that constantly affect war and peace decisions, and (b) contingent historical conditions or changes that vary from place to place and from war to war. In turn, the latter consist of risk factors that tend to stabilize relations among belligerents, and factors that do the opposite and escalate war risk. I call these three war forces "anarchy", "stability" and "escalation." My theory of war is falsifiable by testing the theoretical tri-force model of anarchy, stability and escalation directly on the empirical data of observed distributions. The model detects the significant presence of anarchy, stability and escalation, and is better supported than the earlier, less theoretical Poisson and Weibull models. Preliminary tests have been conducted with the pilot dataset, with results mostly supporting the presence of the three forces with varying intensities (diachronic change) for different belligerents across epochs ("War and Politics in Ancient China: 722 - 222 BC", Internation6mal Studies Association, Vancouver, March 1991, with Melanie Mason; "War and Politics in Ancient Mesopotamia: 2725 -539 BC", American Political Science Association, Washington, DC, September 1993, with Henrik Sommer; "War and Politics in Ancient China: 2679 BC - AD 184", APSA, New York, September 1994, with David Lai). What do we know today, in terms of scientific cumulation, that we did not know in 1988? 1. It is feasible and productive to collect valid and reliable data on warfare and politics in antiquitythe skeptics were wrong. This has not been easy, but has been made possible through significant advances in many disciplines outside of political science, particularly archaeology, paleontology, anthropology and ethnohistory. Advances in epigraphy and chronometry have been essential in this regard. 2. War originated spontaneously and independently in three separate regions of antiquity: the Near East, China and Mesoamerica. It did not diffuse out of or was learned from a single area--the traditional diffusionists were wrong. 3. Warfare (along with other forms of social violence, such as cannibalism) was occurring in Mesoamerica long before the 16th-century European Conquest--it is an error to assume that Amerindians were peaceful. 4. It is equally feasible and productive to test theoretical models on long-range data, as well as compare results for modern patternsthe skeptics were also wrong on this. 5. Longer, more heterogeneous epochs show a mix of war forces operating on belligerents, thereby explaining why the constant-force Poisson model has been so successful in the pastearlier Poisson successes now have a substantive theoretical explanation, along with an extended empirical basis. 6. Shorter, more homogeneous epochs show more specific, variable forces acting on belligerents, thereby explaining the superior fit of non-Poisson models such as the Weibull--a set of war forces that we did not know existed or that they could be measured. 7. The combined tri-force model of anarchy, escalation and stability forces is superior to most other models tested so far, by a wide margin in some cases, lending support to the theoretical proposition that these are indeed very real and measurable forces acting on belligerentsthe realist-idealist debate is ill-founded, because all three forces coexist and future measurements will detail the intensity and dynamics of each. 8. Periods of large-scale political change (i.e., major dynastic transitions, changes in major governmental organizations, or changes in international order) seem to be preceded by measurable changes in the three forces (anarchy, escalation and stability) with escalation always undergoing a sharp rise in intensity, another new discovery that may also help monitor international conditions and help enact timely preventive measures. 9. The aforementioned results on the empirical validity of the tri-force model hold for all three "prongs" of the trident model, cross-societally as well as long-range, so the universal validity of anarchy, escalation and stability as basic forces of war is thus far strongly supported. FINIS >>>>>>>>>>>>>>>>>>>>>