Scenario modelling with morphological analysis

Technological Forecasting & Social Change 126 (2018) 116–125 Contents lists available at ScienceDirect Technological Forecasting & Social Change journal homepage: www.elsevier.com/locate/techfore T Scenario modelling with morphological analysis Iver Johansen Norwegian Defence Research Establishment, FFI, Pb 25, 2027 Kjeller, Norway A R T I C L E I N F O A B S T R A C T Keywords: Scenarios can serve as points of reference in the future for decisions that we have to make today. Morphological Scenario analysis provides a structured method for ensuring consistency and relevance in scenario development. This General Morphological Analysis paper outlines a method for characterizing the entire solution space of future outcomes in a given subject field, Defense planning and suggests a process for classification of an all-encompassing and mutually exclusive set of scenario classes. Foresight The method is illustrated with an example case, taken from Norwegian defense planning, of establishing a scenario set that encompasses all external security challenges to Norway as a security actor. Four parameters are defined – Actor, Goal, Method and Means. Each parameter is defined in terms of an exhaustive set of possible states or values. A Cross Consistency Assessment is conducted to exclude solutions deemed to be impossible on either purely logical grounds (internal consistency) or based on real world assessments (external consistency). Six scenario classes are defined: Strategic Attack, Limited Attack, Coercive Diplomacy, Terrorist Attack, Criminality and Military Peace-time Operations. 1. Introduction set, thus, is not the probability that it will eventually happen, but the fact that it might happen given certain assumptions about the surround- Scenario planners have long dealt with the problem of capturing a ing world. Schwartz (1996 p. 4) consequently defines scenarios as “tool complex and uncertain world within the confines of a limited number of [s] for ordering one's perceptions about alternative future environments scenarios. The problem is fundamental to future oriented studies and in which one's decisions might be played out”. planning, and a number of techniques and methods have been proposed In order to provide a useful tool for thinking about the future, to deal with it (Bryant and Lempert, 2010; Postma and Liebl, 2005; scenarios have to relate to established knowledge about the outside Groves and Lempert, 2007; Nguyen and Dunn, 2009; Kwakkel et al., world, and to a certain conception of what might actually happen in the 2013). Although no single approach will ever be able to transcend the future. In that vein, van der Heijden (2005 p. 225) posits that scenarios gap between scenario models and the real world, this paper suggests must conform to the principles of plausibility – scenarios must build on a that General Morphological Analysis applied to scenario modelling may logically derived cause-and-effect relationship between and within real solve some of the most pressing problems related to established world phenomena; consistency – scenarios must build on assumptions scenario methodology. that are not mutually exclusive; and relevance – a scenario must contain This paper looks at the scenario modelling problem from the sufficient high quality information to make it useful for its purpose. As standpoint of long term defense structure planning. In the defense an additional admonition, he states that the number of scenarios should planning process, scenarios serve vital functions as vehicles for war be restricted to two, three or – at the most – four, since that is the gaming, simulation, and analysis to support the design of a future force. maximum number decision makers are able to relate to in a systematic As an illustrative case and as an example of the morphological process way. in practice, the paper goes into some detail in explaining the develop- A number of schools and traditions have dealt with, and proposed ment of a scenario set for Norwegian defense planning. their own solutions to the complexities of scenario writing. Bradfield A scenario can usefully be defined as a description of a possible et al. (2005) identifies three dominant schools in scenario building: The future state or condition within a subject field. Scenarios are not French La Prospective school; the Probabilistic Modified Trends school predictions of future events, and although they sometimes provide associated with RAND; and the Intuitive Logics school. The Intuitive probabilities, their main function is to present decision makers with a Logics approach, associated with the oil company Royal Dutch Shell and set of alternative futures against which different courses of action might the Global Business Network, is perhaps the best known among them be measured. The basic criterion for inclusion of a scenario in a scenario (Bryant and Lempert, 2010). Being extensively applied for a wide E-mail address: iver.johansen@ffi.no. http://dx.doi.org/10.1016/j.techfore.2017.05.016 Available online 26 May 2017 Received 9 February 2017; Received in revised form 13 May 2017; Accepted 15 May 2017 0040-1625/ © 2017 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/). I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 variety of purposes, it has become something of a convention for extreme scenarios (see also Wright et al., 2013). The backward logic scenario and planning purposes. works by imagining that rare, high impact events have in fact occurred, Wack (1985) describes how the Intuitive Logics method was and then work backwards by disclosing which conditions would have to developed by analysts at the Shell oil company during the 1970s in a be in place for that particular event to materialize. situation with deep uncertainty about the future development in the Intuitive Logics with its recent enhancements go a long way in international energy market. The key element to the intuitive approach remedying the shortcomings of forward causal thinking, still the focus is a firm grip on what constitutes the forces driving the system, and on causal chains tends to unnecessarily restrict and complicate the awareness that there are some outcomes that are predetermined – events scenario process. This paper, therefore, proposes a shift of focus that have occurred, but whose consequences have yet to materialize, towards modelling non-reducible, complex problem spaces through and some that are fundamentally uncertain. Uncertainty in this context the application of General Morphological Analysis. stems from different sources, some purely random, as when an accident The remainder of this paper will briefly describe the use of scenarios causes a halt in the oil production, but also some that are related to in long term defense planning. It goes on to flesh out some of the choices made by social entities – the actors making up the system. fundamental aspects of General Morphological Analysis. It furthermore Under this particular type of uncertainty, outcomes are determined by suggests a process for classification of future outcomes in the context of multiple, interacting self-interested actors (Blanken, 2012). Hence, long term defense planning. Lastly, it provides an example case where scenarios have to incorporate not only probabilistic uncertainties, but the method is applied to develop scenario classes for defense planning also uncertainties of a strategic nature. purposes in Norway. By measuring the uncertain factors to the range of possibilities provided by the predetermined factors, some outcomes can be excluded from the scenario selection, while confidence in others is strengthened. 2. The use of scenarios in defense planning As an example, Wack shows that by 1972 analysis of the oil market had established that demand for oil was outstripping supply by a large Scenarios are basic to planning of military capabilities. A NATO margin, and that ten years of low economic growth would be required study (Campbell, 2010) observed that all of the nine nations contribut- to fit demand to supply. Hence, rising oil prices were seen as ing to the study used scenarios in some way or other for defining future predetermined and a balanced oil market with stable, low prices over force requirements. the long term could effectively be excluded from the company's Long term defense planning can be defined as “the process of defining scenario portfolio (Wack 1985 p. 82). long-term defense objectives and a strategy for their fulfilment” (Stojkovic This approach is well suited to reduce a potentially vast scenario set and Dahl, 2007 p. 9). Thus defined, the planning process must consider down to manageable proportions. Moreover, one of the attractions of politically determined objectives as well as technically defined force conventional scenario methodologies is their focus on causality. requirements. The process aims to outline a future force that can However, application of a particular causal chain to scenario develop- support the achievement of strategic objectives. Furthermore, defense ment might be extremely difficult given the non-linear properties that planning usually takes place in a resource restricted environment. characterize systems governed by human behavior. Misplaced causal- Hence, the final purpose of the planning process is to create a force plan ity, therefore, is a major source of bias in scenario building, and that conforms to budgetary restrictions at the same time as political and restricts the prospect of fitting discontinuities that cannot readily be military requirements are fulfilled. Consequently, what the planner is framed in cause-and-effect terms, to the scenario set. looking for is the most cost-effective solution to the force structure Consequently, according to Kwakkel et al. (2013 p. 1), conventional problem. scenario approaches struggle when dealing with rare events and cases Considering the complexities involved, it seems obvious that the where there are a multiplicity of possible futures. In defense planning, planning problem does not easily lend itself to simple calculation. In in particular, uncertainties related to political shifts and revolutions, reality, the only reasonably reliable method is to execute a systematic misperceptions and accidents characteristically overwhelm the prede- search among predefined force structure alternatives which are then termined elements. To put the matter bluntly, it is impossible to explain tested against a relevant scenario set. war as an incremental and entirely logical process based on observable The process may be – and usually is – implemented in two stages trends. Or, in a more academic phrasing, conventional scenario (Birkemo, 2013). In the first stage, the force structure alternatives are building tends to break down when confronted with possible futures analyzed with respect to their inherent capabilities and costs. In the that combine extremely low probabilities with potentially disastrous second stage, military scenarios are applied as testbeds to derive consequences. capability requirements. Scenario analysis may involve war gaming, This point is borne out by new research that indicates that interstate the use of simulation models and/or considerations of national conflict follows a power law logic where the size and the likelihood of doctrine. In the final analysis force structure capabilities are compared events, including wars, are affected by mechanisms linking micro-level to capability requirements in order to expose gaps and to direct the actions to macro-level outcomes in a strongly nonlinear fashion development of future force plans. (Cederman et al., 2011 p. 621). Conflicts thus may turn into wars, Capability based planning is sometimes set in contrast to threat and small wars into large wars through the operation of highly unstable based planning. This, however, is misleading, because, as Davis (2002) escalation processes governed by positive feedback loops between the points out, capability based planning is also very much concerned with interacting agents. threats Davis (2002 p. 8). What it is not, however, is concerned with These dynamics are not easily integrated into scenario methodolo- one specific threat and one specific scenario.1 Instead, a diverse gies that rest on forward looking causal reasoning. Referring to crisis scenario portfolio is required for an adequate representation of the management, Wright and Goodwin (2009 p. 16) claim that the security environment. requirement to incorporate rare and extreme outlier scenarios to the Scenario based analysis for long term defense planning is considered scenario set stands in stark contrast to the Intuitive Logics method since best practice among NATO nations (NATO, 2003). The NATO Handbook “… the range of focal scenarios is likely to be constrained by on Long Term Defence Planning (2003) also stresses that an adequate components of the construction methodology”. diversity of scenarios must be applied in order for the scenario set to be Framing the challenges posed by an exceedingly complex political and military environment therefore requires a radically different 1 In the NATO context, the shift from Threat Based Planning to Capability Based approach. Wright and Goodwin (2009) propose to apply a “backward Planning came as a result of the collapse of the Soviet Union and the need to prepare for a logic” to the Intuitive Logics method in order to create a range of more wider spectrum of challenges. 117 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 sufficiently descriptive of future threats and challenges. Developing a and Ritchey, 2015 p. 2). The method aims at identifying and structuring specific scenario for every conceivable situation, however, is unprac- all possible aspects and solutions for non-reducible, complex problem tical even if it were not also impossible. On the other hand, scenarios spaces which in most cases involve human behavior and political are representative of classes of situations, and several studies have choice. These problems are typically permeated with normative and approached this problem by implementing a two tiered approach to judgmental processes, making them even less accessible to quantifica- scenario classification (Campbell, 2010 p. 9-8). The first tier consists of tion or causal modelling. Finding solutions to such problems therefore scenarios defined as generic categories, whereas the second tier require an uncompromising openness of mind to new, unexpected scenarios are case studies which can be allocated to either of the possibilities. Zwicky himself does indeed characterize the method as generic scenario categories. This process assumes that the total solution “totality research” (Zwicky, 1969 p 30). space – i.e. all types of situations that can accrue – can be synthesized in The morphological process can be described as a dialectical one overarching set of generic scenarios. progression through repeated sequences of analysis and synthesis. Generic scenarios and specific case scenarios have different func- Regarding the most frequently applied post-Zwicky version of morphol- tions. According to Birkemo (2013 p. 11) generic scenarios are ogy – the method of the Morphological Box2 – it moves ahead in five employed to derive mission types that represent the entire spectrum of distinct steps (Zwicky and Wilson, 1967 p. 285; Zwicky, 1969 p. future challenges, while specific case scenarios are used for defining 115–120). capability requirements. This point can be illustrated by an example The first step requires an exact as possible formulation of the taken from Norwegian defense planning. The generic scenario Strategic problem, admitting that a precise delineation of the matter at hand may Attack represents a certain “class” of military challenges (i.e. a scenario be unlikely. In the next step the problem must be broken down into a class). Scenario classes provide a basis for defining the overall tasks and parameter set that frames the problem. Each parameter must be missions that a defense force is supposed to accomplish. In this precisely defined and an exhaustive and mutually excluding set of particular case those tasks and missions are related to inter alia possible states, or values, pertaining to each parameter, has to be defensive military operations, protecting vital infrastructure and sup- decided. The third step involves the construction of the morphological porting the reception of allied reinforcements. Within the confines of box – or multidimensional matrix – that contains all solutions related to this generic class of challenges, any number of different specific case the problem. A “solution” in this respect denotes a shape or configuration scenarios may be defined having the same basic characteristics, but where one value is designated for each parameter. In the example being more or less dissimilar in terms of overall scope, axes of advance, below (see Table 3.1), the shaded cells represent one particular operational time lines etc. The case scenarios, thus, provide quantita- solution. tive measures as to the specific capabilities that are required to The morphological box contains within itself the entire morphologi- accomplish a mission set. cal field (or the problem space) of the given problem. The problem space This distinction relates to two different problem sets in defense comprises all solutions that can be constructed on the basis of the planning. On the one hand, generic scenarios provide answers to the parameter set. However, the problem space usually consists of a large question of what the defense structure is supposed to accomplish, amount of “noise” in the form of inconsistent – or impossible – whereas case scenarios, on the other hand, establish how – in terms of solutions. The fourth step therefore entails a thorough analysis of the available resources – this can be done. Capabilities, in addition, are entire morphological field in order to reduce the amount of such noise, versatile, so even a many-sided mission set may be served by a limited and to delineate a solution space. The solution space, then, is a refined number of capabilities. Thus, the aim of the analysis is to seek out an subset of the problem space that only contains solutions that are optimal mix of capabilities, i.e. the least costly force structure that considered consistent. solves all missions. Consistency in this context is assessed on the basis of two criteria: This approach to scenario modelling distinguishes itself radically Firstly, logical consistency, i.e. the internal relationships of the concepts from conventional scenario methodology in its focus on the entirety of involved cannot be mutually contradictory; secondly, empirical consis- possible future outcomes instead of narrowly aiming for a reduced tency, i.e. a solution cannot rest on empirically impossible or highly scenario field. Instead of down-selecting an infinite number of potential improbable assumptions. scenarios to the standard four scenarios of the Intuitive Logics method, In the fifth step the remaining solution space is surveyed and the the two tiered approach basically builds a typology consisting of best solutions are selected for practical application. qualitatively different scenario classes that frame the entire space of Even a relatively small morphological field may contain a prohibi- potential outcomes. tively large number of theoretical solutions. This is evidenced by that By implication, this also means that generic scenarios cannot be fact that a matrix consisting of six parameters each with four values, as derived by considering the forces driving the system. Rather, generic in the example above, contains 4 × 4 × 4 × 4 × 4 × 4 = 4096 dif- scenarios are construed as static points in a parametrically defined ferent solutions. An examination of each and every one of the solution space, and not as a process involving cause and effect configurations in the matrix, therefore, would be an insurmountably relationships among variables. The next chapter will dig deeper into complex task even for a moderately large morphological field. Luckily, General Morphological Analysis as a methodological approach for the there is a way around this problem. In most morphological fields, there development of generic scenarios. are numerous value pairs that can be assessed as inconsistent, either on purely logical grounds, or empirically. Since the solution space cannot 3. General Morphological Analysis contain inconsistent value pairs, usually, the vast majority of the configurations in the morphological field can be eliminated from Morphological analysis – strictly speaking the study of forms – is further analysis. The weeding out of inconsistent value pairs and, well established as a method for modelling structural relationships hence, inconsistent configurations, is done using a procedure called between objects and phenomena in a number of scientific fields like Cross Consistency Assessment. It is not unusual that this process reduces botany, linguistics, geology and mathematics. A generalized version of the method was originally proposed by Swiss-American physicist and 2 astronomer Fritz Zwicky (1898–1974) who used it for purposes ranging Zwicky worked with a number of different techniques related to morphological from astronomy to technological forecasting and social/political pro- analysis. In addition to the method of the Morphological Box, these are, among others, Systematic Field Coverage, Negation and Construction, the Method of the Extremes, blem solving. Reasoning and Action by Analogy, Generalization and Approximation, Flexibility of In its generalized formulation, morphological analysis can be Scientific Truth, the Principle of Perfection and Imperfection (Zwicky and Wilson, 1967 p. defined as a general method for non-quantified modelling (Álvarez 284). 118 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 Table 3.1 The Morphological Box can be presented as a matrix with the parameters in the top row and the values in columns under each parameter. PARAMETER A PARAMETER B PARAMETER C PARAMETER D PARAMETER E PARAMETER F A1 B1 C1 D1 E1 F1 A2 B2 C2 D2 E2 F2 A3 B3 C3 D3 E3 F3 A4 B4 C4 D4 E4 F4 the morphological field by more than 90%. Even if this is not so, and it 4.1. Stating the problem sometimes happens that a large solution space remains after the consistency analysis, increased structure and oversight of the morpho- We begin by stating the problem: What are Norway's future security logical field will be gained. challenges and how can they be described? Before we move on, two In practice, the Cross Consistency Assessment is carried out by things need to be clarified. Firstly, how can the referent object of the systematically working through the entire matrix assessing the consis- study be defined, and secondly what do we mean by a security tency of each and every value pair. Since the number of pairs in a challenge? To answer the first question, Norway is defined as a specific matrix increases at a much lower rate than the number of configura- territory (i.e. the Kingdom of Norway), a population, and state institu- tions when new parameters are added, a relatively small number of tions that enable the exercise of sovereign authority over the territory pairwise consistency assessments will suffice to analyze even a large and the population. As for the last question, national security refers to morphological field (for a formal explanation see Ritchey, 2015). For the relationship between the state and its environment. Security instance, the 6 × 4 matrix in the above example, producing 4096 challenges, thus, for the most part, are those that are external to the unique configurations, contains only 240 value pairs. This process, state itself. What the analysis, thus, seeks to clarify and describe are although simple in principle, can be exceedingly time-consuming if challenges to the security to the state of Norway as defined above, that done manually, so a computerized support tool that presents results in primarily emanate from entities within the international system. an orderly fashion usually is required (Ritchey, 2006). The outcome of the morphological process is an abstract description 4.2. Defining parameters and values of the entire solution space, i.e. all possible solutions – or forms – related to a given problem. The solution space thus amounts to a Having stated the overarching problem, there are obviously an classification, or a typology, of any given problem or phenomenon. A infinite number of ways to break it down into a parameter set. Maria typology is generally multidimensional and conceptual (Bailey, 1994 p. Stenström states that good parameters should be meaningful, equally 4). Thus, each solution – or configuration in the morphological field – is important, abstract, straightforward, independent of each other, and have an internally consistent representation of real world phenomena that many internal connections (Stenström 2013 p. 21–22). are similar and that can be grouped together. The work group approached the problem by asking four simple In contrast to some classification techniques in the social sciences questions: (i) which type of actor can be a source for security challenges (see Elman, 2005 on the use of explanatory typologies), the morpholo- to Norway? (ii) what type of goals in the security domain can be seen as gical process does not make any theoretical claims or purport to explain relevant to a given actor type? (iii) which methods are relevant to actors a given phenomenon in terms of cause-and-effect relationships. The in order to reach a specific goal? (iv) what are the means required to only information one can extract from the morphological process is implement a specific method? whether a given solution is consistent or not, i.e. whether it relates to Thus, we have the parameter set of actor, goal, method and means. something that may exist in the real world. Hence, morphological These parameters were assessed to be meaningful to the subject matter analysis is as much a problem structuring tool as it is a means for expert group, and to have an equal importance relative to the problem analysis and modelling. being analyzed. They also are abstract, straightforward, independent, and they have potentially many internal connections. 4. Modelling scenario classes for long term defense planning in 4.2.1. Actor parameter Norway The term “actor” denotes any generic type of actor operating in the international system that can represent a security challenge to a state. In this section we will present General Morphological Analysis Note that we are not looking for specific actors, but rather representa- applied to the problem of creating an all-encompassing typology of tive types of actors. scenario classes. The analysis was originally carried out as an assign- Although there is no established procedure to define a fixed value ment for the Norwegian Chief of Defense in order to provide a relevant set, two criteria must be satisfied: (i) the range of values must cover all scenario set for the Defense Study 2007 (see Johansen, 2006). For this possible states of a parameter, and (ii) there must be as little as possible work, a small group of subject matter experts was assembled to (ideally no) overlap between values, whether they constitute a scale or contribute ideas as well as to assure quality of results. Furthermore, just separate points with no particular internal ordering (Stenström major stakeholders within the Defense organization were consulted at 2013 p. 23). For the practical process of defining values, the same ideals vital turning points during the process. apply as for the parameters. In addition, a useful approach in searching Of course there are no fixed answers to exactly how this problem for a value range is to begin by deciding the end points, i.e. the largest can be framed, what the essential parameters are, or how the internal and most comprehensive vs. the smallest and most constricted value. consistency in the morphological field can be established. What follows When the end points are defined, further analysis will determine what is a concise summary of problems the group had to tackle in the course can be a suitable value range for the in-between. of the analysis. The most comprehensive value on the Actor parameter obviously is a state. Hedley Bull asserts that states are “independent political communities, each of which possesses a government and asserts 119 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 sovereignty in relation to a particular portion of the earth's surface and territory. a particular segment of the human population” (Bull, 1977 p. 8). States As an alternative and more limited method of military control, we are sovereign entities that do not accept any political authority above define the establishment of military control on limited parts of another themselves. A state, thus, is the highest level of organization in the state's territory. “Limited”, in this context, is used in a purely spacious or international system and, consequently, forms one end point on the geographical manner. The degree of “limitness”, however, may range actor parameter. from single points – e.g. a building or a small town – up to larger parts In the other end of the continuum a single individual clearly is the of a country's land territory. most constricted value. Individuals obviously are extremely limited in As an even less comprehensive method for the application of their abilities to operate as actors in the international system, let alone military force, an enemy actor might conduct operations to deny an pose threats to a society and a state. However, technological develop- opponent access to certain areas or domains while not attempting to ment in all its aspects, including the interconnectedness of everything establish control in those areas in a military sense. This is a military that govern people's daily lives, increasingly empowers single indivi- method that lately has been termed Anti-Access/Area Denial (Thomas duals as well as groups to threaten people and states. Consequently, we and Dougherty, 2013 p. 64–71). Anti-Access/Area Denial, thus, is define individuals – including groups of individuals – as the low end defined as a separate value on the method parameter. point on the Actor parameter. Control and denial of access involve the active use of armed force. Between those two extremes, the number of potential intermediate An actor, however, may choose to employ military force purely for values obviously is unlimited. Here, we will settle for two – network and signaling purposes e.g. by engaging in threatening military maneuvers, business enterprise. Network organization is a product of technological conducting aggressive force deployments, putting forces on alert status developments that enable coordinated activities without regard to etc. On the other hand, military deployment may be entirely covert or physical space, formal organization or large and complex infrastruc- concealed, or indeed primarily intended to deceive or to create tures. The network consequently is the preferred form of organization confusion. These activities do not entail the targeted use of weapons. for non-state actors that threaten Western societies. We will designate this method symbolic use of force. A business enterprise is an organized entity that mainly engages in States also routinely deploy military forces for exercises, intelli- the pursuit of profit. While usually being of no particular security gence gathering etc. We will call this particular method for the concern, business enterprises nevertheless might acquire relevance as employment if military force peace time operations. Although not hostile security political actors by e.g. employing economically backed pres- or threatening, this type of operation may anyway be of a concern, sure against a government, or its agents might violate borders or engage depending on factors such as geographical proximity, military posture, in other activities that impinge upon a state's sovereignty. degree of political (dis)trust, etc. In order to complete the value range on this parameter we also 4.2.2. Goal parameter include relevant methods that do not necessarily imply the use of Next is the Goal parameter. Of course, the concept of “a goal” is military force. We define the following three methods: firstly, attack perhaps even more nebulous than the concept of “actor”. However, a against infra-structure and/or civilian population, secondly, the use of workable approach to define a value range might be to focus on things economic force, and lastly criminality (any activity that constitutes a that an actor might seek to achieve that may threaten or weaken the violation of the legal code). security of the targeted state. Again, we start by defining the most comprehensive value in the range. We determine this to be the overthrow of the existing political regime. Thus regime change, which 4.2.4. Means parameter obviously also implies the complete elimination of political sovereignty, The Means parameter define both the type and the scope of is defined as the most extreme value on the Goal parameter. resources available to an actor. Again we start with the most compre- As a counterpart to regime change, at the other end of the spectrum hensive parameter value, which we define to be large scale use of military we define economic gain to be the most trivial goal that still might serve force. This value implies the build-up, preparation and employment of as motivation to actions that can have security implications for a state. sea, air and land forces in joint operations that are led on the strategic Having thus defined the end points of the value range, we move on level of command. to consider which values can be defined in the interval between the On the next level we define limited use of military force. Of course, extremes. First, we focus on enforcement of political concessions. While any use of force which cannot be considered large scale is in some sense both goals – regime change and the attainment of political concessions limited. However, to be more specific we define limited use of military – play on the overarching ambition to impose one's will on (the force to comprise the deployment of ready forces only (e.g. with little or government of) another state, it is presumed that the latter is the less no call up of reserves, transfer of forces etc.) that are led on the demanding strategy both in terms of resources and risk. operational level of command. Second, while being of a less aggressive nature, a foreign actor may On the non-military side we define the concepts large scale use of maintain a high readiness of military forces involving exercise deploy- non-military force, and limited scale use of non-military force respectively. ments and training missions. In addition, that actor may target a Non-military, in this respect, signify that the use of force is carried out specific state in order to gather information for intelligence purposes, or by entities that are not part of any recognized military force. The maintain a general military posture that is more or less openly directed difference in scale may not be entirely clear. For the purposes of against another state. Thus, we define military exercise/intelligence analysis we draw a distinction between uses that require the concerted gathering as the next lower level on the Goal parameter. efforts of larger networks over time, and uses that that may be put into effect by individuals or smaller groups with limited input in terms of 4.2.3. Method parameter planning, materiel and personnel resources. The Method parameter signifies alternative ways an actor can It also seems relevant to include means that primarily belong to the proceed to achieve a goal. In this sense this parameter is analogous to economic domain as a separate value. Thus, we include economic the concept of strategy. Again, moving ahead by delineating the sanctions to the value set. Lastly, there are obviously any number of extremes in the value set, the most comprehensive method that can other means that may be used by an actor to achieve a particular goal or be conceived of in this context is establishing military control over the to support a particular strategy. However, we do not need to specify this entire territory of another state. Territorial control in this context is any further, so the last value in the value range is simply other means. defined as the ability to deploy own forces at will in a given territory, while at the same time denying other actors' forces access to the same 120 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 Table 4.1 Multidimensional matrix. ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Territory Large Scale Use of Military Force Network Political Concessions Military Control over Parts of Territory Limited Scale Use of Military Force Business Enterprise Military Exercise, Intelligence Gathering Anti-Access/Area Denial Large Scale Use of Non-Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non-Military Force Peace Time Operations Economic Sanctions Attack Against Infra-structure/Population Other Means Use of Economic Force Criminality Table 4.2 Consistency matrix. Cells with an “x” indicate an inconsistent value pair. Open cells indicate that the relevant values are assessed to be consistent. Mil. Exercise/Intel. Gathering Mil. Control Parts of Territory Large Scale Non-Mil. Force Mil. Control Entire Territory Large Scale Use of Force Peace Time Operations Att. Infrastr./Population Symbolic Use of Force Limited Non-Mil. Force Political Concessions Limited Use of Force Economic Sanctions Business Enterprise Denial of Access Regime Change Economic Force Economic Gain Other Means Individual(s) Criminality Network State State Network Business Enterprise Individual(s) Regime Change x x x Political Concessions x Mil. Exercise/Intel. Gathering x x Economic Gain x x Mil. Control Entire Territory x x x x x x x Mil. Control Parts of Territory x x x x x x Denial of Access x x x x x x Symbolic Use of Force x x x x x x Peace Time Operations x x x x x x Att. Infrastr./Population x x x x Economic Force x x x x x Criminality x x x x x Large Scale Use of Force x x x x x x x x x x x Limited Use of Force x x x x x x x x Large Scale Non-Mil. Force x x x x x x x x x x x x x Limited Non-Mil. Force x x x x x x x x x x x x Economic Sanctions x x x x x x x x x x x x Other Means x x x x x x x x x x 4.3. Constructing the multidimensional matrix context of this analysis (security challenges to Norway), we will have to find, firstly, at least one real world actor that may have a motive and, Now that all parameters and parameter values are defined, the secondly, assess whether the relevant actor(s) has the capabilities to put multidimensional matrix, or morphological box, can be constructed. In such a strategy into effect. The analysis (see Johansen, 2006 p. 22–30) Table 4.1 the parameters are presented in the top row, while the values identified only one possible candidate – Russia. are placed in the columns under each parameter. This primarily methodological analysis is not the place for an in depth inquiry into Russia's military capabilities, posture or doctrine. 4.4. Cross consistency assessment Suffice it to note that given the extent of the required build-up of forces, the time needed for that to take place and the ability of Norway and the The matrix represents the entire morphological field of the problem, NATO alliance to put a defensive force in place, a strategy that seeks to which is 4 × 4 × 8 × 6 = 768 unique configurations. The Cross establish military control over Norway's entire territory is deemed Consistency Assessment involves a thorough examination of each pair inconsistent with Russia's military capacity today and in a foreseeable in the matrix. Here, the number of pairs is future. All scenarios associated with this value pair, consequently, are 4 × 18 + 4 × 14 + 8 × 6 = 176. also considered inconsistent. Note also that all scenarios containing the The result of the Cross Consistency Assessment is presented in value Regime Change on the Goal parameter fall outside the solution Table 4.2. space since it too presupposes the establishment of full territorial It is not practical to present an in-depth examination of each and military control. every pairwise assessment in the matrix. However, a few comments can On the other hand, the less demanding method of establishing be made. Firstly, and perhaps surprisingly, the value pair State – partial military control can plausibly be put into effect without a Military Control over Entire Territory is assessed to be inconsistent. sizeable force build-up, primarily by enabling the effect of surprise. While this generally can be held to be a “true” value pair, in the Scenarios associated with this value, thus, may be considered consis- 121 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 Table 4.3 Solution space. State Political Concessions Mil. Control Parts of Territory Large Scale Use of Force State Political Concessions Mil. Control Parts of Territory Limited Use of Force State Political Concessions Anti-Access/Area Denial Limited Use of Force State Political Concessions Symbolic Use of Force Limited Use of Force State Political Concessions Att. Infrastr./Population Limited Use of Force State Political Concessions Economic Force Economic Sanctions State Mil. Exercise/Intel. Gathering Peace Time Operations Limited Use of Force Network Political Concessions Att. Infrastr./Population Large Scale Non-Mil. Force Network Political Concessions Att. Infrastr./Population Limited Non-Mil. Force Business Enterprise Economic Gain Criminality Other Means Individual(s) Political Concessions Att. Infrastr./Population Limited Non-Mil. Force Individual(s) Economic Gain Criminality Other Means Table 4.4 Scenario Class I: Strategic Attack (shaded cells). ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Large Scale Use of Military Territory Force Network Political Concessions Military Control over Parts of Limited Scale Use of Territory Military Force Business Military Exercise, Anti–Access/Area Denial Large Scale Use of Non– Enterprise Intelligence Gathering Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non– Military Force Peace Time Operations Economic Sanctions Attack Against Infra– Other Means structure/Population Use of Economic Force Criminality tent. Logically, this would be any configuration containing the parameter The outcome of the cross consistency assessment is a solution space value Large Scale Use of Military Force on the Means parameter. There is consisting of 12 “surviving” configurations, or “scenarios”. This repre- only one configuration in Table 3.1 which contains that value. Together sents a reduction of more than 98% compared to the total morpholo- with the values State – Political Concessions – Military Control over Parts gical field. The solution space is presented in Table 4.3.3 of Territory, this configuration amounts to the most extensive military The solution space contains all possible security challenges to the scenario in the solution space. It indicates the possibility of a state object of the analysis, which in this case is a particular state actor – employing sizeable military force in large scale joint operations, Norway. The analysis does not, however, end here. The solution space establishing some degree of (temporary) military control to enforce matrix gives little meaning unless the result can be synthesized into political concession on another state actor. This scenario class is meaningful categories. We defined the goal of this analysis to be the designated Strategic Attack (Table 4.4). construction of a set of all-encompassing scenario classes. The matrix will aid the formulation of scenario classes by providing a framework 4.5.2. Scenario Class II: Limited Attack for this process. In the next chapter we will define a set of scenario Moving down to Limited Scale Use of Military Force on the Means classes based on the solution space. parameter, this parameter value combines with State, Political Concessions and Military Control over Parts of Territory. In this case, 4.5. Defining scenario classes however, the configuration also combines with Anti-Access/Area Denial and Attack against Infra-structure and/or population on the Method In defining scenario classes, what we are looking for is an parameter. Taken together these solutions represent clearly a more exhaustive typology of security challenges. The process takes as its limited employment of military forces than that portrayed by the point of departure the solution space matrix. However, among the scenario class Strategic Attack. We will call this scenario class Limited twelve unique solutions present in the matrix there may be certain Attack (Table 4.5). solutions that overlap, or there may be sets of solutions where the division between them serves no practical purpose. This process starts 4.5.3. Scenario Class III: Coercive Diplomacy with a thorough examination of the solution space matrix aimed at Still focusing on configurations containing the parameter values collapsing solutions that resemble each other into one scenario class State, Political Concessions and Limited Scale Use of Military Force, we (see Elman (2005) on pragmatic compression p. 300). identify two additional configurations that combine with, in the first case, Symbolic Use of Force and, in the other, Use of Economic Force on 4.5.1. Scenario Class I: Strategic Attack the Method parameter. Both of these configurations signify actions on We start by seeking out the configuration in the solution space the part of a state actor to coerce a targeted actor to alter behavior on matrix that represents the most comprehensive military challenge. some issue. Coercive strategies rely on the threat of force, but do not necessarily involve the use of force. They may, however, involve 3 deployment of military forces or the application of economic sanctions, Changing assessments on any value-pair would change this outcome, and a sensitivity analysis in order to test alternative assessments and their outcomes is recommended. An either in combination or as separate strategies. This particular sort of altogether alternative method is the Battelle approach (see Nguyen and Dunn, 2009) strategy is designated Coercive Diplomacy. which applies a five point consistency scale in order to obtain a more flexible result. Coercive diplomacy is here thought of as an offensive strategy as 122 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 Table 4.5 Scenario Class II: Limited Attack (shaded cells). ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Large Scale Use of Military Territory Force Network Political Concessions Military Control over Parts of Limited Scale Use of Territory Military Force Business Military Exercise, Anti–Access/Area Denial Large Scale Use of Non– Enterprise Intelligence Gathering Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non– Military Force Peace Time Operations Economic Sanctions Attack Against Infra– Other Means structure/Population Use of Economic Force Criminality Table 4.6 Scenario Class III: Coercive Diplomacy (shaded cells). ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Large Scale Use of Military Territory Force Network Political Concessions Military Control over Parts of Limited Scale Use of Territory Military Force Business Military Exercise, Anti–Access/Area Denial Large Scale Use of Non– Enterprise Intelligence Gathering Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non– Military Force Peace Time Operations Economic Sanctions Attack Against Infra– Other Means structure/Population Use of Economic Force Criminality Table 4.7 Scenario Class IV: Terrorist Attack (shaded cells). ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Large Scale Use of Military Territory Force Network Political Concessions Military Control over Parts of Limited Scale Use of Territory Military Force Business Military Exercise, Anti–Access/Area Denial Large Scale Use of Non– Enterprise Intelligence Gathering Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non– Military Force Peace Time Operations Economic Sanctions Attack Against Infra– Other Means structure/Population Use of Economic Force Criminality Table 4.8 Scenario Class V: Criminality (shaded cells). ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Large Scale Use of Military Territory Force Network Political Concessions Military Control over Parts of Limited Scale Use of Territory Military Force Business Military Exercise, Anti–Access/Area Denial Large Scale Use of Non– Enterprise Intelligence Gathering Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non– Military Force Peace Time Operations Economic Sanctions Attack Against Infra– Other Means structure/Population Use of Economic Force Criminality 123 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 Table 4.9 Scenario Class VI: Military Peace Time Operations (shaded cells). ACTOR GOAL METHOD MEANS State Regime Change Military Control over Entire Large Scale Use of Military Territory Force Network Political Concessions Military Control over Parts of Limited Scale Use of Territory Military Force Business Military Exercise, Anti–Access/Area Denial Large Scale Use of Non– Enterprise Intelligence Gathering Military Force Individual(s) Economic Gain Symbolic Use of Force Limited Scale Use of Non– Military Force Peace Time Operations Economic Sanctions Attack Against Infra– Other Means structure/Population Use of Economic Force Criminality when an actor actively seeks to change the status quo. Of course, the Such operations are endemic features of inter-state military relation- term coercive diplomacy in the academic literature as well as in the ships, hence states normally allocate significant resources to monitor political discourse involves a much wider range of actions, not least and, if it is required, intercept another state's military units. This verbal communication, than those defined in the morphological particular class of challenge lies at the base of any state's security analysis (for an in depth analysis, see Schelling, 1966). However, a functions, thus, representing a “base line” scenario to its military narrow focus on the physical aspects of coercion is validated by the establishment (Table 4.9). primarily military and security purpose of this analysis (Table 4.6). 4.5.4. Scenario Class IV: Terrorist Attack 5. Conclusion The analysis of the solution space matrix reveals that both Network and Individuals on the Actor parameter combine with Political We started out stating that conventional scenario methodology may Concessions on the Goal parameter and use of non-military means be less effective in incorporating future shocks and discontinuities in (both Large Scale- and Limited Scale Use of Non-Military Means) on the the scenario set. Two factors in particular inhibit the possibilities to Means parameter. On the Method parameter, the only option combin- span the scenario canvas wide enough. First, the focus on and reliance ing with this value set is Attack Against Infra-Structure/Population. These on modelling the effect of driving forces and, secondly, the need to keep value sets make up three separate configurations in the solution space the scenario portfolio as limited as possible. These two considerations matrix that confirm to Terrorist Attack scenarios. force scenarios to concentrate around observable trends and “most It should be noted, though, that terrorist attacks in most countries, likely” developments. Especially in the case of defense planning a and certainly in Norway, are not in the first instance considered strong trend focus can have negative effects. The defense planner, more relevant to the external security functions of the state. Hence, it is than most, has to take into consideration effects and developments that not a firsthand matter for defense planning. Two factors, however, may lie far beyond the normal operation of a state's security environment. indicate otherwise. For one, a terrorist attack may be so large that Wars, after all, are rare and extreme events, but still they are what military action is required to fight or prevent it. Secondly, terrorist defense planning in the last instance is all about. So, a viable scenario attacks may be prepared with active or passive support of another state methodology must, as a minimum, enable the analyst to identify or in countries with dysfunctional state and legal institutions, requiring developments that lie beyond the most likely trajectories, thus widen- intervention by military forces (Table 4.7). ing the scope of the scenario space. Complex problems by their very nature consist of an infinite number of parameters and may produce an equally infinite number of solutions. 4.5.5. Scenario Class V: Criminality Consequently, there are no answers or solutions that are completely The scenario class Criminality represents scenarios where a private “right” or completely “wrong”. Applying morphological methodology actor (Business Enterprise and/or Individual(s) on the Actor parameter) to the process does not alter that fact. However, solving the problem or further their economic goals by illegal means. For the most part these finding the right answer may not be the essential issue, but rather to are scenarios that have only marginal security political relevance but finding something that “makes sense” – that works in a pragmatic sense are matters for a state's judicial authorities. However, in some cases, – and the morphological approach may provide just that. economically motivated criminality may have a security political As the example case discussed in this article shows, the morpholo- relevance. This might be the case when unlawful actions take place in gical approach is well suited to explore both the possibilities and the remote locations, and it is only military forces that have the capacity to limits to future developments. It also comes with a number of other intervene; when a private actor operates as proxy for another state, or advantages. when enforcement of jurisdiction is delegated to the military by state Firstly, the morphological approach offers the analyst a structured authorities. This is the case for the enforcement of the United Nations process to the development of scenarios that leaves a clear audit trail, Convention on the Law of the Sea pertaining to Norway's economic which is open to investigation by outsiders. Effectively, the final result zone (Table 4.8).4 can be retraced back to exactly which considerations produced a given scenario set. 4.5.6. Scenario Class VI: Military Peace Time Operations Secondly, instead of focusing on a limited number of point scenarios Military Peace Time Operations comprises any military operation on in an unlimited space, the morphological process enables a generic the part of a foreign state including exercises, intelligence collection or classification of the entire solution space. This classification ensures, (i) even covert operations that is unrelated to any other scenario class. that every generic security challenge is represented in the final scenario set, and (ii) that inconsistent scenarios are eliminated from the scenario 4 http://www.fisheries.no/resource_management/Area_management/economic_ selection. zone/#.V9h0GPmLSM8. Thirdly, by systematically exploring all possibilities to a problem, 124 I. Johansen Technological Forecasting & Social Change 126 (2018) 116–125 the morphological approach enables the definition and selection of Bull, H., 1977. The Anarchical Society: A Study of Order in World Politics. Columbia University Press, New York. outlying solutions just as much as those most likely and mainstream. Campbell, A., 2010. Analytic Implications of the NATO Defence Planning Process SAS-081 However, there are also challenges. One is obviously that a full Specialist Team Summary Report. blown morphological process may be demanding in terms of time and Cederman, L.-E., Warren, T.C., Sornette, D., 2011. Testing Clausewitz: nationalism, mass mobilization, and the severity of war. Int. Organ. 65, 605–638. http://dx.doi.org/10. resources, and with – at least in the beginning – uncertain prospects of a 1017/S0020818311000245. useful result. Another is that the morphological process for a very large Davis, P.K., 2002. Analytic Architecture for Capabilities-based Planning, Mission-system part rests on a lot of judgmental evaluations. This goes from the first Analysis, and Transformation. RAND. Elman, C., 2005. Explanatory typologies in qualitative studies of international politics. formulation of a research question, through the definition of parameters Int. Organ. 59, 293–326. and values, to the consistency assessment and selection of final Groves, D.G., Lempert, R.J., 2007. A new analytic method for finding policy-relevant solutions. This, however, is exactly where a sound and structured scenarios. Glob. Environ. Chang. Johansen, I., 2006. Scenarioklasser i Forsvarsstudie 2007: En morfologisk analyse av process has its place. In addition, by establishing firm criteria for the sikkerhetspolitiske utfordringer mot Norge. http://www.ffi.no/no/Rapporter/06- analysis of the morphological field, and making that process transpar- 02664.pdf. ent, it is possible to communicate results to those who may apply results Kwakkel, J.H., Auping, W.L., Pruyt, E., 2013. Dynamic scenario discovery under deep in practice as well as to the wider scientific community. uncertainty: the future of copper. Technol. Forecast. Soc. Chang. 80, 789–800. NATO, 2003. Handbook on Long Term Defence Planning. (doi:RTO-TR-069 AC/323(SAS- This leads to the final overarching challenge – can we be certain that 025)TP/41). the solution space actually is exhausted and that there are not Nguyen, M.-T., Dunn, M., 2009. Some Methods for Scenario Analysis in Defence Strategic unidentified scenarios that lie beyond the defined limits of the Planning. Defence Science and Technology Organisation (DSTO), Canberra. Postma, T.J.B.M., Liebl, F., 2005. How to improve scenario analysis as a strategic morphological space? The answer is, of course, that we cannot. This, management tool? Technol. Forecast. Soc. Chang. 72, 161–173. http://dx.doi.org/10. however, is no different from any scientific process, the hallmark of 1016/j.techfore.2003.11.005. which is that discoveries must be open to challenge when new evidence Ritchey, T., 2006. Problem structuring using computer-aided morphological analysis. J. Oper. Res. Soc. 57, 792–801. is brought to the table. Ritchey, T., 2015. Principles of cross-consistency assessment in general morphological modelling. Acta Morphol. Gen. 4. Acknowledgements Schelling, T.C., 1966. Arms and Influence. Yale University Press, New Haven. Schwartz, P., 1996. The art of the Long View. Planning for the Future in an Uncertain World, 1996th ed. Doubleday, New York. The author would like to thank Leo Blanken, Douglas Borer, Michael Stenström, M., 2013. Morphological Analysis in Groups: A Personal Guide. (FOI-R–3678– Freeman, Jeffrey Appleget and Robert Burks, all at the Naval SE 85 p.). Stojkovic, D., Dahl, B.R., 2007. Methodology for long term defence planning. Kjeller. Postgraduate School, as well as the two anonymous reviewers for http://rapporter.ffi.no/rapporter/2007/00600.pdf. useful comments along the way. A special thanks also to Marie Røsok Thomas, J., Dougherty, T., 2013. Beyond the Ramparts. The Future of US Special who did proof read the final draft of the article. Operations Forces. Center for Strategic and Budgetary Assessment, Washington DC. This research did not receive any specific grant from funding van der Heijden, K., 2005. Scenarios. The art of Strategic Thinking, second ed. John Wiley & Sons, Chichester. agencies in the public, commercial, or not-for-profit sectors. Wack, P., 1985. Scenarios: uncharted waters ahead: EBSCOhost. Harv. Bus. Rev. 73–89. Wright, G., Goodwin, P., 2009. Decision making and planning under low levels of References predictability: enhancing the scenario method. Int. J. Forecast. 25, 813–825. Wright, G., Bradfield, R., Cairns, G., 2013. Does the intuitive logics method - and its recent enhancements - produce “effective” scenarios? Technol. Forecast. Soc. Chang. Álvarez, A., Ritchey, T., 2015. Applications of general morphological analysis from 80. http://dx.doi.org/10.1016/j.techfore.2012.09.003. engineering design to policy analysis. Acta Morphol. Gen. 4. Zwicky, F., 1969. Discovery, Invention, Research through the Morphological Approach, Bailey, K.D., 1994. Typologies and Taxonomies. An Introduction to Classification US Edition. The Macmillan Company, Toronto (ed.). Techniques. Sage Publications, Thousand Oaks. Zwicky, F., Wilson, A.G. (Eds.), 1967. New Methods of Thought and Procedure: Birkemo, G.A., 2013. Is Norwegian long term defence planning risk based? Kjeller. http:// Contributions to the Symposium on Methodologies. Springer Verlag, New York. www.ffi.no/no/Rapporter/12-00923.pdf. Blanken, L.J., 2012. Reconciling strategic studies … with itself: a common framework for Iver Johansen is a senior researcher at the Norwegian Defence Research Establishment choosing among strategies. Def. Secur. Anal. 28, 275–287. http://dx.doi.org/10. (FFI). He has a cand. polit. (MA) degree in political science from the University of Oslo 1080/14751798.2012.730723. (1987) and specializes in military studies and operational analysis. He also has extensive Bradfield, R., Wright, G., Burt, G., Cairns, G., van der Heijden, K., 2005. The origins and experience in managing research projects. Fields of interest include international evolution of scenario techniques in long range business planning. Futures 37, relations in general, long term defense planning, scenario development, morphological 795–812. http://dx.doi.org/10.1016/j.futures.2005.01.003. analysis, and war gaming. Bryant, B., Lempert, R., 2010. Thinking inside the box: a participatory, computer-assisted approach to scenario discovery. Technol. Forecast. Soc. Chang. 1, 34–49. 125