In the best of all worlds, social science methods are tailored to suit the nature of the phenomena under study. In the real world, established research traditions are more likely to apply their preferred methods to various objects of study – often without much regard to their specific substantive characteristics. They do so as an act of faith in their heuristic value, because their existing toolboxes do not seem to allow for anything else, or because they deliberately presume that an unusual feature can be downplayed or subsumed under a more general category. Given that social science approaches are apt to vary widely in their meta-theoretical assumptions (ontology, epistemology, axiomatic view of human nature), the fruits they bear when applied to the same object are equally apt to be at variance. While their respective scientific value is difficult to assess in the abstract, and absolute truth is beyond reach, it remains possible to discriminate between them on the basis of the more or less rationally convincing character of the theories they produce as regards particular objectsFootnote 1 – thus between the methodsFootnote 2 on which they habitually rely.
This is what this article purports to do by examining the social science literature centring on armed conflict, military institutions and civil-military relations. In light of the ambition outlined above, the choice of military topics, both per se and as a test-case, recommends itself in several ways. First, the seeming ubiquity of armed conflict, its litany of casualty tolls, its overriding and long-lasting impact on all compartments of life, the weight of martial organizations on society, culture, politics and the economy, hardly make it a subject of peripheral interest. Second, despite its obvious significance, it has been relatively understudied outside of a narrow community of specialized scholars, and the international literature on it, though substantial over the long term, tends to be ignored by mainstream social scientists. Third and more to the point, the subject-matter, in some of its central aspects – not least war itself –, does not always yield easy definitions or translations of information into facts. Accordingly, its mode of conceptualization tends to be more sensitive to the choice of approach than in most substantive areas. Fourth, though not all aspects of it are by any means unique, it does offer hallmark features – which analysts and theoreticians overlook at their own peril – that are unparalleled in any other department of social activity in kind or degree. This is where some methods, as we shall see, prove unequal to the task, while others seem to exhibit greater measures of penetrating insight. Finally, it is doubtful whether any other substantive field has been affected by change and renewed in its central aspects over the last twenty years to the degree that the military field has – a characteristic that makes scrutiny of it especially worthwhile for the purpose at hand.
The roadmap to the developments to follow is thus clear. Having briefly identified the major research programmes in social science, this article will start by assessing the part they have played so far in the military field. It will then compare those which have dominated it, and venture an evaluation of their respective degree of success in approaching the intimate nature of things military. Based on such cognitive assessment, it will eventually suggest ways to define, delineate and analyse the field's ample subject-matter, hoping all along to bring coherence to it.
Major research programmes and the military field
Research traditions, combining meta-theoretical assumptions in every possible non-contradictory way, are innumerable among and within disciplines. Yet, if one follows J.-M. Berthelot,Footnote 3 such variety ultimately boils down to three main families of what Imre Lakatos has called research programmes: naturalism, intentionalism and symbolicism.
The first has long been equated with classical positivist orientations (segmentation of reality into simple homogeneous components, observation, preference for stable order, causal/functional/structural analysis, nomological theories, etc.). Yet, the last three decades have seen the rise of new breeds of antipositivist naturalism (which Berthelot does not mention), to be found in various strands of complexity theory (critical realism, nonlinear dynamics, second order complexity) derived in large part from recent developments in natural scienceFootnote 4 and cybernetics. Beside a reaffirmation of realism in the face of mounting relativism or extreme forms of idealism, they share the view that aspects of reality that are unobservable (generative mechanisms, intentions/meanings) and/or unpredictable (through context and interaction, positive feedback and endogenous change, etc.) are not outside their purview. Thus is the naturalist ambition to preserve the unity of science satisfied on terms that are less inimical than the positivist tradition to the characteristics of the social world.
The “intentionalist” family is subdivided into a phenomenological/constructionist wing, characterized by more or less radical forms of contextualism, subjectivist interpretation and indifference to general substantive propositions,Footnote 5 and a social action wing which holds both subjective and objective ends together by connecting context, rational or irrational motives and causal analysis of aggregate, anticipated as well as emergent, factual outcomes.
Finally, orbiting around the “symbolicist” pole are research programmes whose defining characteristic is the idea of autonomous meaning structures (either mythical, historical, ideological or societal) waiting to be unveiled so as to supply interpretations of social phenomena. Such a stance may in radical versions – post-modernism foremost among them – result in the rejection of even the mildest forms of realism.
While its chief merits are clarity and exhaustiveness, such a classification is not without its ambiguities.Footnote 6 Yet the task ahead will be greatly simplified by virtue of the quasi-absence of some of the above-mentioned programmes from studies of things military.
Echoes of the post-modern movement in the military field have remained limited. Beginning in the 1980s, authors outside the specialized milieu have followed the lead of Jean Baudrillard or Paul Virilio in denouncing the dehumanising influence of “technoscience” and “hyperreality” (a self-referential web of meaning peculiar to unchecked Western modernity – now gone global and in the process of implosion), and in applying their vision to war,Footnote 7 portrayed as bloody non-events devoid of political significance. Another echo came in the early 1990s from within the specialized community when a few key contributors embraced the label “post-modern” to conceptualize the dramatic shift from Cold War to post-Cold War.Footnote 8 They did so against the advice of some of their colleaguesFootnote 9 who considered the implications of post-modernist discourse (not least objectivity as illusion) at variance with what was basically a continuation of a classical brand of Weberian/pragmatic military sociology in the Chicago tradition of Harold Lasswell and Morris Janowitz. In the present decade, the post-modern label seems to have gradually gone out of fashion in that community of scholars.
Likewise, the phenomenological/constructionist programme has not so far attracted a strong following in the field. In substantive terms, it has mostly been restricted to a critical view of identity formation – notably with regard to gender, sexual orientation and ethnicity – in relation to the military and its impact on society.Footnote 10 In methodological terms, it has drawn attention to the need for “writing researchers into their research” on the military, especially to their own construction of social realityFootnote 11 (on the basis of class, gender, ethnicity or past military experience) in interviews or participant observation. Mainstream military sociology was briefly indebted to a soft version of social constructionism in the 1990’s, in a context of sweeping all-round change where military raison d’être, missions and situations had to be defined anew in the midst of uncertainty.Footnote 12 However, as in the case of post-modernism, such a departure was short-lived and slight.
The same applies to the antipositivist naturalism of the last three decades. Critical realism has yet to focus on military matters, and has barely begun to make its mark on international relations theory;Footnote 13 nonlinear dynamics, surfing on the intellectual disarray which followed from the failure of conventional IR theory to predict the Cold War's demise, made a promising debut in international relations and the military field.Footnote 14 It inspired one of the most illuminating articles on Clausewitz ever written.Footnote 15 Yet, it has so far not produced the mass of scholarship that seemed in the cards, and the promise still has to be fulfilled.Footnote 16
We are thus left with only two major contenders, which between them encompass most of the literature: “positivist naturalism” and the “social action/interaction” framework. The former sees the social world as objective, and social behaviour as determined; it probes it in terms of linkages phrased in the language of variables, quantified or not, and seeks to reveal “covering laws”, Newtonian style. The latter, pioneered by Weber a hundred years ago, is a hybrid of naturalism and intentionalism, weaving nominalism and realism together, in which theory runs in terms of generalized uniformities of social process based on “rules of experience,” in hypothetical situations governed by clear features and one-sided motives – though the irreducible complexity of real contexts leaves the door open for contingency, objective possibility and deviation from type. Closely allied to the action framework, and open to cross-formulation, is the interactionist tradition descended from Georg Simmel and the early Chicago school of pragmatist social science.Footnote 17
The positivist tradition and its handicaps in the military field
Objectivism by far dominates the literature when it comes to what has been termed the “causes of war” issue. While present (if less in evidence) in the subfield of civil-military relations, it fails to figure prominently in studies of military action and structures. This reflects the balance of relative influence among rival research programmes in the disciplines (psychology, political science and sociology) that have contributed most to the treatment of such topics.
Premised on the separation of object and subject, the positivist tradition looks at phenomena from the outside and, in Hempel's famous phrase, “carves nature at its joints.” While this has long seemed unproblematic in the natural sciences, it raises serious issues in the social realm where pre-given, fixed entities are harder to come by. Positivists are thus led to delineate their objects of study by relying on common-sense definitions or formal (legal or administrative) distinctions, and reify them. This is the source of serious problems in a field where substances are elusive and fluid. A few simple examples will suffice to illustrate.
Unsurprisingly perhaps for a phenomenon Clausewitz famously likened to a chameleon, definitions of war are numberless. The issue is further confused by metaphoric usage and the admixture of normative considerations. Historical change has compounded the conundrum: where wars were long considered to be violent inter-state conflicts, 20th century trends have added conflicts between a state and one or more non-state armed groups, and more recently conflicts between collective non-state actors.Footnote 18 Revolutions and wars were often studied separately, with little cross-fertilization despite the central role played by the military dimension in the former. Civil wars, once excluded on the strength of Ancient Greek philosophy's distinction between stasis and polemos, are by far today the majority case, and can no longer be ignored. Another difficulty in discriminating between what is to be counted or not counted as a war is the appreciation of conflict magnitude. All of this raises for researchers engaged in putting together historical data setsFootnote 19 arduous problems which they can only solve by adopting conventional definitions and setting arbitrary battle death thresholds.Footnote 20 The same applies to other substantive areas in the field.
One way of overcoming such difficulties is to tolerate a degree of contradiction. A good example is provided by Gaston Bouthoul's brave attempt, half a century ago, to embrace the “war phenomenon” in functionalist terms.Footnote 21 Having started with a vibrant profession of objectivist faith, he proceeded to outline his methodology and the types of data he would consider. To his reader's surprise, beyond material facts his list includes intentionality, which he declares observable without further comment. Another instance of such inconsistencies is Martin Edmonds’ general study of civil-military relations in the broader (social as well as political) sense.Footnote 22 This author, applying systems theory, chooses to restrict analysis to the regular armed services legally defined, thus differentiating them sharply from reserve components, police forces and irregular armed groups. But this objectivist love of clear boundaries has downsides, that he acknowledges in passing: military forces have internal functions, police forces sometimes hold military roles, hardly anything separates activated reserve from regular components, which brings him to conclude: “It would too arbitrary here to separate totally armed services from these other organizations for purposes of analysis.”
Reification is also responsible, in the same substantive domain, for the fixist (“non dated, non-localized”) view of professional norms offered by Samuel Huntington in his masterful The Soldier and the State, firmly anchored in a structural-functionalist theoretical framework.Footnote 23 More generally, positivists find it difficult to account for, or factor in, endogenous change – a most unfortunate state of affairs in a department of activity where boundaries, identities, aims, structures, cultural ethos and norms are apt to evolve spontaneously as a function of action/interaction processes.
The second major consequence of objectivist methodology resides in the axiomatic conjectures required by analysis to substitute for the role intentionalist theories assign to contextualized meanings. The problem here is that the number of possible theoretical frameworks based on such axiomatics seems limited only by researchers’ imagination.
One (large) category sees war as wholly irrational. Innumerable theories based on that axiom variously find its roots in human nature or evolution (most psychological theories), ecological/biological drives (holistic versions of social darwinism), spiralling public opinion frenzies (nationalism, ethnic hatred), archaic or perverted institutions (Marx, Veblen, Wright Mills), particular cultural patterns, some negative feedback function to relieve demographic pressure (or the social tensions entailed by a male youth bulgeFootnote 24), the contagious quality of armed conflict, the status incongruence of states (Galtung), lateral pressure mechanisms born of rapid industrial growth, etc. Military institutions help the process along due to skewed recruitment and socialization patterns, professional bias, goal displacement, and so forth.
Another (equally large) group of theoretical insights symmetrically pictures war as entirely rational. Economic theories of war, cost-benefit analysis, applications of game theory to armed conflict, theories premised on a general competition for resources and/or status are by-products of that postulate. Special mention should be made here of structural theories. Kenneth Waltz's neorealism, which has become widely influential, regards international relations as characterized by anarchy resulting from the absence of compelling central regulation,Footnote 25 the interstate system as structured by the distribution of power within it, the behaviour of states as governed by survival/security concerns at a minimum (universal domination at a maximum), and the name of the international relations game as balancing to inhibit the rise of a single dominant power.Footnote 26 Wars occur when the balancing becomes ineffective, thus raising the question of stable (parity) and unstable (preponderance) power and alliance configurations. Rivalling Waltzian neorealism are theories which, following Thucydides, see the international arena as hierarchical – as cosmos rather than chaos –, and war as entailed by the challenge of a rising contender to the existing dominant power (thus making parity the necessary condition for major war).Footnote 27
If one adds the consideration that the positivist research programme's nomological ambition and its associated belief in determinism deny contingency a role – except in the form of stochastic disturbance factors in equations –, one begins to see how and why objectivism fails to capture some of the traits that are central to the military field's core issues. Martial behaviour is neither wholly irrational, entirely rational, nor fully determined by external causes. It is at one and the same time rational, irrational and subject to contextual uncertainty – the chance events whose potentially dramatic effects no war historian can afford to ignore, the “friction” and “fog” which Clausewitz saw as powerful inhibitors of predictability where the outcomes of military action are concerned.
Other axioms are concerned with the orderly or disorderly/conflictual nature of society or of the aggregate effects of social behaviour. They also generally result in denying resort to arms its specific, sometimes unique character. Regulation theories, whether psychological or functionalist, optimistically see war as a breach of order that can be remedied. C. H. Cooley, in 1917, contended that nations are socii like any other, and as such can be brought into line through international social control.Footnote 28 This is not far removed from what Durkheim and Mauss (or even, if ties of sentiment are substituted for collective consciousness, Freud) thought on the subject.Footnote 29 That line of reasoning was later resumed by the English school of international relationsFootnote 30 as well as by Karl DeutschFootnote 31 and his disciples. The post-Cold War era has seen those conjectures resurface in light of the powerful globalizing trends which make the one-world hypothesis seem less implausible.
Parsonian structural-functionalism, by assuming equilibrium in social systems, added weight to the idea that war can only result from exogenous change or internal deviance. He went so far as to picture East-West relations in the Cold War as essentially similar to those between Republicans and Democrats on the U.S. political scene: the very fact of ideological polarization implies that there is a common frame of reference, and thus can war be averted. In that view, the sociologist's task is to identify the “mechanisms by which this integrative process can take place.”Footnote 32 More generally, whether derived from Parsons or not, the idea that war results from a momentary disturbance of systemic balance has given rise to numberless variations based on economic, demographic or cultural change, imported modernization, desynchronization of subsystems, conflict among elite groups, etc.
Closely related to such theories is the axiomatic conjecture that war can be subsumed under a more general category, be it conflict,Footnote 33 violence or (today) human security, thus denying war its unique characteristics. Though less frequent and hardly the preserve of positivists, the opposite assumption – war as the normal horizon of politics, and a valid metaphor for all social relationships – is also present.Footnote 34 Typical of that position is the inversion of Clausewitz's Formula: politics is the continuation of war by other means,Footnote 35 a view that cannot account for the military's contemporary peace support missions other than as a front for devious attempts at domination. Though poles apart, those who refuse to see war, real or potential, for what it is anywhere and those who choose to see it everywhere finally all tend to a common conclusion: the rejection of armed conflict and armed forces as a sui generis substantive area.
The problem for positivists is then that, contra regulation thinkers, history is full of intractable conflicts ending in bloodshed, and contra those theorists who postulate wars of all against all, it is also replete with instances where cooperation and at least relative harmony and peace prevail. The task of social science is to think through the presence and the absence of war (or balance in civil-military systems) together, which causal determinism, deductive theories based on rigid axiomatic conjectures and covering laws all too plainly fail to do.
The net result so far when it comes to war studies has been an embarrassing surfeit of internally coherent but mutually contradictory theories, with none producing more than precious few valid and non-trivial empirical generalisations, weak or inconsistent linkages between prediction and fact, or loose statistical fit in quantified models – and no means to arbitrate between them. Kenneth Waltz acknowledged as much in 1979:
Among the depressing features of international-political studies is the small gain in explanatory power that has come from the large amount of work done in recent decades. Nothing seems to accumulate, not even criticism.Footnote 36
Ten years later, a review of the literature on international conflict came to the conclusion that… “if there is one overarching conclusion to be drawn from the foregoing pages it is a bold question mark.”Footnote 37 Strongly suggesting that nothing much has changed recently, the reviewer of a book (co-authored by J. D. Singer, dean of quantified war data analysis) which synthetized research findings from over 500 quantitative studies of war at the level of state, dyad, region and international system,Footnote 38 concurred:
[T]he outcomes of these studies are, for the most part, heterogeneous, ambiguous, not particularly robust, conflicting and sometimes even contradictory. About very few theoretical frameworks and potential polemogenic factors (predictors) some consensus or communis opinio exists.
In light of such paucity of analytical successes, positivism's nomological ambitions create a dilemma: either one sticks to that principle in the hope that over time refining theories will improve the fit between predictions and data, or one focuses on the specifics of contexts, in which case the nomological yields way to an ideographic approach. This is the conclusion arrived at by some students of war as well as by generalists.Footnote 39
While less dramatically conspicuous, the weaknesses of positivist approaches when applied to military-type institutions and civil-military relations are nevertheless real. They generally result in analyses and theories that either see soldiers as mere civilians in uniform or military structures as wholly apart – which, one way or the other, they are not.
The strengths of the Weberian/Simmelian tradition
By contrast, the social action framework seems to have enjoyed a higher degree of achievement in the military field. Taking into account contextualized meanings allows one to dispense with rigid postulates, and to factor in endogenous change as a result of interaction, both between foes and among friends in real or potential armed conflict. Causal comparison of discrepancies between empirical/historical reality and the ideal-typical constructs (typologies or developmental trends) hypothesized to interpret it leaves room for contingency. Such an approach, relying on the structure and definition of situations, enables one to think through behaviour as both rational (political ends, strategic or tactical means) and irrational (sacredness of mission, violent public opinion impulses and reversals, the charismatic elements of leadership in life-and-death situations, the emotional components of loyalty and small group cohesion – to name but a few), peace and conflict, or the internal interplay of institutional and organizational dimensions. It correctly locates what makes martial action unique as well as what does not (see infra). All of which cannot be done when one observes behaviour, structures or process from outside as if they were natural phenomena.
The action framework has no need for legal definitions: collective fighting can be termed “military” whenever the use of potentially deadly force can be traced to political ends calling into question a polity's independence or unity. Hence, revolutions, civil wars and terrorism can be studied (though with specific traits) in the same manner as foreign military conflict – irrespective of casualty numbers. Likewise, soldiers do not need to wear uniforms and draw regular pay from state treasuries: partisans, resistance fighters, militias, are included if their violent action can be shown to be premised on political motives. Conversely, regular soldiers may cease to act militarily if their motives are alien to politics. In other words, the social action research programme makes possible a disjunction, which objectivist approaches cannot even begin to contemplate, between the external attributes of actors and the type of action they conduct. The issue of legitimacy is no longer treated as an independent variable, but seen as an emergent property with possible feedback on morale and political determination.
These benefits are, of course, purchased at the cost of renouncing nomological theories and predictions based on regularities, the former deemed out of reach, the latter too uncertain (political goals are a priori indeterminate), both because real contexts are irreducibly complex: concepts cannot exhaust reality.Footnote 40 But the nominal typologiesFootnote 41 and developmental trendsFootnote 42 produced by the Weberian/Simmelian tradition are nonetheless powerful analytical tools with which to probe historical experience/empirical data.
It is no coincidence that application of this research programme to the military field produces propositions which echo the central tenets of Clausewitz's profound analysis of war. The close proximity of Clausewitz's approach to the methodology developed decades later by Weber has been noted before.Footnote 43 It could be said to be quite as close to Simmel (interaction) and his legacy as cultivated by Chicago pragmatism (reality in a state of flux, pluralist orientations, considerable degree of autonomy among elements whose integration to the whole is problematic). This is where, based on assumptions of interconnectedness and context, chance, complexity, indistinct boundaries, positive feedback and the like, the contribution of nonlinear dynamicsFootnote 44 to the study of given real interaction processes could break new ground and prove invaluable. The same applies to network analysis.Footnote 45
One additional beauty of social action is that far from remaining prisoner of established disciplinary traditions and compartmentalization of the field into areas subject to different analytical treatments, it has transcended disciplinary boundaries and in doing so effectively advanced knowledge in coherent, integrated waysFootnote 46 that sharply contrast with the positivist record of analytical failures.
Rather than go through a list of illustrations of the cognitive successes encountered in the military field by the social action research programme, this writer's best plan at this point is to present a synthesis, restricted to essentials, of what he regards as its main achievements.
Social action and the military field: a brief overview
A. Defining military action
In the beginning were military action (for each side) and interaction (between sides). They have long been equated with war, a polymorph best defined by means of four conceptual dimensions: violence, organization, legitimacy in the service of sovereignty – a formulation which differentiates and recombines the elements of Clausewitz's “trinitarian definition”. In keeping with recent historical experience, mention of states is dispensable: sovereignty is either well-established, or just a group's strong enough aspiration to be deemed worth fighting for with violent means. It can be frontally attacked and seized (or, as the case may be, defended and confirmed), but also to varying extents gnawed at the margins by coercively restricting one or several of its attributes (territory, population, political institutions, freedom of action of those who hold ultimate power). Mere threats against it provide a justification for political/strategic realism (though not necessarily for a hard-line version of it, Schmittian style). War thus defined covers foreign conflicts, revolutions, civil wars, armed rebellions and terrorism. The intensity of war (and the coercive extent of norms derived from it within each camp) depends on meanings in context – definition of the enemy,Footnote 47 perceptions of what is at stake.
As Clausewitz saw, left to its own devices, the pure conflictual logic of war would result in escalation to extremes. A permissive reason for this, as neorealists have seen, is the absence of any agency commanding enough muscle and/or authority to restrain belligerents whenever they wield considerable relative military power. However, the price of total war in blood, treasure and internal political fallout makes it difficult to sustain over the long haul. The majority of historical wars have been limited, either because stakes were not high enough to fight a total war, or sufficient human and material resources could not be mobilized, or both. While escalation processes are an ever-present risk, actually more frequent are the reverse processes of moderation (which initially intrigued Clausewitz, and probably helped him reach his conclusion that war is governed by evolving political conditions and considerations,Footnote 48 though it can escape the control of sovereign leaders and influence politics in return). Finally, military power produces effects short of actual war: it can intimidate (persuade or deter) by altering the calculus of potential gains or losses in political confrontations, and to some extent determines the status of political groupings.
War, however, is not the sole modality of political resort to arms. Armed force has always been the ultimate means of internal political control when constitutional order is seriously challenged. And recent decades have seen the rise of peace support operations. In both cases, the same definition (violence + organization + legitimacy + sovereignty) applies. This raises the question of what separates war from those other uses of force. Part of the answer lies in the differential status of moderation: a frequent but by no means assured process in war, it is present in peace support and internal political control from the beginning as a matter of course. The other part relates to the existence in such situations of overwhelming military superiority on the side of the State or the mandated coalition, capable of restraining protesters or belligerents, and of a basis of political legitimacy that is strong enough in the eyes of a wider majority to transcend the sectional or local interests at stake. The use of force is therefore less deadly, although there again the risk exists of an escalation on both sides. State military institutions have learnt over time to differentiate between those various logics of action and the amounts of force required in each case.Footnote 49 Professionals in the restrained use of force, sometimes referred to as “paramilitary”, have emerged on internal scenes, and soldiers earmarked for peace support operations now frequently train with police forces prior to commitment. Such minimum resort to physical coercion admits of degrees, depending on the level of threats to legitimate order: “muscular” enforcement of peace or internal public order when they are out of control, peacekeeping or maintien de l'ordre when they prove manageable. Military action thus comes in several varieties that can be logically classified by crossing simple dichotomies, as in the following table:
Table i A typology of military action
Such distinctions, the logics of moderation or possible escalation, and each cell's internal differentiations of degree, account for the division of labour found among military-type institutions (armed services, paramilitary, police forces). However, once projected onto the single axis of the rational amounts of force to be used in various situations, they tend to become blurred: there is no natural asperity on the continuum leading from maximum violence to maximum negotiation that could serve as a clear boundary between two or more of the above ideal-types. This is, in all cases, because situations, and for each side the initial stakes that characterize them, are apt to evolve over even short periods of time.Footnote 50 The resurgent globalization, force asymmetries and terrorism of recent years have brought external and internal security much closer than was earlier the case, and added to the blurring. The division of labour is thus increasingly nominal, and role flexibility has become the rule.
The conceptual unity of military action is therefore more strongly affirmed than ever. This holds despite the extreme diversification of military-type units into air, sea and land components, tactical, operational and strategic levels, and a mosaic of specialized functions within them. The motley quality of uniforms and unit traditions – carriers of identity and cohesion – that add to the apparent diversity should not detract attention from what all military subcomponents and actors have in common: sense of mission and flag loyalty.
B. Unique and non-unique characteristics
The unique traits of military action and actors all derive directly or indirectly from resort to arms in the cause of sovereignty when the order goes out. Military distinctiveness is both sociopolitical and functional.
Its sociopolitical aspects of relate to the fact that those to whom defence and promotion of a polity's independence, interests and values are entrusted do not constitute an occupational/professional group just like any other. The human and material resources placed at their disposal, society's efforts to satisfy their moral and social needs, the relative prestige of their uniforms and leaders, exert varying degrees of influence on the polity. While such influence is licit and welcome within limits, the possibility exists that such limits be exceeded. In political terms, its access to arms may give the military more weight than is warranted: some of its members may transform it from the instrument of politics that it ought to be in the nature of things into an autonomous political actor. A worst-case scenario sees it seize power, or act as a supraconstitutional arbiter of internal politics. But there are more subtle ways of exerting undue influence on the sovereign, economy, society or culture, and grey areas where proper balance raises difficult problems. In other words, the issue of civilian control of armed forces so that they do not disfigure the sociopolitical regimeFootnote 51 they are supposed to defend is of more serious concern than with most other groups or lobbies.
However, that issue has been mitigated since the late 19th century by another concern: due to the rising complexity of military art, political leaders have increasingly come to rely on military leaders’ professional expertise. This implies trust between the two groups – never permanently achieved, even when military groups and elites are meaningfully integrated into the parent society, and responsive to its needs, interests and values. It also implies that beside traditional military subordination, relations be oriented to political-military coordination of action as well. Professionalism presupposes a fair margin of autonomy, which is likely to obtain only if political leaders set the objectives of action, secure the resources, indicate the criteria by which the outcomes of action will evaluated, and leave the rest to professionals.Footnote 52 The full panoply of forms that control may assume can be represented as follows:Footnote 53
Table ii Forms of civilian control
In functional terms, those who are charged with the task of applying legitimate organized violence, and in doing so put their own lives or physical and moral integrity at risk, are subject to powerful norms destined to warrant their individual motivation to fight, their collective effectiveness, and the trust placed in them: honour and loyalty to the sovereign are their expected cardinal virtues. Backed by the prospect of drastic sanctions in case of desertion or dereliction of duty, motivation relates to one ultimate meaning: salus populi est suprema lex. It is reinforced by charismatic leadership, the support and recognition of peers, superiors as well as society at large, and the presumptive legitimacy of orders received (which, absent any personal trespasses against the law or basic dictates of conscience, exonerates combatants from personal responsibility – but does not annihilate ethical concerns). Primary cohesion, the emotional and practical support it provides, based on a sense of common destiny, are of paramount importance in the face of danger and the prospect of killing or maiming other humans. In structural terms, the immediacy of danger, the need for swift responses to unforeseen events and economy of force principles translate into unity of command – a degree of centralization seldom matched. The hierarchical relations that this implies are not simply the consequence of bureaucracy: they institute personalized command relationships based on mutual recognition (of which the exchange of military salutes is the symbol). The primacy of group and institution over the individual expresses itself in open-ended availability for service and discipline – a mechanization of behaviour, reinforced by drill and practice, so as to make it predictable, thus calculable. Such traits are all the more pronounced as combat and danger are close in time or distance; they are less in evidence in peace support and action in aid of internal political order (from which the heroic dimension is usually absent) – though military socialization insists on worst-case scenarios, and combat norms are etched in all military minds as the ultimate referent. Another central characteristic of military action is the degree of contingency, hence of uncertainty, liable to affect it. Sheer accident and friction are obvious forms. But the main source of uncertainty stems from the fact that action bears on an opposing party similarly motivated, and intent on achieving surprise – not on inert matter. Hence the realization that absolute control over events is unlikely in hostile interaction, and the frequent reference to providence or fateFootnote 54 despite painstakingly meticulous preparation for the treatment of unexpected adversity so typical of military settings. Finally, as with any institutional activity (but probably more so in military action than elsewhere), effectiveness can only be measured synthetically on the basis of qualitative criteria: the opposing side has renounced its goals, or it has not.
Such characteristics, made necessary or unavoidable by hostile interaction, conspire to deny military action the status of pure engineering: to limit the place occupied in it by instrumental rationality. They could be termed irrational if some of them (not least cohesion and recognition) were not precious functional assets, at least in part manageable by those in authority. They reflect a traditional, communitarian/hierarchical type of social relations that is far removed from the structures and ideals of modernity.Footnote 55
The institutional dimension, however, is only half the story. It is counterbalanced by a second functional dimension, oriented to cooperation with friendly parties internal or external to the military, in which instrumental rationality enjoys pride of place. It resides in aspects of action relating to planning, command and control, coordination and support. Its prevalence is proportional to the distance or the amount of protection from hostile forces. Contrary to combat situations, where fear of casualties among those filling crucial specialized roles implies a degree of skill versatility, contexts where organization dominates are marked by a division of labour limited only by the numbers available. It draws on all the resources of modernity – notably technology, management and communication – and thus constitutes a powerful agent of rationalization. Everything in that logic of action can be concerted among specialized actors and units, subject only to optimisation under capacitary or coordination constraints (and the kinds of dysfunction that are known to develop in interstitial spaces where informal relations and repertoires get in the way of rational organization). Its degree of success is ascertained in terms of efficiency rather than effectiveness, and analytically measured using quantified technical, management or even (as in morale surveys) sociopolitical criteria. In pure principle, this organizational dimension is devoid of any unique content.
Yet, precisely, in military action and structures, neither of the institutional and organizational dimensions can fully express itself at the expense of the other. They are always found together in varying doses, locked in an endless dialogical relationship in which one is restrained by the inescapable influence of the other, and whose local/temporary arbiter is the logic of the situation or problem in hand. The locus of either dimension cannot be located with any precision in formal charts: even combatants in the thick of combat have to think of providing for rest, food, ammunition, fuel, medical evacuation of casualties, and their own reputation; even service support units must provide for their self-defence in case of surprise attack. In other words, resulting from a contrast between institution and organization that is stronger than in other departments of activity,Footnote 56 military uniqueness is essentially relative and fluctuating on the basis of contextualized meanings.
C. Developmental trends
The timeless perspective briefly sketched above is affected by powerful historical trends. One is the 20th century rise of technology; the other relates to long-term normative change.
Following closely on the emergence of mass citizen armies and bureaucratic military structures, the harnessing of industrial power to military action from the late 19th century onwards profoundly altered the face of war and military systems. Morris Janowitz masterfully depicted the social consequences of such trends at mid-(20th) century: quantitative marginalization of fighters, deformation of hierarchical structures from pyramidal to double diamond-shaped, counter-balancing of traditional top-down control by lateral coordination of action, liberalization of discipline, the rise of managers and pragmatism among military elites, etc.Footnote 57 After 1960, new waves of technological change pushed rationalization to new heights, allowing the organizational dimension to grow further, and altering traditional martial identities. More recently, instant communications, computerization, the Internet and like innovations have turned martial structures and action processes into something approximating neural networks and their functioning, with the result that hierarchical channels are often bypassed, short-circuits occur between lower-echelon field units and top leadership, and the three classical levels (tactical, operational, strategic) of action become compressed by the overriding need to keep up with and respond to up-to-the minute developments and media reporting.
But the most dramatic effects of technological change came with the emergence of strategic weapons – first, long-range heavy artillery and air power (which considerably increased the numbers of civilian casualties in the two world wars), then nuclear arms (which promised even more). The latter were the source of a hitherto unheard-of paradox: the neutralization of overwhelming power through mutual deterrence.
With the advent of mass destruction weapons, it was soon recognised after 1945 that the function of armed forces was no longer to win total wars pitting coalitions of polities against one another, but to avert them. This was an unanticipated consequence of purposive action since superpowers now leading the new coalitions of East and West had acquired such weapons in order to obtain a decisive strategic advantage. But the prospect of large-scale destruction of civilized life drove expectancy of strategic gain below zero and effectively inhibited major war: while not altogether impossible through accidents, defective perceptions of enemy intentions, or irrationality on the part of political leaders, it was increasingly seen as unlikely as time went by. The consequence was that while one could not dispense with war readiness and huge arsenals, armed forces were on stand-by and mostly remained so to the Cold War's end. Limited wars, of course, were still part of the landscape at the periphery of areas of strategic interest to both coalitions, but for fear of escalation to extremes they had to be restricted in terms of goals, resources and – ideally – duration (as America learned in Vietnam, the Soviet Union in Afghanistan). After the mechanics of nuclear deterrence became familiar to strategists, only a small fraction of service members took part in them. Due to this change affecting their roles, armed forces turned into what Janowitz termed “constabulary forces”.
The post-Cold War era has seen considerable change. Nuclear deterrence was placed on the back burner, invasion defence plans were shelved. Collective security could resume under the aegis of the UN, subject only to international consensus. The constabulary concept not only survived (it was one of very few characteristics of the late Cold War to do so), but actually prospered: action, which for decades had been so scarce, suddenly became plentiful, from coercion of international deviants (Iraq, Serbia) to a long list of peace support operations around the globe. Technology, promising “digital battlefield awareness” in real time and a “revolution of military affairs”, seemed set to take care of more serious (but unlikely) contingencies, and to guarantee status quo powers overwhelming conventional superiority. After hardly more than a decade, 9/11 again changed the equation (of which more anon).
To assess the place of normative change with regard to military action, one has to go back to the turn of the 20th century. At that time, a few leading personalities anticipated that industrialized warfare between coalitions of large nation-states – a distinct possibility since the emergence in the middle of Europe (then the dominant continent) of Germany as a major power, and the humiliation it had inflicted on Denmark, Austria and France three decades earlier – would bring brutalization and mass misery, and had tried to restrain war through international conventions (The Hague 1899, 1907). Their efforts could not avert the loss of collective control over events in the calamitous month of July 1914 which sealed the 20th century's fate. When the Great War and its horrors ended, the creation of a League of Nations belatedly inspired by Kant's essay on perpetual peace, the Locarno Agreement of 1925 and the outlawing of war as a normal instrument of policy (in the so-called Briand-Kellog Pact of 1928) resumed a fledgling tradition which proved no match for the ideological passions and tensions of the 1930s. But it benefited greatly from the general revulsion against armed conflict after 1945, and the advent of nuclear weapons made war's legitimacy highly conditional as far too dangerous for civilization itself. The UN charter codified that widespread sentiment: interstate disputes were now to be settled non-violently, and in the face of threats against world security, the “international community” would fight as a team in what an American president named a “police action”. Though the Cold War quickly dashed those hopes, the Zeitgeist could no longer be the same.
As a result, the charisma of the nation-state, so strong between 1870 and 1945, came to an abrupt end in the West. Nationalism was retrospectively held responsible for past world wars; even sovereignty and patriotism became suspect. Colonial wars and the U.S. intervention in Vietnam went through a natural history of indifference followed by unpopularity. This drove the prestige of the military down, at a time when new-found economic affluence, the personal security afforded by the welfare state, and more critical outlooks generated by rising average educational levels, gradually fostered individualism among the young. By the mid-sixties, serving under arms as a conscript was no longer regarded as an honour by increasing fractions of successive age-cohorts. Worse still, armed forces came to be seen symbolically as the epitome of everything youths now rejected – strict norms, primacy of society over the individual, duty to the state, etc. –, seeking as they were less conformist, more laid-back mores than their parents had consented to. Conscription became more problematic as nuclear weapons and high technology lessened the need for mass armed forces: fewer young people were inducted in each age-cohort, creating inequities that the system was often unable to justify. In countries where it had shallow historical roots, it was abolished in the 1960s (United Kingdom) and 1970s (United States); it persisted elsewhere as best it could through the Cold War's last decades. But citizen service under arms did not long survive its demise: all-volunteer formats are now the norm the world over.
D. The contemporary scene
The post-9/11 context has seen the world's premier military power, whose defence budget exceeds 40 % of all military expenditures around the planet declare war on a shadowy group of latter-day barbarians combining fanaticism with high technology, whose exact followership, degree of organization and even real identity are unknown. Combined with neo-conservative ideology – a strange mix of the Hamiltonian, Jacksonian and Wilsonian traditions in U.S. foreign policy – and its declared ambition to turn military power into political clout to reshape and influence the Greater Middle East, the “War on Terror” led to the twin land interventions in Afghanistan and Iraq. The problem, after five or six years of fighting and nation- or state-building, is that outcomes have hardly matched expectations. The conditions that prevail as a result on the contemporary international scene are extraordinarily ambiguous.
Globalization, the interdependence it generates, and the emergence of new economic heavyweights (China, India, Russia) dependent on external markets, produce a mix of cooperation, competition and tension. This tends to lessen the chances of serious conflict among major powers. Should serious military tensions (say, between China or Russia and the U.S.) nonetheless come to dominate, nuclear weapons would in all probability usher in a new situation of mutual deterrence. Confrontations would thus more likely approximate non-zero-sum, mixed motive games of the Cold War type than to lead to major conflict. This is made all the more plausible as the West will see its relative demographic and economic weight decline as the century wears on. It will thus feel little incentive to antagonize the fast emerging half of the planet at the risk of reactivating resentment for five centuries of Western domination, and turning Huntington's “West against the Rest” into a self-fulfilling prophecy.
At the same time, there is no shortage of religious or ethnic tensions, myths, armed prophets and fanaticized disciples, uncontrolled social and cultural trends, or resentments that are bound to erupt into violent trouble in a number of areas.Footnote 58 There will be more ample opportunity for consensual collective security initiatives to deal with them than the so-called international community can readily absorb. Where no such consensus exists, local flashpoints may induce wider confrontations (as is the case with Georgia as of this writing). Much will turn on events in Pakistan. And what remains uncertain for the longer-term future are the attitudes – bourgeois pacifism, power politics, or a mix of the two? – of China, India, Russia and others when they eventually substitute internal growth engines for dependence on Western markets.
The potential for destabilization thus exists, but the odds for international stability look no less favourable, if not better. All in all, despite inherent difficulties and high costs, terrorism has been remarkably contained.Footnote 59 For the time being, contrary to illusions created by media reporting, current trends in armed conflict and casualty numbers are resolutely pointing downward.Footnote 60 And the great powers to date show no taste for major war.
The influence of military factors goes in the same direction. Contrary to expectations of a decade or so ago, the use of military force has not turned into a fully controlled, rational and civilized exercise occasioning little collateral damage and rare casualties. Digital battlespace awareness is of little help whenever the enemy avoids military encounters of the conventional type, and hides among the population. The memory of what happened to Saddam Hussein and Slobodan Milosevic is enough to ensure that no revisionist states or groups in their right minds will imitate them again. What actually has taken place since 9/11, is a true “counter-revolution in military affairs”.Footnote 61 In the face of overwhelming military superiority, the predictable response of weak but determined antagonists has been to resort to terrorism and fine-tune the techniques used in the 1950s and 1960s in Algeria and Vietnam, adjusting them to the means afforded by accessible high technology to take advantage of new propaganda effects. The weak can now strike (and have done so on a handful of occasions after 9/11) the territory and population of the strong; they use their own population as human shields to paralyze repressive action, banking on the strong's moral dilemmas, or on the risk of public diplomacy disasters when innocents are killed. Success or failure in that asymmetric type of warfare does not revolve around occupation and control of a territory after defeating regular forces, but around both long-term allegiances of the local population, and the patience – tried by casualties, cost, loss of face and frustration – of public opinions back home.Footnote 62 Set operational modes such as “search and destroy”, and the quest for maximum force protection at the expense of meaningful rapport with local populations are ill-adapted to the political goal of winning their hearts and minds, and may account for the distinct failure to transform overwhelming military superiority into political effectiveness. The functionality of conventional military force in such circumstances is in doubt. If such trends are not reversed soon in both Iraq and Afghanistan, the chances are that the U.S., having learned its lesson the hard way, will prudently return to prior practice, i.e., containment plus indirect influence or pressure, and – when needed – interventions with limited objectives, resources and duration.
Another important factor for the future is the mechanism known as “structural disarmament”. It first made itself felt in the provision of major items of military equipment.Footnote 63 As research and development costs rise faster than defence budgets, each new generation of aircraft, tanks or naval vessels sees numbers of them decrease. The price tag of each is so high that their tactical use is inhibited for fear of losing them in combat against targets of much less value. The same mechanism now applies to human resources in all-volunteer forces that are hard and expensive to recruit and train. The issue as far as they are concerned is further complicated by avoidance of friendly casualties, due to public compassion and the political costs they entail in situations in which stakes are seen as low. Persistent difficulties in the recruitment of volunteersFootnote 64 limits the pool of deployable troops, making force regeneration in the field problematic. They invite significant increases in the compensation and insurance packages of military personnel, as well as substitution of technology for combatants, driving up direct and indirect costs to unforeseen heights that economists are beginning to judge ruinous even for the most affluent nation.Footnote 65 The long-term prospect is thus of the strong pricing themselves out of battlefields (or recruiting their soldiers in poorer nations) unless they perceive major threats against central attributes of their sovereignty, and in the face of such danger restore conscription – hardly the most probable outcome of the ambiguous situation described above.
It is therefore premature to think in terms of a return to the state of warFootnote 66 – barring accidents…
Concluding remarks
This article has tried to assess the relative cognitive merits of various research programmes in the treatment of military issues. It has contended that, on such substantive topics probably more than on many others, method strongly influences conceptualization and modes of theory construction. Having noted the infrequent showings of some programmes in the military field, it went on to demonstrate, based on the literature produced on military topics over a century, that the blames classically laid at positivism's door are especially justified in it. An objectivist outlook leads to definitions which miss central characteristics of the subject-matter and, due to a diversity of axiomatic conjectures, generates a cacophony of theories that only imperfectly fit the data defined and selected on that basis. The Weberian/Simmelian tradition, introduced to the field from the 1930s onwards, has shown more promise. Its methodology enables analysis to avoid definitional distortion, to identify the substantive traits which parallel or diverge from those found in other types of activity, and to embrace the various compartments of the military field – armed conflict, military structures, civil-military relations – in coherent, transdisciplinary fashion. Its achievements, detailed in a brief synthetic formulation of findings, suggest that it alone fully accords with Clausewitzian thought, and shows sufficient flexibility to account for recent changes and future prospects which so baffle other research programmes, as well as to open avenues of collaboration with promising new approaches (notably, nonlinear dynamics).
The late Jean-Michel Berthelot was rightFootnote 67: some programmes are better adapted than others to specific fields of study, and show their worth in doing justice to their subject-matter. Unless this writer is very much mistaken, such is the case of the social action framework in the military field.
