Front Page Titles (by Subject) CHAPTER XII.: military systems. - Political Institutions, being Part V of the Principles of Sociology
The Online Library of Liberty
A project of Liberty Fund, Inc.
CHAPTER XII.: military systems. - Herbert Spencer, Political Institutions, being Part V of the Principles of Sociology 
Political Institutions, being Part V of the Principles of Sociology (The Concluding Portion of Vol. II) (London: Williams and Norgate, 1882).
About Liberty Fund:
Liberty Fund, Inc. is a private, educational foundation established to encourage the study of the ideal of a society of free and responsible individuals.
The text is in the public domain.
Fair use statement:
This material is put online to further the educational goals of Liberty Fund, Inc. Unless otherwise stated in the Copyright Information section above, this material may be used freely for educational and academic purposes. It may not be used in any way for profit.
§ 515. Indirectly, much has already been said concerning the subject now to be dealt with. Originally identical as is the political organization with the military organization, it has been impossible to treat of the first without touching on the second. After exhibiting the facts under one aspect we have here to exhibit another aspect of them; and at the same time to bring into view classes of related facts thus far unobserved. But, first, let us dwell a moment on the alleged original identity.
In rude societies all adult males are warriors; and, consequently, the army is the mobilized community, and the community is the army at rest, as was remarked in § 259.
With this general truth we may join the general truth that the primitive military gathering is also the primitive political gathering. Alike in savage tribes and in communities like those of our rude ancestors, the assemblies which are summoned for purposes of defence and offence, are the assemblies in which public questions at large are decided.
Next stands the fact, so often named, that in the normal course of social evolution, the military head grows into the political head. This double character of leading warrior and civil ruler, early arising, ordinarily continues through long stages; and where, as not unfrequently happens, military headship becomes in a measure separated from political headship, continued warfare is apt to cause a re-identification of them.
As societies become compounded and re-compounded, coincidence of military authority with political authority is shown in detail as well as in general—in the parts as in the whole. The minor war-chiefs are also minor civil rulers in their several localities; and the commanding of their respective groups of soldiers in the field, is of like nature with the governing of their respective groups of dependents at home.
Once more, there is the general fact that the economic organizations of primitive communities, coincide with their military organizations. In savage tribes war and hunting are carried on by the same men; while their wives (and their slaves where they have any) do the drudgery of domestic life. And, similarly, in rude societies that have become settled, the military unit and the economic unit are the same. The soldier is also the landowner.
Such, then, being the primitive identity of the political organization with military organization, we have in this chapter to note the ways in which the two differentiate.
§ 516. We may most conveniently initiate the inquiry by observing the change which, during social evolution, takes place in the incidence of military obligations; and by recognizing the accompanying separation of the fighting body from the rest of the community.
Though there are some tribes in which military service (for aggressive war at any rate) is not compulsory, as the Comanches, Dakotas, Chippewas, whose war-chiefs go about enlisting volunteers for their expeditions; yet habitually where political subordination is established, every man not privately possessed as a chattel is bound to fight when called on. There have been, and are, some societies of considerably-advanced structures in which this state of things continues. In ancient Peru the common men were all either actually in the army or formed a reserve occupied in labour; and in modern Siam the people “are all soldiers, and owe six months’ service yearly to their prince.” But, usually, social progress is accompanied by a narrowed incidence of military obligation.
When the enslavement of captives is followed by the rearing of their children as slaves, as well as by the consigning of criminals and debtors to slavery—when, as in some cases, there is joined with the slave-class a serf-class composed of subjugated people not detached from their homes; the community becomes divided into two parts, on one of which only does military duty fall. Whereas, in previous stages, the division of the whole society had been into men as fighters and women as workers, the division of workers now begins to include men; and these continue to form an increasing part of the total male population. Though we are told that in Ashantee (where everyone is in fact owned by the king) the slave-population “principally constitutes the military force,” and that in Rabbah (among the Fúlahs) the army is composed of slaves liberated “on consideration of their taking up arms;” yet, generally, those in bondage are not liable to military service: the causes being partly distrust of them (as was shown among the Spartans when forced to employ the helots) partly contempt for them as defeated men or the offspring of defeated men, and partly a desire to devolve on others, labours at once necessary and repugnant. Causes aside, however, the evidence proves that the army at this early stage usually coincides with the body of freemen; who are also the body of landowners. This, as before shown in § 458, was the case in Egypt, Greece, Rome, and Germany. How natural is this incidence of military obligation, we see in the facts that in ancient Japan and mediæval India, there were systems of military tenure like that of the middle ages in Europe; and that a kindred connexion had arisen even in societies like those of Tahiti and Samoa.
Extent of estate being a measure of its owner’s ability to bear burdens, there grows up a connexion between the amount of land held and the amount of military aid to be rendered. Thus in Greece under Solon, those whose properties yielded less than a certain revenue were exempt from duty as soldiers, save in emergencies. In Rome, with a view to better adjustment of the relation between means and requirements, there was a periodic “revision of the register of landed property, which was at the same time the levy-roll.” Throughout the middle ages this principle was acted upon by proportioning the numbers of warriors demanded to the sizes of the fiefs; and again, afterwards, by requiring from parishes their respective contingents.
A dissociation of military duty from land-ownership begins when land ceases to be the only source of wealth. The growth of a class of free workers, accumulating property by trade, is followed by the imposing on them, also, of obligations to fight or to provide fighters. Though, as apparently in the cases of Greece and Rome, the possessions in virtue of which citizens of this order at first become liable, are lands in which they have invested; yet, at later stages, they become liable as possessors of other property. Such, at least, is the interpretation we may give to the practice of making industrial populations furnish their specified numbers of warriors; whether, as during the Roman conquests, it took the shape of requiring “rich and populous” towns to maintain cohorts of infantry or divisions of cavalry, or whether, as with chartered towns in mediæval days, there was a contract with the king as suzerain, to supply him with stated numbers of men duly armed.
Later on, the same cause initiates a further change. As fast as industry increases the relative quantity of transferable property, it becomes more easy to compound for service in war; either by providing a deputy or by paying to the ruler a sum which enables him to provide one. Originally the penalty for non-fulfilment of military obligation was loss of lands; then a heavy fine, which, once accepted, it became more frequently the custom to bear; then an habitual compounding for the special services demanded; then a levying of dues, such as those called scutages, in place of special compositions. Evidently, industrial growth made this change possible; both by increasing the population from which the required numbers of substitutes could be obtained, and by producing the needful floating capital.
So that whereas in savage and semi-civilized communities of warlike kinds, the incidence of military obligation is such that each free man has to serve personally, and also to provide his own arms and provisions; the progress from this state in which industry does but occupy the intervals between wars, to a state in which war does but occasionally break the habitual industry, brings an increasing dissociation of military obligation from free citizenship: military obligation at the same time tending to become a pecuniary burden levied in proportion to property of whatever kind. Though where there is a conscription, personal service is theoretically due from each on whom the lot falls, yet the ability to buy a substitute brings the obligation back to a pecuniary one. And though we have an instance in our own day of universal military obligation not thus to be compounded for, we see that it is part of a reversion to the condition of predominant militancy.
§ 517. An aspect of this change not yet noted, is the simultaneous decrease in the ratio which the fighting part of the community bears to the rest. With the transition from nomadic habits to settled habits, there begins an economic resistance to militant action, which increases as industrial life develops, and diminishes the relative size of the military body.
Though in tribes of hunters the men are as ready for war at one time as at another, yet in agricultural societies there obviously exists an impediment to unceasing warfare. In the exceptional case of the Spartans, the carrying on of rural industry was not allowed to prevent daily occupation of all freemen in warlike exercises; but, speaking generally, the sowing and reaping of crops hinder the gathering together of freemen for offensive or defensive purposes. Hence in course of time come decreased calls on them. The ancient Suevi divided themselves so as alternately to share warduties and farm-work: each season the active warriors returned to till the land, while their places were “supplied by the husbandmen of the previous year.” Alfred established in England a kindred alternation between military service and cultivation of the soil. In feudal times, again, the same tendency was shown by restrictions on the duration and amount of the armed aid which a feudal tenant and his retainers had to give—now for sixty, for forty, for twenty days, down even to four; now alone, and again with specified numbers of followers; here without limit of distance, and there within the bounds of a county. Doubtless, insubordination often caused resistances to service, and consequent limitations of this kind. But manifestly, absorption of the energies in industry, directly and indirectly antagonized militant action; with the result that separation of the fighting body from the general body of citizens was accompanied by a decrease in its relative mass.
There are two cooperating causes for this decrease of its relative mass, which are of much significance. One is the increasing costliness of the soldier, and of war appliances, which goes along with that social progress made possible by industrial growth. In the savage state each warrior provides his own weapons; and, on war-excursions, depends on himself for sustenance. At a higher stage this ceases to be the case. When chariots of war, and armour, and siege-implements come to be used, there are presupposed sundry specialized and skilled artizan-classes; implying a higher ratio of the industrial part of the community to the militant part. And when, later on, there are introduced fire-arms, artillery, ironclads, torpedoes, and the like, we see that there must co-exist a large and highly-organized body of producers and distributors; alike to furnish the required powers and bear the entailed cost. That is to say, the war-machinery, both living and dead, cannot be raised in efficiency without lowering the ratio it bears to those sustaining structures which give it efficiency.
The other cooperating cause which simultaneously comes into play, is directly due to the compounding and re-compounding of societies. The larger nations become, and the greater the distances over which their military actions range, the more expensive do those actions grow. It is with an army as with a limb, the effort put forth is costly in proportion to the remoteness of the acting parts from the base of operations. Though it is true that a body of victorious invaders may raise some, or the whole, of its supplies from the conquered society, yet before it has effected conquest it cannot do this, but is dependent for maintenance on its own society, of which it then forms an integral part: where it ceases to form an integral part and wanders far away, living on spoils, like Tatar hordes in past ages, we are no longer dealing with social organization and its laws, but with social destruction. Limiting ourselves to societies which, permanently localized, preserve their individualities, it is clear that the larger the integrations formed, the greater is the social strain consequent on the distances at which fighting has to be done; and the greater the amount of industrial population required to bear the strain. Doubtless, improved means of communication may all at once alter the ratio; but this does not conflict with the proposition when qualified by saying—other things equal.
In three ways, therefore, does settled life, and the development of civilization, so increase the economic resistance to militant action, as to cause decrease of the ratio borne by the militant part to the non-militant part.
§ 518. With those changes in the incidence of military obligation which tend to separate the body of soldiers from the body of workers, and with those other changes which tend to diminish its relative size, there go changes which tend to differentiate it in a further way. The first of these to be noted is the parting of military headship from political headship.
We have seen that the commencement of social organization is the growth of the leading warrior into the civil governor. To illustrative facts before named may be added the fact that an old English ruler, as instance Hengist, was originally called “Here-toga” —literally army-leader; and the office developed into that of king only after settlement in Britain. But with establishment of hereditary succession to political headship, there comes into play an influence which tends to make the chief of the State distinct from the chief of the army. That antagonism between the principle of inheritance and the principle of efficiency, everywhere at work, has from the beginning been conspicuous in this relation, because of the imperative need for efficient generalship. Often, as shown in § 473, there is an endeavour to unite the two qualifications; as, for example, in ancient Mexico, where the king, before being crowned, had to fill successfully the position of commander-in-chief. But from quite early stages we find that where hereditary succession has been established, and there does not happen to be inheritance of military capacity along with political supremacy, it is common for headship of the warriors to become a separate post filled by election. Says Waitz, “among the Guaranis the chieftainship generally goes from father to first-born son. The leader in war is, however, elected.” In Ancient Nicaragua “the war-chief was elected by the warriors to lead them, on account of his ability and bravery in battle; but the civil or hereditary chief often accompanies the army.” Of the New Zealanders we read that “hereditary chiefs were generally the leaders,” but not always: others being chosen on account of bravery. And among the Sakarran Dyaks there is a war chief, in addition to the ordinary chief. In the case of the Bedouins the original motive has been defeated in a curious way.
“During a campaign in actual warfare, the authority of the sheikh of the tribe is completely set aside, and the soldiers are wholly under the command of the agyd. … The office of agyd is hereditary in a certain family, from father to son; and the Arabs submit to the commands of an agyd, whom they know to be deficient both in bravery and judgment, rather than yield to the orders of their sheikh during the actual expedition; for they say that expeditions headed by the sheikh, are always unsuccessful.”
It should be added that in some cases we see coming into play further motives. Forster tells us that in Tahiti the king sometimes resigns the post of commander-in-chief of the fighting force, to one of his chiefs: conscious either of his own unfitness or desirous of avoiding danger. And then in some cases the anxiety of subjects to escape the evils following loss of the political head, leads to this separation; as when, among the Hebrews, “the men of David sware unto him, saying, Thou shalt go no more out with us to battle, that thou quench not the light of Israel;” or as when, in France in 923, the king was besought by the ecclesiastics and nobles who surrounded him, to take no part in the impending fight.
At the same time the ruler, conscious that military command gives great power to its holder, frequently appoints as army-leader his son or other near relative: thus trying to prevent the usurpation so apt to occur (as, to add another instance, it occurred among the Hebrews, whose throne was several times seized by captains of the host). The Iliad shows that it was usual for a Greek king to delegate to his heir the duty of commanding his troops. In Merovingian times kings’ sons frequently led their fathers’ armies; and of the Carolingians we read that while the king commanded the main levy, “over other armies his sons were placed, and to them the business of commanding was afterwards increasingly transferred.” It was thus in ancient Japan. When the emperor did not himself command his troops, “this charge was only committed to members of the Imperial house,” and “the power thus remained with the sovereign.” In ancient Peru there was a like alternative. “The army was put under the direction of some experienced chief of the royal blood, or, more frequently, headed by the Ynca in person.”
The widening civil functions of the political head, obviously prompt this delegation of military functions. But while the discharge of both becomes increasingly difficult as the nation enlarges; and while the attempt to discharge both is dangerous; there is also danger in doing either by deputy. At the same time that there is risk in giving supreme command of a distant army to a general, there is also risk in going with the army and leaving the government in the hands of a vicegerent; and the catastrophes from the one or the other cause, which, spite of precautions, have taken place, show us alike that there is, during social evolution, an inevitable tendency to the differentiation of the military headship from the political headship, but that this differentiation can become permanent only under certain conditions.
The general fact would appear to be that while militant activity is great, and the whole society has the organization appropriate to it, the state of equilibrium is one in which the political head continues to be also the militant head; that in proportion as there grows up, along with industrial life, a civil administration distinguishable from the military administration, the political head tends to become increasingly civil in his functions, and to delegate, now occasionally, now generally, his militant functions; that if there is a return to great militant activity, with consequent reversion to militant structure, there is liable to occur a re-establishment of the primitive type of headship, by usurpation on the part of the successful general—either practical usurpation, where the king is too sacred to be displaced, or complete usurpation where he is not too sacred; but that where, along with decreasing militancy, there goes increasing civil life and administration, headship of the army becomes permanently differentiated from political headship, and subordinated to it.
§ 519. While, in the course of social evolution, there has been going on this separation of the fighting body from the community at large, this diminution in its relative mass, and this establishment of a distinct headship to it, there has been going on an internal organization of it.
The fighting body is at first wholly without structure. Among savages a battle is a number of single combats: the chief, if there is one, being but the warrior of most mark, who fights like the rest. Through long stages this disunited action continues. The Iliad tells of little more than the personal encounters of heroes, which were doubtless multiplied in detail by their unmentioned followers; and after the decay of that higher military organization which accompanied Greek and Roman civilization, this chaotic kind of fighting recurred throughout mediæval Europe. During the early feudal period everything turned on the prowess of individuals. War, says Gautier, consisted of “bloody duels;” and even much later the idea of personal action dominated over that of combined action. But along with political progress, the subjection of individuals to their chief is increasingly shown by fulfilling his commands in battle. Action in the field becomes in a higher degree concerted, by the absorption of their wills in his will.
A like change presently shows itself on a larger scale. While the members of each component group have their actions more and more combined, the groups themselves, of which an army is composed, pass from disunited action to united action. When small societies are compounded into a larger one, their joint body of warriors at first consists of the tribal clusters and family-clusters assembled together, but retaining their respective individualities. The head of each Hottentot krall, “has the command, under the chief of his nation, of the troops furnished out by his kraal.” Similarly, the Malagasy “kept their own respective clans, and every clan had its own leader.” Among the Chibchas, “each cazique and tribe came with different signs on their tents, fitted out with the mantles by which they distinguished themselves from each other.” A kindred arrangement existed in early Roman times: the city-army was “distributed into tribes, curiæ, and families.” It was so, too, with the Germanic peoples, who, in the field, “arranged themselves, when not otherwise tied, in families and affinities;” or, as is said by Kemble of our ancestors in old English times, “each kindred was drawn up under an officer of its own lineage and appointment, and the several members of the family served together.” This organization, or lack of organization, continued throughout the feudal period. In France, in the 14th century, the army was a “horde of independent chiefs, each with his own following, each doing his own will;” and, according to Froissart, the different groups “were so ill-informed” that they did not always know of a discomfiture of the main body.
Besides that increased subordination of local heads to the general head which accompanies political integration, and which must of course precede a more centralized and combined mode of military action, two special causes may be recognized as preparing the way for it.
One of these is unlikeness of kinds in the arms used. Sometimes the cooperating tribes, having habituated themselves to different weapons, come to battle already marked off from one another. In such cases the divisions by weapons correspond with the tribal divisions; as seems to have been to some extent the case with the Hebrews, among whom the men of Benjamin, of Gad, and of Judah, were partially thus distinguished. But, usually, the unlikenesses of arms consequent on unlikenesses of rank, initiate these military divisions which tend to traverse the divisions arising from tribal organization. The army of the ancient Egyptians included bodies of charioteers, of cavalry, and of foot; and the respective accoutrements of the men forming these bodies, differing in their costliness, implied differences of social position. The like may be said of the Assyrians. Similarly, the Iliad shows us among the early Greeks a state in which the contrasts in weapons due to contrasts in wealth, had not yet resulted in differently-armed bodies, such as are formed at later stages with decreasing regard for tribal or local divisions. And it was so in Western Europe during times when each feudal superior led his own knights, and his followers of inferior grades and weapons. Though within each group there were men differing alike in their rank and in their arms, yet what we may call the vertical divisions between groups were not traversed by those horizontal divisions throughout the whole army, which unite all who are similarly armed. This wider segregation it is, however, which we observe taking place with the advance of military organization. The supremacy acquired by the Spartans was largely due to the fact that Lykurgus “established military divisions quite distinct from the civil divisions, whereas in the other states of Greece, until a period much later…the two were confounded—the hoplites or horsemen of the same tribe or ward being marshalled together on the field of battle.” With the progress of the Roman arms there occurred kindred changes. The divisions came to be related less to rank as dependent on tribal organization, and more to social position as determined by property; so that the kinds of arms to be borne and the services to be rendered, were regulated by the sizes of estates, with the result of “merging all distinctions of a gentile and local nature in the one common levy of the community.” In the field, divisions so established stood thus:—
“The four first ranks of each phalanx were formed of the full-armed hoplites of the first class, the holders of an entire hide [?]; in the fifth and sixth were placed the less completely equipped farmers of the second and third class; the two last classes were annexed as rear ranks to the phalanx.’
And though political distinctions of clan-origin were not thus directly disregarded in the cavalry, yet they were indirectly interfered with by the addition of a larger troop of non-burgess cavalry. That a system of divisions which tends to obliterate those of rank and locality, has been reproduced during the re-development of military organization in modern times, is a familiar fact.
A concomitant cause of this change has all along been that interfusion of the gentile and tribal groups entailed by aggregation of large numbers. As before pointed out, the Kleisthenian re-organization in Attica, and the Servian re-organization in Rome, were largely determined by the impracticability of maintaining the correspondence between tribal divisions and military obligations; and a redistribution of military obligations naturally proceeded on a numerical basis. By various peoples, we find this step in organization taken for civil purposes or military purposes, or both. To cases named in § 512, may be added that of the Hebrews, who were grouped into tens, fifties, hundreds and thousands. Even the barbarous Araucanians divided themselves into regiments of a thousand, sub-divided into companies of a hundred. Evidently numerical grouping conspires with classing by arms to obliterate the primitive divisions.
This transition from the state of incoherent clusters, each having its own rude organization, to the state of a coherent whole, held together by an elaborate organization running throughout it, of course implies a concomitant progress in the centralization of command. As the primitive horde becomes more efficient for war in proportion as its members grow obedient to the orders of its chief; so, the army formed of aggregated hordes becomes more efficient in proportion as the chiefs of the hordes fall under the power of one supreme chief. And the above-described transition from aggregated tribal and local groups to an army formed of regular divisions and sub-divisions, goes along with the development of grades of commanders, successively subordinated one to another. A controlling system of this kind is developed by the uncivilized, where considerable military efficiency has been reached; as at present among the Araucanians, the Zulus, the Uganda people, who have severally three grades of officers; as in the past among the ancient Peruvians and ancient Mexicans, who had respectively several grades; and as also among the ancient Hebrews.
§ 520. One further general change has to be noticed—the change from a state in which the army now assembles and now disperses, as required, to a state in which it becomes permanently established.
While, as among savages, the male adults are all warriors, the fighting body, existing in its combined form only during war, becomes during peace a dispersed body carrying on in parties or separately, hunting and other occupations; and similarly, as we have seen, during early stages of settled life the armed freemen, owning land jointly or separately, all having to serve as soldiers when called on, return to their farming when war is over: there is no standing army. But though after the compounding of small societies into larger ones by war, and the rise of a central power, a kindred system long continues, there come the beginnings of another system. Of course, irrespective of form of government, frequent wars generate permanent military forces; as they did in early times among the Spartans; as later among the Athenians; and as among the Romans, when extension of territory brought frequent needs for repressing rebellions. Recognizing these cases, we may pass to the more usual cases, in which a permanent military force originates from the body of armed attendants surrounding the ruler. Early stages show us this nucleus. In Tahiti the king or chief had warriors among his attendants; and the king of Ashantee has a bodyguard clad in skins of wild beasts—leopards, panthers, &c. As was pointed out when tracing the process of political differentiation, there tend everywhere to gather round a predominant chieftain, refugees and others who exchange armed service for support and protection; and so enable the predominant chieftain to become more predominant. Hence the comites attached to the princeps in the early German community, the húscarlas or housecarls surrounding old English kings, and the antrustions of the Merovingian rulers. These armed followers displayed in little, the characters of a standing army; not simply as being permanently united, but also as being severally bound to their prince or lord by relations of personal fealty, and as being subject to internal government under a code of martial law, apart from the government of the freemen; as was especially shown in the large assemblage of them, amounting to 6,000, which was formed by Cnut.
In this last case we see how small body-guards, growing as the conquering chief or king draws to his standard adventurers, fugitive criminals, men who have fled from injustice, &c., pass unobtrusively into troops of soldiers who fight for pay. The employment of mercenaries goes back to the earliest times—being traceable in the records of the Egyptains at all periods; and it continues to re-appear under certain conditions: a primary condition being that the ruler shall have acquired a considerable revenue. Whether of home origin or foreign origin, these large bodies of professional soldiers can be maintained only by large pecuniary means; and, ordinarily, possession of these means goes along with such power as enables the king to exact dues and fines. In early stages the members of the fighting body, when summoned for service, have severally to provide themselves not only with their appropriate arms, but also with the needful supplies of all kinds: there being, while political organization is little developed, neither the resources nor the administrative machinery required for another system. But the economic resistance to militant action, which, as we have seen, increases as agricultural life spreads, leading to occasional non-attendance, to confiscations, to heavy fines in place of confiscations, then to fixed money-payments in place of personal services, results in the growth of a revenue which serves to pay professional soldiers in place of the vassals who have compounded. And it then becomes possible, instead of hiring many such substitutes for short times, to hire a smaller number continuously—so adding to the original nucleus of a permanent armed force. Every further increase of royal power, increasing the ability to raise money, furthers this differentiation. As Ranke remarks of France, “standing armies, imposts, and loans, all originated together.”
Of course the primitive military obligation falling on all freemen, long continues to be shown in modified ways. Among ourselves, for instance, there were the various laws under which men were bound, according to their incomes, to have in readiness specified supplies of horses, weapons, and accoutrements, for themselves and others when demanded. Afterwards came the militia-laws, under which there fell on men in proportion to their means, the obligations to provide duly armed horse-soldiers or foot-soldiers, personally or by substitute, to be called out for exercise at specified intervals for specified numbers of days, and to be provided with subsistence. There may be instanced, again, such laws as those under which in France, in the 15th century, a corps of horsemen was formed by requiring all the parishes to furnish one each. And there are the various more modern forms of conscription, used, now to raise temporary forces, and now to maintain a permanent army. Everywhere, indeed, freemen remain potential soldiers when not actual soldiers.
§ 521. Setting out with that undifferentiated state of the body politic in which the army is co-extensive with the adult male population, we thus observe several ways in which there goes on the evolution which makes it a specialized part.
There is the restriction in relative mass, which, first seen in the growth of a slave-population, engaged in work instead of war, becomes more decided as a settled agricultural life occupies freemen, and increases the obstacles to military service. There is, again, the restriction caused by that growing costliness of the individual soldier accompanying the development of arms, accoutrements, and ancillary appliances of warfare. And there is the yet additional restriction caused by the intenser strain which military action puts on the resources of a nation, in proportion as it is carried on at a greater distance.
With separation of the fighting body from the body-politic at large, there very generally goes acquirement of a separate head. Active militancy ever tends to maintain union of civil rule with military rule, and often causes re-union of them where they have become separate; but with the primary differentiation of civil from military structures, is commonly associated a tendency to the rise of distinct controlling centres for them. This tendency, often defeated by usurpation where wars are frequent, takes effect under opposite conditions; and then produces a military head subordinate to the civil head.
While the whole society is being developed by differentiation of the army from the rest, there goes on a development within the army itself. As in the primitive horde the progress is from the uncombined fighting of individuals to combined fighting under direction of a chief; so, on a larger scale, when small societies are united into great ones, the progress is from the independent fighting of tribal and local groups, to fighting under direction of a general commander. And to effect a centralized control, there arises a graduated system of officers, replacing the set of primitive heads of groups, and a system of divisions which, traversing the original divisions of groups, establish regularly-organized masses having different functions.
With developed structure of the fighting body comes permanence of it. While, as in early times, men are gathered together for small wars and then again dispersed, efficient organization of them is impracticable. It becomes practicable only among men who are constantly kept together by wars or preparations for wars; and bodies of such men growing up, replace the temporarily-summoned bodies.
Lastly, we must not omit to note that while the army becomes otherwise distinguished, it becomes distinguished by retaining and elaborating the system of status; though in the rest of the community, as it advances, the system of contract is spreading and growing definite. Compulsory cooperation continues to be the principle of the military part, however widely the principle of voluntary cooperation comes into play throughout the civil part.