We do not intend to go further, however, without replying to two objections, which are not without weight, and which we do not wish to leave behind us unanswered.
In a certain number of Latin books of the Middle Ages we find, to describe chivalry, an expression which the "Romanists" oppose triumphantly to us, and of which the Romish origin cannot seriously be doubted. When it is intended to signify that a knight has been created, it is stated that the individual has been girt with the cingulum militare. Here we find ourselves in full Roman parlance, and the word signified certain terms which described admission into military service, the release from this service, and the degradation of the legionary. When St. Martin left the militia, his action was qualified as solutio cinguli, and at all those who act like him the insulting expression militaribus zonis discincti is cast. The girdle which sustains the sword of the Roman officer—cingulum zona, or rather cinctorium—as also the baldric, from balteus, passed over the shoulder and was intended to support the weapon of the common soldier. "You perceive quite well," say our adversaries, "that we have to do with a Roman costume." Two very simple observations will, perhaps, suffice to get to the bottom of such a specious argument: The first is that the Germans in early times wore, in imitation of the Romans, "a wide belt ornamented with bosses of metal," a baldric, by which their swords were suspended on the left side; and the second is that the chroniclers of old days, who wrote in Latin and affected the classic style, very naturally adopted the word cingulum in all its acceptations, and made use of this Latin paraphrasis—cingulo militari decorare—to express this solemn adoption of the sword. This evidently German custom was always one of the principal rites of the collation of chivalry. There is then nothing more in it than a somewhat vague reminiscence of a Roman custom with a very natural conjunction of terms which has always been the habit of a literary people.
To sum up, the word is Roman, but the thing itself is German. Between the militia of the Romans and the chivalry of the Middle Ages there is really nothing in common but the military profession considered generally. The official admittance of the Roman soldier to an army hierarchically organized in no way resembled the admission of a new knight into a sort of military college and the "pink of society." As we read further the singularly primitive and barbarous ritual of the service of knightly reception in the twelfth century, one is persuaded that the words exhale a German odor, and have nothing Roman about them. But there is another argument, and one which would appear decisive. The Roman legionary could not, as a rule, withdraw from the service; he could not avoid the baldric. The youthful knight of the Middle Ages, on the contrary, was always free to arm himself or not as he pleased, just as other cavaliers are at liberty to leave or join their ranks. The principal characteristic of the knightly service, and one which separates it most decidedly from the Roman militia, was its freedom of action.
One very specious objection is made as regards feudalism, which some clear-minded people obstinately confound with chivalry. This was the favorite theory of Montalembert. Now there are two kinds of feudalism, which the old feudalists put down very clearly in two words now out of date—"fiefs of dignity" and "fiefs simple." About the middle of the ninth century, the dukes and counts made themselves independent of the central power, and declared that people owed the same allegiance to them as they did to the emperor or the king. Such were the acts of the "fiefs of dignity," and we may at once allow that they had nothing in common with chivalry. The "fiefs simple," then, remained.
In the Merovingian period we find a certain number of small proprietors, called vassi, commending themselves to other men more powerful and more rich, who were called seniores. To his senior who made him a present of land the vassus owed assistance and fidelity. It is true that as early as the reign of Charlemagne he followed him to war, but it must be noted that it was to the emperor, to the central power, that he actually rendered military service. There was nothing very particular in this, but the time was approaching when things would be altered. Toward the middle of the ninth century we find a large number of men falling "on their knees" before other men! What are they about? They are "recommending" themselves, but, in plainer terms, "Protect us and we will be your men." And they added: "It is to you and to you only that we intend in future to render military service; but in exchange you must protect the land we possess—defend what you will in time concede to us; and defend us ourselves." These people on their knees were "vassals" at the feet of their "lords"; and the fief was generally only a grant of land conceded in exchange for military service.
Feudalism of this nature has nothing in common with chivalry.
If we consider chivalry in fact as a kind of privileged body into which men were received on certain conditions and with a certain ritual, it is important to observe that every vassal is not necessarily a cavalier. There were vassals who, with the object of averting the cost of initiation or for other reasons, remained damoiseaux, or pages, all their lives. The majority, of course, did nothing of the kind; but all could do so, and a great many did.
On the other hand we see conferred the dignity of chivalry upon insignificant people who had never held fiefs, who owed to no one any fealty, and to whom no one owed any.
We cannot repeat too often that it was not the cavalier (or knight), it was the vassal who owed military service, or ost, to the seigneur, or lord; and the service in curte or court: it was the vassal, not the knight, who owed to the "lord" relief, "aid," homage.
The feudal system soon became hereditary. Chivalry, on the contrary, has never been hereditary, and a special rite has always been necessary to create a knight. In default of all other arguments this would be sufficient.