expedient of making even a narrow zone truly impenetrable by means of radical destructions, adopted by the Germans in France in the spring of 1917, requires both time and an elaborate labour programme, neither of which is available in a battle crisis. In some cases, inundations serve the purpose, but even inundations take time to spread. On the Yser front in Oct. 1914, five days elapsed between the order to open the sluices and the creation of an effective barrier thereby. Moreover, in a generally penetrable country the area of decision might turn out to lie in an unex- pected direction, and, if so, a system of fortifications designed to protect regrouping by likly lateral routes might prove to be useless for covering those which actually were required to be used. Thus, a speculative element began to come into schemes of forti- fication. It was no longer possible to justify heavy capital ex- penditure on works by reference to plain and definite needs. Already it was admitted that if the unexpected happened, it would be necessary to make shift with improvised works; already a considerable body of opinion held that permanent works, if admitted at all, should be similar in general design to field works; and the tendency in these circumstances was inevitably to trust to the latter, which were cheaper, could be built just where they were wanted, and according to many experts were just as good in principle as permanent works, and indeed better than most of the expensive fortifications already in existence.
This underground growth of the linear principle was fostered by another cause. The time-honoured relation of the town and its defences was altered. Under the influence of tradition theory continued to conceive of the fortress as a circular defence round a town. But the town had ceased to interest the military engi- neer, who now called it the " nucleus." He disposed his ring of defences at such a distance as to protect the nucleus from bom- bardment, a convention which was imposing a larger and larger perimeter with every improvement in the range of guns, but his works were now meant primarily to take a share in the operations in the field. That being so, except in such cases as that of Port Arthur, where the nucleus contained establishments regarded as essential to the conduct of the war, it was almost a matter of indifference whether a particular town should be protected or not. The old relation of the town and its walls had been based on the fact that the walls preserved the town from pillage and murder. In the course of time this had changed to a great extent, and in the era of " cabinet wars " it had almost vanished. But the old mediaeval spirit of the towns came to life again in such instances as Zaragoza, Colberg, Venice, and even if the towns-people were indifferent or sullen, the governor could usually resist pressure from them, because he was strong in the conception of his plain duty as a soldier to defend the post entrusted to him. But when the defence perimeter had advanced out of sight of the town, and the enceinte had either been turned into a public garden or retained, demilitarized, as a historical monument only; when, further, in peace-time no military barrier whatever differentiated the defended area from the open country; when, lastly, two gen- erations of railway traffic had destroyed the self-centred econo- mic life of the town and blurred its particularism then from the point of view of the town it was in much the same position as any undefended town or village in the theatre of war. It might come within the ambit of military operations or it might not; if it did, it might either resist the invader heroically in the manner of Belfort in 1870-1, or agitate for demilitarization as Lille did in 1914, but the fact that a ring of forts lay out in the country around it had very little influence either way on its conduct as a town. Open towns in modern times have behaved like fort- resses of old, and fortresses like open towns of old.
Correspondingly, the position and outlook of the governor has changed. Formerly the town was his charge, and almost his viceroyalty. His troops were his own; organized for sedentary warfare and not for campaigning, they were not at the disposal of a field army which happened to be operating in the neigh- bourhood, and the town was in practice defended sometimes with spirit, sometimes feebly whatever course operations took outside. Up to the very eve of the World War a French fortress governor was responsible to the Government only, and took no
orders from the commander-in-chief of the field armies. The era of " cabinet wars " made little difference to this state of things; the population might be indifferent to the war in the towns as in the country, but to the governor and his troops the fortress was still a charge to be defended. Moreover, it was a real base for the armies in the field, in that the stores and supplies for those armies were accumulated in the fortress, and a real strategic aid in that it commanded routes that were obligatory for both sides. But when, in our own times, the governor had become simply the commander of a certain group of forces destined like other forces to take their part in a general scheme of battle; when the area within his defences had to a great extent ceased to be the source of stores and supplies for the field army, and when railways, needing protection at all points and not merely at a focus, became the principal lines of communication, the choice between evacuation and defence came to be governed by larger considerations of strategy. The governor's decisions therefore were assimilated in principle to those of any tactical executant of the strategist's instructions. He might defend or evacuate as a field commander might hold his ground or retire. But the peculiar character of his responsibility was gone. Even in France, the country which has been most tenacious of the fortress tradition, the old regulation, already quoted, was modi- fied in the 1913 " Regulations for the conduct of Higher Forma- tions," which empowered the commandcr-in-chief to assume con- trol of any fortress and its forces if he thought fit. 1
On the German side, units made up from fortress garrisons formed quite one- third of the Eastern armies during the first cam- paignsof 1914 operating sometimes a hundred miles away from their fortress of origin and in the sequel, never returning to it.
In sum, therefore, causes of a general character operating before 1914 produced these tendencies: (a) to divorce fortifica- tions from their nucleus or central town, (b) to make them rather linear than circular in trace, (c) to bring them into conformity with the battle-scheme of the field armies (with dtdassemtnt as the alternative), and (d) to construct them as far as possible according to the principles of field fortification.
The theory of fortification, on the other hand, was still bound by the notion of a nucleus, and unable, therefore, for the mo- ment to employ its stock of ideas and methods to the best ad- vantage. The practical technique of fortification and siegecraft was, meantime, progressing in details; reinforced concrete had come into normal use, armour was improving in quality, the defence had it in its power no less than the attack to profit by developments in the design of quick-firing guns and howitzers of medium calibres. Observation balloons and kites were available, superior to the old spherical types; wireless telegraphy removed some of the dangers of investment and made it possible to co- ordinate the activity of a besieged garrison with that of a relieving army. The technique of bored mines developed, and trench- mortars and grenades reappeared. The lessons of Port Arthur in matters of detail-tactics and design were assimilated in the various armies. The enormous defensive power of the machine- gun was realized and to some extent exploited. It remained to synthesize the application of these elements, old and new, in an art of fortification that responded to the new demands and con- ditions of warfare.
This art began to take shape with the introduction of the " group " principle. Advocated by several theoretical writers in the period of controversy, it was applied practically, and on a large scale, by the Germans in the celebrated Feste constructed on the Moselle and the Rhine in the last ten years before the
1 It was in virtue of this new regulation that Gallieni's Paris forces were brought under Joffre's command in the battle of the Marne; and in accordance with the spirit of it that Sarrail acted in the same crisis, when, although only an army commander, he sent imperative orders to the governor of Verdun to despatch his mobile reserves to the battle-field of Revigny. The fact that the governor, General Coutanceau, though himself under attack, complied with this requisition instead of standing on his undoubted legal rights, is itself evidence of the changed outlook of the fortress governor in modern warfare. In a somewhat different way, the confused story of the declassement of Lille in Aug. 1914 points the same moral.
