During the early morning of the 10th our position became yet worse; on the right flank the Japanese drove back Borisoff’s force to Hsiao-kou-tzu and opposite San-tai-tzu, and penetrated as far as the grove of the Imperial tombs. On the east large bodies of them appeared in sight of the Mandarin road. One was opposite Levestam’s force, while another began shelling the Mandarin road near Ta-wa from the heights near Hsin-chia-kou. The orders given on March 5 for the baggage to be sent back in good time had not been carried out, and part of the impedimenta of the 2nd and 3rd Armies, which was stretching along the road near Mukden early on the 10th, blocked the passage of the 5th and 6th Siberians and 17th Corps. On this morning also the Japanese, who had broken through near Chiu-tien on the 9th, began to press our left flank under Meyendorff. The troops sent as reinforcements did not act together, and were driven back north-west. By 10 a.m. Meyendorff was in full retreat—not north-east, but north-west towards the Mandarin road, which he crossed between Ta-wa and Pu-ho. The 6th Siberians now began to retire prematurely, and by so doing exposed the right of the 1st Corps and the left of the 17th. This unnecessarily sudden retirement of more than forty battalions under Meyendorff and Soboleff placed the 17th Corps and the 5th Siberians in a difficult position. Instead of fronting south, they had to front south-east. After a hot fight this force, consisting of thirty battalions, was also obliged to move to the rear prematurely. They did not go to Ta-wa, but west and south of the Mandarin road. This opened out a way for the enemy to that road, and also to the railway north—further on the portion between Mukden and Wen-ken-tun. By seizing this section about 2 p.m., before the rearguards or even the tail of the main body had passed Wa-tzu, they took our troops in flank. We had evacuated the village of San-tai-tzu prematurely, and it was quickly occupied by the Japanese. Between Wa-tzu and this village there is a defile, less than three miles long, through which a large part of the 2nd Army had to force its way under attack from both sides. Portions of the rearguards under Hanenfeld and Sollogub, which tried to get round to the east of it, were captured or destroyed.
I instructed General Dembovski to organize the defence of the Mandarin road at Ta-wa, and for that purpose to utilize the troops retiring along it. By 10 a.m. the distance between the portions of the enemy on the west and east of the railway was only seven miles. It was vital to stop any further contraction of the area of retirement of the 2nd Army. This might be done by blocking the Japanese advance to the railway from the west and north-west. As I was more anxious about the latter direction than any other, I moved out the eighteen battalions under Zarubaeff, which had joined my reserve from the 1st Army, on to the line Ma-kou-chia-tzu–Yang-tzu-tun, and ten battalions of the 72nd Division on the front Tung-shan-tzu–Hsiao-hsin-tun. The first force covered the railway between Hu-shih-tai and San-tai-tzu, and the second barred the enemy’s advance and supported the right flank of Artamonoff’s column. As a reserve to these troops, in case of pressure from the east, a brigade of the 1st Siberian Division was left near Hu-shih-tai station. By 4 p.m. the state of affairs on the Mandarin road became worse, as, immediately after General Levestam’s force had retired behind Pu-ho, Dembovski also abandoned his positions near Ta-wa, and moved off to the west. The fighting ceased as darkness came on. The last of the 2nd Army to fall back were portions of the 1st, 2nd, and 3rd Rifle Regiments under Lieutenant-Colonel Korniloff; they broke through near Wa-tzu in the pitch dark, though hemmed in by the enemy on three sides.
We continued to retire during the night, covered by the rearguard under Muiloff and that of Zarubaeff’s column. On the 11th several units of the 1st and 3rd Armies collected at the village of Yi-lu; but the greater part of the 3rd Army fell back direct on Tieh-ling. Bilderling was unable to carry out his proposal of remaining on the River Yi-lu till the 12th, and, having taken command of Shileiko’s force, after slight opposition retired northwards from Yi-lu village. By doing this he placed the rearguards of the 2nd Army that were still south of this point in a very precarious position. The main bodies of all the armies began on the 11th to occupy a position eight miles south of Tieh-ling on the Fan Ho. The 2nd Army took up a line to the west and the first one to the east of the Mandarin road, the 3rd remaining in reserve. Everything possible was done to restore order amongst the troops, transport, and parks. On the 13th the enemy’s advanced troops reached our positions, and on the 14th they attacked, directing their main effort on the line between the sections held by the 2nd Siberians and 72nd Division. All their attacks were repulsed with great loss, and many hundreds of dead were left in front of our position. Our losses were 900.
The two-weeks battle had badly disorganized several units, especially those of the 2nd and 3rd Armies. The men who had got separated from their own units and attached to others had to be sorted out and restored, baggage, transport, and parks had to be separated, and ammunition replenished. To carry this out made it essential that we should not be in direct touch with the enemy—that there should be some space between us. For this reason, and on account of the turning movement against our right flank along the River Liao, discovered by the cavalry, I decided not to accept battle at Tieh-ling, but to order a general retirement of all the armies on the 14th to the Hsi-ping-kai position, which was the best one between Tieh-ling and the River Sungari. The 1st and 2nd Armies began to move out of Tieh-ling on March 16, and by the 22nd were on the heights of Hsi-ping-kai.
CONCLUSIONS UPON THE BATTLE OF MUKDEN[93]
Both the nearness of the events related above and our ignorance about the enemy make it impossible for any detailed and absolutely impartial judgment to be formed upon the reasons for our defeat in this great battle. The records that have been collected so far, however, are sufficient to throw light upon a few facts—upon certain of our dispositions that did not correspond to the requirements of the case. Those made by the commander of the 2nd Army, to which force was entrusted the duty of stopping Nogi’s turning movement towards our rear, are of particular interest, and certain of them which had a very important bearing on the issue of the operations are now described.
General Kaulbars made neither a sufficient nor a clever use of his cavalry. This fact, coupled with the unfortunate selection of its leaders, was the reason why the mounted branch did such bad work,[94] and behaved in a manner that can hardly be called “devoted” during the Mukden operations. In the instructions given on March 1 to Grekoff’s cavalry to operate against Nogi, the object to be attained was plainly set forth, but how it was to be attained was not clearly defined. The execution of its most important task was also made the more difficult by the fact that Grekoff’s force was, on the same day as the orders were issued, split up into two almost equal groups, of which the eastern was found to be fighting Oku instead of Nogi. To rectify this, the cavalry under Pavloff was ordered on the same day by Kaulbars to undertake a special task against the turning columns, but on the 2nd the order was changed, and eight of Pavloff’s sotnias were put under the command of Launits, who was operating against Oku. No touch was maintained between their different groups, and the greater part of the mounted forces clung to the infantry, and did practically no fighting (the losses suffered by this Arm during the twenty-three days’ operations in February and March were quite insignificant). Yet most of our regiments were quite capable of performing the most difficult tasks of war. The action of the infantry of the 2nd Army on the positions which they had taken up was completely passive. They did not try to get into touch with the enemy to ascertain their strength and dispositions (by taking prisoners), or to occupy advanced posts where these would be advantageous. The reconnoitring patrols of this army also did but little work. The consequence of such unsatisfactory performance of their duties by the cavalry and advanced infantry units of the 2nd Army was that information of the enemy was so meagre that the appearance of a great mass of Nogi’s army on and to the east of the Hsin-min-tun road came as a complete surprise to Kaulbars.
Owing to the appearance of large hostile bodies near Ka-liao-ma, I had on February 28 already ordered him[95] to take immediate steps to ascertain their exact strength, the direction in which they were moving, and their intentions. I repeated this order[96] on March 2, instructing him to find out their strength and dispositions more accurately if possible, and to frame some plan of action. I pointed out the necessity for energetic steps to ascertain the whereabouts of Nogi’s main body—whether it was opposite Sha-ling-pu, or whether it was executing a wider turning movement. On the morning of March 5 I for the third time[97] asked Kaulbars to find out where Nogi’s left flank was. Not one of these orders was carried out, with the result that I had inadequate and incorrect information upon which to form a decision as to the strength and whereabouts of the enemy operating on the right bank of the Hun. Tserpitski’s alarmist reports to the effect that more than three divisions were opposed to him made the fog worse. Kaulbars, who had been ordered to stop Nogi’s flanking movement, on the strength of incorrect information, all the time turned his chief attention towards the western front to Oku, whom he took for Nogi. The latter, owing to the 2nd Army’s inaction on March 3, 4, 5 and 6, was made a present of four days in which to complete his sweeping movement to the north-east,[98] and Kaulbars continued to see danger only on the west, paying insufficient attention to what was happening on the Hsin-min-tun road, north-west of Mukden. On March 1 he conceived a most complicated “castling” manœuvre, which he endeavoured to carry out when in direct touch with the enemy. The Composite Rifle Corps was ordered to cross from the right bank of the Hun on to the left, and the 8th Corps from the left to the right. The Rifle regiments crossed over the river, and by so doing evacuated the most important section near Chang-tan, but the 8th Corps was unable to get across. The enemy at once took advantage of this, and, rapidly throwing their 8th Division forward along the right bank of the river, drove back the relatively weak force of ours still on that side. Kaulbars, moreover, stopped the movement on Sha-ling-pu (of the Composite Division under Golembatovski), which had already been started, and by so doing deprived us of the possibility of checking the heads of the enemy’s columns on March 2. Finally, the 5th Rifle Brigade under Churin—which was moving by my orders to operate against Nogi—was stopped on March 3 by Kaulbars in the valley on the right bank of the Hun, and found itself among the troops opposing Oku.
After weakening Topornin by sixteen battalions, Kaulbars, on reaching his force, countermanded the advance on Sha-ling-pu, which had been begun on the morning of the 3rd, and suddenly withdrew thirty-two battalions to Mukden without fighting. This made our position distinctly worse. He took no steps to establish and maintain touch with Birger’s brigade on the Hsin-min-tun road, and never informed the latter of the order to retire he had given to Topornin on the 3rd. In telling Launits on the morning of March 3 of his decision (to withdraw Topornin’s force to Mukden), he stated that “Grekoff’s column and Birger’s brigade are probably cut off from Mukden,” but he made no attempt to help Birger. And yet up to 2 p.m. on the 3rd Birger’s brigade was not even engaged. Our attempt to retake Ssu-hu-chia-pu on March 4 was stopped by Launits, owing to the receipt of orders from Kaulbars not to attack if it was likely to be a costly operation. Kaulbars did nothing that day, although he had under his command 119 battalions[99] on the right bank of the Hun, and although I had ordered him to assume the offensive. Moreover, he did not even know the whereabouts of the troops under him. Although he had 113 battalions under his command on the right bank on March 5, he again did nothing. He did not carry out my orders to attack the enemy’s left energetically, and permitted these troops, which were at Khou-kha—next to Gerngross’s force—to deploy very slowly, and stopped their advance before they had got in touch with the enemy. Moreover, yielding to the preconceived idea of the main danger lying in the west, he moved sixteen splendid battalions of the 10th Corps from Gerngross’s force, operating towards Hsin-min-tun, on to the left flank of the army. Yet again on the 6th, although he had 116 battalions on the right bank, he effected scarcely anything, for our active operations towards Hsin-min-tun were conducted with an insufficient force, and therefore failed.