The utility of cavalry has always been doubted, Colonel Ardant Du Picq notes:
That is because its cost is high. It is little used, just because it does cost. The question of economy is vital in peace times. When we set a high value upon certain men, they are not slow to follow suit, and to guard themselves against being broken. Look at staff officers who are almost never broken (reduced), even when their general himself is.
[...]
With the power of modern weapons, which forces you to slow down if it does not stop you, the advance under fire becomes almost impossible. The advantage is with the defensive. This is so evident that only a madman could dispute it. What then is to be done? Halt, to shoot at random and cannonade at long range until ammunition is exhausted? Perhaps. But what is sure, is that such a state of affairs makes maneuver necessary. There is more need than ever for maneuver at a long distance in an attempt to force the enemy to shift, to quit his position. What maneuver is swifter than that of cavalry? Therein is its role.
The extreme perfection of weapons permits only individual action in combat, that is action by scattered forces. At the same time it permits the effective employment of mass action out of range, of maneuvers on the flank or in the rear of the enemy in force imposing enough to frighten him.
Can the cavalry maneuver on the battle field? Why not? It can maneuver rapidly, and above all beyond the range of infantry fire, if not of artillery fire. Maneuver being a threat, of great moral effect, the cavalry general who knows how to use it, can contribute largely to success. He arrests the enemy in movement, doubtful as to what the cavalry is going to attempt. He makes the enemy take some formation that keeps him under artillery fire for a while, above all that of light artillery if the general knows how to use it. He increases the enemy’s demoralization and thus is able to rejoin his command.
Rifled cannon and accurate rifles do not change cavalry tactics at all. These weapons of precision, as the word precision indicates, are effective only when all battle conditions, all conditions of aiming, are ideal. If the necessary condition of suitable range is lacking, effect is lacking. Accuracy of fire at a distance is impossible against a troop in movement, and movement is the essence of cavalry action. Rifled weapons fire on them of course, but they fire on everybody.
In short, cavalry is in the same situation as anybody else.
What response is there to this argument? Since weapons have been improved, does not the infantryman have to march under fire to attack a position? Is the cavalryman not of the same flesh? Has he less heart than the infantryman? If one can march under fire, cannot the other gallop under it?
When the cavalryman cannot gallop under fire, the infantryman cannot march under it. Battles will consist of exchanges of rifle shots by concealed men, at long range. The battle will end only when the ammunition is exhausted.
The cavalryman gallops through danger, the infantryman walks. That is why, if he learns, as it is probable he will, to keep at the proper distance, the cavalryman will never see his battle rôle diminished by the perfection of long range fire. An infantryman will never succeed by himself. The cavalryman will threaten, create diversions, worry, scatter the enemy’s fire, often even get to close quarters if he is properly supported. The infantryman will act as usual. But more than ever will he need the aid of cavalry in the attack. He who knows how to use his cavalry with audacity will inevitably be the victor. Even though the cavalryman offers a larger target, long range weapons will paralyze him no more than another.
Wishful thinking?
Not really. Cavalry did well in the modern period (WWI and WWII) when used in accord with its strengths, largely those outlined by Col. du Picq. Hence we see cavalry divisions and even corps making significant contributions on the Eastern Front of both those wars.
The Western fronts were simply too dense with men (said density enabled by the density of the logistics network of France and Belgium) to allow cavalry units the freedom to maneuver. Every attack is a frontal attack if there are no flanks.
I suppose this was meant as a reductio ad absurdum, but in retrospect it looks like a correct prediction about what happened in WW1.
Once armies obtained something akin to cavalry capable of galloping under heavy fire (i.e. tanks), war indeed became a dynamic game of maneuver much like what Du Picq imagined here. But it certainly seems like he was carried away by wishful thinking about the prospects of cavalry (and chivalric?) warfare in the industrial era.
If you include mounted rifles under the umbrella of “cavalry”, WWI also saw the successes of the Australian Light Horse at Beersheba during the Palestine campaign.
On Vladimir’s comment: one element that substantially deepened Australian gains in the August 8 1918 battle near the Somme was a battalion of 8 armoured cars. Even limited to road travel, the armoured cars were able to push into what had been German rear areas and disrupt communications, supplies, and morale.