Military history, widely used in this sense by our Anglo-Saxon allies, remains, for the future military leader, a remarkable source of reflection, applied as much to the formation of judgment and reasoning, as to the illustration and knowledge of the major tactical and strategic principles. Its main merit lies in the supply of a reference system of concrete data constituting a general fund, on which the assessment and resolution of an operational problem must be based. The importance of the principles of warfare and the art of operations, which remains at the heart of the training of chiefs and staff officers, is thus rarely contradicted by facts and historical analysis.

These principles retain all their strength and relevance in current operations, but with a somewhat "erased" relief due to the insidious nature of the confrontations, the absence of vital or dramatic issues (for Western forces at least), and finally by a certain dilution of responsibilities between political, civilian or military actors. As a result, military leaders rarely find themselves subject to a clear dialectic of wills with the enemy or to the pressure of decisive choices involving a campaign, the future and role of a nation or the lives of thousands of men. And it is indeed in the testing of paroxysmal situations that the most clear-cut data of apprehension emerge.ciation on the value of a military leader and his structures for conducting operations, intended to become a reference point in the future.

This is why I have chosen, in the context of a reflection on the principles of war, to evoke what was the last great battle of the French army. The battle of Dien Bien Phû is indeed exemplary in two ways: first of all with regard to these famous principles of war, declined in a positive way by the adversary and, unfortunately, rather in a negative way on our side; but, in another register, it is also exemplary for the French army.But, on another level, it is also possible to think that the memory of this battle, in which the soldier's ethics were literally sublimated by the defenders, must continue to soak the soul of every officer.

Let us therefore project ourselves a little more than 50 years back in time, to this Indochina which still nourishes certain myths of the French army, and more precisely to the year 1952, because it is impossible to explain Dien Bien Phû without evoking the battle of Na-San, which gave substance to the concept of the air-land base.

An initial "good example", Na San, deviant matrix of Dien-Bien-Phû

In the spring of 1952, Giap had to consider that he had temporarily failed in his objectives: the massive offensives on the delta were broken, and the battle of Hoa-Binh ended in a battle of attrition unfavorable to Vietminh. He therefore decided to postpone the effort to the upper Tongan region and Laos in order to attract the French forces in an unfavourable situation, far from their capabilities.s capacity for support, and to inflict a psychological failure on France by demonstrating its inability to defend an area whose populations were its own. The extension of this strategy aims to seize the capitals of Laos (Luang-Prabang and Vientiane), and to reach the Mekong in order to extend the embrace southwards and completely isolate the Franco-Vietnamese forces[1] from the Indochinese North and South.

Faced with the offensive of 3 Vietminh divisions (VM) in October 1952, the initial reaction of General Salan, commander-in-chief, was to withdraw all the French posts of the Haute-Région to the Black River to avoid their successive destruction. The regrouping zone was set at Na-San, crossroads of several communication routes, and provided with a "dakotable" ground (in the immediate vicinity of Son-La). Salan reserves the right to determine the future purpose of this grouping (withdrawal, on-site resistance, redistribution) according to the evolution of the VM manoeuvre. Very quickly, the "Mandarin" [2] formed the opinion that Giap would not resist the opportunity to liquidate the French forces of the Haute-Région, offered concentrated in Na-San, and would throw his entire autumn campaign 52 into it. It was therefore necessary to transform the maneuver and give himself the means to break the attack of divisions 308, 312 and 316.

• First idea: gain time to maintain its freedom of actionIn parallel, a large-scale operation, Operation LORRAINE, was launched from the delta to the VM logistics bases located in the Phu-Doan region. The time saved will enable the air-land base to be set up defensively. Another factor in the freedom of action is to prohibit the VM from taking a long-prepared action on the Na-San defenses, and, in the last phase of the delaying action, to lead Giap to attack almost immediately, without having recognized in detail the solidity of the base. Finally, the very choice of Na-San, by its location 40 minutes flight from the delta's aerodromes (and therefore allowing for sustained aviation action) illustrates the need for freedom of action.
• Second idea: concentrate the French effort and try to dilute the opponent's effort.The materialization of the effort is first perceptible through the involvement of the command: it is clearly the battle of the "genechef", in which he will weigh with all his weight and which he will follow directly from his flying CP or from Hanoi. From then on, an uninterrupted air bridge supplied Na-San, 14 battalions were deployed there, supported by an artillery group and 3 SMLs. The Engineers perform installation wonders with very solidly designed structures, which is facilitated by the laterite soil (very hard) and the abundance of wood immediately available on site. The fire plans and the flanking between structures were also very elaborate, under the impetus of some battalion leaders who had served, as young officers, on the Maginot Line.

As a counterpoint, Operation LORRAINE will have achieved one of its objectives by diverting a reinforced VM regiment from Na-San, while a conspicuous reinforcement of the Laï-Chau base will fix three other VM battalions.

• Third Idea: Maintaining Freedom of Action During the Battle, by the Arrangements TakenThe base is thus designed as a double belt of support points surrounding the runway. The counter-attack reserves (three battalions) were entirely dedicated to this mission, which was the subject of numerous reconnaissance and rehearsals. In addition, it is planned that all support points will be immediately retaken in the event of loss (immediately for the inner belt, from the next dawn for the outer belt). In this logic, the runway must not be closed for more than a few hours.

Let us add to these elements that Giap has at this time neither flak guns, nor 105 (only 75) artillery, and one will understand that, after 3 massive VM attacks, Na-San ends up with an indisputable, great victory: 3.3. 500 enemy dead and wounded for 30 killed and 55 wounded on the Franco-Vietnamese side, and the offensive and logistic potential of divisions 308, 312, 316 sufficiently reached to end the 1952 campaign.

The battle also served as a laboratory, highlighting the interest of the air-land base concept as an answer to the problem of controlling the "great voids" far from the delta. In Salan's mind, however, this concept was only valid through a set of principles:

• the system must be evolutionary, from the hedgehog to the basis of reappropriation of space according to the threat;
• Above all, it must be "retractable" so as not to immobilise too many forces, bearing in mind that for Salan, the priority remains first and foremost the defence and control of the Tonkin Delta;
• it must be part of a whole that avoids too great an enemy concentration and must therefore be supplemented by other bases and peripheral operations;
• it requires a major air effort and therefore a reinforcement of means by the French government and American aid.

In March-April 1953, the effectiveness of the concept was again demonstrated during the new spring VM offensive towards Laos. The bases of Na-San and Laï-Chau set up about fifteen VM battalions, while Salan made the vacuum in front of divisions 304 and 312, in a sparsely populated region, very poor in sources of supplies; Their offensive would then "run out of steam" and come to "die" on the defenses of two new air-land bases in Luang-Prabang and Plaine des Jarres.

It was at this point that the Commander-in-Chief had the idea of what would then be called the "Salan Archipelago". A triangle of air-land bases Laï-Chau, Na-San, Dien Bien Phû, which, surrounded by an action of the GCMA's maquis [3], had to prohibit any massive offensive towards Laos. The occupation of Dien Bien Phû thus appears in the will left to General Navarre, and will be the object of an appropriation by his staff, but, unfortunately, by neglecting the overall action which accompanied it.

Dien-Bien-Phû seen from the strategic level: a battle that was not really chosen, with the stakes poorly appreciated and with a relative or ambiguous priority.

On his arrival in Indochina, of which he knew almost nothing, General Navarre was left to his own devices: no "war goal" was assigned to him, except to do the best he could with the means at his disposal! Faced with a balanced but worrying strategic situation, he was nevertheless decided to relaunch the action and give a new impetus to the Expeditionary Force by taking the initiative and developing a coherent campaign plan. Navarre's dynamic asserted itself in the raids on Lang-Son and Lao-Kay, then in the destruction of the bases of operation of division 320, stopping an offensive projected on the delta. Its campaign plan(s), largely derived from the Salan plan, planned to contain the adversary in Tonkin and to make an effort in Central Annam for the 1953/1954 campaign; after transfers of responsibilities and sectors to After the transfer of responsibilities and sectors to the Vietnamese army, the effort will be carried back to the North, from the new French reserves thus released, during the 1954/1955 campaign, to break up the entire VM battle group.

The strategic gear

As the 316 Division was once again moving towards Laos, the "North-West question" was a recurring wound in the flank of the Expeditionary Force...especially since the decision was made to evacuate Na-San who, in the eyes of the new command team, was no longer of interest. The defense of Laos thus became a "parallel" strategic objective which was to impose itself on the "genechef", without his being able to obtain the slightest guidance from the government. To cover Luang-Prabang, the question of the occupation of Dien Bien Phû will impose itself within the Navarra General Staff and finally lead to the CASTOR operation on November 20th 1953. More than 4,500 men were deployed in Dien Bien Phû a few days later; the idea of maneuver was then to constitute an anchorage point, a base of political-military operations that should radiate in the Thai country. From then on, the infernal spiral began and Navarre, in spite of all its qualities, allowed itself to be dragged into a succession of mistakes.

• The first breach of principle: a decision-making process biased by the general staff. The head of the ^3rd bureau and his officers only knew Na-San in its final phase and, in particular, did not experience the battle and its preparation. They will draw a distorted analysis from it and consider that this success can be reproduced at Dien BienPhû. As soon as the VM offensive towards Laos is confirmed, the temptation will rise, until it becomes irresistible, to succeed in a "nice move" and break the VM battle corps at Dien Bien Phû, when it actually contradicts the "genechef's" plan. The latter will insensitively let himself be influenced, all the more so since the same general staff will seriously underestimate the enemy's capabilities, despite the accuracy of the intelligence (yes!) which goes back and confirms from the beginning of December the extent of the VM offensive toward Dien Bien Phû. One will thus successively consider that Giap will be unable to maintain more than two divisions in the Haute-Région, that his artillery consumption will remain limited to a few thousand shells, finally that the terrain prohibits a masked deployment of VM artillery which will have no choice but to be silent or immediately defeated. The technical or military arguments leading to this assessment are real, but they were not confronted with an effort of imagination that should have been nourished by what was known about VM determination and the human effort it could produce.

The same lack of imagination affects the conception of friendly modes of action, and, when studying the operation on Dien Bien Phû, one does not know how to use it.the potential offered by the French maquis who, from September to November, controlled most of RP 41, before being repelled by the 316 Division offensive was not taken into account. If Operation Castor had been brought forward by one month, the establishment of the air-land base, combined with the action of the maquis, would have strongly thwarted theIf Operation Castor had been advanced by one month, the establishment of the airbase, combined with the action of the maquis, would have strongly thwarted the 316 offensive and would have made it possible to really play the radiation raids around Dien Bien Phû; moreover, it would have been very difficult for the VM to set up its logistics base in Tuan-Giao.

In Navarra, the status of Dien Bien Phû is therefore evolving: from the initial "stake" to save Laos, the base is becoming the tool of an important success improving the map of possible negotiations, obtained on a line of operations that must nevertheless remain secondary.

• The second departure from the principles is the loss of unity of action and concentration of effort.

By agreeing to engage in a real test of strength in the Haute-Région, when its plan was to simply "parry the blows" in the North and to place the effort of the 1953/1954 campaign in the Centre and South Annam, Navarre is committing itself to a major contradiction, which can be overcome once a clear reorientation of efforts is agreed.

However, priority was maintained over Operation ATLANTE, the culmination of the Navarra Plan, which aimed to destroy the VM settlement in Central Annam and hand over political and military control to the Vietnamese state. Planned in 3 phases (ARETHUSE, AXELLE and ATTILA), the operation will mobilize up to 45 Franco-Vietnamese battalions and 8 artillery groups!

At the beginning of December, the 2nd Bureau ^confirmed the movement of two other divisions and of the heavy division 351 towards Dien Bien Phû and the beginning of the installation of a logistics base at Tuan-Giao, Navarre did not change its judgement and decided not to move. was in fact preparing to engage in an intensive battle elsewhere than at the point of application of its effort for 1953/1954.

• Third infringement of principles: a freedom of action from the outset altered.

The battle that was about to begin was very quickly to be fought under the sign of a compromised freedom of action. The Air Force had expressed strong reservations about this choice, because of the distance of the bases (400 km) which only allows a single selection of the air force.Moreover, the weather is very often unfavourable in the Upper Region. Due to air transport constraints, the garrison is fixed (for its support) at 12,000 men.

The "Salan Archipelago" system held because of the enemy dispersion it implied and, above all, because of the virtual impossibility for the VM to establish a base of operations at Tuan-Giao, which would have been too threatened. The isolation of Dien Bien Phû condemned it to be an adventurous base, in an area where the VM had complete freedom of action to prepare the investment. In December, the lace tightens on the French freedom of action. Attempts to get out around the base come up against a numerous and biting opponent, in difficult terrain and very dense vegetation, and result in failures. The base of operations concept disappears and turns into a besieged entrenched camp.

Until 23 December, the terms of the choice remain open for the "genechef". From this date, the complete encirclement by 4 divisions is achieved. It is no longer possible to evacuate the operational group concentrated in Dien Bien Phû without serious breakage. Navarre has lost its freedom of action and finds itself forced to engage here in the major battle of the campaign.

• The fourth breach of principle is that there is no alternative solution that takes into account the fact that the planned manoeuvre is being called into question, in other words, it is a further obstacle to freedom of action.

General Cogny, commander in North Vietnam, worried about the fate of the base (after having supported the principle) proposed an additional manoeuvre from the delta with several mobile groups: to strike the VM logistic depots in the Yen-Bay area, before the Tuan-Giao logistic base is organized. The operation is possible (100 km from the delta) and relevant since the VM battle group with its guns and vehicles now presents classic logistical vulnerabilities. But Navarre and its staff consider it to be inefficient and potentially costly.

There is still one last chance to dodge the battle and, in doing so, inflict a military failure (by default, but very real in terms of capacity for action) on the VM. In February, 308 division leaves Dien Bien Phû for a raid on Laos (still towards Luang-Prabang) and breaks part of the encirclement. At the same time, Air Force General Fay, CAEMC, visiting Dien Bien Phû, declared that if the place was not evacuated quickly, the garrison was lost and proposed to remain at Hanoi to take charge of the evacuation operation, carried out at night in a week with the entire military and civilian transport fleet requisitioned. Significant losses would have been likely (the last square of the defense had to escape through the bush), but even abandoning some of the equipment, what a tremendous slap in the face to the VM battle corps that ofWhat a tremendous slap in the face to the VM battle group to escape the major part of the French grouping and leave it in the void after having concentrated all its divisions far from its bases, and imposed a colossal effort on the controlled populations, finally for nothing or almost nothing. The window of opportunity was centred between the 20th and 28th February before, returning to forced march after the Geneva conference was announced, the 308 took its place again...

Although he now had very strong doubts about the success at Dien Bien Phû, General Navarre let himself be influenced by the garrison command, who felt that a retreat would be morally disastrous and would deprive us of "breaking" Giap's divisions.

• Fifth infringement of principles: a politico-strategic divorce, which contributes to compromise the freedom of action of the "genechef" and perverts the effects of the campaign.

The first pitfall marking the complete dissociation between the conduct (?) of the war by the government and the strategy applied by the commander-in-chief is materialized by the total absence of clear directives (in spite of repeated requests) concerning the defense of Laos [4]. 4] When in doubt, Navarre considered that it had to ensure it, hence the installation of Dien Bien Phû as a "parasitic" element of its campaign.

The most serious factor resulting from this total failure of the policy in front of its strategic responsibilities remains however the decision taken (and prompted by the French government!) to evacuate the operational group concentrated in Dien Bien Phû.) to deal with the fate of Indochina in a large-scale international conference (in addition to the Westerners and the Indochinese states, Vietminh, China and the USSR will also be present), without consulting Navarre in any way on its appropriateness or impact on operations. As soon as the Geneva conference was announced on February 18th, Dien Bien Phû became, more than ever, for the VM the major, decisive stake, justifying all the sacrifices (and immediately the 308 division returned to Dien Bien Phû). General Navarre is now in a real trap.

Finally, after the defeat...which was still only a tactical defeat, the government will turn it into a strategic defeat by its panic and the psychosis ofabandonment at almost any price that seizes it, against the advice of Navarra, which has kept a cool head and measures the new situation (which is neither irremediable nor desperate).

This politico-strategic divorce has been a terrible fatality for General Navarra, and is only attributable to the inability of the politicians of the moment to express a will and make a decision. Thus, the Minister of War (Pleven), accompanied by the Secretaries of State and other ministers involved in the war, as well as the Chief of Staff of the Armed Forceses (General Ely) and the Chiefs of Army Staff, was in Indochina in February 1954, with virtually full government powers to act. He gathered on the spot what in fact constituted a real council of war, listened to the very well-founded objections formulated by the CEMAT (General Blanc) and the CEMAA (General Fay) on Dien Bien Phû (in this case an artilleryman and an aviator, functions on which the resistance of the base rests) and does not take any decision, does not formulate any opinion...(as for General Ely - chief of staff of the armies! - very uncomfortable on any operational question, he confines himself to a total silence). One can however think that it would have been the occasion to re-evaluate the strategic options!

1] It must be recalled that the Vietnamese Army, set up in 1951, fought alongside the Expeditionary Force, which itself remained strongly "yellowed", all its units comprising 1/3 of natives. It is therefore necessary to speak of "Franco-Vietnamese" forces.

2] Having spent 15 years in Indochina, where he knew all the languages, deeply penetrated by Asian psychology, knowing Giap and his adversary in general well, General Salan is the general officer who understands this war best. He is nicknamed "the Chinese" or "the Mandarin".

3] Group of Mixed Airborne Commandos: the French maquis rely on the populations of the high regions of Tonkin and Laos, allies since the conquest, violently opposed to communism and secular enemies of the Annamites.

4] Laos is now a State admitted to the UN, and the only one to have fully recognized its membership in the French Union, unlike Vietnam and Cambodia, which are much more demanding on the "field" of their independence.