The mainforce war has certainly attracted plenty of argument; but this has tended to be about morality or strategy rather than about tactical efficiency. In a curious way both sides in the Vietnam debate seem to agree that in battle the American forces were very effective indeed. Both supporters and opponents of the war stand in awe of the mobility, firepower and futuristic electronic instruments which were used. Perhaps it is true that the enemy was hard to find, or to distinguish from friendly civilians; but once found there seems a wide consensus of opinion that he could be dispatched very quickly. In this sense the mainforce war has been surprisingly uncontroversial, although it is possible that the consensus itself is misplaced. We will return to examine its basis in greater detail at a later point.
The four wars in Indo-China, 1965–73.
The mainforce war has also been less controversial than the village war because it was fought largely in uninhabited border areas.2 It spilt over into only relatively few of the heavily populated provinces; mostly in the northern quarter of the country and in the Mekong Delta. Only in the Tet Offensive and in some of the battles of 1972 did it enter the towns. In the main it had the character of conventional jungle warfare between two regular armies, with few of the complexities occasioned by the presence of civilians.
In both the mainforce and village wars the communists employed guerilla tactics; fighting a war without fronts, moving in small groups and using all the vegetal and human camouflage they could. This similarity of tactics has sometimes led commentators to imply that the two wars were essentially the same; but in fact they were not. The village war was a local affair, run mainly by the Viet Cong and depending heavily upon Mao Tse Tung’s principles of protracted revolutionary war. The mainforce war, by contrast, was an externally organised attack by the North Vietnamese Army. It depended far less upon the dynamics of political subversion than upon the traditional military principles of regular armies. For the village fighter there was essentially no alternative to fighting as a guerilla. It was inherent in his relationship with his village and his local political cell. For the mainforce fighter, by contrast, ‘guerilla warfare was a tactic, not a genre’.3 It was a military stratagem adopted by choice, which could be abandoned, if circumstances were appropriate, in favour of more overt or close-order fighting.
If we take the mainforce war in isolation, therefore, we will be reflecting an important reality. We will also be following in the footsteps of General Westmoreland himself, since he seems to have done his best to keep American troops away from the pacification effort in the villages. His ideal was to concentrate the Americans for large-unit battles in the uninhabited border areas, while the South Vietnamese forces looked after the ‘internal security’ of their own people. He thus decreed that there should be ‘two wars’ within South Vietnam, and not just one.
The rationale behind Westmoreland’s strategy was twofold. In the first place it inserted a screen or ‘shield’ of American troops between the Viet Cong in the coastal villages and their NVA colleagues coming in through Laos and Cambodia. This screen would ‘hold the ring’ for the South Vietnamese pacification effort in the villages. Secondly, Westmoreland aimed to inflict prohibitive attrition upon the NVA so that it would be dissuaded from further adventures in the south. For political reasons the Americans were by and large prevented from making decisive or direct attacks upon NVA bases beyond the frontiers of South Vietnam itself; so the only option remaining was the much less decisive one of attrition. By killing as many NVA soldiers as possible within South Vietnam, it was hoped that the northern leaders could eventually be persuaded of the futility of their campaign, up to the point at which they would call it off.
In the event Westmoreland’s policy seemed almost to have been mirrored by the NVA’s own pattern of operations. As the Americans built up their strength between 1965 and 1968, so did the NVA. Northern formations grew in both size and sophistication, and sought to fight big battles of an ever more conventional nature. They still tried to pass reinforcements to the village guerillas; but this increasingly became secondary to the task of fighting big battles in the border areas. It was a prime aim of the NVA to inflict as many casualties as possible on the Americans, and they were quite ready to meet Westmoreland’s mainforce challenge with a mainforce of their own.
It is arguable that Westmoreland’s emphasis upon the mainforce war was a cardinal error, insofar as it diverted American energy and resources away from the ultimately critical task of building up the South Vietnamese. The village war tended to be neglected, and it has been calculated4 that less than 10% of total American resources were allocated to programmes relating to it. It was only after the Viet Cong had suffered heavy casualties in the 1968 Tet Offensive that progress could be made in this field. Tet shocked the South Vietnamese into setting their house in order, so for the first time some of the vital but much-misunderstood lessons of the Malayan Emergency came to be applied to pacification. Another result of Tet, however, was to precipitate American withdrawal. This meant that just as they seemed to be making genuine progress, the Americans had to change gear and belatedly hand over the mainforce war to the South Vietnamese. Despite early signs that this could be successfully achieved, it was eventually shown to be an unrealistic policy. The final victory of the North in 1975 was the result of a South Vietnamese collapse in the mainforce battle, following denial of US support.
The reason why the Americans failed to build up South Vietnamese forces in good time was that they were distracted by their own mainforce war. This war was far from a walkover, and we may be entitled to question whether the supposed battlefield supremacy of the Americans was really as great as is usually assumed. It did, after all, take them over three years to win their mainforce victory. It was only around 1969–70 that they had genuinely defeated the NVA inside South Vietnam. Surely this is considerably longer than they had hoped, much longer than they predicted at the time, and longer than is normal in conventional wars elsewhere. What went wrong?
There are many structural factors which the Americans can invoke to explain their relatively long delay in mastering the enemy mainforce. It took much longer than they had predicted to build up their logistic and support facilities, and it took longer than they had hoped to obtain the release of sufficient combat units. There were a number of other political decisions which may also have impaired efficiency – e.g. the one year tour of duty, the refusal to send reservists to Vietnam, and the insistence on a high standard of comfort in the rear base areas. The North Vietnamese, for their part, continued to send large formations south long after they had been expected to desist. Although they never remotely approached the total numbers of free-world troops in South Vietnam, they could often field as many men in mainforce manouevre battalions. It has been calculated that the true ratio of ‘foxhole strength’, excluding logistic or auxiliary troops, fluctuated between 1.2 and 0.7 free-world soldiers to every one NVA.5 The Americans and their allies did not therefore enjoy a significant numerical superiority in the mainforce battles.
It is usual to claim that despite all these deficiencies and delays, American forces nevertheless exhibited a devastating ability to smash enemy formations wherever they were found. Both sides in the debate seem to accept that, as far as the purely mainforce battles were concerned, the Americans had more than adequate skill and resources; their tactical victory was a foregone conclusion.
The statistics seem to speak for themselves. Over three times the tonnage of bombs was dropped in Vietnam as in the Second World War.6 More armoured vehicles were deployed in South Vietnam alone than the French had deployed in the whole of Indochina.7 Where the French had eventually been able to call upon a total of 42 helicopters, the Americans deployed about one hundred times as many. Finally, and perhaps most tellingly of all, the Americans seem to have achieved a ‘kill ratio’ in their battles of between six and ten NVA soldiers to every one American (and by 1972, when the American contribution was largely aerial, a ratio of 34:1 was claimed8).
All this appears to be a conclusive proof that, once they got into their stride, the Americans were incomparably and overwhelmingly superior in tactics. There seems to be no room for doubting that the mobility, firepower and electronic surveillance advantages of a modern mechanised force can destroy a primitive infantry opponent at will.
This argument is persuasive, but on reflection we should perhaps exclude ‘electronic surveillance’ from the list of important advantages which the American war machine enjoyed in Vietnam. It is true that they deployed many impressive devices in this field during the war, and that they advanced in leaps and bounds as the war continued. The ‘McNamara Wall’ project to construct a barrier of unmanned sensors along the Demilitarised Zone, for example, stands as a monument to the accelerated research and development of which a rich technological power is capable. It was almost comparable to the ‘Manhattan’ programme of the Second World War, in its sheer speed of achievement and massive mobilisation of brainpower and industrial capability. In little more than one year an entirely new weapon system was conceived, developed and put into production. The so-called ‘electronic battlefield’ was converted almost overnight from a cottage craft into a massive industry. It developed its own enthusiasts, its own logistic base, and most important of all, its own bureaucracy and interest-group which could guarantee its continuity. It took its place alongside the ‘armour’, ‘airmobility’, and ‘artillery’ lobbies as an arm of the service. In practice, however, the achievements of the electronic battlefield were as patchy and uncertain as had been those of the tank in the First World War. It suffered from being in its first, experimental generation.
Surveillance was always of paramount importance to the war in Vietnam, since in fighting against a regular army which opted to use guerilla tactics it was always very difficult to find the enemy. Even the best-equipped and most tactically efficient defensive force was powerless against an enemy it could not see. The same problem faced the American main forces in Vietnam which had faced defensive airforces before the advent of radar, or anti-submarine flotillas before the invention of sonar. In all these cases the enemy was able to make undetected raids which could be countered only after the damage had already been done.
There were two main limitations to visibility for the free-world forces in Vietnam. The first was the dense jungle and forest terrain through which the enemy could move almost at will, concealed by his remoteness from free-world bases, and by the canopy overhead. The second was the darkness of the night. It was a maxim in this war, as in the French war in Indochina, that the defenders controlled the daylight hours; but the insurgents controlled the night. Both of these factors imposed enormous frustrations upon the American striking forces, who could venture only cautiously into the jungle, and formed ‘Night Defensive Positions’ or ‘Custer Rings’ during the hours of darkness.
To meet the challenge of the jungle, one answer was to physically remove the jungle itself. A proportion of the chemical defoliation operations had this objective, while many experiments were carried out on the ground to find a mechanical technique which would be neither as controversial nor as complicated. In the event neither fire nor gigantic instruments such as the ‘Le Tourneau Tree-crusher’ (or ‘Transphibian Tactical Crusher’) were found to be as satisfactory as a simple bulldozer with a specially shaped and reinforced blade known as the ‘Rome Plow’. The army found the Rome Plow to be a convenient method of clearing jungle even though its operation was costly in both casualties and mechanical maintenance. Six Rome Plow companies were eventually deployed, and they made extensive clearances along roads and in enemy base areas such as Warzones ‘C’ and ‘D’ north of Saigon.9 Their cuts afforded greater route security and wider fields of fire, although cynics commented on the speed with which the jungle grew again, and on the fact that a cleared road deprived a convoy of its own cover as much as the enemy’s. By and large land-clearing did offer a genuine answer to the problem of jungle visibility; but it was slow and confined to relatively small areas.
A second solution to the problem of visibility seemed to lie in the use of aerial surveillance. At the start of the war great hopes were placed in the OV-1 Mohawk surveillance aircraft, a relatively quiet fixed-wing machine which carried advanced surveillance equipment including side-looking radar and infra-red scanners. These aircraft could patrol wide areas in a night, and their early use seemed to promise great things. They suffered, however, from several problems.
The Mohawk’s mechanical reliability was often less than perfect especially since payloads tended to be increased above the designed limits. The processing speed of information gathered by the aircraft also turned out to be rather longer than originally hoped. This meant that its immediate tactical ‘responsiveness’ was impaired, and it came to be regarded as more a strategic than a tactical asset. The commander of the First Cavalry Division (Airmobile) felt that the Mohawk should have been deployed at Corps level instead of absorbing Divisional resources.10 It was retained at Divisional level, perhaps, because for institutional reasons the army was very anxious to demonstrate its organic need for this aircraft, in the face of strong airforce criticism. The Mohawk was for a long time an important bone of contention between the two services, and this to some extent distorted its practical employment.
Another aerial surveillance platform which was deployed early in Vietnam was the ‘flareship’ or ‘fixed wing gunship’ – a cargo aircraft fitted with flare dispensers and electrical Gatling guns. These aircraft were in relatively short supply, but they were highly successful and underwent many improvements as the war progressed. By 1972 a C-130 variant was in service with a 105mm howitzer mounting, as well as auxiliary armament and numerous surveillance aids. Their employment, however, was less for reconnaissance than for supporting fire. They would typically be called upon to provide artificial daylight and suppressive fire around the perimeter of a unit beleaguered at night. They were therefore used to assist in the ground battle once it had started, rather than to find the enemy in the first instance. Over jungle, moreover, their vision was very limited indeed.11
Perhaps the most effective aerial reconnaissance was provided by the light helicopter scouting teams from the so-called ‘air-cavalry’. These relied at first entirely upon the naked eye, so their usefulness was restricted to relatively clear terrain and daylight. As the war continued, however, their very high tactical flexibility was increasingly appreciated, and attempts were made to extend their vision into the hours of darkness. Especially in the Ninth Division in the delta, but also elsewhere, experimental searchlight attachments (both white light and infra-red) were rigged and image-intensifiers carried. These fittings helped to remove the cloak of darkness from the enemy, while a chemical ‘people-sniffing’ device could in some circumstances detect a human presence through cover. It was unreliable, however, and fairly ineffective over heavy jungle or in wet weather.
The use of these surveillance aids in light helicopter teams did give a commander additional means of seeing into the night; but they were by no means a complete answer and were not deployed in every operation. They had been unavailable during the early days of the war, and came gradually to the fore only around 1967–8. It is therefore misleading to think of them as a really powerful influence upon the course of the war as a whole, particularly over jungle terrain.
1967 was also the year of the great crash programme to develop unmanned sensors for the ‘McNamara Wall’. These were to be dropped by aircraft into the jungle, from where they would send back radio messages if anyone moved near them. Artillery fire or air bombardment could then be placed upon the intruders, or ground forces sent out to intercept them. This system had the great advantage that it at last offered a method of penetrating below the jungle canopy, and could ‘watch’ wide areas without the need for a garrison permanently on the ground. After three years of war in which the Americans had often been unable to find their enemy, the unmanned sensor seemed to promise a tactical revolution.
The or
iginal scheme had been to deploy the McNamara Wall along the whole length of the Demilitarised Zone; but this plan was changed as the siege developed around the Khe Sanh Combat Base in early 1968. The sensors were hastily redeployed around the base and enemy movements beyond the garrison’s visibility started to be monitored. The NVA had about two divisions lurking in the area, and the sensors made a contribution to locating them precisely enough for bombardments to be brought down upon their heads. In this role the sensors had the effect of extending the lethal range of the defenders’ guns: the cleared glacis around a fortification no longer stopped at the jungle edge.
The early sensors at Khe Sanh suffered from many operational defects, although greater reliability was achieved in subsequent models. Many different sensors gradually became available and were deployed in other areas, particularly for the strategic interdiction of the Ho Chi Minh trail through Laos and Cambodia (the ‘Igloo White’ programme). In a more strictly tactical role they continued to be used as an auxiliary in the defence of firebases, as they had been at Khe Sanh, and also for ‘mechanical ambushes’ along trails thought to be used by the enemy. Of particular note was the line of such traps laid out and maintained in 1970 around Loc Ninh by the 11th Armoured Cavalry Regiment. This made an apparently very effective contribution to the interdiction of enemy supplies, although it is worth noting that the mechanical ambushes themselves needed fairly constant patrolling on the ground, in order to check results.12 They were not a complete substitute for ground forces, but really only an auxiliary.
Enthusiasts of the unmanned sensor delight in recounting the tale of FSB Crook, a firebase constructed deep in enemy territory with the specific intention of provoking attack. The base was ringed with sensors and well supported by artillery and aircraft. Everyone hoped that the unfindable enemy would oblige by attacking the base; and sure enough he did so between 5th and 7th June 1969. Early electronic warning helped the defenders to massacre about 400 enemy soldiers for the loss of only one American.13 The whole idea of battlefield sensors appeared to have been triumphantly vindicated.
Forward into Battle Page 21