‘Life is a rock but the radio rolled me…’

tumblr_nb88ij9V0e1qivon6o1_1280

The war in Vietnam was often heralded as ‘a new kind of war’, one that reached its awful climax in what James Gibson brilliantly criticised as ‘technowar’.  It had many dimensions, including the reintroduction of chemical warfare (Agent Orange and all the other herbicides) and the development of the ‘electronic battlefield’.

But at the time the US Army made the most of its commitment to what it called ‘air mobility’.

As you can see from this contemporary Army video, the concept was claimed as revolutionary (and, in its way of course, counter-revolutionary).  ‘An entirely new concept of warfare known as heli-borne or air mobile operations has been developed by the United States Army,’ claims the commentator, ‘and has been successfully employed to meet the problems posed by South East Asia’s hostile wilderness and bye enemy who hides there.’

In fact, it wasn’t invented in Vietnam, but it was a dominant mode of army operations there: you can download the US Army’s historical report on Air mobility 1961-1971 here, for example, the Vietnam Center and Archive has a helpful page on ‘The helicopter war’ here, and you can read an extract from Walter Boyne‘s How the helicopter changed modern warfare here.  There is also a considerable literature on ‘dust-off‘ that I’m working through for my new research project on casualty evacuation in war zones.

Airmobility 1961-1971

And don’t forget its role in popular culture.  The helicopter loomed large in the iconography of Francis Ford Coppola‘s Apocalypse Now (1979):

apocalypsenowbdcap4_original

More recently Pete Adey provided a summary of a more general concept of ‘aeromobilities’ [Geography Compass 2/5 (2008) 1318-1336] though his emphasis, perhaps not surprisingly, is on vision.

What interests me here, though, is another capacity, and one without which the potential of air mobility would have remained unrealised.

I’m talking about the voice on the radio.  I’ve written about the role of the forward air controller before, and the parallels between the air strikes they directed in Vietnam and today’s remote operations in Afghanistan (see ‘Lines of descent’, DOWNLOADS tab).

GREGORY From a view to a kill Shock and Awe Extract.001

This destructive power was captured with extraordinary economy by Phil Caputo in A Rumor of War:

Simply by speaking a few words into a two-way radio, I had performed magical feats of destruction. Summoned by my voice, jet fighters appeared in the sky to loose their lethal droppings on villages and men. High-explosive bombs blasted houses to fragments, napalm sucked air from lungs and turned human flesh to ashes. All this just by saying a few words into a radio transmitter. Like magic.

But the radio was part of a much wider network of military violence and military logistics in Vietnam.  Here is Frederick Downs in The killing zone:

With the radio, we grunts could make use of modern weapons. Without it, everything stayed put. We used the radio to call in artillery, naval gun support if it was close enough, air strikes, gunships, dustoffs, Puff the Magic Dragon [the AC-47 gunship], mortars, tanks, APCs and other rifle platoons. The radio kept us supplied. One day our order went in; the next day the chopper flew out with a delivery. We found each other by using grid coordinates and radioing them back and forth. A pilot knew he had the right location when we popped smoke and he identified it over the radio. By this method, we received C rations, ammo, new weapons, grenades, parts for our equipment, shoes, new clothes, underwear, socks, medicine, personnel replacements, beer, iodine tablets for use in the water, mail, and once in a while even a chaplain. To complete the cycle, the radio was used to extract us from trouble. Saving a life was often a matter of seconds. The radio was also a comfort at night. The periodic radio checks assured us that friends and help were always near.

Here too, incidentally, there are lessons for contemporary analysis: satellite imagery is clearly of vital significance for today’s advanced militaries, but so too are satellite communications.  I’ll discuss this is another post, but without those communication links there would be no full-motion video-feeds from all those Predators and Reapers – and their operating range would be dramatically constricted.  Ground operations in Afghanistan and elsewhere would also be virtually impossible without their radio links.

The-new-RTO

From 1965 the main field radio in Vietnam was the PRC-25 (‘Prick 25’).  You can find a detailed technical specification here, but here are the key passages:

‘The PRC-25 was about the size and weight of a case of soda. With its battery “can” included, call it a case of soda sitting on top of a six-pack. (It actually weighed slightly more than that, 23.5 pounds) There was a handle on each side at the top to carry it. The radio consisted of two parts, both in metal boxes, called “cans.” The upper can held the radio itself, the lower can held its battery pack. Metal buckles held the two together. The radio was tough and would easily survive a 50 foot fall from a helicopter onto a metal-planked runway. You could throw the whole thing in the water for an hour, completely submerged, then pull it out and expect it to work…

‘The radio antenna was exactly like a metal tape measure, but the bottom foot or so was a round flexible tube that screwed into the radio. There were actually two antennas, a regular one and a long-range antenna, carried in a canvas bag strapped to the side of the radio The radio had a transmission range, with the short antenna, of about 3-4 miles, but various terrain factors could influence this, of course. It helped to be higher up. The long range antenna was supposed to be good for up to 18 miles.

The rule of thumb was that the battery was good for about a day of casual operation, listening mostly, some occasional transmissions. In a period of intense use, transmitting/receiving all the time, it was good for perhaps 2-3 hours. The way the LRRPs [Long Range Reconnaissance Patrols] and SF [Special Forces] used it, shutting it off and only coming up at scheduled times to briefly transmit or listen, it was good for perhaps four days. Spare batteries were usually kept in a spare .30 caliber ammo can. When expended, the battery pack had to be physically destroyed. Inside were flashlight-type batteries which the Viet Cong could use in booby traps or to ignite bombardment rockets.’

Hoffman Humping HeavyNotice first the extraordinary weight of the thing.  I’ve detailed the loads humped by soldiers and Marines in Vietnam before, but you can see from this that the radio operator (RTO) was even more heavily burdened.  He also had to contend with a difficult load distribution: according to Rodger Jacobs ‘radiomen had to wear their radios on their chests because if worn on their backs the thickness of the jungle and the vines would constantly catch on the controls and change the frequencies.’  The best account of the trials and tribulations of an RTO that I know is Phillip Hoffman‘s appropriately titled Humping Heavy (right).

RTO2

Then notice the size of the aerial (above); RTOs carried a ten-feet rigid mast in sections but much of the time used a three-feet whip antenna.  This made the RTO extremely vulnerable: not only was he a marked target, but he was always close to the platoon or patrol commander and so both were hi-vis priorities in an enemy attack or ambush.   For that reason the most prized possession of many RTOs was a North Vietnamese Army rucksack: ‘They’re better than anything the Corps has,’ Jeff Kelly was told.  ‘It’ll hang lower on your back and won’t catch on branches. But the big thing is you won’t be giving off that radioman silhouette.’  Most RTOs taped the antenna down, but Hoffman made the mistake of questioning the wisdom:

Right away he demanded I stuff the flexible, three-foot whip antenna down my shirt to limit me (and by extension him) as a target. He knew a priority of the enemy was to knock out communication, and our commo was located on my back. I complied with his directive but made the mistake of telling him our signal strength would suffer. In no uncertain words he told me never to question him again.

Even with a network of relay stations and airborne retransmissions, communications were uneven and intermittent: the terrain could block signals, especially in the Central Highlands, and rain (especially during the monsoon) could play havoc with reception.  At night even a whisper was dangerous; here is John Edmund Delezen:

Hourly situation reports are sent to the radio relay atop of Hill 950 some three kilometers north of the Khe Sanh airstrip. The “sit-reps” are not sent in the form of words-we dare not speak in the black void; when the relay asks us to acknowledge his call, there are just the two distinct pauses in the constant squelch as the handset is keyed twice. The two small audible clicks are all that connects us with the world, and all that assures the relay that we have not disappeared into the liquid black night.

Artillery fire direction center Vietnam

It could be strangely remote for those receiving the transmissions too (above, an artillery fire support center).  Kenneth Sympson, an artillery officer, explained:

‘Our only contact with the men of the patrols was from radio transmissions—the infrequent call-in at a checkpoint or request for fire on some pretargeted location on their route. They were a noise on the radio and a trace on a map of the region. When you fired an artillery round in support, it was almost as if you were simply throwing it into the night and it just disappeared. It would later strike a place on the map, but there was no life there; there was only some representation for crossing trails or the contour lines indicative of the slope of a hill or a pin hole named Registration Point 3.’

Downs says much the same, describing gun crews anticipating ‘the release of their impersonal death into a grid square.’

But for those beyond the wire those staccato messages were far from abstract or impersonal.  ‘The radio was our link with literally everything outside our platoon,’ said Downs, ‘from supplies to survival’.  And without it, as he also said, ‘everything stayed put.’

Hence my title: and for those too young (or too old) to remember it, listen to this (YouTube).

‘Stop, hey, what’s that sound?’

I’ve used this Buffalo Springfield song before

there’s something happenin’ here
what it is ain’t exactly clear
there’s a man with a gun over there…

stop, hey, what’s that sound?

– and I’ve posted about the sound of war before too, here and here.  This post is connected to both of them, but it also follows directly from my more recent post on bodies

McCARTHY CAn age ago Trevor Barnes recommended Tom McCarthy‘s novel C to me, and I’ve been re-reading it these last few days. There’s an extraordinary passage where McCarthy’s protagonist Serge, an observer with the Royal Flying Corps during World War I, is being driven to Nieppe when his truck  detours ‘to drop off some piano wire’ to a special unit in the woods north of Vitriers.

Inside the main [hut], he finds a huge square harp whose six strings are extended out beyond their wooden frame by finer wires that run through the hut’s air before breaching its boundary as well, cutting through little mouse-holes in the east-facing wall. In front of the harp, like an interrogation lamp, a powerful bulb shines straight onto it; behind it, lined up with each string, a row of prisms capture and deflect the light at right angles, through yet another hole cut in the hut’s wall, into an unlit room adjoining this one. There’s a noise coming from the adjoining room: sounds like a small propeller on a stalled plane turning from the wind’s pressure alone. “What is this place?” Serge asks. “You’re an observer, right?” the slender-fingered man says. Serge nods. “Well, you know how, when you’re doing Battery Location flights, you send down K.K. calls each time you see an enemy gun flash?” “Oh yes,” Serge answers. “I’ve always wondered why we have to do that …” “Wonder no more,” the man says with an elfin smile. “The receiving operator presses a relay button each time he gets one of those; this starts the camera in the next room rolling; and the camera captures the sound of the battery whose flash you’ve just K.K.’d to us. You with me?” “No,” Serge answers. “How can it do that?” “Each gun-boom, when it’s picked up by a mike, sends a current down the wires you just pissed on,” the man continues, “and the current makes the piano wire inside this room heat up and give a little kick, which gets diffracted through the prisms into the next room, and straight into the camera.” “So you’re filming sound?” Serge asks. “You could say that, I suppose,” the man concurs.

Sound ranging traces

Still puzzled, he follows the man into another hut:

The interior’s suffused with red light. At a trough propped up against the far wall, a man with rolled-up sleeves is dunking yards of film into developing liquid, then feeding it on from there into a fixing tank. As the film’s end emerges from this tank in turn, he holds it up, inspects it and tears off sections, clipping these with clothes pegs to a short stretch of washing line, from where they drip onto the discarded strips on the room’s floor below them. “Yuk,” Serge whispers beneath his breath. “What?” the slender-fingered man asks. “Nothing,” he replies. “Look here,” Serge’s guide says, unclipping a strip of the developed film and pointing at dark lines that run, lengthways and continuous, along its surface. The lines—six of them—are for the most part flat; occasionally, though, they erupt suddenly, and rise and fall in jagged waves, like some strange Persian script, for half an inch, before settling down and running flat again. On the film’s bottom edge, beside the punch-holes, a time-code is marked, one inch or so for every second. The jagged eruptions appear at different points along each line: staggered, each wave the same shape as the one on the line below it, but occurring a quarter of an inch (or three-tenths of a second) later. “So,” Serge’s elfin guide continues, “these kicks are made by the sound hitting each mike; and they get laid out on the film at intervals that correspond to each mike’s distance from the sound. You see them?” “Yes,” Serge answers. “But I still don’t—” “These ones ready to take through?” the guide asks the developer. The other man nods; with his piano-player’s fingers, the guide unclips the other drip-dried strips, then leads Serge out to yet another hut. This one’s wall has a large-scale map taped to it; stuck in the map in a neat semi-circle are six pins. Two men are going through a pile of torn-off, line-streaked film-strips, measuring the gaps between the kicks with lengths of string; then, moving the string over to the map slowly, careful to preserve the intervals, they transfer the latter onto its surface by fixing one end of the string to the pin and holding a pencil to the other, swinging it from side to side to mark a broad arc on the map. “Each pin’s a microphone,” the slender-fingered man explains. “Where the arcs intersect, the gun site must be.” “So the strings are time, or space?” Serge asks. “You could say either,” the man answers with a smile. “The film-strip knows no difference. The mathematical answer to your question, though, is that the strings represent the asymptote of the hyperbola on which the gun lies.”

Serge has stumbled into what Peter Liddle called ‘the Manhattan Project of the 1914-18 war’: sound ranging.  It was a technique first developed by the French and German militaries  – a German sound ranging analysis section is shown below – taken up (warily) by the British (Berton claims their  old-school gunners at first thought it radical nonsense) and refined by the Canadians. As McCarthy’s brilliant reconstruction shows, it involved using sound to locate enemy artillery batteries.

The usual configuration was to have six ‘Tucker’ microphone stations at carefully surveyed intervals along an arc 4000 yards behind the front line with two observation posts in front of them, all linked to a recording station in the rear by 40 miles of wire.  When the observers saw a gun flash or heard its boom they sent a signal that activated the oscillograph and film recorder.  In the course of 1916 the British established eight of these sound-ranging sections, each plotting battery positions on base maps supplied by ‘Maps GHQ’.  In ideal conditions (which were rare) the operation could be completed for a single battery within three minutes (using graphical rather than computational methods) and with an accuracy of 25-100 yards (for more, see J.S. Finan and W.J. Hurley, ‘McNaughton and Canadian Operational Research at Vimy’, Jnl. of the Operational Research Society 48 (1) (1997) 10-14).

Analysis section of German sound ranging troop 1917

The single best source on this – though the title sounds like Flash Gordon – is John Innes‘s Flash spotters and sound rangers: how they lived, worked and fought in the Great War (1935), but Pierre Berton‘s Vimy (1986) has some useful summary pages on the Canadian role in its development. There is also a truly excellent survey that places sound-ranging in the wider context of the ‘battlefield laboratory’ in Roy MacLeod, ‘Sight and sound on the Western Front: surveyors, scientists and the “battlefield laboratory”, 1915-1918’, War & Society 18 (1) (2000) 23-46.  For more technical discussions, Peter Chasseaud‘s Artillery’s astrologers: a history of British survey and mapping on the Wester Front 1914-1918 (1999) is a key source, but there is also an exemplary (short) explanation here, from which I’ve borrowed the simplified summary diagram below.

Sound ranging

The Germans were evidently impressed by the efficacy of the Allied system, as this directive issued by General Ludendorff shows:

According to a captured English document the English have a well- developed system of sound-ranging which in theory corresponds to our own. Precautions are accordingly to be taken to camouflage the sound: e.g. registration when the wind is contrary, and when there is considerable artillery activity, many batteries firing at the same time, simultaneous firing from false positions, etc. The English have an objective method (self-recording apparatus). It is important to capture such an apparatus. The same holds good on the French front.

Since most of the killing and destruction was the work of the artillery, these counter-battery operations were a crucial part of the ground war.  Multiple sources were used, including aerial reconnaissance, balloon observation and flash spotting as well as sound ranging, and when combined these could eventually locate enemy guns to within 5-25 yards.  This map of German artillery intelligence for Vimy in March 1917 shows something of the power and complexity of this ‘acoustic cartography’:

German artiller intelligence map Enemy batteries known with certainty to be firing and spotted 15 to 22 March 1917 Vimy

Maps showing the location of enemy batteries (‘Positions maps’ like the British one below) were issued on a regular and eventually even a daily basis.

Positions map. December 1917-January 1918

This was all part of a more general metricisation of the battlefield – hideously appropriate for the killing machine that was industrialized warfare – which you can also see in the ‘barrage maps’ that choreographed the moving curtain of artillery fire behind which troops were supposed to advance on the enemy lines.  John Keegan is very good on what he called ‘the mathematics of the barrage’ in his The Face of Battle, and from his aircraft high above No Man’s Land, Billy Bishop described it as ‘clockwork warfare’:

The waves of attacking infantry as they came out of their trenches and trudged forward behind the curtain of shells laid down by the artillery, were an amazing sight. The men seemed to wander across No Man’s Land, and into the enemy trenches, as if the battle was a great bore to them. From the air it looked as though they did not realise that they were at war and were taking it all entirely too quietly. That is the way with clock-work warfare. These troops had been drilled to move forward at a given pace. They had been timed over and over again in marching a certain distance, and from this timing the “creeping,” or rolling barrage which moved in front of them had been mathematically worked out.

BARRAGE MAP bataille-ypres-passchendaele-carte-secret

Here’s part of the barrage map for Vimy, which shows the calibration even more clearly:

Vimy barrage map 1917 (extract)

McCarthy captures this ‘clockwork’ movement perfectly when Serge looks down from his aircraft while spotting for artillery:

As the second-hand needle moves across the final quarter-segment of his watch’s face, Serge feels an almost sacred tingling, as though he himself had become godlike, elevated by machinery and signal code to a higher post within the overall structure of things, a vantage point from which the vectors and control lines linking earth and heaven, the hermetic language of the invocations, its very lettering and script, have become visible, tangible even, all concentrated at a spot just underneath the index finger of his right hand which is tapping out, right now, the sequence C3E MX12 G … Almost immediately, a white rip appears amidst the wood’s green cover on the English side. A small jet of smoke spills up into the air from this like cushion stuffing; out of it, a shell rises. It arcs above the trench-meshes and track-marked open ground, then dips and falls into the copse beneath Serge, blossoming there in vibrant red and yellow flame. A second follows it, then a third. The same is happening in the two-mile strip between Battery I and its target, and Battery M and its one, right on down the line: whole swathes of space becoming animated by the plumed trajectories of plans and orders metamorphosed into steel and cordite, speed and noise. Everything seems connected: disparate locations twitch and burst into activity like limbs reacting to impulses sent from elsewhere in the body, booms and jibs obeying levers at the far end of a complex set of ropes and cogs and relays. The salvos pause; Serge plots the points of impact on his clock-code chart, then sends adjustments back to Battery E, which fires new salvos that land slightly to the north of the first ones. 

There’s a lot more to say about this, which I plan to do in the full presentation, but what interests me at least as much is the way in which this precision (Gabriel’s ‘order and reasonableness’) – so clear and crystalline when plotted on maps or seen from the air – was confounded on the ground, not least by the shattered landscape of craters, trenches and barbed wire and by the vile agency of what Siegfried Sassoon called the ‘plastering slime’ that clawed and dragged at the soldiers’ bodies and which was erased from what Edmund Blunden called the ‘innocuous arrows’ and ‘matter-of-fact symbols’ of the maps and aerial photographs.

More particularly, I’m interested (here) in the production of an altogether more sensuous soundscape, part of the corpography that danced such a deadly gavotte with cartography that I sketched out in my previous post.  In Touch and intimacy Santanu Das argues that

‘The mechanised nature of the First World War severed the link between sight, space and danger, a connection that had traditionally been used to structure perception in wartime.  This disjunction resulted in an exaggerated investment in sound.’

The memoirs, letters and diaries from the Western Front confirm that the hideous noise of battle worked its way inside the very body of the soldier.  I’ve got pages of examples, but here is just one, Edward Lynch in Somme Mud (and, for that matter, in Somme mud):

‘The shells are missing us by a matter of yards.  Noise is everywhere.  We lie on the shuddering ground, rocking to the vibrations, under a shower of solid noise we feel we could reach out and touch.  The shells come, burst and are gone, but that invisible noise keeps on – now near, now far, now near, now far again.  Flat, unceasing noise.’

I’ve emphasised the passage that simply resonates with what I described previously as a haptic geography.  And so, not surprisingly, the same sources also show that, just as the landscape was inhabited, so too was knowledge of it; that knowing those deadly sounds – ‘knowing the score’, I suppose – was a vital part of staying alive.  Here is an extended passage from A.M. Burrage‘s War is war:

We are becoming acclimatised to trench warfare. We know by the singing of a shell when it is going to drop near us, when it is politic to duck and when one may treat the sound with contempt. We are becoming soldiers. We know the calibres of the shells which are sent over in search of us. The brute that explodes with a crash like that of much crockery being broken, and afterwards makes a “cheering” noise like the distant echoes of a football match, is a five-point-nine. The very sudden brute that you don’t hear until it has passed you, and rushes with the hiss of escaping steam, is a whizz-bang. For a perfect imitation of a whizz-bang, sit by the open window of a railway compartment and wait until an express train passes you at sixty miles an hour. The funny little chap who goes tonk-phew-bong is a little high-velocity shell which doesn’t do much harm. “Minnies” and “flying pigs” which are visible by day and night come sailing over like fat aunts turning slow somersaults in mid-air. Wherever one may be, and wherever they may be going to drop, they always look as if they are going to fall straight on top of one. They are visible at night because they have luminous tails, like comets. The thing which, without warning, suddenly utters a hissing sneeze behind us is one of our own trench-mortars. The dull bump which follows, and comes from the middle distance out in front, tells us that the ammunition is “dud.” The German shell which arrives with the sound of a woman with a hare-lip trying to whistle, and makes very little sound when it bursts, almost certainly contains gas.

 We know when to ignore machine-gun and rifle bullets and when to take an interest in them. A steady phew-phew-phew means that they are not dangerously near. When on the other hand we get a sensation of whips being slashed in our ears we know that it is time to seek the embrace of Mother Earth.

The end of the war was apprehended in both registers, and this frontispiece from Benedict Crowell’s Demobilization (1920) shows how an American sound-ranging station captured the moment the guns fell silent:

The end of the war: 11 a.m. on 11 November 1918

‘Stop, hey, what’s that sound?’