Journey of a wounded soldier

I’ve written before about Harry Parker‘s Anatomy of a soldier – an extraordinary novel(for multiple reasons) that reconstructs the journey of a British soldier who steps on an IED in Afghanistan through the evacuation chain to Camp Bastion and on to Selly Oak in Birmingham (see also ‘Object Lessons’, DOWNLOADS tab).  I’ve also sketched out an ‘anatomy of another soldier‘, describing in similar terms the precarious journey of a soldier wounded on the Western Front in the First World War back to Blighty.  It’s part of my project on medical care and casualty evacuation from war zones – the Western Front, the Western Desert, Vietnam, and now Afghanistan, Syria and Iraq.

Much of my archival work (on the First and Second World Wars) has been done at the Imperial War Museum and the Wellcome Library in London, and now the IWM has provided a series of short but sharp insights into the journey of a wounded soldier from Afghanistan back to Britain.

It’s not the experiment that Harry conducts – which isn’t to disparage either of them, and in fact Harry did a reading from ‘Anatomy’ at the IWM – but works through the IWM’s signature mix of objects, documentary and interview.  It includes an interview with Corporal Harry Reid, recalling his experience of being wounded;

‘… a vague recollection of spinning round in the air, not sure if I did or not…  I lay on my back, looked down, I couldn’t see my legs at that stage, a big dust cloud all around, so I couldn’t really see anything, and I couldn’t hear anything…  I weren’t in any pain at that particular time, I just felt like shock and numbness, as if I’d walked into a door…

I looked across to this left hand, thinking right, I need to get a first aid kit out here, because your training kicks in straight away, in your right-hand pouch you’ve got your tourniquets, your first field-dressing, and your morphine…  I knew something violently had just happened… I looked across and this finger was hanging off … so I kept hold of that and I thought I’m not losing that as well…  I looked across at my right arm and it were twisted up around my back so then I shouted for a medic … but obviously I shouted but I couldn’t hear myself shouting, which was quite strange…

He crawled back towards me, risking his own life … and he gave me some morphine and started putting tourniquets on.  He put  a tourniquet on my arm, pulled it obviously really tight to stop the blood flow but I felt it pinch my skin … that felt painful, I couldn’t really feel anything else, so I told him not very politely to calm down a bit because it was pinching my skin…

Then I remember being in and out of consciousness..’

That last sentence is crucial; it turns out that one of the most traumatic after-effects of blast injuries is the inability to remember what happened between the initial shock and recovering consciousness in hospital.  Many of those wounded in the First and Second World Wars recalled only too well what they suffered during their evacuation, but later modern war is accompanied by powerful narcotics that combine analgesics with amnesia.    Here is Emily Mayhew in A Heavy Burden:

As ITUs [Intensive Therapy Unit] became more advanced, so did a condition known as ITU-PTSD –the stress induced, post-traumatically, by not knowing what has happened to the patient during the hours and days that are missing from their memory.

How much worse … would this be for the soldier who fell in the desert, was swooped away by MERT {Medical Emergency Response Team], saved and nursed at Bastion, flown half a continent away and then woken, not with their unit around them dusty and shouting, but their family, strained and weeping.

Recovering those lost hours, days and even weeks is a central part of my own work (see also ‘The Geographies of Sixty Minutes’ here).

So it’s good that the web page for Journey of a Wounded Soldier also features a triptych of images from the brilliant work of David Cotterell showing evacuation from Bastion to Britain (above), and interview clips addressing treatment and rehabilitation at Birmingham.

Anatomy of another soldier

I’ve drawn attention to Harry Parker‘s Anatomy of a soldier before: see here and here (and especially ‘Object lessons’: DOWNLOADS tab).  Most of the reviews of the novel were highly favourable, applauding Parker’s experimental attempt to tell the story of a soldier seriously wounded by an IED in Afghanistan through the objects with which he becomess entangled.

But writing in The Spectator Louis Amis saw it as an object lesson in ‘How not to tell a soldier’s story‘.  He complained that Parker’s device produced a narrative

‘as if the war were composed only of its inanimate processes, either accidental or inevitable. It’s a different planet to the bloody, profane, outlaw Iraq of [Phil] Klay’s Redeployment, radiating shame, PTSD and suicide, and the unbearable awkwardness of transmitting such truths to an alienated civilian world.

Parker’s device gestures aptly towards a spreading out of consciousness, a transmutation, the scattering of the individual along some plane at the threshold of death; the sensations of depersonalisation and hyper-perceptivity associated with traumatic experience; and the soothing quiddity of simple objects, as opposed to abstract thought, for a recovering victim. But it is also a method of averting the gaze from a war’s futility and waste, and worse — and probably, therefore, too, from the true nature of any saving grace.’

I do think Parker’s narrative accomplishes more than Amis allows. It succeeds in making the war in Afghanistan at once strange and familiar; and its strangeness comes not from the people of Afghanistan, that ‘exotic tableau of queerness’ exhibited in so many conventional accounts, but through the activation of objects saturated with the soldier’s sweat, blood and flesh.  It’s also instructive to read the novel alongside Jane Bennett‘s Vibrant matter: a political ecology of things or Robert Esposito‘s Persons and things, as I’ve done elsewhere, and to think through the corpo-materialities of modern war and its production of the battle space as an object-space: but neither of these has much to say about how their suggestive ideas might be turned to substantive account.

Still, Amis’s point remains a sharp one; Scottt Beauchamp says something very similar:

Harry Parker goes further than [Tim] O’Brien [in The things they carried] in giving equal narrative play to nonhuman things. Not only do they make the plot of Parker’s novel possible, they also bear semiconscious witness to our shared reality, corroborating it. Their inability to pass moral judgment comes off as a silent accusation. If this ontological shift toward objects is the most honest way we have of talking about war, it’s still limiting: it turned its weakness—its inability to fully articulate the moral significance of war—into a defining characteristic.

But I haven’t been able to let Parker’s experiment go.  So, for one of my presentations in Durham last month – on the parallels and differences between combat medical care and casualty evacuation on the Western Front in the First World War and Afghanistan a century later – I sketched out an Anatomy of another soldier.  It’s based on my ongoing archival work; earlier in the presentation I had used diaries, letters, memoirs, sketches and photographs to describe what Emily Mayhew calls the ‘precarious journey’ of British and colonial troops through the evacuation chain – you can see a preliminary version in ‘Divisions of life’ here – so this experiment was a supplement not a substitute.  But I wanted to see where it would take me.

So here are the slides; they ought to be self-explanatory – or at any rate, sufficiently clear – but I’ve added some additional notes.  I should probably also explain that in each case the object in question appeared on the slide at the end of its associated narrative.

***

I discuss aerial photography and trench mapping on the Western Front – and the difficulty of navigating the shattered landscapes of trench warfare – in ‘Gabriel’s map: cartography and corpography in modern war’ (DOWNLOADS tab).

You can find a short account of the synchronisation of officers’ watches on the Western Front in ‘Homogeneous (war) time’ here.

A shortage of cotton (combined with its relatively high cost) together with the extraordinary demand for wound dressings prompted the War Office to use sphagnum moss – the British were years behind the Germans and the French in appreciating its antiseptic and absorbent qualities, which also required dressings to be changed less often.  You can get the full story from Peter Ayres, ‘Wound dressing in World War I: the kindly sphagnum moss’, Field Bryology 110 (2013) 27-34 here.

But one RAMC veteran [in ‘Field Ambulance Sketches’, published in 1919] insisted on the restorative power of the white bandage, administered not by regimental stretcher bearers but by the experts of the Royal Army Medical Corps’s Field Ambulance:

The brown first field dressing, admirable as it is from a scientific point of view, always looks a desperate measure; and if it slips, as it generally does on a leg wound, it becomes for the patient merely a depressing reminder of his plight. A clean white dressing, though it may not be nearly so satisfactory in the surgeon’s eyes, seems to reassure a wounded man strangely. It makes him feel that he is being taken care of, gives him a kind of status, and stimulates his sense of personal responsibility. With a white bandage wound in a neat spiral round his leg, he will walk a distance which five minutes earlier, under the dismal suggestion of a first field dressing, he has declared to be utterly beyond his powers.

I borrowed the white maggots (and some of the other details of the wounds) from John Stafford‘s extraordinary, detailed recollection of being wounded on the Somme in August 1916 available here.

Carrying a stretcher across a mud-splattered, shell-blasted landscape was immensely tiring and it was all too easy to lose one’s bearings.  From ‘A stretcher-Bearer’s Diary’, 17 September 1916:

‘The shell fire, and the mud, are simply beyond description, and it is a miracle that any escape being hit. We have to carry the wounded shoulder high, the only way it can be done, because of the mud. Our shoulders are made raw by the chafing of the stretcher handles, although we wear folded sandbags under our shoulder straps. Sweat runs into our eyes, until we can hardly see. When a barrage comes we must keep on and take no notice, as even if we could find cover, there is none for the man on the stretcher….

‘…The rain has made the ground a sea of mud, and we have to carry the wounded three miles to the Dressing Stations, as the wheeled stretchers cannot be used at all. Two men using stretcher slings could not carry a man thirty yards, and I have seen four bearers up to their knees in mud, unable to move without further assistance.

By the time of the 3rd Battle of Ypres, it could take eight men to carry a stretcher half a mile to an aid post – and it could take them two hours to do it.

Even in ideal circumstances, manoeuvering a stretcher down a narrow, crowded trench was extremely difficult, ‘like trying to move a piano down an avenue of turnstiles.’  During major offensives a one-way system was in operation, and stretcher bearers were supposed to use only the ‘down’ trenches.  From the Aid Posts the RAMC stretcher-bearers of the Field Ambulance would take over from the regimental stretcher-bearers.  Here is one young novice, Private A.F. Young with the 2n3/4th London Field Ambulance:

Step by step we picked our way over the duckboards. It is useless to try and maintain the regulation broken step to avoid swaying the stretcher. Slowly we wind our way along the trenches, our only guide our feet, forcing ourselves through the black wall of night and helped occasionally by the flash of the torch in front. Soon our arms begin to grow tired and the whole weight is thrown on to the slings, which begin to bite into our shoulders; our shoulders sag forward, the sling finds its way on to the back of our necks; we feel half-suffocated. A twelve-stone man, rolled up in several blankets on a stretcher, is no mean load to carry, and on that very first trip we found that the job had little to do with the disciplined stretcher-bearing we had spent so many weary hours practising. We are automatons wound up and propelled by one fixed idea, the necessity of struggling forward. The form on the stretcher makes not a sound; the jolts, the shakings seem to have no effect on him. An injection of morphine has drawn the veil. Lucky for him.  

Stretcher-bearers changed – they worked in relays close to the front – but the stretcher remained the same.  Ideally the wounded soldier would remain on his stretcher only as far as the Casualty Clearing Station, from where used stretchers would be returned to dressing stations and aid posts by now empty ambulances.  

Twelve stretchers were supposed to be kept at every Regimental Aid Post, but supplies could easily run out.  When Major Sidney Greenfield was wounded, he remembered:

… the call ‘stretcher-bearers’, ‘stretcher-bearers’, the reply ‘No stretchers’. ‘Find one, it’s an officer.’

And it was not uncommon for those evacuated ‘in a rush’ to remain on their stretcher until the base hospital; and since ambulance trains heading to the coast were less urgent than troop trains and supply trains heading in the opposite direction the journey was usually a slow one.  If the nearest hospital turned out to be full, a not uncommon occurrence, the train would be sent on to the next available one, thus prolonging the journey still more.    

H.G. Hartnett recalled the sheer pleasure of finally being put to bed at the base hospital at Wimereux:

After being washed and changed into clean pyjamas I was lifted off the stretcher on which I had lain for five days and nights into a soft bed—between sheets.

The contrast, of course, was not only with the canvas stretcher but with sleeping in the trenches wrapped in a groundsheet.

Before the widespread introduction of the Thomas splint (above), ordinary or even improvised splints were used.  Here is Sister Kate Luard on board an ambulance train in October 1914:

The compound-fractured femurs were put up with rifles and pick-handles for splints, padded with bits of kilts and straw; nearly all the men had more than one wound – some had ten; one man with a huge compound fracture above the elbow had tied on a bit of string with a bullet in it as a tourniquet above the wound himself.

A fractured femur would turn out to be one of the most common injuries, described by Robert Jones as ‘the tragedy of the war’: if fractures were not properly splinted the soldier would arrive at the Casualty Clearing Station in a state of shock caused by excessive blood loss and pain:

‘These men required radical surgery to save their limbs and lives… Entry and exit wounds would have to be extended widely, removing all dead skin and fat… The bone ends of the femur at the fracture site would then have to be pulled out of the wound and be inspected directly [for loose fragments of bone, clothing and debris]… Wounded soldiers arriving at casualty clearing stations with a weak pulse and low blood pressure secondary to excess blood loss due to inadequately splinted fractures would be unlikely to survive the major procedure’ – let alone the amputations that were often administered.

Mortality rates in such circumstances were around 50 per cent. The Thomas splint was specifically designed to immobilise a fractured femur, and by April/May 1917 its use during the battle of Arras had reduced the mortality rate to 15 per cent, and far fewer men lost their legs: see Thomas Scotland, ‘Developments in orthopaedic surgery’, in Thomas Scotland and Stephen Heys (eds) War surgery 1914-1918.

Stretcher bearers were trained to apply the splint in the field, as in this case, but one senior officer made it clear that in any event it had to be applied no later than the Regimental Aid Post:

The Thomas thigh splint should be applied with the boot and trousers on, the latter being cut at the seam to enable the wound to be dressed. The method of obtaining extension by means of a triangular bandage has been sketched and circulated to all MOs in the Divn. After the splint is adjusted it should be suspended both at the foot and at the ring by two tapes at either end tied to the iron supports one of which is fitted to the stretcher opposite the foot and one opposite the hip.

More information on this truly vital innovation: P.M. Robinson and M. J. O’Meara, ‘The Thomas splint: its origins and use in trauma’, Journal of bone and joint surgery 91 (2009) 540-3: never in my wildest dreams did I imagine reading or referencing such a journal – but it is an excellent and thoroughly accessible account.  See for yourself here.

It was vital not to leave a tourniquet on for long.  Here is one RAMC officer, Captain Maberly Esler, recalling his service on the Somme in June 1915:

If a limb had been virtually shot off and they were bleeding profusely you could stop the whole thing by putting a tourniquet on, but you couldn’t keep it on longer than an hour without them losing the leg altogether. So it was necessary to get the field ambulance as soon as possible so they could ligature the vessels, and the quicker that was done the better.

Lt Col Henderson‘s pencilled notes on the treatment of the wounded (1916-16) urged stretcher bearers to make every effort to stop bleeding with a compress or bandage: ‘ A tourniquet should only be applied if this response fails and where a tourniquet is applied the [Regimental Medical Officer] should be at once informed on the arrival of the case at the [Regimental Aid Post].’  By May 1916 Medical Officers were being warned ‘against too frequent use of the tourniquet, on the grounds that the dreaded gas bacillus (perfringens) is most likely to thrive in closed tissues.’

A tourniquet could aggravate damaged tissues and did indeed increase the risk of gangrene; 80 per cent of those whose limbs had a tourniquet applied for more than three hours required amputation.

This was a major responsibility; sometimes the card was filled in at a Dressing Station, sometimes at the Casualty Clearing Station.  George Carter‘s diary entry for 31 August 1915 explains its importance:

‘My work consists of nailing every patient and getting his number, rank, name, initial, service, service in France, age, religion, battalion and company. That is usually fairly plain sailing, I find, but entails a certain amount of searching [extracting paybook or diary, for example] when a patient is too ill to be bothered with questions. Then I have to find out what is the matter with him, what treatment he has had, and what is going to be done with him… The reason for taking these particulars and making out forms is to prevent any man being lost sight of, whatever happens to him. If he finishes in England after taking a week on the journey, he has got all his partics on him, everywhere he has stopped, the RAMC have been able to see at a glance all about him and can turn up all about him if called on.’

But things could easily go awry.  Here is one young soldier, Henry Ogle:

I think it must have been here [at the CCS] that orderlies tied Casualty Labels on our top tunic buttons, and got mine wrong, though it may have been at Louvencourt or even Hébuterne. Wherever it had happened, it was here that I first noticed it and called the attention of an orderly to it. I had been wounded in the right calf by part of a rifle bullet which penetrated deeply and remained in but I had been labelled for superficial something or other, while Frank Wallsgrove was GSW for gunshot wound. I said, ‘Mine’s wrong, for we two were hit by the same bullet.’ ‘Can’t alter your label, chum. Anyhow it doesn’t matter. It’ll get proper attention.’ We were already being packed into a train so nothing could be done and I didn’t worry about it.

At the base hospital he tried again:

An orderly came along (it was then dark night) and threw a nightgown and a towel at me. ‘Bathroom. Down that passage. On the right. Any of them.’ ‘Don’t think I can get there. Can’t walk.’ ‘Let’s see your label.’ ‘Label’s wrong.’ ‘What do you know about that? Go on.’ ‘I know a bloody sight more about it than you do, chum, but I’ll see what I can do.’ It was not easy as the leg was quite out of action and my orderly friend had no time to watch…  On crawling back I found Frank tucked into bed. Our case-sheets were clipped to boards which hung on the wall behind our beds and, so far, the items from our tunic labels had been copied out on the case-sheets. The next morning the customary round of visits was made by the Medical Officer on duty with Matron and Sister of Ward and an orderly or two. I tried to explain that my label was wrong and Frank backed me up but we were simply ignored. My wound was dressed as a surface wound.

It was only after the swelling of his leg alarmed Matron that Henry was shipped off for an X-ray that revealed the need for an operation to remove the bullet.

‘T’ for anti-tetanus serum.  In the first weeks of the war tetanus threatened to become a serious problem: on 19 October 1915 Sister Kate Luard recorded ‘a great many deaths from tetanus’ in her diary, but two months later she was able to note ‘The anti-tetanus serum injection that every wounded man gets with his first dressing has done a great deal to keep the tetanus under.’  In A Surgeon in Khaki, published in 1915, Arthur Andersen Martin confirmed that ‘every man wounded in France or Flanders today gets an injection of this serum within twenty-four hours of the receipt of the wound’ – at least, if he had been recovered in that time – and ‘no deaths from tetanus have occurred since these measures were adopted.’

More information: Peter Cornelis Wever and Leo van Bergen, ‘Prevention of tetanus during the First World War’, Medical Humanities 38 (2012) 78-82.

Morphine was administered for pain relief, but it still awaits its medical-military historian (unless I’ve missed something).

This was Boyle’s anaesthetic apparatus, but before the widespread availability of these machines a variety of systems was in use and, in the heat of the moment, the administration of anaesthesia was often far removed from the clinical, calibrated procedures the machine made possible. Here is a chaplain who served at No 44 Casualty Clearing Station:

I spent most of my time giving anaesthetics. I had no right to be doing this, of course, but we were simply so rushed. We couldn’t get the wounded into the hospital quickly enough, and the journey from the battlefield was terrible for these poor lads. It was a question of operating as quickly as possible. If they had had to wait their turn in the normal way, until the surgeon was able to perform an operation with another doctor giving the anaesthetic, it would have been too late for many of them. As it was, many died.

The most fortunate patients were those who had little or no recollection of the procedure.  Here is H.G. Hartnett on his experience at No 15 Casualty Clearing Station (the second occasion he was wounded):

 I was destined for surgery and lay in agony on my stretcher until near 9.00 pm, when orderlies carried me into a brilliantly lit operating theatre. I was placed on the centre one of three operating tables where I lay watching doctors and nurses completing an operation on another patient only a few feet from where I lay. When my turn came my wound was uncovered and a doctor placed a mask over my face. Then he asked me the name of the colonel of my battalion as he administered the anaesthetic. I remember no more about the operation or the theatre. When I returned to brief consciousness about 4.00 am the next morning I was lying on a stretcher on the ground in a large canvas marquee, in the third position on my side of it. Others had been carried in during the night, all from the operating theatre. The fumes of the anaesthetic from their clothes and blankets continued to put us off to sleep again. The day was well advanced when I finally returned to full consciousness.  

In the early years of the war anaesthesia was not a recognised speciality – and chloroform was the most widely used agent – but as the tide of wounded surged, operative care became more demanding and Casualty Clearing Stations assumed an increasing operative load so it became necessary to refine both its application and the skills of those who administered it.   In the British Army advances in anaesthesia were pioneered by Captain Geoffrey Marshall at No 17 Casualty Clearing Station at Remy Siding near Ypres from 1915.  By then nitrous oxide and oxygen were commonly used for short operations (which did not mean they were minor: they included guillotine amputations) but longer procedures typically relied on chloroform and ether.  A crucial disadvantage of chloroform was that it lowered blood pressure in patients who had often already lost a lot of blood.  ‘If chloroform be used,’ Marshall warned, ‘the patient’s condition will deteriorate during the administration, and he will not rally afterwards.’  And while ether would often produce an improvement during the operation, this was typically temporary: ‘the after-collapse [would be] more profound and more often fatal.’   His achievement was to show that a combination of nitrous oxide, oxygen and ether significantly improved survival rates for complex procedures – from 10 per cent to 75 per cent for leg amputations – and to have a machine made to regulate the combination of the three agents.  His design was copied and modified by Captain Henry Boyle, whose name became attached to the device.  

More information: Geoffrey Marshall, ‘The administration of anaesthetics at the front’, in British medicine in the war, 1914-1917N.H. Metcalfe, ‘The effect of the First World War (1914-1918) on the development of British anaesthesia’, European Journal of Anaesthesiology 24 (2007) 649-57; E. Ann Robertson, ‘Anaesthesia, shock and resuscitation’, in Thomas Scotland and Steven Heys (eds) War surgery, 1914-1918.

Bovril was advertised in all these ways; the company used a sketch of the Gallipoli campaign to claim that Bovril would ‘give strength to win’ and that it was a ‘bodybuilder of astonishing power’.  In 1916 the company even published an extract from a letter purported to come from the Western Front, accompanied by an image of an RAMC Field Ambulance tending a wounded soldier: 

But for a plentiful supply of Bovril I don’t know what we should have done.  During Neuve Chapelle and other engagements we had big cauldrons going over log fires, and as we collected and brought in the wounded we gave each man a good drink of hot Bovril and I cannot tell you how grateful they were.

Oxo seems to have been less popular, and least for any supposed medicinal or restorative properties, but it was often sent to soldiers by their families at home.  One advertising campaign enjoined them to ‘be sure to send Oxo’, and in one ad a Tommy writes home to say that when he returned to his billet to find the parcel, ‘the first thing I did was to make a cup of OXO and I and my chums declared on the spot this cup of OXO was the best drink we had ever tasted.’  

The image shows a surgeon using a fluoroscope to locate the fragments of the bullet:

An early Crookes x-ray tube visible under the table emits a beam of x-rays vertically through the patient’s body. The surgeon wears a large fluoroscope on his face, a screen coated with a fluorescent chemical such as calcium tungstate which glows when x-rays strike it. The x-ray image of the patient’s body appears on the screen, with the bullet fragments appearing dark.

The ‘partner’ referred to was the Hirtz compass (visible on the left of the image).  According to one standard military-medical history:

The essential feature of the H[i]rtz compass is the possibility of adjustment of the movable legs that support the instrument, so that when resting on fixed marks on the body of the patient the foreign body will be at the center of asphere, a meridian arc of which is carried by the compass. This arc is capable of adjustment in any position about a central axis. An indicating rod passes through a slider attached to the movable arc in such a way as to coincide in all positions with a radius of the sphere, and whether it actually reaches the center or not it is always directed toward that point. If its movement to the center of the sphere is obstructed by the body of the patient, the amount it lacks of reaching the center will be the depth of the projectile in the direction indicated by the pointer.

The value of the compass lies in its wide possibility as a surgical guide, in that it does not confine the attention of  the surgeon to a single point marked on the skin, with a possible uncertainty as to the direction in which he should proceed in order to reach the projectile, but gives him a wide latitude of approach and explicit information as to depth in a direction of his own selection.

The compass built on Gaston Contremoulins‘ attempts at ‘radiographic stereotaxis’; it could usually locate foreign objects to within 1-2 mm: much more than you could possibly want here.

The reassuring scientificity of all this is tempered by a cautionary observation from a wounded officer, Major Sidney Greenfield, who was X-rayed at a Casualty Clearing Station: 

My next recollection was the x-ray machine and two young fellows who were operating it. Apparently the operator had been killed the previous night by a bomb on the site and these two were standing in with little or no experience of an x-ray machine. Their conversation was far from encouraging and was roughly like this: ‘Now we have got to find where it is … is it this knob?’ ‘No.’ ‘Try that one.’ ‘Try turning that one.’ ‘No, that doesn’t seem to be right.’ ‘Ah, There it is.’ ‘Where’s the pencil. We must mark where it is. Now we have to find out how deep it is.’ After some time they seemed to be satisfied. In my condition and knowing little about electrical machines such as x-ray I wondered whether I should be electrocuted and was more relaxed when I was taken back to bed.

Incidentally, X-rays were called Roentgen rays (after the scientist Wilhelm Roentgen who discovered them in 1895) but the British antipathy towards all things German saw them re-named ‘X-rays’ from 1915: Alexander MacDonald, ‘X-Rays during the Great War’, in Thomas Scotland and Steven Heys (eds) War surgery, 1914-1918.

In addition to these terse communications, nurses and chaplains usually wrote to relatives on behalf of their patients. It was seen as a sacred duty, but it often seemed to be a never-ending task.  On 1 August 1917 Sister Kate Luard confided in her diary: ‘I don’t see how the “break-the-news” letters are going to be written, because the moment for sitting down literally never comes from 7 a.m. to midnight.’  In the case shown here, Sister Kathleen Mary Latham had written to Lt Hopkins’s wife on 12 November 1917 from a Casualty Clearing Station to say that

‘your husband has been brought to this hospital with wounds of the legs, arms, hand and face.  He has had an operation and is going on well. Unfortunately it was found necessary to remove the left eye as it was badly damaged, but he can see with the other though the lid is swollen and he cannot use it yet.  No bones are broken.  It will not be advisable for you to write to this address as he will probably be going on to the base in a day or two.’

The telegram from the War Office is dated three days later, by which time Hopkins had reached the base hospital at Le Touquet.  Sister Latham’s earlier account of her work at Casualty Clearing Station No. 3 at Poperinghe in 1915 is here.

***

In Durham, Louise Amoore pressed me on the anthropomorphism that seems inescapable in a narrative like this; it worries me too (I’ve always been leery of Bruno Latour‘s Aramis for that very reason).  I tried removing the ‘I’ and substituting an ‘it’ but I found doing so destroyed both the operative agency of the objects and, perhaps more important, the transient, enforced intimacy between them and the soldier’s body.  That intimacy was more than physical, I think.  I’ve already cited the reassurance provided by the prick of a needle, the whiteness of a new bandage; but the mundanity of objects could also be disorientating, intensifying an already intense sur-reality.  Here, for example, is Gabriel Chevallier recalling the moment when he and his comrades went over the top:

The feeling of being suddenly naked, the feeling that there is nothing to protect you. A rumbling vastness, a dark ocean with waves of earth and fire, chemical clouds that suffocate. Through it can be seen ordinary, everyday objects, a rifle, a mess tin, ammunition belts, a fence post, inexplicable presences in this zone of unreality.

Aramis also alerted me to another, and perhaps even more debilitating dilemma: a latent functionalism in which everything that is pressed into service works to carry the soldier through the evacuation chain.  That seems unavoidable in a narrative whose telos is precisely the base hospital and Blighty beyond.  Yet we know that, for all the Taylorist efficiency that was supposed to orchestrate the evacuation system in this profoundly industrial war, in many cases the chain was broken, another life was lost or permanently, devastatingly transformed.  As you can see, I’ve tried to do something about that with some of the objects I’ve selected.

I’ll probably add more objects: this is very much a work in progress, and I’m not sure where it will go – so as always, I’d welcome any constructive comments or suggestions.  Any written version would involve longer descriptions, I think, and would probably dispense with most of the scaffolding of notes I’ve erected here (though some of it could and probably should be incorporated into the descriptions).

A heavy reckoning

Emily Mayhew‘s Wounded was one of the catalysts for my present research project on medical care and casualty evacuation in war zones.  The original idea was to complete four case studies – the Western Front in the First World War, the deserts of North Africa in the Second, Vietnam and Afghanistan – but since then it has expanded to include a detailed analysis of attacks on hospitals and healthcare in Syria and elsewhere.  But running throughout these investigations is an interest in what Emily called ‘precarious journeys‘ – and a determination to break away from the usual academic voice (see here) –  so an announcement of her new book is extremely welcome.

It’s due from Profile in May:

What happens when you reach the threshold of life and death – and come back? As long as humans have lived together on the planet, there have been wars, and injured soldiers and civilians. But today, as we engage in wars across the globe with increasingly sophisticated technology, we are able to bring people back from ever closer encounters with death. But how do we do it, and what happens next? Here, historian Emily Mayhew explores the modern reality of medicine and injury in wartime, from the trenches of World War One to the dusty plains of Afghanistan and the rehabilitation wards of Headley Court in Surrey. Mixing vivid and compelling stories of unexpected survival and giving astonishing insights into the frontline of medicine, A Heavy Reckoning is a book about how far we have come in saving, healing and restoring the human body. But what are the costs involved in this hardest of journeys back from the brink? From the plastic surgeon battling to restore function to a blasted hand to the double amputee learning to walk again on prosthetic legs, Mayhew gives us a new understanding of the limits of human life and the extraordinary costs paid both physically and mentally by casualties all over the world in the twenty-first century.

The book is published in conjunction with the Wellcome Collection, and I should note that Emily has also co-curated the current exhibition Wounded: Conflict, Casualties and Care at London’s Science Museum (more information here and here).  I spent a fascinated couple of hours there when I was in London last month – it is well worth a visit, though the parallels it draws between the Western Front and Afghanistan were too abbreviated for me.

Journeys from No Man’s Land

Stretcher-bearers

I’ve agreed to join a panel organised by Noam Leshem on Remnants of No Man’s Land: history, theory and excess at the Annual Meeting of the Association of American Geographers in Chicago next April (I imagine this is a follow-up to the session at the RGS/IBG in September).

The no-man’s lands of the First World War were never limited to the killing fields between the trenches. Their impact was never fully confined by the time and space of the battles: it lingered on the bodies of soldiers, in contaminated ecologies and in the radically altered post-war intellectual landscape. The violence that is unleashed in the no-man’s land and the destruction it wrought does not result in emptiness, in a terra nullius, but in excess that can never be fully contained.

This session invites additional reflections on the excessive quality of no-man’s land: its materialities, ecologies, cultural expressions and political-ideological articulations. It aims to deepen the theoretical import and conceptual power of ‘no-man’s land’, and move beyond its use as merely a convenient colloquialism. Similarly, we seek to engagements with other histories of no-man’s lands that are not solely confined to the Western Front during WWI.

LOBLEY Dugouts in the embankment near Le Cateau

Despite that last sentence, this is what I’ve come up with; these abstracts are always promissory notes, of course, written so far in advance that they can provide little real indication of what eventually transpires.  Fortunately we are now no longer lumbered with the Yellow Pages-style book of abstracts so I doubt anybody will actually read this on the day.  But here goes:

Journeys From No Man’s Land, 1914-1918

During the First World War on the Western Front a central logistical preoccupation of military planners was the deployment of troops to the front line and the evacuation of casualties from the battlefield. These priorities were closely connected – the aim was to provide medical treatment as close to the site of the wound as possible so that troops could be returned expeditiously to the line – but they also often confounded one another as hospital trains headed for the coast were shunted into sidings to allow troop trains to move up. In this presentation I address three questions. First, what it was possible to know about the ‘lie of the land’, particularly in the deadly spaces between the front-line trenches? Here I focus on the connections between aerial reconnaissance, night patrols and trench maps. A second question concerns the arrangements made in advance of major offensives – the disposition of stretcher bearers and aid posts, field ambulances and casualty clearing stations – and the ways in which these visible geometries of the medical-military machine affected the sensibilities of soldiers waiting to go ‘over the top’. Finally, how did the wounded apprehend and navigate No Man’s Land, and how did they make what Emily Mayhew calls their precarious journeys away from the fighting?

There’ll be more posts on this as I circle in towards the presentation.  It’s part of my new research project which explores military-medical machines and the casualties of war 1914-2014, but which is now widening to include other aspects of medical care in contemporary conflict zones like Gaza and Iraq/Syria and the militarisation of medical intervention in West Africa.

Medical-military machines and casualties of war 1914-2014

SABER, Desert Rat Sketchbook

As promised, I’ve added the outline of my proposed new research programme on the casualties of war to the DOWNLOADS page (scroll way down).  I’ve omitted all the pages in the formal application that drove me to distraction – knowledge mobilisation plan (sic), budget justification, ‘expected outcomes’ and the rest – but if you do look at this ‘Detailed Description’ (as it has to be called), please bear in mind that:

(a) this is still very preliminary, and my work is in the earliest of stages (this is a grant application, after all);

(b) everything had to fit into a prescribed, very limited space (leaving no room for nuance); and

(c) I was more or less required to use the Harvard reference system, the enemy of all good writing.

I’ve added a few lines of clarification and a series of illustrations to liven things up (no room for those in the original), but I hope this gives you some idea of what I’m up to.  And, as always, I’d welcome any comments or suggestions – preferably by e-mail to avoid the spam filters.

In working on this, I’ve stumbled into a series of unusually rich primary sources and secondary literatures, and since I’ve mentioned good writing I want to flag two texts that have kept me going throughout this process.

MAYHEW WoundedThe first is Emily Mayhew‘s Wounded: from Battlefield to Blighty 1914-1918, which I noted in an earlier post; for editions outside the UK the subtitle becomes A new history of the Western Front.  I’ve now read it, and admire its substance but also its style enormously.  It’s based on painstaking research, literally so, and yet the main chapters read like a novel, and the analytical-bibliographic apparatus has been artfully moved to the Notes where it becomes a model of clear, concise and thought-provoking commentary rather than a cage that hobbles the narrative.

MacLeish Making war at Fort HoodThe second is Ken MacLeish‘s Making war at Fort Hood: life and uncertainty in a military community.  I met Ken at a workshop in Paris last year, and if you read anything better this year – in style and substance – than his Chapter 2 (‘Heat, weight, metal, gore, exposure’) I’ d like to know about it.  The combination of ethnographic sensitivity, elegant prose, and a theoretical sensibility that Ken wears with confidence and displays with the lightest of touch is simply stunning.

Since I’m off to York next week I’ll be spending the next few days preparing a new presentation on ‘Drones and the everywhere war’.  I hope this will also give me  time to return to my posts on Grégoire Chamayou‘s Théorie du drone which I had to put on hold while I trekked from Flanders to Afghanistan for my grant application…

Precarious journeys

Much of last week was taken up with working out a new project for the next round of the Social Science and Humanities Research Council’s Insight Grant programme.  A ‘Notification of Intent’ to apply is required (I’m deliberately not saying ‘needed’) before you can actually apply in October – but since the NOI requires a plain-language summary and a figure for the total budget most of the planning has to be done months before the application.  I could fill a whole blog – and other non-digital receptacles – about the sense in all that; suffice to say I hit the button ten seconds (sic) before the electronic shutters came down.

The application is for a project called Medical-Military Machines and the Casualties of War: Genealogies and Geographies of Care.

859084-dust-off-inside-afghanistan-039-s-medevac

One of the central claims made by protagonists of later modern war is that its conduct is accurate and proportionate, legal and ethical, thereby raising the bar for ‘just’ or, as James Der Derian has it, ‘virtuous’ war (and as most readers will know, he would insist on those scare-quotes).  It has done so, its advocates argue, by limiting casualties through new modes of intelligence, surveillance and reconnaissance, new weapons systems, and new modes of accountability.  I explore these issues in my ongoing SSHRC project, Killing Space (DOWNLOADS tab)not least through my continuing study of drones (much more to come!) and this project maps its other, vitally important dimension – a sort of ‘Caring Space’ – in order to provide an indispensable substantive test for these claims.

The project concerns the provision of medical care for those wounded by military and paramilitary violence, casualties who are often overlooked in vexed but vital debates over ‘body counts’ and what constitutes (following Judith Butler) a grievable life.  I’m not going to ignore those matters, far from it, but my main concern will be on the survivors of military violence.  As I’ll explain in a moment, I want to analyse both combatant and civilian casualties, and so confound the simplistic politics in which the right is supposed to care about the one and the left about the other.

The project will involve both genealogy and geography.  I’m using ‘genealogy’ in something like the Foucauldian sense, but all I’ll say here is that historical depth is plainly essential to specify what is (and is not) novel about the ways in which advanced militaries wage war.  So the project will involve four case studies focusing on the United States and its allies.  The first three are the Western Front in World War I, North Africa in World War II, and South Vietnam (1963-1975) .  In this traverse from ‘total war’ to James Gibson’s ‘techno-war’ I’m planning to leverage my work on ‘The natures of war’.  While researching that presentation and long-form essay – which will eventually appear in War Material – I found  a treasure-trove of sources that I want to explore in much more depth and detail for this new project.  The fourth case study will involve the cluster of wars in the Greater Middle East post 9/11, and while much of this has been familiar ground for me ever since I started writing The colonial present, there are many new issues to address – including the deliberate targeting of hospitals and medical doctors by some factions and what Omar Dewachti calls the ‘therapeutic geographies’ involved in the transnational movement of war casualties from (say) Iraq, Libya and Syria to hospitals in Lebanon, Jordan and India.

The project has three components that address different geographies of casualty care.

Stretcher-bearers

MAYHEW Wounded(1) Modern military medicine has sought to provide immediate care for troops injured in combat as close to the site of the injury as possible by deploying medical personnel and equipment in forward positions, and establishing evacuation routes for more seriously injured patients to higher-order medical facilities in the rear.  These systems have been transformed by technical advances designed to increase the time-space compression of treatment: the more widespread use of motorized ambulances in the Western Desert, for example, and helicopters for medical evacuation (‘dust-off’) in Vietnam and later conflicts. I plan to reconstruct these networks and their transnational extensions and to calibrate the changing transit times, and then to turn these skeletal geometries into human geographies through diaries, letters and, as we near the present, interviews, that I hope will bring into view the multiple people involved in these precarious, fleshy, and profoundly intimate journeys.  My inspiration for this is a series of thought-provoking essays in the Journal of the Royal Army Medical Corps (really), which provide a way in to the geometries and networks, and (very different) Emily Mayhew‘s Wounded: from Battlefield to Blighty, 1914-1918, due out next month, which uses the idea of a ‘journey’ in what could develop into a sort of phenomenology of care; I’ll say some more about some of this in a later post.

AEF Evacuation system WWI

The other two components follow from a remark made by Michel Foucault in ‘The Eye of Power’.  There he suggested that ‘doctors, along with the military, were the first managers of collective space’, but he assigned them to different spaces (‘campaigns’ versus ‘habitations’). Instead I want to explore what happens when military and medicine are called upon to imagine and manage the same space and install what, following the example of Mark Harrison, I’m calling a ‘medical-military machine’ in a war-zone.  So I’ll be following two tracks that are usually kept separate – civilian and combatant casualties (and here I want to extend the ongoing debates over their distinction from an abstract legal to a substantive therapeutic terrain) – and tracing the junctions where they intersect, in order to establish two other, complementary and sometimes countervailing geographies of care.

Medevac

(2) There is an important sense in which modern war has always been ‘war amongst the people’: this is not a late twentieth-century preoccupation.  Images of ‘No Man’s Land’ on the Western Front distract attention from the injuries suffered by civilian populations who continued to inhabit houses and work farms behind the front lines, for example, while ground and air offensives in South Vietnam produced hundreds of thousands of civilian casualties.  So a second question is this: in what ways and in what places have militaries assumed medical responsibility for civilian casualties before and beyond the parameters of the Medical Civic Action Programs of contemporary counterinsurgency?

Secours Quaker

REDFIELD Life in Crisis MSF(3) Conversely, the military has not been the only agency making medical interventions in war-zones, and this is not a late twentieth century development either.  Civilian hospitals are increasingly important in today’s urban wars (where they often become targets too), but I want to pay particular attention to the work of international agencies.  I plan to analyse two voluntary organisations, the Friends’ Ambulance Unit and the American Field Service in the two world wars, and (I hope) two contemporary NGOs, the most obvious candidates being the International Committee of the Red Cross and Médecins sans Frontières.  I’m not assuming any direct filiations, and I’ll no doubt find all sorts of differences between them (particularly between the earlier and the later ones), but I’m particularly interested in the tensions between what at the moment I see as a common, more or less cosmopolitan engagement and the imperative to provide place-specific casualty care (and the logistics of doing so).   So a third question revolves around the rise of a ‘militarized humanism’ and the emergence of what Didier Fassin calls  ‘humanitarian reason’ as, perhaps, a form of governmentality.

http://vimeo.com/66342865

This really is just a bare-bones summary, and since I have another two months to flesh it out I’d really welcome any advice, suggestions or criticisms.  As I’ve described the project here you can see, I hope, that my case-studies and the questions I think they’ll enable me to address arise at the intersections of medical and military geography but also involve political, cultural and legal geographies.  And, as ever, those geographies all have a stubbornly little g: this really isn’t a disciplinary project.