Cross-posted from the Men, Women and Care blog:
In a Twitter discussion sparked by the BBC’s screening of Peter Jackson’s ‘They Shall Not Grow Old’ in November, Greg Jenner quoted the statistic that nearly 90% of British servicemen returned home and that half of these were injured. The first statistic didn’t worry me. 88% is the generally accepted rate of survival for the conflict, although it appears high in the context of a cultural narrative which emphasises the death and destruction of the war above all else. The second figure, however, was not one I recognised from my research into disable ex-servicemen of the conflict, so I asked Greg where he had got his figures from, as my own estimate is that about 30% of men returned from service did so suffering from illness or injury. What resulted was a conversation that, as I noted at the time, was one I wasn’t in a position to engage in properly, given that I was lying in bed, attempting to tweet on my iPad after a very long day listening to my son sing in choir at a large-scale Remembrance Service. This, then, is my attempt to think through the issues raised by Greg’s tweet, and my querying of it.
To start with the problem, which is that I believe 50% of all men returning from the war as injured to be an overestimate. Greg’s figure, which comes from the Imperial War Museums and has been cited by a number of historians over the years, is based on the official Statistics of the Military Effort of the British Empire During the Great War: 1914-1920. The summary table of the ‘Total per cent of Battle Casualties and Deaths’, which is tabulated by theatre, gives the rate as 5 casualties to every 9 men sent out to France, 2 to every 9 men sent to the Dardanelles, 2 to every 12.5 men sent to Mesopotamia and smaller ratios for other theatres, including Salonika, East Africa, Egypt and Italy. In percentage terms, this works out at 56% for France, 23% for the Dardanelles, 16% for Mesopotamia and smaller percentages for other theatres. The total percentage of casualties across all theatres has not been calculated, although the History of the Great War Based on Official Documents: Medical Services: Casualties and Medical Statistics of the Great War states that ‘the total numbers affected by war service, in the sense of death or some form of war disablement for which State compensation was given, may be estimated at approximately 2,414,000 or 40.2 per cent. of those who served.’
There are some problems with these figures, however. First, figures from the Statistics of the British War Effort include casualties not only from Britain but from the colonies and dominions as well. The statistics cannot, therefore, be considered reliable in discussions of the effect of casualty numbers on British society alone. They force us to think about Britain as a colonial actor in this period, but do make local and regional analyses much harder.
Second, the inclusion deaths in the count of battle casualties makes it much harder to evaluate how many men returned from the war injured or ill. If we could calculate the total percentage of casualties due to death, injury and illness across all theatres, we could subtract the accepted 12% mortality rate that is the generally accepted figure for the British armed forces. But we don’t have that figure. The percentage of battle casualties does distinguish between ‘killed and died’, ‘wounded’ and ‘missing and prisoners’ for each theatre, but specifically excludes casualties due to sickness. The medical services statistics, meanwhile, only counts those in acknowledged as ill or wounded by the Ministry of Pensions, not all those who became casualties during the war.
This raises the question of what is meant by casualty even when deaths are discounted from the numbers. I have focussed on the word injury throughout my discussion so far, reflecting the omission of the sick from the official calculations. But men did not become casualties only through injury, that is damage caused by accident or attack. Many casualties, particularly in the Middle East, were the result of illnesses, including those which could persist long after demobilisation, such as malaria or the rather general ‘debility’, or physical weakness resulting from other illness.  Some illnesses, perhaps most notability heart conditions, were the result of aggravation of pre-existing conditions by the experience of war, conditions that had either been missed in hurried medical inspections  or had been dismissed as not severe enough to affect a man’s service in light of the ever-increasing military demands for manpower. To say that these men returned from the war suffering from injuries would be inaccurate, yet they undoubtedly returned from the war suffering from physical consequences of their service, whether attributable to or aggravated by it.
Sickness rates are analysed in the official statistics, as part of the wastage rates. Also tabulated under ‘wastage’ are statistics relating to multiple injuries. A single injury does not equate to a single man injured. Because of the military priority for manpower, the work of the medical services aimed at conserving manpower through not only the treatment of wounded men but also their rehabilitation for return to active service where possible. Wounded men were thus often recycled through the military medical system, some multiple times for multiple injuries. This repetition and overlap is at least partially captured in the statistics, but not in a way that makes clear how this affected the headline figure.
Finally, there is the problem of the invisible injury and underdiagnosis in the casualty rates as defined by the official statistics. The most obvious example of this is, of course, psychological traumas which are generally accepted as being underdiagnosed by the British military medical authorities throughout the war, in part due to the social stigma attached to such traumas and in part due to lack of medical understanding of the conditions. (These two things are, of course, related.) It is worth noting, however, that underdiagnosed invisible injury also includes conditions such as hearing loss, where diagnosis had to be based on highly subjective observations. The existence of both invisible injury and the stigma attached to it has led several historians, most notably Jay Winter, in a lecture at the British Academy in 2014, to suggest that might higher numbers of men suffered from disabling traumas than is reflected in official statistics. While the statistics do, undoubtedly, underestimate the number of men so injured in war, the methodology used to arrive at the figure of 25% quoted by Jay remains somewhat opaque.
Nonetheless, it is possible that the numbers of invisibly injured undiagnosed and therefore uncounted in the official statistics cancels out the overcounting of individual men who suffered multiple injuries and illnesses. There is no way, from the statistics available, to demonstrate this, so that 50% feels like a guess rather than solid estimate. Which brings us to the figure that I have been using in my research, also drawn from the official statistics, this time those relating to the men in receipt of some sort of support from the Ministry of Pensions up to 31st March, 1930. This figure is given as 1,664,000 or 27.7% of those who served.  Based on the mortality rate of 12%, this gives a figure of 31.5% of those who returned from the war suffering from an attributable illness or injury sufficiently serious enough to warrant a pension or gratuity.
Again, we have to qualify what is being counted here. Men in receipt of either a gratuity or pension from the Ministry of Pensions could do so for either an illness or an injury, so they cannot be described solely as injured men. To receive support they had to show that the illness or injury was either attributable or aggravated by the war, with aggravated conditions often disallowed for support. Other exclusions included injuries caused by carelessness or those deemed to be self-inflicted. Then there is the issue of self-reporting. Pension had to be applied for; ill or wounded servicemen were not automatically assessed for them, although Article 9 of the Pensions Bill gave them the right to apply for a pension following discharge if attributability or aggravation could be demonstrated. Some men did not apply for a war pension because they had sufficient income not to need additional support; other viewed any sort of state support as associated with the heavily stigmatised aid of the New Poor Law and the workhouse, or else as a form of dependence that was at odds with their subjective sense of self as independent male breadwinners. Of course, the willingness to define oneself as disabled by war might change over time and with circumstances. A wound or illness might deteriorate; financial circumstances might change for the worse. A man reluctant to apply for a pension in 1919 might feel he had no other option in 1927, although he might have a harder time proving attributability at that distance from the point of illness or injury.
So just over 30% of men returning from the war injured or ill is almost certainly an underestimate, but by as much as 20% or 1,056,400 men? Even in the circumstances, that seems too big a margin of error to me. So I would conclude, not very helpfully, perhaps, that the actual figure falls somewhere between the two, probably more than the 1.7 million men noted by the official statistics but not as many as 2.7 million. If we wanted to talk in round figures, about 2 million men would seem as close a guestimate as can be made on the figures currently available. According to the Office for National Statistics, the UK population in 1929 was estimated at 45.7 million. Disabled ex-servicemen, it can therefore be argued, made up approximately 4.5% of the British population. But this figure comes, as we have seen, with all sorts of qualifications.
In his statistical examination of casualty records, Those We Forget (2014), David Noonan is sharply critical of the statistical methodologies of all the official histories of the war efforts of Britain and her Dominions. Noonan has attempted a statistical revision of the Australian official record. A similar exercise has yet to be carried out in relation to the British armed services from the First World War, although, in comparison to Noonan, who used the well-preserved embarkation records of Australian First World War servicemen for his re-evaluation, historians of the British armed forces face greater difficulties due to the loss of relevant documentation over the years. Nonetheless, it behoves us to treat the statistics of the war effort, official or otherwise, with critical care. As I hope this post has demonstrated, quantitative analysis can throw up a range of interesting nuances in relation to questions about the social impact of mortality, survival and illness, even if the numbers can’t answer our questions as clearly or easily as we might hope.
 T.J. Mitchell and G. M. Smith, History of the Great War Based on Official Documents: Medical Services: Casualties and Medical Statistics of the Great War (Imperial War Museum, 1997; originally released 1931), p.315.
 David Noonan, in his statistical analysis of Australian casualty records, defines debility as a euphemism for psychological trauma in the records. There is, however, no evidence that this was, in fact, how it was used at the time.
 See J.M. Winter The Great War and the British People (Basingstoke: Macmillan Eduction, 1985), pp.48-64.
 Mitchell and Smith, Medical Services, p.315.
 This discrepancy can, perhaps, be seen in the estimate of the the International Labour Office of 1.7 million British ex-servicemen suffering from war-attributable disability in 1921, a year after the official statistics recorded 1.4 million men being granted a pension or gratuity for war-atributable disability. ‘Studies and Reports: The Compulsory Employment of Disabled Men’, Series E, no.2, International Labour Office (Geneva, 1921), p.2. I am indebted to Bethany Rowley for providing me with this figure and reference.