LAND AND WATER. 
Fobruarj' 17, 1916. 
however ignorant one may be of the metliods whereby such 
things are computed. 
But when we look into those metliods we shall, I 
think, be amply satisfied that they arc imperfect and 
indeed, almost valueless. 
The writer begins with the foundation for all these 
calculations, the lists pubhshed officially by the German 
authorities. 
He adds these together, including those which 
appeared during the month of January, and arrives at a 
total of 2,627,085 casualties, up to and before February 
ist, 1916. 
But when he comes to the criticism of these figures, 
he breaks down altogether. 
The Four Griticalx Points. 
There are four essential departments in this criticism : 
(i) We have to find out what proportion of wounded 
and dead these hsts either delay in publishing or omit 
altogether. 
(2) We have to find out what proportion of those 
appearing in the lists return to active service of the same 
sort as that which they left when they were wounded or 
invalided. 
(3) We have to estimate what proportion over and 
above those mentioned in the casualty lists are men off 
the strength at any moment from sickness, because that 
category, as we have seen, is not mentioned at all in the 
casualty lists. 
(4) We liave to estimate the " permanent margin of 
temporary losses." 
Unless we can get somewhere near a rough estimate 
of these four points om: calculations are obviously 
worthless. 
Now the writer of the article makes no sort of 
attempt to arrive at and to prove any one of these four 
fundamental estimates. 
Suppose you want to know what a man's available 
cash is at any moment. He gives you an account dated 
upon the ver.y day of your enquiry. He admits, however, 
that the account does not include his last transactions, 
but is, in all its items, more or less belated ; Of one set 
of items ttiere is no record for three months past, of another 
for two Baonths, etc. He further admits that one whole 
category of expenditure is never mentioned at all. Finally, 
he does not tell you in his accounts what proportion of 
his cxp< -nditure is in the form of loans subject to repay- 
ment, Irmt only tells you that " a large part of it " is of 
this SOI t. 
It is obvious that liis accounts so stated are, for the 
purpoJ^e f»f an exact estimate, worthless. You could only 
arrive at such an estimate by judging from other of the 
man's actions or from the analogy of other men similarly 
placed : (i) What is the average delay in the appearance 
of an ii.cm upon the accounts ; (2) What proportion of the 
expcn($ture is in the shape of good debts which have been 
repaid ; (3) what proportion of the whole is formed by that 
category of expense which he has refused to mention ; 
and (4) what money is still out on loan. 
It is perfectly clear that if you do not know anything 
about these four things the accounts he has rendered you 
are worthless. 
It is precisely the same with a set of casualty lists. 
The delay between any financial transaction and its 
mention in the accounts corresponds to the delay in the 
appearance of names upon the casualty lists. 
The proportion of expenditure consisting of loans 
wliich are repaid corresponds to the sick and wounded 
vho come back to full ictive sers'ice again. 
The categorv- of expenditure which your informan 
refuses to mention and which he admits does not appea 
in his accounts at all, ccrrcsponds to the cases of sickness 
as distinguished from wounds. 
WTiile the amount of cash which is not at the moment 
available because it is still out on loan and has not been 
returned, corresponds to that number of sick and wounded 
men which, at any gn-cm moment, arc off the strength 
although thev will at so«ne fut>'— time bo back on the 
strength again after their <urc. I is this which is called 
i.' ike -permanent margin of temporary losses." 
Now when we turn to the article in the Times of 
February Oth we find no sort of argument upon these 
four fundamental essentials. 
On the first point, the average delay m the mention 
of the names, we have the exceedingly vague sentence : 
" They are often belated — but so are ours." 
The sentence is not only worthless as a piece of 
exact calculation, but it is obviously bad in logic. What 
has our rate of delay got to do with an estimate of the 
enemy's losses ? Though we should pubhsh no casualty 
lists at all there would yet remain the problem of ascer- 
taining what his losses were. As a way of excusing the 
enemy from deliberate bad faith such a remark may have 
some purpose, but as a method of belittling the enemy's 
losses it is meaningless. If our own lists are very much 
belated the only conclusion useful to the pres nt purpose 
which could be drawn from that fact would be that the 
German losses were even larger than was supposed. For 
instance, if the average British .'Vrmy in the field were a 
fifth of the average German Army in the field and if we 
were working only on the analogy of our own figures, wc 
should multiply our casualties by 5 to arrive at theirs. 
But if our real casualties (on account of delay in publica- 
tion) were at any given date half as much again as the 
numbers published to that date, then, on the same 
analogy the real German losses to date would be half as 
much again as the published ones. 
The first part of the sentence is clearly useless. 
Everybody knows that the German lists, like all lists, are 
somewhat belated. The whole point is how much belated. 
Unless you can answer that question within certain 
approximate limits you are not calculating at all, but only 
talking at large. 
Now the readers of this journal are familiar with the 
methods by which the average amount of delay can be 
arrived at. It varies, of course, with the amount and 
the severity of the fighting, with the distance over which 
the information must travel, with the opportunities for 
ascertaining and checking the results — opportunities that 
differ with ground, climate, and a hundred other things. 
But the average rate of delay we know to be from six to 
eight weeks. 
How do we know this ? We know it by noting af tci 
what delay certain losses, the exact date of which the 
Allies can determine, appear in the lists. 
For instance, a particular German company was 
captured almost entire on the 26th of September by 
the French. It appears in the German lists on the 29th 
of October. Certain German dead identified by the 
French in the first days of October do not appear until 
December. Even in the month of January there is a 
respectable number of names appearing, the casualties 
referring to which took place more than three months 
earlier. 
The work has been done with minute care all over 
Europe. Its results are fairly well known. The general 
conclusions are published from time to time, particularly 
by the authorities in Paris. There is no excuse for 
ignoring exact results of this kind, and if one does ignore 
them one's conclusions have no value at all. 
The very figures given in the article to which I am 
referring are amply sufficient to prove so obvious a truth. 
For instance, the Times gives the German losses in 
August 1914 — killed, missing, severely wounded, lightly 
wounded — all — at 9,213 ! August was the month of the 
great assaults on Lif-ge, of the cavalry skirmishes througli 
Belgium, of the tremendous struggle on the Sambrc, ol 
Guise, of Le Cateau, of Sarrail's violent and successful 
stroke against the Crown Prince which saved Verdun ; 
the first two days' fighting of the bloodiest battle of the 
lot, the Grand Couronne, took place on the last two days 
of August ; finally, August saw the smashing of the twc 
.\rmy Corps defending East Prussia and the tremendous 
counter-stroke of Tannenberg. 
The total German losses in that month may have 
been 15 times 9,000 or may have been 20 times or 25 
times 9,000 ; but the figure 9,000 for all those August 
losses is obviously nonsense. 
What then does it mean ? It means that the lists 
compiled and checked up to and including the 31st of 
August, 1914, were only the first tlriblcts and referred 
only to the very first stages of the fighting. 
