RWF Regular Reservists

[Muerrisch]

We know from Martin G's patient research that the Royal Welsh Fusiliers had fewer than 1400 regular reservists [of all three categories A, B, and D] available to mobilise. We also infer that some 3800 men enlisted during the relevant period mid 1902 to mid 1914. Paul Nixon confirms my figure of 3800 and assures me that he knows of no significant numbering gaps.

So how was it that these 3800 men were whittled down to 1400 during those "piping times of peace"? [W.Shakespeare, Richard III].

The Terms of Engagement for Line Infantry from mid-1902 were successively three years colours and nine reserve, nine years colours and three reserve, and seven years colours and five reserve. Any man enlisted after 4 August 1902 and before 5 August 1914 should either be a regular reservist or:

1. still serving, either the balance of twelve years, or having signed to go for pension after 21 years. Estimating this total is difficult but will be attempted.

2. Dead, either in service or during the reserve period. A reasonable estimate can be made.

3. Invalided, either in service or during the reserve period. A reasonable estimate can be made.

4. Discharged, either in service or during the reserve period, for a variety of reasons: by purchase, by indulgence, incorrigible behaviour, committing a felony and being in jail, and a few others. A reasonable estimate can be made.

5. Deserted during service or disappeared during the reserve period. A reasonable estimate can be made.

6. Transfer to another Corps

I WOULD WELCOME ADDITIONS TO THIS WORK LIST.

A few questions need to be addressed by those knowing more than I.

a. I have assumed that a soldier was enlisted [and committed to serve having been accepted] before his number was issued. Am I correct? My available sources are KR 1912/14, and Regs Recruiting 1900. The former requires a soldier to be allocated a number after posting or transfer [para 1897], the latter makes it clear that attestation should precede posting.

b. on discharge from the colours to the reserve, was a soldier held to be a reservist in his own regiment, or [perhaps living in Scotland or Ireland after discharge] could he go "on the books" of a corps near to home? Given the imperative of occasional musketry refreshers during the reserve period, this would make sense for the Treasury economically, even if not militarily.

[Guest]

I hope to have more granular data soon from the WO 114 files. If the data is there as the file titles promise, I should be able to obtain at Regimental and Battalion level

1. Monthly Returns Rank and File 1890-1913

2. Weekly Returns of the British Army at Home 1914

3. Militia Monthly Returns 1890-1907

4. Militia Establishment and Strength 1908

5. Militia Enrolled strength 1903-1906

6. Special Reserve Establishment and Strength 1909-1915

If, (and it is a big 'if') the data is there, it should be possible to reconstruct the regimental and battalion data and while details on wastage and causes might not be there, it will be possible to see just how many men fell out of the Service and also fell out of their future obligations.

Numbers leaving the Infantry but Not Reaching the Reserves. There is some evidence in the hard data that a high proportion of men leaving the Regulars did not in fact reach the Reserves. The reasons are wide and varied. Here are two snapshots from 1911 and 1904.

1911. In 1910-11 (the most recent data I have) the Line Infantry saw a Total Decrease* of 17,224 ORs of which 7,284 had transferred to the Army Reserve, meaning only 42 out of every 100 men leaving the infantry did not have reserve obligations. The missing men are explained by a long list: death, discharged to pension (21, 18 and 12 year categories), Invalids, Free, Not likely to become efficient, Services no longer required, with disgrace, Miscellaneous, Struck off as deserters, net transfers to other Arms, to Permanent Staff TF, Struck Off, other Causes etc. 9,940 in total or 58 in every 100 men leaving the Regulars. Distributing these 9,940 men across 74 paired battalions, that represents (on average) 134 men per paired battalion in this single year. The numbers leaving in 1911 were the lowest for a decade.

1904. At the other end of the time-line, in 1903-1904 the data is: Effectives: 171,116.....Total Increase of 22,868, Total Decrease of 41,577 of which 24,584 transferred to the Reserves representing 59 men in every 100. The numbers not making it to the reserves for similar reasons to 1911 were 18,709 or the equivalent of 252 per paired battalion (read Line Infantry regiment).

As a General rule all numbers gradually declined as the Army continued to contract post the South African War. Using these two data points as a rough guide, it is easy to see how somewhere between 252 and 134 men per annum did not end up in the Reserves. A simple average across those two data points is 193. Across those 8 years that would be approximately 1,544. Add say, 120 for each year 1912, 1913 and 1914 that would approximate to 1,900 from 1904-1914. I will dig up the other historical data that I have (190-1906). I realise this is a simplistic extrapolation and service obligations varied. It needs a more thorough examination of the annual data.

Of the 2,400 'missing' Reservists, the above might explain a large proportion. Add to that some men allowed to live and serve out obligations abroad, deaths, incapacity, desertion from Reserve obligations after they had started in the Reserves would all add up over a decade. .....Just some initial thoughts.

MG

* Note this is not a Net Increase. In 1911 the Line Infantry saw an offsetting Increase of 36,838 the vast majority being recruits who would still be serving with the regulars in 1914.

Edited for clarity.

[spof]

What would happen if a man in Class B emigrated?

[Muerrisch]

What would happen if a man in Class B emigrated?

He was required to seek permission to emigrate or "pursue a seafaring life" and a record of the reservists with this permission was kept, together with a note of their nearest British Garrison. He would be included in the totals of reservists, as being liable to mobilisation. Obviously he could take some time aririving, but that is a different matter.

It is likely that the nearest garrison or consulate would be required to arrange his passage ....... other ranks were assumed to be feckless and incapable of going from A to B without an NCO or clear directions.

[Guest]

The Terms of Engagement for Line Infantry from mid-1902 were successively three years colours and nine reserve, nine years colours and three reserve, and seven years colours and five reserve. Any man enlisted after 4 August 1902 and before 5 August 1914 should either be a regular reservist or:

1. still serving, either the balance of twelve years, or having signed to go for pension after 21 years. Estimating this total is difficult but will be attempted.

2. Dead, either in service or during the reserve period. A reasonable estimate can be made.

3. Invalided, either in service or during the reserve period. A reasonable estimate can be made.

4. Discharged, either in service or during the reserve period, for a variety of reasons: by purchase, by indulgence, incorrigible behaviour, committing a felony and being in jail, and a few others. A reasonable estimate can be made.

5. Deserted during service or disappeared during the reserve period. A reasonable estimate can be made.

6. Transfer to another Corps

I WOULD WELCOME ADDITIONS TO THIS WORK LIST.

Grumpy -

Edit. You can calculate a proxy for the wastage in a single Line Infantry Regiment by taking the Line Infantry data for each year and dividing it by 74 (for the 74 paired battalions) ..........

The relevant page from the 1912 (with 1911 data) General Annual Returns etc are attached. They show the list of additional reasons. I have asterisked the parts showing the numbers transferred to the Army Reserve. I have assumed all others in the 'Decrease' section did not get transferred to the Reserve. This might be incorrect but my logic is that if they were they would appear in this section. As you will see there are a few esoteric reasons for 'decrease' which appear small, but when added up over 10 years do come to meaningful numbers. MG

[Guest]

..and the historical data for the Army as a whole. Note in 1910-11 the Line Infantry accounted for 49.0% of the Decrease of the Army.

[Guest]

This just about exhausts my data...here is the same data as above, but for 1903, 1904, 1905, 1906 and 1911 (I dont have the annual reports for the other dates). If anyone has them, the data would be gratefully received. . What is clear is that in some years (1904 for example) only 37% of the men leaving the regulars made it to the reserves. The data is incomplete but for the five years of data the 941 men made it to the Reserves and 1,105 didn't. I would guesstimate the numbers in the missing years might be in a similar order of magnitude, so we are looking at ball-park 2,000 men per regiment between 1903 and 1914 who didn't make it as far as the Reserves.

Note on methodology: The numbers were divided by 74 - the number of paired-battalions Regiments in the British Army at the time. In theory this should generate an approximation for the number of men reaching the Reserve each year and the numbers not reaching the reserve of a typical Line Infantry Regiment.

Any mistakes are mine. MG

[Muerrisch]

Martin has solved the puzzle as far as the infantry as a whole is concerned, in that there were so many ways whereby a soldier never reached the reserve, and so many years in which these ways could happen. I await eagerly the "granular" data of individual regiments or battalions, so that we can compare patterns rather than apply crude averages.

A few very minor points to make regarding the above.

The year in which the numbers reaching the reserve almost reached parity with those reaching the reserve was the end of the ill-thought-out "three and nine" colour service. This does not surprise me at all, in that the malign influence of the "many ways" had only three years in which to work. By extension, we might [all other things being equal] expect the converse when the 1904 cohort, signed for "nine and three" reached its nine year mark. We shall never know, because Gavrilo Princip started the Great War, and all bets were off.

Another point is that perhaps 1902 and 1903 should be disregarded when number crunching, as the end of the Boer War caused a huge "Peace Dividend" and distorted the figures.

And lastly many of the causes of not reaching the reserve continued to act on those who achieved reserve status [death and invalidity for example], so that fewer than half the reserve potential was achieved on mobilisation.

Well done Martin indeed. As Michael Caine is reputed to have said "not a lot of people know that!"

[Guest]

Dear Michael Caine

There are notes in one of the Annual Reports (1912) on the distortions in the data which you might find interesting. Rather than transcribe it I thought it would be easier to post the pages. It might be relevant for the future calculations. I will attempt to trick the system to post all the pages continuously... let's see what happens... Hmmmmm 628 KB on one post. Huzzah.

Note para 14 and 15 below clearly indicates the Army was acutely aware of the imminent swings in numbers. MG

[Muerrisch]

Thanks very much: I had read about the crisis elsewhere but never seen it acknowledged officially.

Good at book-keeping, our ancestors!

[Muerrisch]

I have got well into the number crunching assuming that RWF was "typical" regarding total "army" percentage annual gross wastage.

Suffice it to say pro tem that the army said goodbye each year to 7.56% of its strength at the beginning of the period, falling slowly to just under 5% at the end of the period.

The causes that I took to apply to SERVING soldiers [as opposed to men who reached the reserve] were:

death

invaliding

purchase after three months

purchase early

free discharge

inefficient

disgrace

miscellaneous

desertion

other

all of which applied almost 100% as a "when they've gone they've gone" factor.

So far my treatment has not considered men signing for twelve years or more, and drop-out whilst on the reserve. Even so, a quick back-of-the -envelope ball-park figure based on 6% every year does very nasty things to say 100 men joining in year ending 1904 who, it may be recalled, signed for nine years colour service.

After nine years we end with about 6 gone first year, 6 the next, 5 the next, 5 the next, 5 the next,4, 4, 4, 3, 3 total 45 gone, 55 remaining to pass to the reserve. This is blatantly crude and does not represent a thought-out solution but it does demonstrate that I was wrong to be surprised by how small the reserve was compared to the intake.

I have not checked the above, I think/hope any errors are small

[Muerrisch]

please is anyone out there keeping up with these musings and mumblings?

I have crunched the cadre of 1904-5 using the annual army wastage and only 197/317 are likely to have passed to the reserve.

[Guest]

Grumpy... My instincts tell me that unless we use the hard data for each factor for each of the ten years, the compounding effect of small errors over ten years over multiple factors might induce exponential margins of error. It is quite possible that your calculations so far are very close, but unless we start with the available data, I fear the margin of error might be large.

In my mind I see every tranche of recruits from every year prior to 1914 'decaying' over time to leave a residual number who made it into the reserves - rather like the profiles of various radioactive isotopes' 'half-lifes' ...one could draw similar charts for each tranche.

To help me understand your methodology - I assume you for example done separate calcs for the each of the years. Have you also done separate calcs for each of the alternative terms of engagement? e.g in 1906 Infantry men could join for 2 years, 7 years or 9 years if I am reading the tables correctly). Each would enter the Reserves at different times ( I know you know this!) but does your calc separate these tranches in 1906 and separate them in other years) .

Looking at the Returns for 1903-1906 the period of engagement seemed to vary a lot and in finite tranches (as I think you have highlighted in previous posts on other threads). Based on the tables, outside the 12 year engagement, the 'Short Service' for the Line Infantry seems to have been;

1903-04....3 years only

1904-05....3 years only [ ditto]

1905-06....3 year or 9 Years [ditto. And note the numbers joining for 9 years was 90.8% of the Short Service men]

1906-07....2 years, 7 years or 9 years [ditto. Split: 1.2% 16.4%, 82.4%]

As you know I don't have the data for the between years

1910-11....7 years only [ditto]

In there should be separate calcs for

1903-1904 men joining for 3 years

1904-1905 men joining for 3 years

1905-1906 men joining for 3 years

1905-1906 men joining for 9 years

1906-1907 men joining for 2 years

1906-1907 men joining for 7 years

1906-1907 men joining for 9 years

etc.... through to 1913.

I know you have forensic knowledge of the Reserve requirements. Just curious to know if your calc splits of the tranches?

Additional Complexity. Also I note that the returns also show the numbers who re-engaged after desertion, joined from the Militia or (later) the Reserve. Although these are small numbers, when added back at the start of each period, when compounded over 10 years the numbers might be meaningful. Not sure if your calcs have considered this element.

To further complicate matters, there are tables of Re-Engagements and Extensions of Service in each year. So for example, in the Line Infantry there are the following 20 columns in the 1910-11 tables for example which are populated with data;

Re-engagements to complete 21 years service with the colours....after 9 years Service

Re-engagements to complete 21 years service with the colours....after 10 years Service

Re-engagements to complete 21 years service with the colours....after 11 years Service

Re-engagements to complete 21 years service with the colours....after 12 years Service

Extension of Short Service From 3 (or 2) years to complete 6, 7 or 8 years ....under 1 year's Service

Extension of Short Service From 3 (or 2) years to complete 6, 7 or 8 years ....after 1 year's Service

Extension of Short Service From 3 (or 2) years to complete 6, 7 or 8 years ....after 2 years' Service

Extension of Short Service From 3 (or 2) years to complete 6, 7 or 8 years ....after 3 years' Service

Extension of Short Service From 3 (or 2) years to complete 6, 7 or 8 years ....after 4 years' Service

Extension of Short Service to Complete 12 Years......under 1 year's Service

Extension of Short Service to Complete 12 Years......after 1 year's Service

Extension of Short Service to Complete 12 Years......after 2 years' Service

Extension of Short Service to Complete 12 Years......after 3 years' Service

Extension of Short Service to Complete 12 Years......after 4 years' Service

Extension of Short Service to Complete 12 Years......after 5 years' Service

Extension of Short Service to Complete 12 Years......after 6 years' Service

Extension of Short Service to Complete 12 Years......after 7 years' Service

Extension of Short Service to Complete 12 Years......after 8 years' Service

Extension of Short Service to Complete 12 Years......after 9 years' Service

Extension of Short Service to Complete 12 Years......after 10 years' Service

And this is just for ONE YEAR. Going back as far as the 1903-04 Returns there are similar tables that also give the breakdown by Arm, so it is possible with every annual return to rebuild the wastage data at a very fine level. Just on Extension of Service over 10 years there could be 200 different calculations. Then multiply this by the other 10 or so factors. The re-engagements are not small numbers and again, when compounded over a decade will distort the data considerably. In near years (close to 1914) these re-engagements might have a significant reduction on the number of men entering the Reserves.

The bottom line (if you will excuse the pun) a bottom-up approach is going to be extremely complex and given the extraordinary power of mathematical compounding* over 10 years, I think the margins for wide error are large unless one uses the data from every year. With only half the data, it is a terribly difficult job.

If I can locate the correct WO Files at TNA for the missing volumes I hope to have the data very soon. I am extremely confident this can be done. While the complexity of the necessary calc looks large, once the data is in place, scaling the calculations is rather simple.

MG

* By way of illustration If we started with two samples (A and B ) of 100 each and grew Sample A by 1% each year and shrank Sample B by 1% each year for ten years, Sample A would be 92 and Sample B would be 108 at the end of the period...... If we were to do the same exercise with +3% and -3% respectively the Samples would, (after ten years) be 73 and 134. A huge differential.

More pertinently, for this exercise, an annual wastage of -1% versus a wastage of say -3% would reduce a sample of 100 to 92 or 78 respectively. Compounding very small percentages over multiple years can have large impacts. If the assumption in data are say just a few percentage points out, the conclusions might be significantly off. This is further exacerbated by the number of factors we are considering; the margins of error simply widen exponentially. For this reason given the complexity of the necessary calculations I think it is essential to get the available data. This will not eliminate errors but will significantly reduce the compounding effect.

A rather long winded way of saying the margin of error is hyper-sensitive to very small changes in assumptions. ...and where we don't yet have the hard data we are making assumptions. MG

[Muerrisch]

Martin thank you. The exercise may well be pointless, but it is a way of getting to understand the factors and the complexity that reduces potential reservists to 1/3 of input.

I know of no periods of LINE INFANTRY initial engagement [as opposed to optional extensions after engaging] than:

7 & 5 until 1902 then 3 & 9 until 1904 then 9 & 3 until 1906 when reverted 7 & 5.

1902 May
3 years colours and 9 on reserve.
Army Orders (AO) 117/02

1902 Jul
Extensions for those on 3 years or 7 years initial engagement, can extend to 8 years or 12 years
AO 159/02

1904
Proposed, not implemented: a ‘2 years with colours plus 6 years reserve’ engagement for Home, or ‘ 9 years with colours plus 3 year reserve’ for general service
The Development of the British Army

1904 Nov
Terms of service 9 years with colours and 3 years reserve
AO 189/04

1905
2 years with colours and 10 years reserve for certain large regiments tentatively examined
AO/204/05

1906 Sep to 1914
7 years with colours and 5 years reserve for all. Extensions to 7 years allowed
AO 209/06

War
3 years or the duration, and also as above

Curious to know where your variations come from please!

As for: A rather long winded way of saying the margin of error is hyper-sensitive to very small changes in assumptions. ...and where we don't yet have the hard data we are making assumptions. MG

don't I know it! For 41 years I was a Scientific Civil Servant, as a meteorologist, ending up as Chief Met Officer British Forces Germany as a Principal Scientific Officer. All of my woes were the weather being hypersensitive to changes in initial states.

[Guest]

Grumpy.

Meteorology. Brilliant. So you'll know all about Chaos Theory...... I was not trying to discourage your very noble attempt to solve this enigma. I will send you the relevant pages and you will see the tranches. It is the very reason I highlighted them as they didn't quite fit with the terms of engagements I had seen from previous posts or threads .... and I double checked it was Infantry and not another Arm.

The pursuit is certainly not pointless but it is important to be aware (as I know you are) of just how slippery this data is. We grind slowly onwards..... MG.

[Muerrisch]

One source of error which might not ever be eliminated is having to use estimates of annual enlistments into the regiment, when all I have to go on is plotting regimental number against enlistment for the data points that I and Paul Nixon have dug out of the records. Now if the annual regimental data have the total per regiment per annum, that is one source of error sorted!.

Another consideration regarding men who opted to extend their service. These come into the "when they are gone they are gone" category as far as ever reaching the reserve is concerned UNLESS THE EXTENSION, WHEN ADDED TO SERVICE ALREADY ACHIEVED, FALLS SHORT OF TOTAL 12 YEARS. The long term and readily checkable army-wide % for men serving 12 years of more is 1.5%. I Have not looked at an infantry value yet but I would be surprised if very different.

Look at Section II Part VII Table 5, Comparative ages of serving soldiers. 1.54% men were 30 or over in 1911, and the value every year before hovers at 1.4% or above. The vast majority of these men would have enlisted between ages 18 and 22 or so, and either "went for it" or subsequently extended.

What I am trying to do is join ends to middle: I know more-or-less how many enlisted 1902 to 1914 August, and I know exactly how many reservists were available immediately before the war. What I/ we are trying to do is explain the wastage in one regiment by number crunching.

I performed a comparable analysis of subterranean mining by the British in the Great War [and previously], using the RE formulae for how much bang was needed for each 'ole, and seeing if the 'ole came in on spec. Stand To! was kind enough to publish it several years ago. The formulae were progressively refined over the period by the engineers. I ended up by having to plot the results on a log/log scatter diagram because of the disparity in scale of early Chatham experiments compared with the biggest [accidental] bang at RAF Fauld underground bomb dump in WW II.