EDITOR'S NOTE: This section of EEL edited by Paddy Griffith is intended to: (1) debate all sides (positive or negative) questions and/or comments on the Penisula War, etc. (2) publish our readers articles and comments on the same (see Paddy's statement of policy in EEL 78).
The section can only be successful with our readers active participation. Please send your contributions, comments, etc.
Readers will recall my "Havanitos" guessing game (EEL 81 p.17) which was later revealed (EEL 82 p.39) to be based on the second battle of Castalla, 13th April 1813. In this article I hope to round out the series by discussing the replies I have received from readers, and drawing whatever conclusions seem appropriate.
Four readers sent in their thoughts - Howard Whitehouse from Georgia, James C. O'Neil from Arizona, and the battling Zuparko brothers - Ned and Bob - from California. (For convenience in what follows Z have coded these names according to their - and my - initials). Although this is scarcely a large sample, I found it diverse and thoughtful enough to give me plenty to chew on. I am especially grateful to all four of these gentlemen for playing my game. Thank you:
The first question concerned the identity of the battle. This was spotted only by HW, while BZ fell into my crafty trap by mistaking the battle for that at San Marcial - another notable Spanish victory in 1813. Not that it really mattered which battle it was, but the 75% failure to identify it did reassure me that respondents were answering on the merits of the case rather than on remembered reading about the specific event.
The second question concerned casualties, for which the results came out as follows:
( * Note that only two of the four respondents differentiated between the various combats within the overall. results of the 1st Div. assault).
I concluded from these results that respondents seemed to under-estimate the mortality on the Napoleonic battlefield. In this particular case my own estimates had been fairly solidly based on historical evidence, I felt, so 1 was persuaded that respondents were genuinely erring on the 'humane'side. a number of provisions must be made, however:
- I had accidentally misled them about the proportion of casualties at Biar, when I said the allies suffered "relatively small loss to themselves". Of course this loss was "relatively small" only in the context of the 3:1 odds against them and the ; hours' duration of the combat. The allied loss was actually about equal to the French loss in numerical. terms.
- JCO stated that his figure= for losses were the numbers of killed and wounded only, and therefore excluded the significant numbers of combat skulkers - those who "just : an away or got lost in the shufflle. He indicated that there might be 2 skulkers to every casualty, thereby adding approx. 100% again to the overalll average of responses. if James' skukers are included as 'losses', then the respondents' overall estimates are pretty close to the mark, and indeed soars beyond it. This raises a very important question of whether or not skulkers can indeed count as casualties. After a burst of incoming fire the skulkers may not be immediately available to resume the attack; but they will nonetheless be available to answer the roll-call a day or two later.
- The respondents apparently shed their inhibitions about casualties in two specific cases - ie they overestimated both Spanish and French losses in the fighting on the western flank of the main attack (where the Spanish were most heavily engaged), and they placed losses in the rearguard right at about equality whereas 1 had guessed (and this time it was only a guess) that the victorious side would have done better than the loser.
I was struck that respondents seemed to think of 'British vs French' clashes as relatively gentlemanly affairs - without too many French losses and hardly any British - while they assessed 'Spanish vs French' combats as very bloody indeed. I detected in this a certain 'hangover effect' from strategic and even political stereotypes into the much more technical field of small-unit tactics. There did seem to be a "racial interpretation" straining away in there, somewhere, baffled only by the inescapable fact that Castalla was actually a Spanish victory.
There's nothing anachronistic about all that, of course, since most commentators the time of the Napoleonic Wars went out of their way to stress racial differences. We may even ask ourselves how far the contemporary casualty returns themselves were affected (unconsciously) by such considerations. It may well have been the case, for all I know, that Whittinghmm counted in his skulkers when he reported the Spanish casualties, while Adam and Reeves made every effort not to do so when they reported theirs. It was at any rate easier for a Spanish skulker to escape and evade in Spain than it was for a Briton or a Frenchman.
Another general tendency which struck me was that the respondents seemed to downgrade the casualties caused by skirmish fire as opposed to volley fire. NZ actually stated in his letter that he felt the number of hits per round fired should be halved for skirmishers as opposed to line volleys. Clearly this is an important and contentious area for discussion, but the historical record at Castalla does seem to indicate that skirmish lire was every bit as lethal as other types.
The third question concerned the timings, and from here on my 'control' answers started to become progressively less firmly based on historical evidence than they had been for the casualties. The table of results came out as follows (expressed in numbers of minutes):
| Action | Average of Responses | PG (in EEL 82) |
|---|---|---|
| In 1st Div. Assault | ||
| Flank Skirmish Action | 65 | 210 |
| Attacks on Spanish | 47 | 180 |
| Attack on British | 20 (5-45) | "5-60" (!) |
| Rearguard Action: | 75 | 180 |
In this case there was an even greater divergence than with the casualties between the responses and my own estimates based on the evidence I had before me. I must confess that the responses seemed to me to reflect pretty accurately what I, as a long-term wargamer and student of the period, would have expected for the levels of action portrayed. Yet the scraps of evidence available seemed to indicate that each action really lasted about three times as long as we war-gamers might have conceived. Maybe there were long unreported pauses in the action (eg for the putative duel on the 2/27th's front) - and perhaps in our wargames we should make far greater allowances for such pauses than is usually the case.
But on the other hand it is also conceivable that the combats themselves really took far longer than we tend to imagine, especially when it. was a matter of skirmishing or manoeuvering at long range from the enemy. My own inclination (reinforced by my recent researches into the American Civil War) is to posit far more indecisive, and relatively bloodless, firing at long range than is generally imagined. Admittedly this does not adequately meet all the time problems raised by the close-range action of 2/27th at Castalla; but it may go far to cover all the rest of the battle.
My inclination is to assume (albeit without specific evidence) that the French vs Spanish combats at Castalla were pretty much stand-off affairs. I would be delighted and intrigued if any reader could suggest some scenario in which there was serious close-in scrapping in these combats, within the level of casualties which apparently applied.
The fourth question was all about ammunition expenditure, which NZ immediately realised was intimately connected with casualty rates and timings. (His estimates of casualties were based on an assumed duration of fire, hence assumed number of rounds fired, hence assumed casualties. He used an assumption of 'one round in 600 hits anyone' and compared it with the Hughes/Lauerma assumption of 0.2-0.5% hits for musketry. His final estimates were based on juggling these two figures against the traditional notion that "a winner suffers 10% casualties while a loser suffers 20%". All a bit 'hearsay' I suppose - but at least his reasoning seemed to be more scientifically explained than that of any other respondent).
The table for rounds fired came out as follows, expressed as infantry rounds fired per man (artillery is discounted due to lack of response):
Once again my respondents guessed at a lower level of action than I had, allowing only 57 shots to be fired for every 100 of mine. This proportion very neatly matches the different casualty scores assumed by them and me, as it happens, so it emerges that all of us were assuming approximately the same sort of lethality per shot fired (i.e. neatly within NZ's 0.2-0.5% hits). They assumed about 1400 casualties with about 340,000 rounds, while I assumed about 1800 with 600,000, For the record this averages around 0.3% hits in both cases.
The main difference between us seems to be, once again, the question of duration of fire. The respondents allowed less fire because they envisaged less prolonged action. None of their soldiers came near to firing off all their cartridges, while quite a lot of mine appear to have done so. (Its a pity, in a way, that Castalla didn't go on longer, because then we might have started to hear complaints of ammo shortages in the documents!)
The trouble with the 'rounds fired' question at Castalla is that (to my knowledge) we have no documentary evidence for the number of rounds fired, nor even any very firm indications on the timings. Hence, a variety of different interpretations can be placed upon the known casualties.
Either they were caused by the sort of protracted tiraillement that I have assumed, and the 10.3% hit rule is vindicated, or they were caused in a much shorter time (as the respondents assumed) but with considerably higher lethality than 0.3% -- let's say 0.5 or 0.6%. Admittedly this miniscule percentage doesn't sound a very dramatic difference from the original. assumption; but in fact it does represent an increase in firepower of 70-100% which is surely a significant overall revision and affects the whole way that we visualise this battle. The readers now have all my evidence in their hands, and are in a position to decide for themselves which model they prefer.
The fifth question concerned the fortifications, which left respondents totally divided. Two of them did at least agree that the effect of fortifications must have been significant but not decisive, since the Spanish surely derived most of their advantage from the terrain itself rather than from artificial improvements to it.
The sixth question was to find the 'man of the match'. If I may descend into a points system to record the votes. I find the final table is as. follows: Whittingham - 7 votes; McKenzie & Adam 3 votes each. Adam would have had more had he not been specifically voted against by one respondent, who felt that his achievements were no more than the normal line of duty and hence less impressive than Whittingham's inspired manoeuvres with dubious troops. Reeves, alas, came bottom with only 2 votes, both from me. So Whittinqham wins the gold.
Finally, there was some disagreement over whether or not Castalla would be any good as a wargame. Several people pointed out that it would be very hard to ensure an historical result, given the efficiency ratings normally attributed to Spanish troops, while NZ made the pertinent point that there were too few decisions in the battle for it to make an interesting game.
Equally, 'hidden movement' and 'psychology' played too much of a part for Castalla to be susceptible to treatment in a conventional table-top format., he thought, so some advanced ("WD Style") concepts would have to be used. Why is it that conventional table-top games always seem to be ideal so long as no one tries to apply them to a real battle?
Back to Empire, Eagles, & Lions Table of Contents Vol. 1 No. 84
Back to EEL List of Issues
Back to MagWeb Master Magazine List
© Copyright 1985 by Emperor's Headquarters
This article appears in MagWeb.com (Magazine Web) on the Internet World Wide Web.
Other articles from military history and related magazines are available at http://www.magweb.com