During World War 2 the USAAF looked into the concept of placing armour on their planes. Numerous airmen were tasked with the job of counting and noting the location of bullet holes on aircraft returned form missions. This study has become a classic of bad data collection. The airmen were only recording damage that allowed a plane to make it back to the airfield. According to their data pilots were never, ever shot through the head.
My recent blog posts have triggered an unexpected discussion on the value of bayonets. I find myself in the unusual position of agreeing with certain points of both sides of the argument. One argument that commonly crops up is what might be termed “the bayonet wound fallacy”.
You will see a statement that “very few wounds are inflicted by bayonets”. Read a few military histories and you will find a number of accounts of bayonets being used when charging the enemy. These are most common in the nineteenth century and earlier when single shot weapons were the norm. There are accounts, however, of bayonet use in more recent conflicts such as Vietnam, the Falklands and Afghanistan. There is a clear discrepancy here.
A related claim is “very few bayonet charges are completed. Usually one side or the other loses its nerve and retreats”. It is worth noting that a retreat may not be due to cowardice. If the contested ground has no strategic or tactical value it is good military sense to fall back and subject the new occupants to a mortar and artillery barrage on the previously registered coordinates. Clearly some bayonet charges did fizzle out, but military memoirs also describe encounters where the two sides do clash.
The reason for this discrepancy is also bad data collection. Two things are likely to happen to the losers in a close combat fight. If they are relatively unscathed they may escape. Sometimes they will flee only to be shot down by covering forces. Alternately the loser will be stabbed, struck and stomped on until they stop moving. Surveying how many bayonet wounds are being treated in a field hospital does not give a true picture of bayonet use. The majority of bayonet victims will never make it to the aid centre.
If you applied the same research criteria to the aftermath of a battle in Roman times you would probably get a similarly erroneous result and conclude that Roman legionaries seldom used their swords! It would be likely that there were very few sword wounds in the field hospital. The rear ranks of a cohort routinely killed fallen enemies as they advanced. Enemies that tried to run were often cut down by cavalry.