Cynical things to learn first in Arms and Armor

Posted on 4th October 2018

I’ve started to write a second portion of my “rules of quality research”, this time dedicated to what I consider most fabulous failures of historical science, but then it became obvious that it is easier to start with Arms and Armor – which as it is day is ridden with failures. So here is my short list of things a beginning researcher must learn to doubt or even dismiss out of hand.

First is when someone presents a mathematical formula that can estimate weapon’s “efficiency”. Basically once a year you always have a college graduate rediscovering that mechanical moments or something else from Physics 101 can be applied to swords, and then it gives you 1000 N*meter moment per Katana and 876 for a broadsword, which means there is roughly 15% gain… Complete nonsense through and through, as the problem of weapon’s actual performance is intrinsically multi-dimensional (the speed with which the required energy is delivered to the weapon versus the amount of such energy, penetration ability against armor versus such against flesh, reliability, weapon’s cost etc. etc.) and nonlinear – especially when one also considers amount of skill required and whether in a given engagement a swordbearer would actually have time/space to perform the required technique. No single number can capture all of those; while some formulas can be useful, everything seen so far in the sword world is uneducated folks trying to prove that the weapon they like (and usually the only one they know something about) is in fact objectively superior.

Same goes for all and every claim of “super steel”, the one that cuts through everything and never breaks, produced by “secret techniques” sometime in the 14th century. Those are constant and consistent, the only thing that change is which steel or weapon is claimed to be an unprecedented marvel. European ones seldom are, because they tend to be researched by more cynical people (with real Ph.D. and metallurgy experience) and after WWII white nationalism is out of fashion in Western universities, while non-white nationalism is vividly encouraged. The said  claims in 99% of the cases are supported by exactly zero evidence and rely on circular citations – Yamazaki sensei quoting Britannica article, the article – Baron von Rosen’s “Travel in Tartaria and Orient”, and von Rosen’s original statement is typically something along the lines “in all countries I never found a steel equal to that of Tartars”. Which is about as vague and pointless as one can imagine.

One should also tread carefully when someone attributes changes in weapon’s shape to regional environment, or development of either new armor or new fighting techniques. These arguments are not to be dismissed out of hand, but one needs to keep in mind that to be valid they have to evolve from speculations to actual comparisons against period records (wound types, casualty figures etc.), or in the very least against parallels in countries with similar environment or armor (the rule of quality research – if we postulate that A causes B, it must cause it everywhere). What one often sees instead is 20th century martial artists developing very complex fighting techniques that use daggers or throwing knives or something much more exotic – and in historical records the total number of those actually killed by such weapons on a battlefield being basically 0.01 percent overall. Typically at all times you have 2-3 weapons accounting for more than 90% of casualties; the level of optimization of the rest therefore is something highly theoretical in nature. And even the evolution of these 2-3 weapons is driven by many economical and even random factors (for example, blind and mistaken imitation of someone), so that the exact impact of functionality can be quite uncertain. One can say that much of the very basic functional development occurred already in the Bronze Age, what Iron Age saw is a variety of implementations of more complex and advanced forms.

Related – any statement that a war was lost because of disparity in technological advancement requires fact check. If you look at the works of military modelers who try to understand the efficiency of WWII militaries, you’ll find that even with similar (or even inferior, as was the case with Germans in 1944-45) technology employed by both sides there is still an order (!) of magnitude difference in efficiency, i.e. ability to inflict casualties and win the battle. When the militaries compared are from far away regions, this number can jump to 30-40 (!). At the same time total air superiority in WWII gave you only a 1.5 maximum boost. A better sword at best would give you a substantially smaller increase, and always-popular-choice – spear, did not even evolve that much in the past 2500 years. And then the arms market was surprisingly open and efficient even in Roman times – and eager to provide weapons to countries having no native production at all. Actually it is very surprising how many of large and powerful states remained comparatively speaking insignificant in manufacturing!

One must doubt any sentence which compares the facts separated from one another by more than 400 years. History is in general non-linear (spiral, circle and other oscillatory behavior is common) and any proof that involves both Bronze Age and 19th century but nothing in between usually does mean that nothing in between actually adhered to this “universal” trend.

Last but not least – almost any conclusion reached on the basis of linguistic, metallurgical or other analysis which does not take into account that the object being studied is a medieval sword – has to be taken with many grains of salt.

Linguists live in the world largely confined to modern ethnographic research and a select few chronicles. The latter were written by so few people that small changes in mannerism do reflect differences in time and space – while swords were quite often signed based on highly conservative and persistent standards that existed specifically in smithing community. As a rule these signatures are not datable solely based on script or grammar used. With metallurgy the issue is more fundamental, as all non-invasive methods rely on aggressive fitting of surface scattering profile to a combination of pre-computed profiles of specific elements. “Press a button to get result” systems will not give alternative fits nor will elaborate on how the fit’s quality can be impacted by various assumption. The result is that very often one finds Uranium or Rhodium in ancient steels, as these are expensive materials that modern fitting software made for “ready to use” market, i.e. scrapeyard, wants to identify. In the very least one needs to carefully repeat the fitting allowing only for a typical composition of medieval steel – C, Fe and various “issues” – S, P, aluminum oxides etc, possibly Va, Ni, Mb or Ti if the ores are known to naturally contain them. Of those C is the most elusive as its scattering properties and spatial distribution in some steels can easily create a false reading. Actually, all steel companies today test for C content using destructive mass-spectrometry. A couple of bonus points. First, there is no such thing as non-invasive technique that “ignores” patina, at best we are talking about 1um of analized layer thickness. Second, the further back in time we go what changes is not so much the average composition, but its sigma. First century AD sword whether in Rome or East Asia can be pure Iron, but can sometimes be a high class multi-part steel with differential heat treatment and 0.7% carbon. You start seeing more uniformity towards 11-15th centuries, but until then whether with iron or soft metals one should do statistics rather than a single measurement.

Last Articles

Nihonto in numbers

Posted on 18th December 2018

After posting the statistics regarding “Oriental” weapons I’ve decided to do a similar study on Japanese swords, i.e. combining the results of sales I monitored for the past 10 years. As before, in quite a few cases those had to be extrapolated (hence the round numbers) since some months were simply...

How rare are the Eastern Swords?

Posted on 18th December 2018

How rare are historical edged weapons when it comes to things we call Middle Eastern, Islamic, Eastern, Caucasian – basically the whole region we attempted to cover in “A study of the Eastern Swords”? I’ve looked through my database of sales, compiled for the past 10 years and calculated the total number...

Kiyomaro, modern papers and 100% guarantee

Posted on 23rd October 2018

At each assembly of American collectors there is always plenty of private talks about supposed issues with this or that kind of papers and various conspiracies involving NBTHK, yakuza or “major Japanese dealers”. It usually boils down to that there are bad papers (green, early Hozon, late TH, NTHK-NPO etc....

Cynical things to learn first in Arms and Armor

Posted on 4th October 2018

I’ve started to write a second portion of my “rules of quality research”, this time dedicated to what I consider most fabulous failures of historical science, but then it became obvious that it is easier to start with Arms and Armor – which as it is day is ridden with...