Obviously, swords vary in quality. Ideally, I would like to know what a standard army issue sword would have cost, if I bought it privately. But I realised we may only be able to assert what this sword cost the army. Failing that, I would be interested in any recorded sword prices.

I have access to PART of „The Economic History of Byzantium: From the Seventh through the Fifteenth Century“. Therein I found no answer.

Another paper that will help to put your answers into perspective is this one. by B. Milanovic. It asserts that:

The average “modest” salary (including benefits in kind) for an unskilled worker was estimated by Morrisson and Cheynet (2002, p. 869) at about 1 nomisma (N) per month or between 10 and 12 N per year.

  • 4
    What I know is that an axe for wood cutting, in midieval Russia around 1400s costed slightly more than 1 year income of an average peasant.
    – Zmur
    Apr 10, 2021 at 9:18

2 Answers 2


I am using some information that I found in Uzbrojenie w Polsce średniowiecznej 1450-1500, edited by A. Nowakowski (2003).

It is, obviously, concentrates on that period of time, but there are also other information, to put some equivalence between different periods and regions.

So, for example, a sword would cost - depending on quality and additional accessories and ornamentation - anything between 15 to 70 grossi pragenses in about 1472 in Poland. Cheapest would not be of arms quality, but acceptable "standard issue" - for actual, true combat - sword would go for about 40 gr. For a comparison a cow was worth 40 gr, as well.

Same source mentions, for example that price of a greatsword in 1320s England was 3 shillings 4 pence. Which calculates to about 40 Novus denarius, thus about 40 grossi pragenses.

One pence can be equated to Carolingian Novus denarius (following Carolingian system of 1 pound = 20 shillings = 12 pennies), which was equivalent to 1.3g of pure silver. One can roughly convert between silver and gold by 14:1 ratio, and in the OP period one can say 1 solidus (nomisma) would be worth about 24 Novus denarius. Of course, due to all kinds of shenanigans those are more like theoretical numbers, because - for example - nomisma dropped from about 95% purity to about 30% purity in just 50 years between 1034 and 1081...

So, while the idea of the penny as 1/240 of a pound of silver kept constant for a long time, the value also was debased. For example, when established in 1300, grossi pragenses was 3.5 g of pure silver. 100 years later it was 2g of pure silver and in 16th century it was less than 1.2g...

Anyway, between 1000 AD Constantinople and 1470s Poznan in Polonia Major there are major differences... Mostly population increase and improvement in productivity and food production. Poland would have then an abundance of foodstuffs, compared to 1000s Constantinople.

My estimate is that standard issue military sword in Byzantine army would go for about 40 novus denarius calculated based on silver content, which would be about 2 nomismata (gold coins).

It is very hard to estimate prices, because, as mentioned, quality and additions of a weapon can triple or even quadruple the value, with work needed would be the greatest impact (that is better sword takes longer to make). In addition, there are factors that will impact it further. For example, 40 gr for a cow in 1470s Poland is way, way cheaper than 3 nomismata in 1000s Byzantium.

If you want, you can look up how prices in Byzantium changed over time here: Prices and Wages in the Byzantine World


After reading comments from @justCal (on the topic unrelated), I realized something: in 21st century, some of the words and idioms now carry completely different context.

For example, OP asks for a "standard army issue sword cost if I bought it privately". This wording is highly misleading for two reasons: there was no "main contractor" back then and military-grade was considered to be "of the style and size", but not quality.

It is true that Roman legionares were, at least in later times, equipped from State-owned armories, but it is unclear who actually produced these weapons. Some researchers even mention actual factories producing equipment and arms, but that is late Roman Empire, not Byzantine Empire 800 years later. Because of the limited sources on that time and place, it is really hard to even be sure that there was a "standard", let alone a "standard issue", because due to the time-and-resource-cost of producing a sword back then, that production would be scattered across the Empire.

Of course there would be large workshops - or possibly even factories; Byzantium was built on the foundation of Roman Empire, so it would inherit a lot - which would specialize in producing weapons for the state-organized military, but literally one would not be able to produce ALL of the arms needed (it was not just sword, but armor and throwed weapons, larger weapons - onagers, scorpions, ballistas...), and even then it would be by hand-made all the way. Which, in turn, would mean that no two spathion would be the same.

It is safe to say that "military-issue" weapons at that time could be bought privately without any trouble and they would be, in fact, same thing - one of the swords available for purchase in any given workshop would be bought by a bodyguard for a citizen and literally one next to it could be sold to a kentarch leaving for a campaign, or contracted by local garrison.

Which brings me to the second point - for much of the history "military-issue" would be synonim of "well-made, but not exceptional". Up until early 1950s what was available for military was available for citizens and quality of the privately owned arms were of much higher order. This is, for example, a source of advantage of the American Revolutionaries - their hunting rifles were more accurate than standard Bessie musket provided to the British troops.

The notion that military-grade equipment even today is misleading, because - to quote an old military axiom - every piece of equipment issued to a GI was made by the lowest bidder. Which, by definition, doesn't apply to a private purchase and even if in earlier times state did issue such equipment to it's military personnel, it was NEVER better than something similar (in utility and purpose) available on the market. It flipped only in recent times, and even then, it's still not the issue of availability, but of price. For example - I know of at least one person that owns SEVERAL F/A-18 fighter jets. Obviously, the reason why there isn't a lot more is basically price and operation costs... Same with M1 Abrams tanks - there are several people in and outside of US that own at least one of them...

  • None of this looked like ChatGPT to me. None of the hallmark traits, and that second sentence, @justCal, has a grammatical error that ChatGPT wouldn't make. This looks fine to me.
    – cmw
    Apr 20 at 14:28
  • @justCal Oops, wrong site: "gptzero.me" is the one I used. I chose the "one pence" paragraph, and it still showed 1% chance it's AI. My bets would be on the detector you're using, potentially. I wasn't familiar with the troll here, but of course, they're bound to rear their ugly heads. I'm pro-removal of GPT here for sure, so my comments are really limited only to saying that this looked legit to my eye and that better detection methods are needed (though I worry it will be impossible to do it fully by machine). Also, all this might need to be scooted off into a chat room somewhere.
    – cmw
    Apr 20 at 16:27
  • @cwm - the problem with gpt detectors are exactly what I pointed out earlier: just look at syntax and grammar and you can immediately know GPT-generated content, but this is also a way to defeat the detector... Just delete or move a comma, replace one word with another and probability a detector would assign to a text drops like a stone in a well. And last but not least - what is wrong with writing an answer and then polishing it with GPT? I have added another answer here - which I will mark as such and delete soon - so that you and justCal have a comparison.
    – AcePL
    Apr 21 at 9:03

This is next to an impossible task unless there are some remaining archives with military expenses at the time that could bring us closer to the answer. It's good that you have mentioned a relatively specific type of sword but even so the quality of materials would have varied and prices the same. If we take into consideration the wages at that time I would estimate between half a year to a year's pay depending on the quality and/or any other craftsmanship requested by the buyer at the time. However, please consider that the price limit on any weapon can be limitless, depending on the one making the purchase, same as today. If we also consider the period, during Basil II and all the campaigns during his reign, the price of a sword may be lower as production would have been high even before his reign started due to civil strife.

  • 4
    It seems to me that a reasonable estimate can be given in terms of the number of days of skilled craftsmen's time even if not in currency. How many swords could a Byzantine production worker of the time produce in a year? Was materials cost a significant fraction of the total cost?
    – Mark Olson
    Apr 3, 2021 at 0:22
  • 1
    @MarkOlson - I can answer part of that, a Spathion had 12-15 blacksmith work hours per sword, a Paramerion had 10-12 hours per sword, from "Logistical considerations for the arms production industry in the Middle Byzantine Empire" in Byzantine and Modern Greek Studies Vol 44 issue 2.
    – ed.hank
    Aug 3, 2021 at 21:04
  • @ed.hank - I would dispute the numbers, because swordsmiths I asked dropped numbers of at least 3-5 days, depending on the quality of the sword. On average, they said they would need about 50 hours to make a combat-capable sword... But that's just a side note, because even when counting 10 hours for a paramerion, it translates to about 12 years to complete an average order for military equipment of 5000 swords...
    – AcePL
    May 5 at 21:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.