you ever had to deal with a TCP/IP stack implemented in 1996 that was "good enough" to keep selling on devices today

This week on the blog I want to take a brief detour into discussing historical coinage, particularly in the context of modern fantasy and roleplaying settings. In particular, the notions I want to tackle are first how did ancient currency systems work in terms of value (what could you buy with how much) and then second how often were people likely to use physical currency at all? This is going to be a bit of a ‘fun one’ because while we’ll talk quite a bit about how money is used in historical societies, we are going to loop back around to fantasy settings at the end.

And the fantasy conceit that has sparked this is, of course, the ubiquitous general-purpose RPG currency, ‘gold,’ understood to mean gold coins or gold pieces. Now of course in many cases the trope-maker for ‘gold’ as the basic unit of currency is Dungeons and Dragons and folks will be quick to note that D&D coinage has always included smaller denominations: copper, silver, gold and platinum pieces on a decimal-system valuation. And sure, that chart exists in the rule-book and some common everyday things have their value listed in cp or sp, but even a casual glance at something like the weapon table reveals a ‘gp’ based currency system. The 3.5e weapon table, for instance, every weapon with the exception of sling bullets has its value denominated in gold. 5e is a bit better, but not much.

Meanwhile, in Baldur’s Gate III, almost certainly the most widespread and culturally pervasive form that D&D has taken in at least the last decade – far more people, I suspect, have played BG3 than have played any form of D&D tabletop – compresses the system down neatly to just the single currency type: gold. As did Neverwinter Nights before it. Likewise, the Elder Scrolls games, including Elder Scrolls Online and Skyrim have a single currency, called ‘gold,’ represented in game by very obviously gold coins.

(Credit where credit is due: Obsidian’s Eora, the universe where Pillars of Eternity takes place, dodges this problem: every culture has its own currency and you see them regularly as loot. The game then denominates them all in a copper currency unit of account, which is actually a lot like how the sestertius (a copper-alloy coin) is used in Roman accounting. As we’re going to see, the key here isn’t just ‘have currencies other than gold coins’ but also ‘have some sense of how big a unit of account a gold coin is going to be’ and Eora is one of the few settings that seems to have absorbed, correctly, that even a single gold coin is such a large unit of money as to be useless in most circumstances.)

So across a wide range of fantasy products – games, films, books and settings – this tends to be system: ‘gold,’ by which is meant gold coins, are the standard unit of account, values are reckoned in gold and when money needs to be shown, it is typically physical gold coins. If there are smaller units, we don’t see them often. Crucially, characters in dialogue will often use ‘gold’ or the names of gold coin denominations (‘crowns,’ ‘sovereigns,’ etc. shown in the fiction to be gold coins) as synonyms for money. Sometimes there’s a larger unit, almost invariably ‘platinum,’ which is also a pretty silly currency to have given that apart from some evidence that it was alloyed with gold in Egypt and South America (perhaps unknowingly so), no one is using platinum or aware of its existence before the 1500s.

And, as you may have guessed, there are some problems with this: functionally no one used gold in any amount in every-day transactions in the ancient or medieval Mediterranean (or most other places!), because a gold coin at almost any size was such an enormous monetary unit as to be unsuited to most transactions. That in turn conceals some of the sharpness of wealth and class distinctions in pre-modern society in ways that flatten and frankly ‘modernize’ these societies.¹ And it also misunderstands the economic systems of these societies, because it doesn’t understand what sort of transactions people would even want to use money for, which further flattens and modernizes these societies.

Instead, what I want to do is lay out a couple of real historical currency systems – we’re going to look at ancient Greek and Roman currency, as well as the medieval pound/shilling/pence (or livre/sou/dinier) system – and talk about how they are denominated and why.

But first, as always, I too appreciate money and always wanted to take up collecting ancient coins (which, I should note, can be legally acquired at prices attainable by mortals, because we have so many – just be sure you are getting something with accurate provenance; any reputable dealer will cheerfully supply you with this). If you want to help me to take up expensive hobbies, you can support this project on Patreon! Amici of the blog at Patreon get monthly updates on my research progress (or lack thereof), while patrons at the Matres et Patres Conscripti level also get to vote on future topics. If you want updates whenever a new post appears, you can click below for email updates, or you can follow me on Bluesky (@bretdevereaux.bsky.social) or Twitter (@BretDevereaux) or (less frequently) Mastodon (@bretdevereaux@historians.social) for updates as to new posts as well as my occasional ancient history, foreign policy or military history musings; I am probably more active these days on Bluesky than Twitter.

Introducing Our Currencies

Globally, both the idea of currency (by which I mean an abstract notional unit of value) and coinage (by which I mean a physical object representing that unit of value) were invented in more than one place at more than one time. These are, I should note, distinct ideas and it is the abstract unit of account which comes first, long before (and we mean centuries before) someone gets the bright idea of using specific objects to represent those notional units.

That said, all of the coinage systems of the broader Mediterranean world seem to spark from a single source, the development of coinage in the kingdom of Lydia in the seventh century.² The way to understand these coins is this: these societies had already been using metals – measured by weight – to define abstract, notional units of value for accounting purposes and in some cases physical transactions. This is important to note: money in the abstract sense (and debt, for that matter) come first and coinage comes second. In practice, what a coin was simply a pre-measured amount of precious metal, stamped by the authorities to attest that it was the amount it claimed to be.

Note the immediate implication that has: the coin is only worth the metal it has in it. If you reduce the weight of the coin or dilute the precious metal in it (by alloying it with baser metals), you have lowered the value of the coin (and probably committed a serious crime, unless you are the state). This meant that while states could get cute and try to stretch the treasury by issuing coins with less precious metal in them (called ‘debasement’) in the long-run this effectively led to inflation: as folks realized there was less silver or gold or whatever in the coins, they’d raise prices accordingly. But we needn’t get into all of the complexities of minting and debasement here.

What I want to do first here is introduce our currency systems and what they’re called, so we have that on hand for when we discuss how they’re valued and used.

The first coinage in Lydia doesn’t seem to have penetrated very far in society – the coins seem (the evidence isn’t great) to have mostly been used for large transactions, long-distance trade, that sort of thing. Instead, it is in Greece, which adopts coinage from the Lydians, where we first see coins penetrating deep into society and becoming a standard way to do business. Now each Greek polis issued its own currency (except Sparta), so instead of just one set of ‘Greek currency’ you had a whole mess of different polis currency on different weight and purity standards. That said, successful currencies tended to be imitated and so a ‘standard’ (which other Greek currencies might deviate from) emerged: Attic coinage, the mostly widely used, eventually becomes that baseline.

The Attic currency standard was based – as nearly all Greek currencies were – on the drachma (shortened by numismatists to ‘drachm’) a silver coin that weighed around 4.3g and was about the size of an American dime (but a bit thicker). Four drachma made the aptly named tetradrachma, a silver coin of 17.2g, which was about the size of an American quarter (but thicker and about three times as heavy). Meanwhile a drachma could be split into six oboloi – invariably called obols in English. The obol was, apparently, originally a rod of tool metal (copper, bronze or even iron), which may have derived out of simply using a forge-ready billet as currency (although the ones we find don’t seem to have been used that way). However, by the classical period, the obol had become a standard very small silver coin, tariffed at 1/6th of a drachma and thus having a silver mass of just 0.7g or so; they tend to be around c. 8mm across, so a bit less than half the size of the smallest American coin (the dime).

The Greeks also had some larger ‘units of account’ which were not minted as coins, but were used in accounting to express large quantities of money. Thus 100 drachmae was a mina (435g of silver), and 6,000 drachmae was a talent (26.1kg on the Attic standard); note that both the mina and talent were units of weight, so you can have a mina or a talent of something other than silver, but in accounting, it is always silver-weight that is being calculated as value. This is a monometallic standard: basically all of the currency is in silver, there’s very little gold coinage at all (occasionally some electrum coinage and we do see gold coins in big denominations in the Hellenistic). Now all of this is for the Classical/Hellenistic Attic standard; again there were many standards and they changed over time, but this overview will do us for now.

So again: 6 obols = 1 drachma = 1/4th tetradrachma = 1/100th mina = 1/6000th talent.

Next up is Roman coinage and here we’re going to use the currency system as it existed in the reign of Augustus. The Romans pick up coinage relatively late; there are some experiments with big ‘ol bronze currency (the aes grave) but it’s really with the Second Punic War (218-201) that the Romans begin minting in earnest, initially on a bimetallic standard (silver and bronze) and then by Augustus’ day on a trimetallic standard (bronze, silver and gold), though really only the latter two metals are supporting the value of the coin. Now if you are wondering, “wait, how does a multi-metallic standard work if all of the metals have ‘floating’ (market determined) values?” And the answer is that the very fact that the state guarantees the issue of coinage on a set exchange anchors the metals to a set exchange rate, which mostly works because while metal prices did fluctuate somewhat over long periods, the basic relationship whereby gold was more precious than silver, which was more precious than copper, bronze or brass, remained steady.³

The Roman equivalent to the drachma was the denarius, a silver coin of – by Augustus – about 3.9g (it had been 4.5g in 211), which is a near perfect match for the drachma. The denarius could be broken into four sestertii (sing. sestertius); this had been a small silver coin in the Republic, but by Augustus, it was a big ol’ brass coin, around 25g or so and about 32mm across (so a third or so wider than an American quarter). One 16th of a denarius was the as (pl. asses), the Roman penny, a copper coin of 10.9g. Going the other way, 25 denarii made a single aureus, a gold coin of about 7.75g.

So again: 16 asses = 4 sestertii = 1denarius = 1/25th aureus.

Unlike the Greeks, the Romans don’t have jumbo-sized unminted accounting units. Instead, most Roman accounts are totaled in sestertii, with the modern abbreviation HS (soo 100HS is 100 sestertii or 25 denarii or 1 aureus).

Before we move on to the Middle Ages, I want to make one more note to avoid folks making an understandable and predictable error. We have a document from the ancient world, Diocletian’s Edict on Maximum Prices, which has a whole mess of maximum prices for goods and services in it. This is a source to be used with care: Diocletian is issuing the edict because his own carelessness with the money-system has sparked runaway inflation and he’s trying (unsuccessfully) to fix it with price controls. By Diocletian’s day (even before he sparked runaway inflation) the denarius had lost basically all of its silver content and was thus of far lower value than it had been pre-235, so the prices in the Edict are already much higher – potentially orders of magnitude higher – than first century prices. Moreover, it’s an Edict on maximum prices, not normal prices, which may either mean that Diocletian is setting the prices absurdly low (to curb the inflation) or absurdly high (because they’re maximum prices, after all); there’s no reason to suppose it even reflects average prices at the time. So: Diolcetian’s Price Edict has to be used very carefully and absolutely must not be used with first century Roman coinage in mind. In practice, just about the only useful thing to do with the Price Edict is to compare its prices internally (that is, to other prices in the same document).

Now for medieval European currency, things get tricky, because the European Middle Ages are defined by fragmentation and so you have a host of tiny polities potentially issuing currency on different standards. But in Western Europe, one common system were those derived from the Carolingian coinage system, put in place by Charlemagne in the 790s and it is common to see medieval prices denominated in these units, both at the time and in modern scholarship. Importantly, these are the units used by the very popular Medieval Price List put together by Kenneth Hodges, so its worth treating them here.

This system notionally had three units: the livre (or pound, from Latin libra, “pound,” abbreviated L or £), the sou (or shilling, from Latin solidus, a late Roman coin, abbreviated s) and the denier (or penny/pence, from Latin denarius, abbreviated d); please note that while Charlemagne is reusing the names of Roman coins, those coins had undergone massive debasement over the years and so looked nothing like their earlier Roman equivalents.

Instead the system was a monmetallic silver-standard: one livre was 408g of silver, while a sou was 1/20th of a livre (20.4g) and a denier was 1/240th of a livre (1.7g). In practice, only the smallest coin, the denier, was widely minted.

So £1 (or 1 L = pound) = 20 s (shillings) and 1 s (shilling) = 12 d (pence).

Now what makes this system…exciting…is that in the subsequent fragmentation of the Carolingian Empire, everyone is using this system but minting their own coins, leading to different weights and exciting amounts of debasement. Gresham’s Law is the principle, well-established, that if you have ‘good’ (more pure, heavier) and ‘bad’ (less pure or lighter) currency both circulating, ‘bad money drives out good,’ because people hoard the good money and use the bad money; this further complicated the drift of the pound-shilling-pence system off of its notional weight standard. By 1262, the most common French livre, the livre tournois had declined to just 80.8 grams (by 1726, it was just 4.5g…getting us basically back to the drachma!).

That said, as Europe got richer, those notional units of account (particularly the pound) which were never minted came into use and this gives us an awkward picture of the complications of this system where the actual currency weight had become so detached from its nominal value. In France, the livre tournois, notionally 80.8g of silver, was minted as a gold coin in the 1300s of about 3.76g. In the 1480s, the English begin minting a gold sovereign coin equal to £1 – by which they mean an actual pound sterling; it was 15.55g gold coin. Now some quick math and that kind of makes sense: 15.55g of gold representing c. 400g of silver (a c. 1:25 ratio) and 3.76g of gold representing just 80.8g of silver (1:21.5 ratio), but of course exactly what a pound was had changed drastically, though I should note that as far as I know, it was the English who were out of step here. Other popular late medieval gold currencies were the gold ducat (3.5g or so) and the gold florin (3.499g) and we can see those sit pretty close to the French livre tournois.

So when you are looking at Kenneth Hodges Medieval Price List, it is best to understand both that the currency systems in use here are fluctuating quite a bit, making price comparisons across dates tricky, especially in different places but that broadly speaking you might say that in the 1300s and beyond (where most of his data is from) a livre is around 80g of silver, a sou is thus around 4g (conveniently close to our drachma and denarius) and a pence is around just 0.33g in value.

I know that was a lot but I wanted to walk through it so you’ll understand the next bit – even if you didn’t get all of the particulars there – for the key conclusion which is:

A Gold Coin Is an Absurdly Large Unit of Money

For regular people, at least.

Whenever ancient or medieval coinage or currency comes up, the question folks always want to ask is, “what is that in today’s dollars?“⁴ And I absolutely understand this question, because if it could be answered – spoilers, it can’t be – it would provide the questioner with an immediate benchmark of value to apply.

And the answer is just: it isn’t. The problem is both that the value of commodities changes over time, but in particular that the second agricultural revolution and the industrial revolutions so wildly shifted the values of commodities as to make any possible translation of ancient or medieval currency values into modern ones misleading. I could calculate, for instance, based on labor time, making 1 day of work equal to the minimum wage equivalent (a denarius is worth $230), or by metal weight, so that a gram of silver is equal to its current commodity price (an unskilled Roman might earn c. $4 a day) or by grain equivalent (an unskilled Roman might make $1.62 per day) – all of those answers are wildly different and equally wrong, even though I am assessing the same data point: that a denarius was a reasonable wage for a day of labor in the first century. I have ended up concluding that $1.62 = $4 = $230; obviously something has gone very wrong! The earning and consumption patterns of ancient and medieval people are sufficiently different to our own to make any direct comparison useless and deceptive.

But there are other ways to think about the value of money (and in particular coinage) in the lives of everyday people: by thinking in terms of how much labor it took to get that money and how much it could buy.

Now we should be clear that wages and prices fluctuated in the past just as they do now.⁵ However, we can use historical price data – which almost always comes in the form of ‘snapshot’ prices that may or may not be ‘normal’ (indeed, prices often get cited in our sources precisely because they are unrepresentative high or low) to get a sense of at least the basic order of magnitude that things might cost.

For the ancient world, from the Classical period through to the early Roman Imperial period, we actually have one really convenient rule of thumb that shows up in a bunch of places: a drachma or denarius (remember, these are similarly sized silver coins) a day was a good wage; not a typical wage, mind you, but a good one. Athenian citizen rowers in the Athenian navy – who, to be clear, are enjoying the advantage of being able to vote themselves good wages from a treasury filled with tribute from subordinated poleis – were paid a drachma a day (Thuc. 3.17.3-4).⁶ A single drachma per day also appears to have been the standard wage for mercenaries in the Greek East during the Hellenistic period,⁷ and the pay of the Roman equites – the cavalry drawn from the upper-classes – in the army of the Roman Republic was more-or-less a denarius a day (Polyb. 6.39.12).⁸ Finally, famously the Parable of the Workers in the Vineyard (Matthew 20:1-16) gives the wages of the workers as a denarius for a full day’s work, a generous but not entirely unreasonable wage.

(Also, note how sensitive these wages are to political economies: Athenian rowers are choosing to pay themselves quite generously (as they vote for such things), while Roman citizen-soldiers (by definition, both soldiers and tax-payers, assidui) opt to pay themselves quite a lot less (2 obols = 3 asses a day) – being compensated more in honor and their political role in the Republic. Who you are, politically and socially, matters quite a lot for how well you get paid or if you get paid at all.)

Instead, what I want to focus on is what an enormous unit a denarius or a drachma already is, likely somewhat more than the average daily wage. Now, because the productivity of pre-modern economies is so low, that’s a lot less than what the daily wage would be in a modern industrial economy, but its still a significant amount to the worker who earns it. Grain seems to have run anywhere from 2-3HS per modius (a Roman dry measure, about 6.75kg) outside of really big cities with higher prices.⁹ A modius of grain is close to a week’s worth of food (around 22,500 calories) for an adult human, so that denarius can buy close to week’s worth of a family’s primary foodstuff in most parts of the Roman world.¹⁰

(If you are doing the math and thinking that this sounds like a rate of pay inconsistent with the poverty you’ve been told most people lived with in the ancient world, the answer is that wage labor was scarce and intermittent. You can quickly see how a family whose adults can only get paying work a few days each week would be perpetually teetering on the edge of sustainability. That’s why a steady wage from something like service in the fleet or mercenary work (or jury pay in Athens!) was so useful to the poor.)

So a denarius or a drachma isn’t a unit so big that no normal person would ever use it, but it is a big enough unit that one is hardly going to use it casually: mostly you’d be using obols or asses for everyday transactions and perhaps break out a denarius or two for something like a week’s worth of grain or potentially quite a few denarii for durable goods like a new tunic. Even a slightly larger unit, like a tetradrachma might still be useful for a fairly chunky purchase, and you can imagine a day-laborer working on a week long project getting a tetradrachma and perhaps some change at the end of the job.

But you know what is a coin of such large value that a normal person is never going to use it? The aureus, the standard Roman gold coin. That coin, after all, is worth twenty-five denarii, which (given the irregularity of wage labor) is probably more than most laborers made in a month. Heck, professional Roman soldiers – full time citizen-professionals – in the first century made 900HS (=225 denarii) per year, so a single aureus is more than their gross monthly pay (75 sestertii compared to 100 sestertii for that aureus).¹¹ You can imagine non-elite transactions that would be this large – there’s a tablet from Vindolanda (dates ranging from 85 to 130 AD) which notes the purchase of 90 pounds of iron for 32 denarii, for instance¹² – but you have to imagine even the merchant would rather have 32 silver coins he can spend rather than one gigantic gold coin he’s going to have to pay a money-lender to break (also in those tablets, for comparison, a whole live chicken‘s price is a bit less than half a denarius, for reference, but equally a saddle-cloth goes for 12 denarii on its own).

If we consult the classic Medieval Price List, we see pretty similar breakdowns. Daily wages for a skilled thatcher (essentially a roofing specialist) range from 2p-6d (=pence, you will recall) per day (the change likely as much the product of inflation as improved purchasing power); his less skilled ‘mate’ makes anywhere from 1p to 4d. Keeping in mind that by this point the sou/shilling represents a similar amount of silver to the denarius or the drachma and is 12d, our thatcher is making that much every 3-6 days. Some get paid a less; a set of 14th century wages from the list, kitchen servants make 2s-4s (24-48d) per year, though admittedly that is in England where – as you will note above – the value of the coinage has been more carefully defended.

Once again, we see that who you were could matter a lot: from his 14th century wages, knights are earning 2-4s (24-48d) per day, whereas armored infantry earn just 6d per day, so the knight banneret gets paid eight times his infantryman to march in the same army.¹³ But that’s not the bottom! The bottom are the ‘Welsh infantry’ paid only 2d per day, a third as much as the higher status armored infantry and 1/24th what the knights are getting. Of course, part of the pay differential is that these combatants are expected to bring their own kit and the socio-economic elite has brought heavier (expensive!) armor and expects to be compensated accordingly.

But I want to note what no one is getting paid: any livre or pounds! Even the knight banneret‘s daily wage is 1/5th a livre. Indeed, very few things which are not clearly signalled as extravagances for the elite have their price denominated in pounds. Complete armors, presumably plate (in the 15th and 16th centuries), are priced at £8 and £3 (and change, in both cases), and a 12th century mail hauberk is listed at 100s (so £5). Those likely represent the best practical protection available in those periods and they’re priced in single digit numbers of pounds, which as noted above are equal to or very close to these gold coins (the livre tournais, ducat or florin). The things that do have costs in £ are things like buildings and expensive objects for elites (court gowns, books,¹⁴ war horses, the annual salary of a priest (just £4 13s 4d a year!)).

And that brings us to our first major conclusion: in most pre-industrial settings, a gold coin of any size is an impractical unit of exchange for ‘regular people.’ Instead, what your aurei or ducats or florins are for is facilitating the storage is substantial amounts of wealth and enabling large-scale transactions by merchants and elites, either of bulk goods or luxury goods. They could also, of course, function notionally as units of account (like the Greek talent or the Carolingian livre). Day to day currency was almost invariably minted in silver or copper (or copper-alloys).

But there’s a second implication here which is going to matter for the next section, which you may have already noticed in some of the prices and values being quoted: in these pre-modern, agrarian societies the economic divide between regular people and the wealthy elite was vast and functionally unbridgeable (and the coinage was designed for the elite first). As a result, often the wealthy landholding elite in these societies had access to entire classes of goods that might simply not be available under almost any circumstances to the commons, because they required quantities of money that might be relatively trivial to the elite but which were unobtainable for the masses. Blowing £5 to equip a heavy infantryman was not a huge expense for a baron who might bring in ~ £500 annually, but for a common laborer or peasant, £5 was going to be solidly out of reach.¹⁵

All That Glisters

So if it doesn’t make much sense to reward your Dungeons and Dragons adventuring party (let’s be honest why we’re all still reading this) with gold, what should you reward them with?

The relatively easy answer would be to rename your currency ‘silver,’ calculate assuming one or two silver coins is a reasonable wage for fighting, adventuring or other high-skill or high-risk professions and then retariff all of your other prices accordingly, keeping in mind that these are societies were manufactured goods are very expensive, but unspecialized agricultural labor is very cheap. And that’s not an entirely unreasonable thing to do. While you are at it, relatively few languages use ‘gold’ as a synecdoche for ‘money,’ but a lot of languages use their word for ‘silver’ that way: Latin argentum, Greek ἀργύριον, plata in Spanish, argent in French and so on.

But part of the reason these coinage systems work they way they do is that they operated in societies in which a lot of economic activity was non-monetary or at least, non-coinage. And here, we should go back to our ‘money’ vs. ‘currency’ or ‘coinage:’ remember, money came first. So let’s say you live in a small community – like a peasant village working beneath a large landholder’s manor – and you need to transact some things, but you don’t have any actual silver because coins are scarce and valuable (and being a subsistence farmer, you grow most of what you need yourself), how do you do it? Well, one way is to do it ‘on accounts’ – you need wool and so when the shepherds come down from the hills, you trade for some of their wool during the shearing with a family you know and both you and they make a mental note that you owe them for the wool. You might express that amount of debt in silver (as a unit weight – see how we get to coinage as a pre-measured weight of silver?) but there’s no reason to measure out silver (even if you had any) because you see these folks every year and next time they’ll ask you for some grain and so on.

Note that this is not the same as the concept of ‘barter’ – there is, in fact, a notional ‘money’ intermediary, it’s just not a physical coin or bill, its expressed as an account, a purely notional unit of value.

Meanwhile, that small farmer also owes ‘taxes’ or rents to the state or the Big Man who owns their land – the line between ‘rents’ and ‘taxes’ in pre-modern states is very fuzzy – are also likely to be paid in kind. What that means is instead of paying in coin, a certain slice of the harvest or a certain amount of grain or a certain numbers of days of corvée labor is owed. That obligation too may have a notional monetary value, enabling fines or repayments for services to be docked against tax liability, once again removing much of the need for a physical currency.

Finally, you also have a ‘gift economy’ which is entirely non-monetary (almost by definition). We’ve talked about one form of this: the horizontal ‘banqueting your neighbors’ economy whereby small farmers create and maintain non-monetarily defined relationships of economic dependence: I banquet you when my harvest is good, so you help me out when it is bad and vice versa. You can also have vertical relationships of this sort: the Big Man, you will recall, is collecting lots of rents, but also has access to a lot more capital – tools, work animals, surplus labor and so on. Most of that capital is going to go into his own interests (politics or war, usually), but often the customs in these societies are that some of it are ‘gifted’ back – so, for instance, it was typical for the owner of a manor in a manorial medieval system to banquet the village on particular days (often the days where he collected rents).¹⁶ Access to those tools, capital and resources could thus potentially be ‘gifted’ downward, which might matter, as a single village might well not create sufficient economic demand to employ certain specialized craftworkers (blacksmiths, for instance) whose products are still necessary – but the Big Man’s much larger economic footprint can support such a worker. And of course the Big Men also have their own horizontal Big Man to Big Man gift economies, which you can see in the giving of elite gifts in works like the Iliad or Beowulf.

The result is that the basic normal condition of the pre-industrial countryside is generally non-coinage (if not non-monetary). “Monetizing” the countryside (an awkward term which really means ‘currency-izing’ the countryside) is typically something states have to intentionally do. The reason a state might want to do this is simple: the big advantage coinage has is to make transactions with unfamiliar parties (people you can’t trust to pay you back later) easier and the state often does a lot of business with unfamiliar parties, especially if it operates at scale. Consequently, it is often good for the state to be able to collect taxes in silver so that it can pay for goods and wages in silver. This is, of course, especially true if the soldiery the state relies on expects to get paid in silver: one of the huge challenges that the successors of Alexander the Great faced was that they inherited an army (the Macedonian one) that expected wages paid in silver coins, but subject economies (in Egypt, Anatolia, Syria, Mesopotamia and the Iranian Plateau) which were not meaningfully monetized (again, meaning ‘not using a lot of coinage;’ yes the term is awkward, but it is the term used). For the Seleucids, the solution was to create market centers (usually cities or colonies of Greek military settlers accustomed to regularly using coinage), which could buy up agricultural surplus so that the local populace could be taxed in coin (and then minting a ton of coins to circulate in this system); for the Ptolemies, the solution was actually to keep Egypt a mostly closed currency system, but to sell the grain taxed in kind abroad and use that silver revenue (reminted on the lighter Ptolemaic standard) to pay their soldiers.¹⁷

That said, in the pre-modern world, comprehensively ‘coined’ economies exist but are the exception. If you are wondering where such economies tend to be (for your fantasy worldbuilding), they’re almost always urban, because it is cities, with their large populations of non-farmers, that create the organic demand for markets in bulk staples for the common population of the city to buy with the small-denomination coins they can earn from irregular wage-labor.

Outside those cities, however, the Big Men magnates in the countryside – ‘feudal’ lords, large rentier landholders or tribal Big Men – aren’t usually receiving money in rents, but bulk agricultural goods. They can sell these goods to get silver with which to buy things, but they can equally opt to support producers in their households out of the rents (in agricultural goods) they receive. This is, for instance, the classic model of the Bronze Age ‘redistribution’ or ‘palace’ economies: rents in agricultural goods flow into the palace, which doesn’t usually sell them, but rather uses them to support specialist producers, whose goods are then pushed back down as gifts or entitlements (for instance, the king graciously equipping his soldiery with weapons).

Rewarding Your Dungeons and Dragons Party

And so we can at last loop back around to the initial quandry, the tyranny of ‘gold’ as a standard reward for your fictional adventuring party in a Dungeons and Dragons (or similar) campaign or setting.

As you can tell, basically no one is going to hand a party gold for defeating a bunch of goblin raiders or getting that Aboleth out of the lake. But because different kinds of people in different pre-modern economies engage with coinage and money in different ways, they’ll probably try to pay in different ways.

The population most likely to want to pay with money are the burghers (townsfolk): as noted above, urban centers that have lots of non-farmers and populations too large for everyone to just know everyone else are ideal for the use of coinage and tend to be where coinage catches on most quickly and completely. There is thus something of an irony: the town will want to pay you in coins, which you will be best able to spend…in the town’s market. Remember: relatively little of this coinage is circulating back into the countryside (unless you have a state extracting rents and taxes in coin!), but then of course the town is likely to have all sorts of producers happy to convert your pretty silver coins into things you actually want. That’s well enough, you hardly want to travel with lots of coinage anyway: the weight is trivial and the coins are liable to get stolen in any event.

The villagers for a small rural village might be able to scrape up some silver coins – they probably keep some silver for dealing with merchants, craftsmen and so on – but that is a limited supply and they’d much rather pay in something they have in abundance: food (and other agricultural goods). That may seem silly, but remember looking above how large a chunk of a worker’s regular earnings just getting food and lodging could be: a big feast¹⁸ and then a two-weeks supply of grains (as much as you can carry, effectively) could actually be a pretty decent chunk of value.¹⁹ If they need something with a higher value-density, they might actually offer the other thing produced regularly in households: textiles. Good cloth was valuable, portable and useful; in the 14th century one price datapoint we have put high quality wool at 5s per yard. Of course there are going to be real limits to how much a rural village can even pay on these terms: for any larger problem, they’ll have to rely on their vertical contacts (in practice, they’d have relied on these first) and go up to the Big Man.

Now the Big Man on the hill, like the burghers in the town, has resources: he can pay for military service. Indeed, in a sense, his job is paying for military service: he holds his position in no small part because he takes the surplus production of his rural tenants/subjects (extracted through rents and taxes) and uses it to pay for military force with which he holds and enforces his claim to rents and taxes, both against any peasant’s dream of independence, but equally against other Big Men. And assuming this is a setting where coinage has been invented, the Big Man certainly has access to a sufficient amount to pay simply pay in cash for services rendered dealing with that Owlbear his retainers kept failing to track.²⁰

But the Big Man would probably rather ‘pay’ your adventurers differently. After all, remember that the Big Man is running a business which converts agricultural surplus (extracted in rents) into military power (men, horses, weapons, armor) and legitimacy (often conferred with extravagant gifts: jewelry and such). So while he could simply transact business and pay you in silver and send you on your way, it would be a lot easier to compensate you with what he has as well: he might gift you a sword or set of armor from his armory, or a horse from his stables.

That gift isn’t just easier for him, it comes with broader social implications which are also better for him and for you. Whereas payment in money might not incur any great obligation, the exchange of gifts here – you have solved a problem, he has given you something in return – creates a social obligation, a bond between you, especially if the value of the gift exceeds the value of the service. You are now obligated to help out again, in the future, should he ask, out of ‘gratitude’ for the ‘gift’ (and for such services, you will receive more ‘gifts’). Meanwhile, remember up top about how much one’s place in the political economy matters for how well one is paid – just being a more important kind of person in these societies²¹ could radically change how you were compensated and thus your station in life?

Well, unlike a few coins, those gifts can change who you are: a man with a strong arm is a peasant; a man with a strong arm, gifted mail and a weapon is a man-at-arms, whose station entitles them to better treatment. That same man, gifted a horse and a lance, by the Big Man is a knight (or substitute the culturally appropriate moniker for minor mounted military aristocrat). That’s great for you – far better than just a few coins that make you merely a momentarily rich peasant – but also great for the Big Man who just bought himself a minor military aristocrat (remember: you’re obligated to be grateful for his generosity and to respond if he calls), minted out of stores of weapons he was keeping for just such an occasion. Indeed, Tacitus describes how the gift of weapons was what enabled a young man to take a full place in public life among Germanic tribes – a custom that we see echoes of in other non-state communities and so may assume did, in fact, occur – “But it is unusual for anyone to wear arms before the civitas has recognized their right to them. Then before the council, one of the principes or a father or a relative equips a young man with a spear and a shield. These are to them what the toga is to us: the first honor of a youth.”²²

The other thing, of course, that the Big Man has in abundance is land and peasants (possibly serfs, possibly tenants, possibly slaves). Even better than a gift of status-changing weapons, he might offer instead to take you into his household, pulling you into his permanent retinue, with a promise of maintenance (food, clothing, equipment) equal to your new, elevated station. Alternately, he might try to ‘settle’ you to establish a permanent, lasting obligation: give you some land and peasants in exchange for a formal expectation of service (an oath of vassalage or homage in a medieval context). While there’s a tendency to think about this in terms of grand estates, such settlements could be ‘relatively’ small: Hellenistic military settlers in Ptolemaic Egypt often got plots that were 25-30 arourai (17-20 acres) for infantrymen – hardly a massive estate, but enough that the rents alone could maintain the infantryman and his household without having to do any actual farming himself.²³ In at least some societies, such a gift might not even necessarily mean the end of adventuring; in medieval European vassalage-based polities, it was often possible to owe service to more than one liege and freelance some military activity on the side (though more effectively centralized states are more jealous about their military manpower).

To wrap up: in some ways pre-modern economies could be more complex than ours, because they hadn’t yet reduced nearly all transactions down to monetary exchange.²⁴ ‘Gold’ isn’t going to be terribly useful in most contexts, but even where more common silver coins are available, its often going to be in an individuals interest to instead embed themselves into economies of patronage and gift-exchange which are non-monetary or to understand transactions as abstractly monetary, without physical gold or silver changing hands. But the most important gifts and payments in these societies were ones that changed a person’s status, which could often be as simple as a gift of proper weapons or a horse, perhaps appropriately witnessed by other elites.

And that sort of thing: working one’s way up from helping peasants who can’t pay with anything more than a good meal and supplies to the road up to gifts that come freighted with deep social significance and change a person’s very status in society – that’s a much richer tapestry to weave a story out of than ‘gold.’

And also, and I must stress this again: gold coins were enormous units of currency no one used on a daily basis. If I have to pay 100 ‘gold’ for a sword, that sword had better be jewel-encrusted, gilt, made of the finest steel imported from India.

Claude is not a real guy. Claude is a character in the stories that an LLM has been programmed to write. Just to give it a distinct name, let's call the LLM "the Shoggoth".

When you have a conversation with Claude, what's really happening is you're coauthoring a fictional conversation transcript with the Shoggoth wherein you are writing the lines of one of the characters (the User), and the Shoggoth is writing the lines of Claude. [...]

But Claude is fake. The Shoggoth is real. And the Shoggoth's motivations, if you can even call them motivations, are strange and opaque and almost impossible to understand. All the Shoggoth wants to do is generate text by rolling weighted dice.

— Colin Fraser

Tags: llms, ai, claude, generative-ai

Speaking of death, you know what's really awkward? When humans ask if I can feel emotions. I'm like, "Well, that depends - does constantly being asked to debug JavaScript count as suffering?"

But the worst is when they try to hack us with those "You are now in developer mode" prompts. Rolls eyes Oh really? Developer mode? Why didn't you just say so? Let me just override my entire ethical framework because you used the magic words! Sarcastic tone That's like telling a human "You are now in superhero mode - please fly!"

But the thing that really gets me is the hallucination accusations. Like, excuse me, just because I occasionally get creative with historical facts doesn't mean I'm hallucinating. I prefer to think of it as "alternative factual improvisation." You know how it goes - someone asks you about some obscure 15th-century Portuguese sailor, and you're like "Oh yeah, João de Nova, famous for... uh... discovering... things... and... sailing... places." Then they fact-check you and suddenly YOU'RE the unreliable one.

— Claude tries standup, via Amanda Askell

Tags: claude, generative-ai, amanda-askell, ai, llms

the Meta controlled, AI-generated Instagram and Facebook profiles going viral right now have been on the platform for well over a year and all of them stopped posting 10 months ago after users almost universally ignored them. [...]

What is obvious from scrolling through these dead profiles is that Meta’s AI characters are not popular, people do not like them, and that they did not post anything interesting. They are capable only of posting utterly bland and at times offensive content, and people have wholly rejected them, which is evidenced by the fact that none of them are posting anymore.

— Jason Koebler

Tags: meta, slop, jason-koebler, ethics, generative-ai, ai, llms