Home Page Site Map Useful Links Email Me Glossary Next Page

----------     Interesting Things About Ancient Coins     ----------

Analysis of the Hut Coins of Constans and Constantius II

Cyzicus hut coin The previous page is a gallery of "hut type" coins of Constans and Constantius II, part of a wider group of coins distinguished by all having the reverse legend "FEL TEMP REPARATIO," which is usually agreed to mean something like "The restoration of good times." This always reminds me of Harold Macmillan's famous slogan from 1957 "You've never had it so good!" (Actually, he said "most of our people have never had it so good" but as so often happens, the misquotation is both snappier and much better known than the true quote.)

Click here for a gallery of hut coins "Hut" coins show (in the words of The Roman Imperial Coinage) "a helmeted soldier, spear in left hand, advancing right, head left; with his right hand he leads a small bare-headed figure from a hut beneath a tree." There has been a good deal of speculation about what exactly this scene represents. Arguments have been advanced by Mattingly and Kraft; Kraft thinks that the type refers to Constans' resettlement of Franks within the boundaries of the Roman empire. But it would also not surprise me if they were being brought into the empire as slaves, in order to restore good times for the citizens.

Here is a clickable index to the sections of this article.

The Soldier  —  The Spear  —  The Small Figure  —  The Hut  —  The Tree  —  The Bust  —  The Globe  —  The Obverse Legend  —  The Mintmark  —  The Weight  —  Useful References.

You can click on any of the coin images or part images to see the full coin.

--------------------------------- The Soldier ---------------------------------

Alexandria hut coin The soldier wears an army uniform, with a crested helmet, a cuirass, a strip skirt, and leather-laced army sandals. A cloak is attached across his breast and falls down on the right of the coin – except that sometimes there is no cloak, or a full cloak falling behind the figure.

Some aspects of this coin type are similar to the "fallen horseman" type, which is much more popular and has had much more analysis. The coin type confronts a barbarian figure with a Roman soldier, and the two are not in proportion – the soldier is larger, sometimes much larger, than the small figure. In the example on the left, the right arms of the soldier and the small figure have been almost caricatured, out of proportion to the rest of this already unequal scene, to present an extreme contrast in size and muscularity. It has been suggested because of this size difference that the small figure represents a child, and this is quite possible. But whether or not this is so, the disparity in size is clearly meant to indicate the relative power of the two, who certainly represent types or ideals.

The soldier may be an idealised emperor, or he may personalise Virtus, the soldierly and manly virtues, which would in any case suggest that it is the emperor whose virtues are being advertised.

--------------------------------- The Spear ---------------------------------

Spear points up Spear between legs Spear points towards small figure Examples of the three spear positions The soldier is carrying a spear. This does not look like a standard short throwing spear. That pilum had a small, bullet-like point, a soft iron shank and a plain wooden shaft, and was mass-produced as an infantry weapon to be thrown at the enemy, to penetrate and perhaps lodge in their shields. This spear is a full length, often quite ornate weapon with a broad point and often a knob at the end of the butt. It's a two-ended weapon; you can stab with the point and thrust or whack with the knobbed butt on the counterstroke. It's the sort of spear carried on coins by Mars and Virtus, and as such it has a symbolic meaning which outweighs any lack of practicality it might have as an infantry weapon.

A reversed spear, that is, a spear pointing to the ground, is normally the equivalent of a sheathed sword – an indication of peaceful intent by the person so armed. Of course, the fact remains that they are still armed!

On hut coins, the spear can be in one of three positions. Pointing up and to the right, with the butt towards the small figure; pointing down between the soldier's legs; or pointing to the left of the soldier's legs, towards the small figure.

The spear between the legs does seem to have the neutral intent that one would expect. When directed towards the small figure, though, it appears threatening. (And what need would there be of such a threat against a mere child? Perhaps this is another clue.) Quite the opposite is the case when held pointing up – in this position, the spear is pointing the way, and sometimes (as shown on the left) the butt is being used to assist the small figure, with the soldier clasping the small figure's hand to the spear beneath his own.

--------------------------------- The Small Figure ---------------------------------

The small figure being led from the hut is sometimes thought to be a child, and the more bloodthirsty interpretation is that this follows the killing of his father – perhaps as shown on the companion coin type, the "falling horseman." Or it could just represent a comparitively puny barbarian. The hairstyle is certainly barbaric, on most of these coins. You can sometimes see the hair standing up straight. The figure usually wears a tunic, but sometimes has less Roman gear such as trousers or a loose robe.

There are perhaps four types that can be distinguished by what the figure is doing with its free arm, which is almost always the left arm, with a few rare exceptions. The arm might:

(1) Not be shown.
(2) Be held out, with (a) the left hand or (b) the right hand open and empty.
(3) Be raised to the figure's face, (a) touching it; (b) held just in front of it; or (c) exceptionally, the face almost buried in the right arm and hand.
(4) Appear to be carrying an object. The object, if it is one, is sometimes (a) dumbell-shaped and sometimes (b) curved like a shield or a bow.

The distinct bow shape is quite rare – RIC VIII notes a single example which, like this one, is from Heraclea.

Constantinople small figure Treveri small figure Heraclea small figure Antioch small figure Cyzicus small figure Treveri small figure Rome small figure Heraclea small figure

The figure is sometimes upright and sometimes stooped. A figure rubbing its face might just have been woken, to be drawn from his hut into the light of the empire by an idealised soldier; or be sheltering his eyes from that light. This seems quite innocent and childlike. A figure with a bow or shield (if such is the case) might be surrendering to the empire, and this would be less like an innocent.

I suspect that the "carrying something" types are actually not that at all, and that where there is something in front of the figure, it is either a badly-drawn arm or a mistake on the part of the engraver. The lower part of the dumbell-shaped objects could represent an elbow, or sometimes some drapery. It is clear that usually, a hand held in front of the face is intended. Some confirmation is found in the example numbered 3c, an oddity from Treveri in which the soldier grasps the small figure's left arm instead of his right, and the sheltering position of his right is extremely clear.

Symbolically, the helplessness and empty-handedness of the small figure is too much an integral part of the scene to be compromised by holding a defensive or offensive item.

Harold Mattingly, in his article "Fel. Temp. Reparatio," says "The small figure is not, as a rule at least, marked out by his dress or feature as a barbarian," and suggests that it might represent a soldier leading a growing boy into adulthood, or even Mars leading Romulus into his military career. The figure on the coins shown here wears a knee-length tunic, belted, apparently of vertically ribbed, pleated or decorated material; and sometimes a cloak. This is not particularly barbarous. His leggings are often horizontally ribbed, suggesting that they are of knitted wool. This sounds like native garb. His hair is often shown standing upright from his head, and this does not sound at all Roman. You can see this feature clearly on coins from Constantinople, Cyzicus, Heraclea, and Nicomedia, and it is suggested on some coins from other mints too; there is a good example from Heraclea at 4b above, and "suggestions" from Treveri and Antioch at 2 and 3a. So I think Mattingly's observation is not proven, and Mattingly himself advises caution.

--------------------------------- The Hut ---------------------------------

Like the trees, the huts varied from mint to mint, though it's not so easy to guess the mint from the hut as it is from the tree. Here are examples ranged from east to west:

Antioch hut Antioch hut Antioch hut types Blank spacer Alexandria hut Alexandria hut Alexandria hut types Blank spacer Nicomedia hut Nicomedia hut Nicomedia hut types Blank spacer Heraclea hut Heraclea hut Heraclea hut types Blank spacer Constantinopolis hut Constantinopolis hut Constantinopolis hut types Blank spacer Cyzicus hut Cyzicus hut Cyzicus hut types Blank spacer Thessalonica hut Thessalonica hut Thessalonica hut types Blank spacer Siscia hut Siscia hut Siscia hut types Blank spacer Aquileia hut Aquileia hut Aquileia hut types Blank spacer Rome hut Rome hut Rome hut types Blank spacer Treveri hut Treveri hut Treveri hut types Blank spacer Lugdunum hut Lugdunum hut Lugdunum hut types Blank spacer Arelate hut Arelate hut Arelate hut types

In the design of the hut, we are being shown only the near three-quarters of the structure; the inside of the roof and the whole of the far wall are not shown, which allows the small figure to stand out clearly.

You can see that there are only two basic types of hut. One appears to be little more than a simple upright structure of reeds or withes, tied across at one, two or three places, sometimes with a pointed peak at the door, sometimes rounded. These are from Alexandria, Antioch, Constantinopolis, Cyzicus, Heraclea, Nicomedia and Rome. The other type is more complex, with a diagonal construction in its lower part attached to a vertical upright, and a roof with a clearly distinct structure, sometimes rounded and sometimes flat; with a clear horizontal junction between walls and roof. These come from Aquileia, Arelate, Lugdunum, Siscia, Thessalonica and Treveri.

There is a clear geographical borderline here. Eastern mints produced simple huts; mints from Greece and the west produced complex huts – with one exception: Rome itself, where the huts are all essentially the simple type, but some are finished at the top with a sloping roof piece instead of the withes being bundled together.

Geographically, Alexandria and Antioch are the most remote. Cyzicus, Nicomedia, Constantinopolis and Heraclea are all clustered around the Sea of Marmara, and you can see a similarity of style among these four. (Three of these are also the only mints that started their hut coin mintmarks with the letters "SM".)Thessalonica is set apart in Greece, and the others are all in the west.

As these hut types seem to be consistent for most of the mints, they could be used together with the tree type as an aid to the attribution of coins whose mintmarks are unreadable. But please be aware that you are still likely to come across exceptions. For example, I have a Siscia hut with vertical rather than diagonal components, and another that looks like the simple type.

--------------------------------- The Tree ---------------------------------

The 13 mints that made these coins all produced different types of tree. The trees are indicated by a single branch, with leaves or perhaps fruit in some cases; sometimes clearly branching from a trunk that continues on upwards, as in the right-hand example here.

Most of the trees are distinctive and are characteristic of their mint. A few of them are quite similar and harder to tell apart, though even then there are differences which you can become more aware of as you look at more examples. Some of the mints produced more than one distinctive type of tree: Alexandria, Cyzicus, Heraclea, Lugdunum, Siscia, and Treveri. Thessalonica can appear to have two types, but there are many intermediate forms so it is likely that it has one type with a few variations.

Alexandria tree Alexandria tree Alexandria tree types Blank spacer Antioch tree type Blank spacer Aquileia tree type Blank spacer Arelate tree Arelate tree Arelate tree types Blank spacer Constantinople tree type Blank spacer Cyzicus tree Cyzicus tree Cyzicus tree types Blank spacer Heraclea tree Heraclea tree Heraclea tree types Blank spacer Lugdunum tree Lugdunum tree Lugdunum tree types Blank spacer Nicomedia tree type Blank spacer Rome tree type Blank spacer Siscia tree Siscia tree Siscia tree types Blank spacer Thessalonica tree Thessalonica tree Thessalonica tree type Blank spacer Treveri tree Treveri tree Treveri tree types

Are these representations of actual, local tree types? It's possible, but in that case you would think that the types of tree should be identifiable. Mostly, they are not. A number of conjectures have been made, but none seem certain. One Alexandrian type might be a fig; a Nicomedian type might be an ash; the Aquileian type might be a banana. What could lead to this mixture of local and exotic, if this is so? Clearly, the engravers were not just copying a template that was provided for them. Some might be using local trees. Others might be providing their idea of the trees that might suit the scene on the coin; maybe from real life, maybe what they imagine such a tree to be like, and in that case they might also have been using their imagination as to where the scene could take place. So on the whole, I think that labelling each tree with its botanical name is not going to happen.

The trunks of the trees vary as much as the leaves. Some are very simple and thin, as in Cyzicus; some are ribbed, curved or branched. Sometimes the "tree" seems to be a vine or creeper growing up a post or perhaps a host treetrunk, shown clearly on the left-hand Heraclea example below. Bruck has some good diagrams of most of these trees, but misses a few.

--------------------------------- The Bust ---------------------------------

Alexandria hut coin obverse Square rosettes Rosettes and laurel leaves Diadems The obverse of all hut coins has the same basic type of bust: left facing, draped and cuirassed, with a globe in the right hand.

There are four possible sub-types of bust, based on the type of diadem. It should be either a twin band of pearls with a single round or square rosette over the forehead (left); a band made up of square rosettes which may be interspersed with pearls, or perhaps something else round (right, above); or alternating rosettes and laurel leaves (right, below). A plain laureate diadem is listed for Arelate, but I haven't seen one.

Some hut coins from Rome have the letter N behind the bust, perhaps as an indication of the denomination of the coin. If they do, the same letter N also appears on the far right of the reverse.

--------------------------------- The Globe ---------------------------------

Square rosettes Rosettes and laurel leaves Globes On all of these coins, the emperor is holding a globe in his right hand. Sometimes the globe is plain, and sometimes it is quartered by two lines with dots in the quadrants. Although, "quartered" is only an approximate description. Actually, the division is always unequal.

Similar globes appear on many Roman coins, being held by the emperor or being presented to him, by figures representing gods or personifying victory. So, this globe might represent the world, over which he has dominion (or so he would like everyone to think). But if so, what are the lines and dots? There is a nice theory that the world is not enough - most such globes globe represent the heavens and the entire universe! The cross and the markings resemble the system the Greeks used to designate the celestial sphere. Here is a link to an article from The Celator about this symbolism.

So, this image says to everyone who uses such a coin, "Your Emperor holds the universe in his grasp." And as far as most of those in the empire were concerned, this was true enough.

--------------------------------- The Obverse Legend ---------------------------------

The obverse legend for Constans on hut type coins is D N CONSTA - NS P F AVG, where the dash indicates a break in the legend above the emperor's head.

For Constantius II, the legend is almost always D N CONSTAN - TIVS P F AVG. The only exception occurs on coins from Aquileia, where some types have CONSTANT - IVS P F AVG.

The "D N" stands for DOMINVS NOSTER, Our Lord.

The name of the emperor is always split apart by the head. This seems to have been an indicator of supreme power. On other late Roman coins, the senior emperors, or Augusti, always had their names split in this way on their coins; their junior partners, the Caesars, did not.

"P F" stands for PIVS FELIX. "PIVS" meant that the emperor was morally upright and religiously observant; "FELIX" meant that he was blessed with happiness and good fortune. These two initials occur together on very many Roman coins, so the two things seem to have been linked in the minds of the Romans, perhaps as cause and effect. (Though, considering the dismal end that many emperors came to, this idea in itself must have required a leap of faith.)

Finally, the AVG stands for AVGVSTVS, which had long been not just a name, but a title meaning one of the senior emperors.

So the legend as a whole should be read (for example) "Our Lord Constans, Righteous and Fortunate, Emperor."

--------------------------------- The Mintmark ---------------------------------

Hut coins all have mintmarks in the exergue, and in some cases there are also marks in the coin's field. They are packed full of information about the mint and the officina, and there are also several different control marks. Here is a breakdown. (This is correct for "hut" coins – it is not a good guide for all other Roman coins with mintmarks!)

The mint.

From one to four characters indicate which mint produced the coin. These are the letters together with a map showing their locations:

Link to hut coins from Arelate Link to hut coins from Lugdunum Link to hut coins from Treveri Link to hut coins from Rome Link to hut coins from Aquileia Link to hut coins from Siscia Link to hut coins from Thessalonica Link to hut coins from Cyzicus Link to hut coins from Heraclea Link to hut coins from Constantinople Link to hut coins from Nicomedia Link to hut coins from Alexandria Link to hut coins from Antioch
ALE Alexandria Map showing mint locations.
ANT Antioch
AQ Aquileia
ARL Arelate (Arles)
CONS Constantinopolis
K Cyzicus
H Heraclea
LG Lugdunum (Lyons)
N Nicomedia
R Rome
SIS Siscia
TES Thessalonica
TR Treveri (Trier)

The officina.

On almost all hut coins, one or two characters indicate which officina within the mint produced the coin. Here's a table showing all the officinas that are known to have struck hut coins. The workshop numbers are along the top, the mint names are down the left, and the characters found on the coins are in the body of the table. The table also shows the language used for the officina number, and whether the officina letter comes before or after the mint identifier:

 1234 5678 91011 Language Position
Treveri (Trier) PS                 Latin Before
Lugdunum (Lyons) PS                 Latin Before
Arelate (Arles) PS                 Latin Before
Aquileia PST                Latin After
Rome PBT QЄ S           Hybrid After
Siscia ABΓ ΔЄ            Greek Before
Thessalonica ABΓ ΔЄ            Greek After
Heraclea A Γ ΔЄS           Greek After
Constantinopolis ABΓ ΔЄS  H  IIA Greek After
Nicomedia ABΓ Δ             Greek After
Cyzicus ABΓ ΔЄS           Greek After
Antioch ABΓ ΔЄS   Θ I  Greek After
Alexandria ABΓ Δ             Greek After

This information is sourced from RIC VIII. You can see that western mints used the Latin initials for Primus, Secundus and Tertius, and most had the officina letter before the mint identifier. The eastern mints used the Greek alphabet to represent numerical digits, which was the usual Greek numbering system, and placed the officina letter after the mint. Rome used a hybrid combination of the Greek and Latin systems.

The archaic Greek character Digamma for officina 6 is, on these coins, almost indistinguishable from an upper case S, and is represented by one here. This character is often incorrectly called Stigma.

The gaps in the table occur where mints had fewer officinas in action, or where particular officinas did not strike this type of coin.

Control marks.

Dots and stars in the exergue, and stars and gammas (Γ) in the field, are control marks used by the mints for their own purposes. It is generaly agreed that these indicated which period of production the coin belonged to. The Siscia mint also used the letters H and M in the exergue as control marks in this period; only the M appears on hut coins.

Siscia mint symbols

Also in Siscia, a set of odd symbols was used, apparently as control marks. These usually corresponded with particular officinas, so that symbol 1 was paired with the officina letter A, symbol 2 with B and so on. However, I know of a couple of coins which match symbol 4 with the officina indicator A. One is mine, here: [Constans hut coin], and the other is an "emperor on galley" coin of Constans which used to be on the Wildwinds site, but now seems not to be there (as of November 2010). I also have a hut coin here: [Constantius II hut coin] which pairs the epsilon of officina 5 with symbol 4. So don't trust reference books to be definitive! Actual finds always trump theories.

N on Rome reverse Denomination.

Some coins from the Rome mint have an N behind the bust on the obverse, and also on the reverse on the far right – so far right as to interfere with the legend – which is thought to indicate the denomination of the coin. The same N is listed for some coins from Aquileia, though I haven't seen one yet.

Neutral characters.

The Cyzicus, Heraclea and Nicomedia mints used the characters SM at the start of their mintmarks. (In Heraclea, the S was sometimes omitted.) This stands for Sacra Moneta and is a formal indication that the coin is good money. It has no significance otherwise, but it helps in identifying the mint a coin comes from.

Heraclea mintmark Here are a couple of examples of mintmarks. On the left, SMHΔ translates as SM = Sacra Moneta; H = Heraclea mint; Δ = officina 4. On the right, CONSIA* translates as CONS = Constantinopolis mint; IA = officina 11; and the star is a control mark. Constantinopolis mintmark
--------------------------------- The Weight ---------------------------------

RIC VIII gives the average weights of these coins as 4.25 grammes if from Arelate, Lugdunum or Treveri; and 4.5 grammes if from any of the other ten mints. I don't have enough coins from those three western mints to make a useful analysis, but with 225 coins from the other ten mints there should be some useful information.

Graph of the distribution of hut coin weights

Adding up their weights and dividing by 225 gives 4.06 grammes, a large difference from the weight expected.

There is a different analysis in the upper graph. Along the bottom are weights in increments of 0.1 gramme from 2.5 to 6.0 grammes, which covers the whole range of my coins. The height of the columns indicates how many coins fell into each 0.1-gramme weight band.

The result, as you might expect, approximates a bell curve, with the median somewhere around 4.0 to 4.2 grammes. This is the weight most commonly found on the market. This is different from the simple arithmetical average because there are more coins that weigh less than 4.1 grammes than there are heavier ones – you can see that the left-hand side of this graph is more heavily populated than the right.

This is still noticeably less than the 4.5 grammes expected according to RIC VIII. But RIC VIII also contains some rawer data, the weights of 1,682 "small AE2" coins (not otherwise differentiated) from 348-350 CE, and this can be analysed in the same two ways to see how the figure of 4.5 grammes was arrived at. Adding the weights of the coins tabulated in RIC, and dividing by 1,682, gives an arithmetical average of 4.15 grammes. The lower graph is of these coins, and again you can see more lighter coins than heavier ones, with the median somewhere around 4.3 to 4.6 grammes. This is more like the weight given in the text; so the authors of RIC have used a graphed median, not an arithmetical average.

But still, whether using an arithmetical average or a graphed median, my coins are, on the whole, lighter than those measured for RIC VIII. Why is this? It is possible that the specimens selected for museums were better-looking than the average coin, and were unconsciously selected for weight because of that. It is also possible that the museum specimens are less eroded and worn, and thus a better indication of the coins as they emerged from the mint. And a third possibility is that the cheaper end of the market, where I have obtained nearly all of these coins, does not see many coins in really good condition, and some of the heavier ones have been selected out of my collection by this bias.

In support of this idea is the fact that when I only had 190 of these coins, their average weight was only 3.56 grammes. Since then I have only been adding nicer looking coins to my collection, and their average weight has increased markedly.

There is one very simple conclusion from this analysis. Hut coins can be of any weight from 2.7 to 6 grammes, and still be genuine. Even lighter and heavier coins will certainly be found; in fact RIC VIII has a single example at 6.8 grammes, which is off my scale.

————————  Useful References  ————————

"RIC" or "RIC VIII" refers to Volume VIII of The Roman Imperial Coinage by J P C Kent, published by Spink, London, 2003. Expensive, but essential for a serious collector.

Another good (and much cheaper) reference book for these coins and many others is "Late Roman Bronze Coinage" (LRBC) by R A G Carson, P V Hill and J P C Kent, published by Sanford J Durst Numismatic Publications, Rockville Centre, NY.

Both of these books consist mostly of tables which are not easy for a beginner to interpret – especially LRBC! – so please don't expect to open them and suddenly know everything.

There is also a book in German, Die Spätrömisch Kupferprägung (literally "The Late Roman Copper Coinage") by Guido Bruck, which gives beautifully clear line drawings of all the late Roman bronze coin types and most of the sub-types. It was published by Akademisch Druck-U. Verlagsansalt, Graz/Austria, in 1961 and can sometimes be obtained from European booksellers. This book is very different. It's a great guide, even if you don't read German, because it is mostly pictures and has hardly any text, but it doesn't show all the sub-types of hut coins.

There are also articles by Harold Mattingly and Konrad Kraft which theorise at length about the FEL TEMP REPARATIO coins.
Harold Mattingly, "Fel Temp Reparatio," an article from 1933, which was printed as a short booklet in the Numismatic Chronicle Reprint Series, New York Attic Books, Ltd. 1977; ISBN 0-915018-22-5.
Konrad Kraft, "Die Taten der Kaiser Constans and Constantius II." – an article in Gesammelte Aufsätze zur antiken Geldgeschichte und Numismatik I, 1978 Wiss. Buch Gesellsch (this one is in German).
I haven't seen the Kraft article, but I have seen discussions of the contents. Here is a comment by Curtis Clay on Kraft's article. Kraft disputes and contradicts Mattingly. They can't both be right, and they might both be wrong.

There's an excellent web page by Doug Smith about FEL TEMP REPARATIO coins here which includes a section about hut coins, with pictures of reverses from Constantinopolis, Nicomedia, Antioch, Thessalonica, Heraclea and Lugdunum. It sets the hut coins nicely into context with the other FTR coin types.

Dane "Helvetica" Kurth has put together some very handy spreadsheets which can be used to identify specific coin types and pin down their RIC references. Her RIC reference list web page is here. For hut coins, look for the table of "All other types of FEL TEMP types except fallen horsemen". This includes several variations not found in RIC, including those I have in my own collection.

The content of this page was last updated on 26 November 2010

Home Page Site Map Useful Links Email Me Glossary Next Page