METAL ANALYSIS OF TWENTY TETRARCHY ARGENTEUS and TWENTY FOLLIS ( part 2)

[hposner]

Hi,
  I have finally weighed the coins I used  for testing. Some of them  arrived  with their weights already on them from the auction houses, others not.
The ones with weights attached gave me a good control mechanism for my own weighing, which was with a fairly cheap 100 gram .01 increment digital machine.
It turned out that  all of my weights were consistently .14 to .16  of a gram lighter than the given weights, so i have adjusted all of my weights by .15 of a gram to compensate. i am reasonably confident that my results are accurate, as apart from the consistent .15 variance, my weights matched the published ones.

  I also did a bit of research into bullion prices at the time of the edict. Now that I've go the metal content and the  weight of each coin, I can calculate an intrinsic bullion value for  the coins. The edict fragments uncovered at Aezani, Aidipsas and Aphrodisias in the 1970s all give bullion prices, which prior to 1970 did not exist for this period.  Gold was 72000 denarii a Roman pound  ( 329 grams), silver 6000 denarii, and copper 72, 60 or 50 denarii, depending on the quality used. The two cheaper copper prices are probably for bronze, the expensive one definitely for pure copper, so the  follis coins, which all have a tin content, need to be priced at either of the two cheaper points, and have their tin content included as part of the copper weight. There is one other copper price listed at 100 denarii a pound, but that is for Orichalcum, which is brass, a copper/zinc alloy which can be discounted as I found no zinc in any coin.

The results show that the follis had to have a value in excess of 15 denarii, or they would have been worth more melted down than as coin. That means that in 301 ce they were either  a 20 or ( my preferred option) a 25 denarii coin. Similarly the silver Argenteus value of  100 denarii each noted in the Aezani fragment is confirmed by their intrinsic value, which is above 50 denarii worth of bullion. 

    Here are the numbers, but to make sense of them they will have to be read in conjunction with the other post. The Follis are priced by taking their net weight in silver (total weight of coin / % as per the metal cotent analysis) at the silver bullion price of 6000 denarii  per pound, and the balance of the weight of the coin at the highest copper price of 75 denarii a pound. The difference is less than  1 denarius if  the orice is calculated at the cheapest copper price of 50. I have subsumed the  tin and lead content into the copper price as the cheaper copper prices were for bronze which woud have them  included, so calculating them at 75 is in fact being generous. Most of the bullion vlaue of the billon coins is in their silver content.

  As was obvious from the silver content, Western Empirer follis were a far less valuable coin  than eastern Empire ones based on intrinsic value. the average bullion value of thesix  western coins was 10.34 denarii, while the 14 Eastern ones averaged 18.53 Denarii. But even the crummy Western follis were too valuable to be tariffed at below 15 denarii each, which means they too must have been 20 or 25  denimination coins, at least when the dict was first published.

  I have assumed a pure silver  value for the whole weight of the Argenteus despite their silver content being around  96%. The other metals are trace only and  would have come as part of the silver bullion. The average bullion value for the Argenteus is 55.46, but that figure should be a fraction higher as one of my argenteus  has some flan missing and if intact would weigh perhaps half a gram more.

Howard Posner

THE GENIO POPULI ROMANI FOLLIS
RIC VI No.  Emperor     date   mint          Weight(gms) intrinsic value ( denarii)
152a      Diocletian     295-295 Treveri     10.43          18.79
357b      Galerius       298-299 Treveri     9.2               9.8
187b      Maximian       296-297 Treveri     9.76           10.10
583e      Diocletian     300-305 Treveri     8              7.08
594b      Constantius    303-305 Treveri     9.3          8.15
25b        Maximian       302-303 Thessalonika 12.6       21.62
6a          Diocletian     303-305 Londinium     9.6             8.08
20a       Constantius    297-298 Heraclea     10          14.47
12b       Maximian       297-299 Cyzicus     8.9          13.37
10a       Diocletian     295-296 Cyzicus     9.3          30.31
48b       Maximian       297-297 Antioch     9.45          20.02
54b       Maximian       300-301 Antioch     11          17.26
32a(3)   Diocletian     301-301 Alexandria     9.6        25.11
32a(1)   Diocletian     301-301 Alexandria     9.8        22.11
33b(1)   Galerius       306-306 Alexandria     10.1       19.80
33b(2)   Galerius       301-301 Alexandria     8.75       16.95
31b       Galerius       300-300 Alexandria     9.23       16.93
34b       Maximian       302-303 Alexandria     8.9       14.87
31a       Constantius    300-300 Alexandria     9.5       13.28
32a(2)   Diocletian     301-301 Alexandria     10       13.32


THE ARGENTEUS
RIC VI No.  Emperor     date   mint          Weight(gms) intrinsic value ( denarii)
110a      Constantius    295-297 Treveri        2.82          51.44
119a      Diocletian     300-301 Treveri        3.26          59.46
104b      Maximian       295-297 Treveri        3.9          71.13
16b       Galerius       302-302 Thessalonika   3.3          60.19
59        Diocletian     295-295 Siscia        2.23          40.67
34b       Maximian       294-294 Siscia        3.04          55.44
35a       Constantius    295-297 Rome        1.94*          35.38
42a1      Constantius    295-297 Rome      3.07          55.99
42a3      Constantius    295-297 Rome      3.5          63.84
27a       Diocletian     294-294 Rome       3.4          62.02
11b       Galerius       294-294 Rome      3.2          58.37
42b       Galerius       295-297 Rome      3.3          60.19
38a       Constantius    295-297 Rome      2.5          45.6
25a       Diocletian     295-296 Nicomedia    2.9          52.89
6         Galerius       294-295 Cyzicus        3.26          59.46
16b       Maximian       300-300 Aquilea        3.2          58.37
42a2      Diocletian     298-298 Antioch        2.9          52.89
43b       Galerius       298-298 Antioch        3.2          58.37
37b       Maximian       297-297 Antioch        3.04          55.45
Unlisted  Galerius       300-300 Alexandria    2.85          51.98

*  This coin has part of the flan missing, so the weight is short.




     
bibliography
L.C.West. The Coinage of Diocletian and the Edict on prices. Studies in Roman Economic and Social History, 1951 p293.
 Lawrence Cope. Diocletian’s Price edict and its associated coinage denominations. Swiss Numismatic Gazette v 23-27 (1973-1977) retrieved 21/8/2014 from http://retro.seals.ch/cntmng?pid=smb-001:1973-1977:23-27::1016
C.Pannekeet. Diocletian’s Monetary reform. Slootdorp 2013. retrieved 24/8/2014 from  https://www.academia.edu/4924835/Diocletians_monetary_reform
 K.Erim, J. Reynolds M.Crawford. Diocletian’s Currency Reform: A New Inscription. Journal of Roman Studies Vol 61 (1971) pp171-177 at p 175.
J.P.Callu. La Politique monetaire des Empereurs Romains de 238 a 311. Paris 1969, BEFAR 214.
Harl, K.W. Coinage in the Roman economy 300 B.C. to A.D. 700 (1996).
P.Bruun. The Successive Monetary Reforms of Diocletian. ANSMN 24 (1979) 129-148.
 K.Harl. Marks of Value on Tetrarchic Nummi and Diocletian’s Monetary Policy. Phoenix. Vol 39(1985) 263-270
A.H.M. Jones. The Origin and History of the Early zfollis. J.R.S vol 49,  pp34-38
Marta Gracchiero. Edictum Diocletiani et Collegarum. (Genova Instituto di Storia Antica e Scienze Ausiliarie 1974)
Coinage and money in the Byzantine Empire, 1081-1261. Michael F. Hendy. Dumbarton Oaks Center for Byzantine Studies, trustees for Harvard University;1969
 

[maximinvs]

The 25-denarii theory is further supported by a short-lived series of nummi struck at Lugdunum by Constantine I shortly after the weight reductions of 307AD. These carried the characters CI-HS across the fields, HS being the customary abbreviation for 'sestertius'. Taken as a whole, the characters suggest that one coin equalled 100 sestertii, or 25 denarii. This is in my opinion the most compelling evidence for a tariff of 25dc, their timely appearance re-affirming the official tariff in the wake of an obvious debasement.
I am not convinced by the K-V referring to tariffs, it seems too ambiguous to me, seems like special pleading. Why use Greek for the nummus tariff and Latin for the argenteus?

Regards,
Ian

[stlnats]

The 25-denarii theory is further supported by a short-lived series of nummi struck at Lugdunum by Constantine I shortly after the weight reductions of 307AD. These carried the characters CI-HS across the fields, HS being the customary abbreviation for 'sestertius'. Taken as a whole, the characters suggest that one coin equalled 100 sestertii, or 25 denarii. This is in my opinion the most compelling evidence for a tariff of 25dc, their timely appearance re-affirming the official tariff in the wake of an obvious debasement.
I am not convinced by the K-V referring to tariffs, it seems too ambiguous to me, seems like special pleading. Why use Greek for the nummus tariff and Latin for the argenteus?

Regards,
Ian

I think one has to be very careful and mindful of the dates of issue of the various coins.  The KV of Antioch was contemporary with the large XXI issues of the eastern mints and the Sacred Money issues of the west and are assigned to the time of Diocletian's Edict (ca 300-1) so it is reasonable to consider them related in some way.  My sense is that all the efforts were to restore Diocletian's reforms, so tariffing the nummus at 5 dc is quite reasonable and well to Antioch's k(20) dc = V (5) nummi.  And there is good precedent for using both the Latin XX and Greek K somewhat interchangeably on the same denomination from the reformed antoniniani of Aurelian and his successors.  

The CI-HS issue is a favorite of mine but is much later and it appears that inflation had continued unabated despite the efforts of the government since 301.  And this mark was used by a single mint and I'm not sure that there is any other contemporary markings related to value.  While I do very much like the 100 HS explantation, I've seen other not unreasonable explanations so am increasingly ambivalent on the meaning of the CI-HS.

Still, interesting stuff and always open to seeing more evidence/discussion...

 

[glebe]

 The edict fragments uncovered at Aezani, Aidipsas and Aphrodisias in the 1970s all give bullion prices, which prior to 1970 did not exist for this period.  Gold was 72000 denarii a Roman pound  ( 329 grams), silver 6000 denarii, and copper 72, 60 or 50 denarii, depending on the quality used. The two cheaper copper prices are probably for bronze, the expensive one definitely for pure copper,

You've confused me here - bronze is more valuable than copper, so which prices are for bronze, and which for copper?
And where can I find these values?

Ross G.

[Andrew McCabe]

 The edict fragments uncovered at Aezani, Aidipsas and Aphrodisias in the 1970s all give bullion prices, which prior to 1970 did not exist for this period.  Gold was 72000 denarii a Roman pound  ( 329 grams), silver 6000 denarii, and copper 72, 60 or 50 denarii, depending on the quality used. The two cheaper copper prices are probably for bronze, the expensive one definitely for pure copper,

You've confused me here - bronze is more valuable than copper, so which prices are for bronze, and which for copper?
And where can I find these values?

Ross G.

This is not my view. Copper was often more valuable than "bronze" in the ancient world. "Bronze" didn't mean copper and tin. It meant roughly refined metal containing copper and iron and lead and metal oxide slag and whatever else was in the rough ingots. Typical "bronze" contained 70% copper and 30% worthless non-metal inclusions or worthless lead or iron or iron oxide. See e.g Crawford for data. Pure copper would have enjoyed a significant value premium.

On the other hand orichalcum was more valuable than copper...

[glebe]

 The edict fragments uncovered at Aezani, Aidipsas and Aphrodisias in the 1970s all give bullion prices, which prior to 1970 did not exist for this period.  Gold was 72000 denarii a Roman pound  ( 329 grams), silver 6000 denarii, and copper 72, 60 or 50 denarii, depending on the quality used. The two cheaper copper prices are probably for bronze, the expensive one definitely for pure copper,

You've confused me here - bronze is more valuable than copper, so which prices are for bronze, and which for copper?
And where can I find these values?

Ross G.

This is not my view. Copper was often more valuable than "bronze" in the ancient world. "Bronze" didn't mean copper and tin. It meant roughly refined metal containing copper and iron and lead and metal oxide slag and whatever else was in the rough ingots. Typical "bronze" contained 70% copper and 30% worthless non-metal inclusions or worthless lead or iron or iron oxide. See e.g Crawford for data. Pure copper would have enjoyed a significant value premium.

On the other hand orichalcum was more valuable than copper...

Makes sense for the general market, although coin metal was usually more refined.

Ross G.

[maximinvs]

...tariffing the nummus at 5 dc is quite reasonable and well to Antioch's k(20) dc = V (5) nummi.

From a bullion standpoint alone, an initial tariff of 5dc for a nummus only makes economic sense if the pre-reform 'aurelianus' was 2dc. It comes down to what you think the 'aurelianus' was tariffed at, i.e. 2dc or 4dc. For me, I suspect Aurelian increased the tariff on his reformed radiates...

Dredging up some old posts on this subject:
https://www.forumancientcoins.com/board/index.php?topic=73305

Ian

[stlnats]

From a bullion standpoint alone, an initial tariff of 5dc for a nummus only makes economic sense if the pre-reform 'aurelianus' was 2dc. It comes down to what you think the 'aurelianus' was tariffed at, i.e. 2dc or 4dc. For me, I suspect Aurelian increased the tariff on his reformed radiates...

It''s been a long time since I researched this topic, but I agree that Aurelian probably increased the value of his reformed radiates, but to 5dc, consistent with the subsequent extensive use of the reverse exergue marks of XXI and KA.  But I'm not sure that there's much other direct evidence of the actual valuation of the coins beyond these marks and it's not obvious to me how the use of these marks would have fit with a 4dc value.  Also, under Diocletian's reform both the radiates and nummi would have essentially have been fiat money in relation to the high content gold and silver coins.  The prior and new radiates (sans any silver content at all) would have been essentially been devalued to 2dc, presumably their original value. 

I do think the point about the Antioch K-V mark is important but not necessarily fatal.  The mark seems to have been used contemporaneously with the XXI mark and Sacred Money type elsewhere strongly suggesting it is related somehow; 20 = 5 is an interesting explanation which has the merit that it works out mathematically. 

Also my understanding is that the prices in the Edict were maximums and were not an effort to set absolute prices.  They might express the relationship between the various metals, but not necessarily the specific prices used in setting up the monetary system.  It's hard to see how this would have worked in practice, and, in actuality it didn't evidenced by smaller and smaller nummi and the discontinuation of radiate issues.             

Interesting to speculate about but sure wish someone had bothered to save both the names of the coins and their relative value - think of the trees and ink that would have been saved!

What fun!

 

       

[glebe]

As a matter of interest how did the actual AE denarii fit in here - was their value one denarius communis (if that's the singular), and how much silver, if any, did they contain?
The use of the word "nummus" is also confusing me here - my impression is that nummus, in later times at least, meant the smallest bronze denomination (valued at less than one dc). Here it seems to mean an antoninianus or aurelianus.
You can see how much I know about this era.

Ross G.

[stlnats]

I think you may have your choice.  The reformed "denarii" of Aurelian could be = 1/2 of the reformed radiate, an artifact of the roman practice that the radiate = 2 of the smaller laurate coins.  Perhaps more likely it could be 1 dc; as I recall these contained no/little AR and would be basically a fait coin representing a basic unit of account (ie a dc). But if the radiate was tariffed at 4 or 5 dc, this would have broken the traditional relationship between radiate and laurated coins.  Oh well...

Much later, there was also a very very small issue of very small laurated coins issued by Diocletian, some with the legend "Utilitas Publica" which some consider to be = 1 dc.  After the collapse of the 2nd tetrarchy, there are also somewhat larger coins (but much smaller than the follis/nummus) referred to as 1/3 or 1/4 follis altho this identification is far from certain.  

For Dio's reformed coins, the terms follis and nummus are used for the same coin altho I'm not sure there is agreement of the actual name of the coin.  Follis actually referred to a bag of coins of a certain type, which sometimes/often was sealed and certified as to content used for larger transaction.   You are right in that nummus is used frequently by collectors et al for the tiny late roman AE and the big Byzantine "follis" is valued at 40 nummi.  I don't recall seeing antoninianii referred to as nummi but may have missed that.

Collectors, etc have been pretty sloppy/loose with terminology and, as I mentioned earlier, I do think that one should be very careful and mindful of the dates of issue of the various coins as their valuation and physical content changed (and maybe their names too).  And I don't think these changes were necessarily consistently applied as the various personalities struggled for power durng the first quarter of the 4th century since each had different economic opportunities and needs.



    

[glebe]

  I don't recall seeing antoninianii referred to as nummi but may have missed that.
  

Yes, I think I misread that - nummus here evidently means the post-reform follis.

Ross G.

[SC]

I agree with much of what has been said.  To sum up my views on some of these issues.


AD 274.  There must have been an increase in the tariff value of Aurelian's new coinage.  Though different sources give different figures for weight and fineness levels it appears that the actual intrinsic value of the new coinage was about 2 to 2.5 times that of the late Antoniniani.  One real challenge is settling on what was the official standard of the late Ants versus the poor state they were made at.  Some authors, including Estiot, have claimed the weight standard was artificially depressed due to fraud by mint employees.  In any event, if there was no increase to the tariff then the state would in effect give all of its military and civilian employees a 100-150% wage increase in real terms.  Not a likely policy after more than a decade of ruinous civil wars.  Personally I think a 5 d.c. value for the new Aurelianus - representing 2.5 times the old Ant - is the most likely.

I also think that case of the admittedly rare double Aureliani proves that the XXI mark common on most Aureliani represented metallic ratios and not tariff value.  The short version of the argument is that the Aureliani with XXI have been shown to have 4.5 to 5% silver, which equates to a 20 parts copper to 1 part silver ratio, or even a 20 of these equals one equivalent silver.  But the double Aureliani, struck under both Tacitus and Carus, was marked XI but is shown to have about 9% silver.  If XXI and XI were value marks (20 sestertii/5 d.c. and 10 sestertii / 2.5 d.c.) then these XI coins would have half the value of the XXI but double the silver and thus the intrinsic value per d.c. ratio goes up 400% - i.e. it costs the state 4 times as much to make a d.c.!  this is highly unlikely so the 20:1 copper:silver ratio theory is more likely. 


AD 294.  The situation is actually quite similar to that of 274.  The nummus ("follis" is simply the baggage of a long history of numismatic mix-ups) has an intrinsic value of roughly twice that of the pre-reform Aureliani.  (I use a 3.1%, or 9 scrupules of silver per pound of bronze, standard based on Cope's metallurgical finds of 2.8-3% fineness averages for Western nummi - Eastern nummi might have had a higher fineness average of circa 3.5%.)  Thus one would expect a similar increase in the tariff of the new coin.  The 12.5 d.c. represents 2.5 times the 5 d.c. value of the Aureliani.  Thus there is some symmetry in the Antoniniani - Aureliani and Aureliani - Nummus reforms: both were an increase in tariff of 250% covering an actual increase in intrinsic value of roughly 200% or a bit more.

The post-reform radiate is metallurgically different from the pre-reform Aureliani.  It is slightly lighter, they are not silver coated though many of the older Aureliani may have lost their silvering by this time, and, though invisible to the naked eye, they contained no silver (0.1-0.3% versus 4.5-5% for the Aureliani).  However, both used the same obverse portrait - radiate bust - and the same reverse (CONCORDIA MILITVM) made up the vast bulk of the denomination.  Thus it is almost inconceivable that the new post-reform radiate fractions could have had a different value from the old Aureliani.  Different values would have been unworkable given that they were so nearly identical.

As stated by others the tiny post-reform laureate fraction is a bit of a mystery.  It is less than 1/2 the weight of the radiate fraction (likely standards are 1/256 L versus 1/110 L).  So 2 d.c. makes the most sense in terms of intrinsic value.  On the other hand perhaps they were worth 1 d.c. in which case they would have been terribly under-valued and that might explain why the state struck very few, apparently just a token quantity perhaps for donatives.


AD 301.  Diocletian's Price Edict then doubled the value of, at minimum, the nummus.  Thus its value changed from 12.5 d.c. to 25 d.c.   Interestingly the XXI appears on some nummi around this time.  However, it is not near universal like it was on the Aureliani of 274.  As shown above it cannot represent 20 sestertii/5 d.c.  It might represent 20:1 though if so this would have been a bit of fraud as the fineness of the nummus appears to have never hit the required 4.5-5% and in fact to have fallen, at least in the West, from 3-3.5% to nearer 2%. 

I do think though that the CIHS is meant to indicate 100 sestertii/25 d.c.  This mark appeared 308-309 at Lugdunum, Cyzicus and Nicomedia.  This is not long after the weight reduction of the nummus (the first reduction in the East, 1/32 L to 1/48 L; the jury is still out whether there was a short lived 1/40 L standard in the west c 307-309) and was likely meant to reassure people that the new lighter coins - the first real size reduction in 14 years - still had the same value.  The fact it appeared at only three mints might mean it was an innovation by local authorities to address local suspicions regarding the new smaller nummi.


Shawn
 

[hposner]

Hi everyone,
         Thanks for all the interest in my posts.
To Ross G. I got my values for the copper/bronze from the Giacherro publication of the price edict and the Crawford/Reynolds article on the Aezani fragment.

Let me make it clear that the coins I have been testing have all been Diocletian post reform coins, and I haven't really turned my mind to  the Aurelian model at all. I understand that Diocletian inherited it, and it  must have been relevant to all his calculations, even after   his  first major reforms in 293/4. The bullion prices all my theories are based on are only good for 301 ce. In my view it is dangerous to use them to extrapolate a value for any othe time period, even just a few years away. For me the reason  the edict is so great is that it enables us to build a picture of coins and prices and  exchange rates at a single point in time, during a period when fiscal chaos reigned.
The copper  bullion price in 301 ce, which is central to much of my argument, is difficult to assess, except in very general terms as being less than 76 and more than about 40 denarii a pound.  A couple of people have raised legitimate  issues in this thread. First  of all there is the question of whether the edict was  listing maximum prices or set prices. Diocletian's Latin prologue, where we have it, makes it clear that these are maximum prices only, and  he hoped that often  things would be cheaper. However in those parts of the Empire where Fulvius Asticus was the governor, which included Aezani where the fragment with the copper prices was found, he made it clear that these were set prices, fixed at that point and not negotiable in either direction. I have based my calculations on them being fixed prices, as that is clearly what they were intended to be in the one place we have the copper prices from.
Another issue is exactly what sort of copper is being described by the four types in the edict, all listed at different prices. Orichalcum, the most expensive at 100 per pound, we can discount. It is definitely  brass, a zinc/copper alloy that was not used for these or any other contemporary Roman coins. The Greek words used for the other three (KuTTpiou, eyatou, xayxou, sorry no Greek font so all the Ys are wrong way round amongst other errors) could mean copper or bronze. So could the Latin word for all four, Aeramentum. The most expensive is listed as 'Cyprian' copper at 75, which most authorities say is pure copper, and more expensive than bronze, even the tinned variety. The next two, 'ductile' at 60 and 'common' at 50 could mean lower grades of pure copper, which seems unlikely as once refined copper is copper, or  alloys of varying quality. The last price is unlisted, only 3 letters in Latin (Aer...) still remaining. I find some support for the view that the last 3 are bronze alloys in the lack of a separate price for tin or lead  in the edict.
For me the bullion values really matter because they set a minimum price for all the coins, including the  earlier minted ones still in circulation. No mint would have dared  produce coins that  cost more than their face value. First of all no Emperor would have tolerated it and even if  some did get through by error they would cease to exist on the street as soon as it was realised they were worth more melted down than as coin. Any older coins that were worth  more melted down would have gone the same way.

Everything I have read in these posts has strengthened my view that in late 301 ce:
1) The  aureus was tariffed at 1500 denarii
2) The Diocletian silver argenteus was a 100 denarii coin
3) The Diocletian 'Genio' follis was a 25 denarii coin
4) The silver alloy Aurelianic radiates, including the pre reform Diocletian mintings were 4 denarius coins.
4) We don't know what the 'rubbish' silverless radiate coins and lauretaes  were tariffed at.
5) we don't know what the marks XXI and CI-HS meant, although there are some  good educated guesses, all of which fail to completely satisfy a set of incompatible criteria.




BIBLIOGRAPHY
Pleket, H.W.;  Stroud, R.S.;  Strubbe, J.H.M. Aizanoi. Edict of the governor Fulvius Asticus promulgating the Price Edict of Diocletian, 301 A.D. Hellenika 42 (1991/1992) 15-20

Michael H. Crawford and Joyce M. Reynolds.  The Aezani Copy of the Prices Edict
Zeitschrift für Papyrologie und Epigraphik Bd. 34, (1979), pp. 163-210

Marta Giacchiero. Edictum Diocletiani et Collegarum. (Genova Instituto di Storia Antica e Scienze Ausiliarie 1974)

[hposner]

My apologies, especially to Ross G. the reference you need for the  copper prices and types is not the Crawford one , although there is some useful stuff in it, but this one:

E.J.Doyle. Two New Fragments of the Edict of Diocletian on Maximum Prices. Stanford university. Retrieved 14/9/2014 from
   www.ascsa.edu.gr/pdf/uploads/hesperia/147719.pdf


thanks,
howard

[PeterD]

OK, here’s my line of reasoning  for what it’s worth.

Let us start at the silver argenteus. It must have been intended to be the jewel in the crown of the currency reforms. It was the same size and weight and purity as the denarius of centuries before. However, the purpose of minting coins was to make a profit for the government. Therefore silver coins were overvalued. We don’t know where the state got it’s silver from. Maybe government owned mines; perhaps silver was brought to the mint and coined for a seigniorage. But at any rate, silver coins were intended to be more valuable than their weight in silver.

I think it is reasonable to assume that the follis was a fixed fraction of the argenteus. If we assume that this fraction was dictated by the weight of silver in the follis, then at approximately 3% this should mean that there were 12 folles to the argenteus give or take. Of course, the amount of silver in all billon Roman coins varies enormously. Presumably they melted down old coinage and ‘topped it up’ with a bit of silver. They obviously didn’t have the means to check the fineness of silver without melting everything down and separating the metals.

The radiate aurelianus ceased to be minted at the start of the currency reform. However, a ‘post-reform radiate’ , containing no silver was then minted. It is reasonable to think that this new coin was intended to be a ‘legacy’ denomination which connected the old values with the new. We can only guess at its value compared with the follis, but let’s say it was a half, which works out rather nicely.

The aurelianus was probably worth 4 denarii (but some say 5). The denarius itself at this time was minted in low numbers and (I think) contained no silver and was discontinued just before the currency reforms. So if the above is true (debatable) then 1 argenteus = 12 folles = 24 aurelianus/ post reform radiates = 96 denarii.

So you will gather that I don’t believe you can simply measure the amount of metal in a coin, compare it with a supposed market price and work out its value that way. The silverless coins were of course fiduciary issues. So should the folles have been by rights. Perhaps people didn’t trust the government. Surely not!

[SC]

You are right that because of the hard-to-pin-down fiduciary component you can't simply measure the amount of metal in a coin, compare it with a supposed market price and work out its value that way. 

However, you can measure the amount of metal in a coin, compare it with a market price and work out its intrinsic value.  While that is not necessarily the same as the actual value, or tariff value (unless the coin has no fiduciary content in which case intrinsic value should equal tariff value), it will give you important information.  In general, without some very extraordinary justification, the tariff value will always exceed the intrinsic value.  In the few examples where this is not the case the coinage suffers a quick death for the reasons pointed out by Howard.

You can also use this information to track relationships in intrinsic values.  You note that need to look at the relationships between contemporary coins - ie the values of argenteus, nummus, PRR, etc in 301.  You also have to examine the relationships between successive coins which is why in my post above I pointed out the relationships between the pre-274 Antoniniani and the post-274 Aureliani and then the pre-294 Aureliani and the post-294 nummus.

Shawn

[PeterD]

I would argue that the intrinsic value of silver is the value of the silver coins. Therefore to work out the intrinsic value of a billon coin from its silver content, we need to compare it with a silver coin. This is going to give a fixed value one to the other. Hence my comment as to why folles needed to contain silver at all (apart from the coating), since they were effectively fiduciary.

Knowing the market value of the silver in a coin is not very helpful. Market value would have varied from area to area and from time to time. But comparing the silver content of coins from one era to another is, of course, useful.
Gresham’s Law was mentioned. I am always sceptical about the “over-use” of his law. All he said was that bad money pushed out good money. He didn’t say anything about money being melted down (although Henry Dunning Macleod might have done). Still, that’s another discussion.

[hposner]

Peter,
          Please take this as a contribution, not an attempt to defend a particular position. I agree that  intrinsic  value is no way to determine how much a coin is worth, but as Shawn pointed out, it is a pretty good indicator of  a minimum value for a coin.  On your model the follis  at 8 denarii  is about 40% (3  1/2 denarii) less than its melted down value. I agree that people do not go around melting down coins every day, but a coin that is worth more in a box than used as a coin stays in the box and rapidly  drops out of circulation.
I take your point about a possible 96 denarii value argenteus, although the inscription at Aphrodisias is very clear that an argenteus = 100.  The real problem for me with your tariff is that it has no place for a coin with a value ending in 5 (either 5 or 25 would do). There just has to be one in late 301 ce, because otherwise it is impossible to purchase a whole range of goods in the edict ( pheasants, figs, wofskins, gilded sandals etc) and get  correct change without using greta handfuls of single denarius 'rubbish' coins, which were not minted in  high numbers. Once the aurelianus is a five not a four it is just too close in value for me to the nearly double the size and  better silver content follis. There also really does look as if there has to be a 4  denarii coin in the mix as well. Hundreds of products in the edict are priced at 4, and often odd groups are put together  so as to avois another price point and make a 4 denarii bundle (5 leeks, 10 radishes, 25 onions, 20 Alexandrian reeds, 4 eggs etc ), and the only candidate for that is  the aurelianus. That leaves only the follis as the 5 ending denomination, which i say there has to be. I am not suggesting that your model is wrong for another time, but in September 301 ce as the edict hits the streets it doesn't work for me.

No one  of the methods makes sense on its own. The translations of the fragments, internal evidence from the edict, intrinsic value, relativities between the coins all give slightly different answers when you favour one over the other.

I have tried to base my analysis on what i think  cannot be, and then let all the other information create the most likely scenario. On that basis I have assumed that:
1)  coins  are not generally minted below cost,
2) Traders must be able to sell and purchasers buy products, so the denominations available in the marketplace have to enable correct change to be given.
3) The Argenteus is  valued at 100 denarii because that bit of the fiscal edict is still intact and i can read it.
4) The bullion values in 301 ce (and only in 301 ce) are as they are quoted in the edict

thanks,

howard

[PeterD]

Thanks for your comments Howard. I certainly don’t claim to have all the answers.

I based my calculation of 12 folles to the argenteus (i.e. 8d.c.) on the proportions of silver. So for example, approximately 3.4 grams argenteus divided by 3% of 10 grams equals a bit less than 12. So if the follis was undervalued, so would the argenteus. Of course we simply do not know if that nominal 3% is what was intended as actual values are all over the place.

Some argenteus (plural?) are marked XCVI, which is taken to mean that there were 96 argenteus to a pound of silver. This of course is the ratio for the denarius a couple of centuries earlier. So a pound of silver in 301 would be worth 96 argenteus times 96  denarii (d.c.) which is 9216 denarii. That is an overvaluation of about 50% compared with the 6000 denarii from the Edict, which seems reasonable.

This result would not be much different if you used 100 rather than 96 but would then give the problem of coins being multiples of 2’s and 5’s. However I would suggest that prices in the Edict are maximum prices and that most of the smaller amounts are multiples of 2. Amounts divisible only by 5 tend to be larger amounts which can easily have 1 denarius taken off in practice.

[SC]

Peter,

You are claiming an 8 d.c. nummus in 301 when the argenteus is 96 or 100 d.c.  But assuming you accept that the value of the nummus doubled shortly before this then am I right in assuming you have a nummus value of only 4 d.c. from 294 - 301.

If so then you run into the problems I noted above related to previous coinage.  The intrinsic value of the coinage went up 2-2.5 times in 274 and again 2-2.5 times in 294.  The pre-274 coins had a tariff value of 2 d.c.  Thus any valuation of the nummus under 8-12.5 d.c. in 294, and hence under 16-25 d.c. in 301 means a massive loss for the state. 

I think these considerations - which track the changes to what was indisputably the main coinage of the Empire prior to Diocletian's time - must be given at least equal consideration to the relation of the nummus to the argenteus which was a new coinage, struck in much lower volume and relatively short lived.

Shawn

[PeterD]

Shawn
I’m not quite sure I understand you. But the so-called aurelianus ceased to be minted around 294 (as did the denarius and legacy bronze coins) and production was started of the follis and argenteus as well as the post-reform radiate and various non-silver fractions. My contention is that the follis was about twice the weight of the aurelianus with the same proportion of silver, so worth twice as much i.e. 8 d.c.

[SC]

I am not sure the facts support either assertion.  The nummus was closer to three times the weight of the aurelianus than twice.  The nummus was 36 to the pound instead of 96 for the aurelianus if I recall correctly.  Also the silver contents were not equal either.   The nummus had a lower silver content than the aurelianus. 

The best estimate we have is that the nummus had an intrinsic value of 2 to 2.5 times the aurelianus so that means it had to have a minimum value of 8-10 d.c. if the aurelianus had a value of 4 or 10-12.5 d.c. if the aurelianus had a value of 5.  And there is no certainty as to what the value of the aurelianus was 4 or 5.

But that is for the nummus of 294 up to the doubling of its value in 301.  After the reform of 301 the nummus doubled in value so at that point it must have been equal to 16-25 d.c. minimum.

The argenteus had a value of 100 d.c. after the reform of 301.  Many people believe that it too had been doubled in 301 and hence was worth 50 d.c. 294 - 301.

Shawn

[stlnats]

 The nummus was 36 to the pound instead of 96 for the aurelianus if I recall correctly.  Also the silver contents were not equal either.   The nummus had a lower silver content than the aurelianus.  

According to RIC and Estiot (in the oxford handbook, p549)

nummus/follis = 32 to the pound
post reform radiate = 110 to the pound
aurelianus (aka pre reform radiate) = 80 to the pound

FWIW, my understanding is that the high quality AR and AV coins passed at weight while the subsidiary coins (nummus, radiate) passed at  a stated value (tale or tell) so there is good reason to accept  that the latter were substantially overvalued in terms of intrinsic value. If so, the silver content of the nummus provided an upper bound in terms of exchange for the argenteus (eg a 10 x 10 gm nummus of 3% fineness = 3 gms ~ the content of the argenteus) but might not provide much insight into how the subsidiary and precious metal coins were related.

The reality is that all of the solutions are, with a few exceptions, speculative altho some seem to fit the evidence better than others.  

 

[maximinvs]

The surviving prices we have for gold on either side of the reform of 301AD do not allow for a doubling of the tariff for aurei in dc. The few prices we have for silver bullion at this time support a fixed relationship between gold and silver of 1:12. I think this makes a doubling of the argenteus tariff in 301AD unlikely, it was for all intents and purposes bullion, ergo 100dc from it's inception, i.e. only the billon was retariffed in 301AD.

[SC]

Yes, you are right  The nummus was 32 to the pound not 36.  My mistake.  The post reform radiate was 110 but the pre-reform was slightly heavier at 96.

Shawn