COIN HOARDS 2005

314

NEAR EAST

A hoard of Byzantine folles from Beirut

Paul Belin
53. Beirut 1994
Dep. 551 or possibly later
Disp. National Museum, Beirut.

34

A small hoard of Byzantine folles was unearthed in 1994 during excavations conducted
by a team from Leiden University in the Beirut Souks area (site BEY 011). It contained
34 folles dating from the reign of Anastasius I (498518) to the reign of Justinian I (527
65). The coins were found very close to each other and several pieces were corroded
together. All the coins with identifiable mintmarks where struck in Constantinople.
TABLE 1 Composition of the Hoard
Issued
Anastasius I (498518)
Justin I (51827)
Justinian I (52765)
Emperor unidentifiable
(498538/9)
Total

Constantinople

498512
51827
52738/9
498538/9

22
4
5
1
32

Mint:
Total Percentage
unidentifiable
22
64.7%
1
5
14.7%
1
6
17.6%
1
2.9%
2

34

The reform of the coinage by Anastasius I in 498 introduced a follis of c.9 grams with
a diameter of 2327 millimetres. A larger size follis of c.18 grams followed in 512. All
the coins of Anastasius in this hoard were of the smaller module and therefore struck
between 498 and 512. Unfortunately the weights of the coins in the hoard were not
recorded. Until 538/9 no significant change was made in the size of the follis but in that
year the weight of the follis was increased to c.22 grams and the diameter to 40
millimeters. No such coins were in the hoard. The folles in the hoard show different
degrees of wear and the older pre-512 coins are considerably more worn than the later
ones especially those of Justinian.
PUNCHMARKS
The small module folles of Anastasius sometimes have a curious punch mark on the

The writer wishes to thank Professor K. Butcher, American University Beirut (AUB) Dr J.P.A.
van der Vin, Money and Banking Museum, Utrecht, and the editor Dr M. Phillips for their remarks
and suggestions, which have greatly improved this article.

MEDIEVAL AND MODERN HOARDS

315

reverse. Two of those in the hoard had a circular mark and one had an mark. Michael
Metcalf analysed a sample of 260 small module folles about one-fifth of which had a
small circular, crescent shaped or cruciform punch mark.1 Among the finds from the
Souks area in Beirut there are also some folles which are marked in this way.2 These
marks are only known on small-module folles of Anastasius, and are almost without
exception placed on the reverse of the coins. Metcalf could not identify any special
feature which set the countermarked coins apart from the coins without a countermark.
The meaning of these marks is still not clear.
DATE OF LOSS
On 9 July 551 a massive earthquake occurred in Lebanon. It was felt over a large area in
the eastern Mediterranean region and caused heavy destruction with great loss of life in
several Lebanese cities. Beirut was destroyed.3 The results from excavations in the Souks
area in the old city centre of Beirut (sites BEY 006 and BEY 045) show that many
houses remained deserted after the disaster.4
The 34 folles were found in a thick layer of coarse rubble, consisting of large stones,
burnt roof debris, lumps of plaster and large quantities of pottery.5 It is thought that this
rubble is the result of destruction caused by the earthquake of 551. There is, however, a
possibility that the layer was mixed with later debris. This means that the coins must
have been lost in, or possibly shortly after, 551.
There are at least two other hoards from Beirut which can also be associated with the
the 551 earthquake. One is a group of 63 small module folles found lying in debris on a
mosaic floor in the House of the Fountains (BEY 006) which was almost certainly
wrecked in the 551 earthquake.6 As well as this hoard, two other Anastasius coins and a
bronze polycandelon supporting at least nine suspended glass lamps were also found
lying on the latest floors. The room in which the coins were discovered was never reoccupied and most of the site was left in ruins.7 The other hoard is not yet published. It
consists of over 60 pre-512 folles, found in a collapsed building also associated with the
earthquake.8 Analyses of the finds show that the small module folles also occur
frequently, together with small bronzes from the previous century, as single finds in
construction phases datable to before the 551 earthquake.9 The evidence of the Beirut
hoards and single coin finds shows that pre-512 small module folles stayed in circulation
in this city long after 512, and at least up to 551.
The hoards and the single coin finds from BEY 006 and BEY 045 both contain a
relatively high proportion of small module Anastasius folles.10 There are, however, a few
marked differences between the composition of the BEY 011 hoard, the site finds and the
other two hoards.
1

Metcalf 1969, pp. 913.

Butcher 20012, p. 108.
3
Darawcheh et al. 2000.
4
Perring 1998, p. 30.
5
Steiner (n.d.) p. 7.
6
For a list of coins in this hoard see Butcher 2001-2, pp. 283-6.
7
Perring 1998. pp. 223, 30; Butcher 1998, pp. 1778; Butcher 20012, p. 283.
8
Personal communication Prof. K. Butcher, AUB, Beirut and Butcher 20012, p. 283.
9
Butcher 20012, p. 105.
10
For the coin finds of the BEY 006 and BEY 045 excavations see Butcher 20012.
2

316

COIN HOARDS 2005

Beirut showing the location of the different Souk sites.

(After Perring 1997-8, p. 15, Fig. 2.)

MEDIEVAL AND MODERN HOARDS

317

TABLE 2
Numbers and percentages for coins, from the souks area for the period 498565
Period
498512
51218
51827
52738/9
538/965

Single coin finds:

BEY 006 and 045 11
150 (76.5 %)
17 (8.7 %)
9 (4.6 %)
2 (1.0 %)
18 (9.2 %)

Hoard BEY 011

22 (66.7%)
0
5 (15.2%)
6 (18.2%)
0

Hoard BEY 006 Unpublished

hoard
63 (100%)
60 +
0
0
0
0
0
0
0
0

The BEY 006 hoard and the unpublished hoard both consist exclusively of small
module folles produced between 498 and 512. Folles produced after 512 seem not to
have been available in Beirut, at least for hoarding in significant numbers, at the time of
the earthquake in 551. The excavations in the Souks area (BEY 006 and BEY 045)
produced only one large module Anastasius follis (see Table 3). Five nummi pieces from
the period between 512 and 518 are less rare but were also not included in the hoard.
This is not unusual because most hoards in the region consist of folles and half folles.12
What is important is that coins produced between 512 and 551 do not appear to have
circulated on any scale in Beirut before 551. So far Kevin Butcher has not found any
coins issued between 498512 and 51251 together in contemporary deposits.13 Because
small module folles are not normally found in Beirut in association with issues from the
period 51251, the occurrence in the BEY 011 hoard of coins which can be dated
between 518 and 538/9 argue for a deposition in or not long after 551, in a period when
the 518538/9 coins might just have been put into circulation in the city and when there
was still an abundance of small module Anastasius folles available.
TABLE 3
Coin denominations on single finds from BEY 006 and BEY 04514
Period
498512
51218
51827
52738/9
538/965

Follis 20 nummi 10 nummi

49
35
66
1
0
0
3
1
2
2
0
0
6
2
2

5 nummi
0
16
1
0
1

Nummus
20(491518)15
2
0
7

12 nummi
0
0
0
1 16

The BEY 011 hoard lacks post538/9 coins. Among the site finds from BEY 006 and
045 however there is a relative high number of post-538/9 issues. Nine of the 18 dated

11

For the number of coins for the different periods see Butcher 20012.
Noeske 2000, I, p. 150.
13
Butcher 20012, p. 107.
14
Figures from Butcher 20012.
15
Because the nummus pieces could not be closer dated than 491518 they have been left out of
table 2.
16
Because the 12 nummi piece could not be closer dated than 52765 it has been left out of table 2.
12

318

COIN HOARDS 2005

Justinian I coins are early ones produced before 542, later coins are less common.17 At
the moment of deposition these coins might not have been available for hoarding because
most dated folles probably arrived in Beirut at some date after the deposition of the
hoard (i.e. 551 or later).18
OTHER HOARDS FROM THE REGION
Hans-Christoph Noeskes thorough survey of coin finds from the dioceses of Egypt and
Oriens (which consisted of modern Lebanon, Syria, Israel and Palestine, southern Turkey
and western Iraq) includes several hoards where the oldest coins are issues of Anastasius.
Most of these hoards have a closing date well into the sixth century or even later. Only
two hoards, one from Sarafand (Sarepta, Lebanon) and one from Djebel Rasa (near
Gaza) have as their latest coins issues which can be dated before the middle of the sixth
century.19 The Sarepta hoard closely resembled the two hoards, the one from BEY 006
and the unpublished hoard mentioned above. Its consisted of 223 small module folles and
half folles struck between 498 and 512. This fits nicely with the finds from Beirut, as
Butcher has already noted.20 Unfortunately there is no detailed information available on
the Djebel Rasa hoard. It started with coins of Anastasius and apparently closed with
coins of Justinian dated 538. It is not known if the Anastasius coins are of the small or
large module type. The Djebel Rasa hoard has a high proportion of post-518 issues, so
the pattern of this hoard seems to differ from that of the hoards found in Beirut, where
the issues of Anastasius prevail. Unfortunately the original report of the hoard does not
seem very reliable. The figures in parcels of 100 seem too neat and the statement that it
ends with a follis dated 538 is impossible since these were not dated until 538/9.
Apart from these two all the other hoards in Noeskes survey which contained coins of
Anastasius are considerably later in date. About 97% of the Anastasius coins in them
were of the large module type. Noeske therefore deduces that the small module series
was effectively withdrawn from circulation in 512 and that the large module folles could
not have been successfully put into circulation unless the small module folles had been
demonetised.21 There are, however, some hoards with a closing date in the seventh
century, which contain a few small module pre-512 coins (together with a few large
module coins!).22 This shows that these coins were still available for hoarding a century
after their production. Noeskes conclusions therefore need to be modified particularly in
the light of the material from Beirut which shows conclusively that the small module
folles were important in this city at least up to 551.
With the Beirut material in mind one can only come to the conclusion that the small
module folles could stay in circulation in some areas (possibly at half the value of the
new large module folles)23 at least until the middle of the sixth century, and disappeared
from circulation only during the second half of the sixth century. The coin patterns from
17

Butcher 20012, pp. 2669: 543/4 (1 coin), 54252 (1 coin), 54952 (1 coin), 55765 (4 coins)
three coins could not be dated very precisely.
18
Butcher 1998, p. 178.
19
Noeske 2000, for Sarafand see I pp. 137-8 and for Djebel Rasa see I, p. 138.
20
Butcher 20012, p. 107.
21
Noeske 2000, I, pp. 1501.
22
Noeske 2000, I, p. 151, these hoards are from: Khirbet Fandaqumya (c.612), Tell Biss (c.612)
and Baalbek (c.631).
23
MIB I, pp. 24 and 36.

MEDIEVAL AND MODERN HOARDS

319

the Beirut hoards might differ from those cited by Noeske (except the Sarafand hoard)
because Noeskes were all deposited considerably later than 550. We have to keep in
mind that the sample is small and geographically very limited, so this needs confirmation
by other data. The hoard evidence to date is still not enough to establish the length of
time for which the small folles remained in circulation in different areas of the Byzantine
empire.
Hoard and site find evidence from Syria, Palestine and Israel point to an absence of
dated Justinianic coinage especially in the southern half, i.e. modern Palestine. One can
conclude that Palestine did not receive these coins in any quantity.24 Noeske suggests
that this gap is due to the coinage reform of 538 when the weight of the follis was
increased from 18.19 grams to 24.95 grams. In 550 the weight of the follis was again
reduced to 18.19 grams. According to Noeske the heavier coins could not be put into
circulation effectively because the lighter coins were still circulating. He thinks that only
a few of the 538/950 folles were produced and that they were quickly withdrawn from
circulation.25 In contrast to the diocese of Oriens, hoard and site find evidence shows that
the Balkans, Anatolia and Syria seem to have received a regular and substantial supply
of dated Justinianic coinage.26 This shows that the size and weight of the post-538/9
coins were not necessarily the reasons for their non-appearance in Syrian finds. Casey
suggested instead that there was an interruption in the supply of coinage, related to the
disbandonment or relocation of limitanei of the Palestine frontier, reflected in the hoard
and site finds of the area.27 Noeskes data show that coins of Justinians successor, Justin
II (56578) are quite common on many sites in the diocese of Oriens.28 If the coin finds
are connected with military activity this can hardly be explained if the limitanei were
permanently disbanded or relocated. Obviously we still need a lot of data before we can
properly substantiate any theory concerning the presence or absence of certain groups of
coins.
CONCLUSIONS
The general pattern of the BEY 011 hoard seems to fit in with the hoard and site find
evidence from Beirut, with a relatively high number of pre-512 Anastasius folles being
used up to 551. Only the Sarafand hoard is similar since the majority of the coins were
well worn.29 Butcher suggests that the differences from other finds in the region might
be the result of local circumstances in Beirut and perhaps other cities on the Phoenician
coast. The municipal authorities possibly might have decided what constituted legal
copper coinage, because the central authorities were unable (or unwilling) to impose a
24

Casey 1996, pp. 21922; Noeske 2000, I, p. 152.

Noeske 2000, I, pp. 1523.
26
Casey 1996, p. 217.
27
Casey 1996, p. 2201.
28
Butcher 20012, pp. 10910: coins of Justin II are relatively scarce in Beirut, only 11 were
recovered during the excavations of BEY 006 and 045. The presence of Justin II (5658) folles in
Palestine is associated with military action against the Persians by Metcalf 1964, p. 34 contra
Bellinger 1938, p. 13: and the attempt to connect it with known events such as the Persian war
which began in the late summer of 572 ends in gratuitous theorizing. In 573 Khusru I (53179) took
Apamea in Syria and Daras in Arabia, but the campaign was in practice confined to Mesopotamia
and Armenia and continued under Tiberius II (57882) whose coins are not found much in Palestine
(see for example Noeske 2000, I, 237 ff.).
29
See the original report by Taylor 1977.
25

COIN HOARDS 2005

320

standardised currency empire wide.30 Shortly after the earthquake the small module
folles possibly circulated together with some 51838/9 coins, as reflected in the
composition of the BEY 011 hoard. The small folles were probably demonetised after a
while because, except in the BEY 011 hoard, they are not found anywhere in association
with post-512 issues, which probably circulated only after 551. The disastrous effects of
the earthquake must have had a great impact. Beirut never recovered for centuries after
551, so this event might also be partly responsible for the differences between coin finds
in this city and other places in the region, which might not have been hit as hard. Maybe
all of the coastal cities in Phoenicia suffered the same fate and when more data become
available the coin patterns for sites on this coast might turn out to be the same as in
Beirut.
Beirut did not receive many new coins in the period between 512 and 551. One can
wonder how the average citizen coped with the seemingly erratic supply of (new) copper
coin.31 In the fifth century the supply of copper coinage became increasingly erratic,
particularly in the west where it all but disappeared. It seems increasingly clear that in
the east people continued to use what was already in circulation for a long time. The
extended circulation of small module coins of Anastasius in Beirut is another example of
this.
BEY 011 CATALOGUE
No.
1

Dia. Dies Mint Officina Reference

(mm)
Anastasius I 26.0 6h CON

DOC 16

2
3
4
5

Anastasius I
Anastasius I
Anastasius I
Anastasius I

7h
7h
6h
6h

CON
CON
CON
CON

6
7

Anastasius I 25.7 7h
Anastasius I 23.4 6h

CON
CON

@?
@

Anastasius I 23.6 6h

CON

9
10
11
12
13
14
15
16

Anastasius I
Anastasius I
Anastasius I
Anastasius I
Anastasius I
Anastasius I
Anastasius I
Anastasius I

CON
CON
CON
CON
CON
CON
CON
CON

30

Emperor

25.0
26.3
24.7
24.8

25.2
25.3
25.1
24.5
25.0
25.9
23.8
25.1

7h
5h
6h
6h
6h
7h
6h
6h

DOC 16
DOC 16
DOC 16
DOC 16

DOC 20a?
DOC 20a;
Metcalf 60
@ ? DOC 20a?;
Metcalf 60
@/ ? DOC 20a/d?
B
DOC 20b
B?
DOC 20b?

DOC 20d

DOC 20d
_
DOC 20e
_
DOC 20e
_
DOC 20e

Observations

Date
498512

Traces of overstriking on obv.

and rev.
Reverse with
wreath border
Reverse with
wreath border

498512
498512
498512
498512

498512
498512
498512
498512
498512
498512
498512
498512
498512
498512
498512

Butcher 20012, p. 107.

Phillips 2004 also addresses the problems of supply, see p. 21 for the decline in coin supply in
seventh century southern Syria and p. 25 for the virtual disappearance of coin use in Syria between
the collapse of the Abbsid administration around 950 and the Byzantine intervention in the
eleventh century.
31

MEDIEVAL AND MODERN HOARDS

17 Anastasius I 25.2 6h
18 Anastasius I 23.4 6h
19 Anastasius I 23.6 6h

CON
CON
CON

_
_
_

DOC 20e
DOC 20f
DOC 20f

20 Anastasius I 25.1 6h

CON ?

DOC 20?

21 Anastasius I 23.5 6h

CON

DOC 20

22
23
24
25

CON
CON
CON
CON

?
?

Anastasius I
Justin I
Justin I
Justin I

26 Justin I

30.1 7h

CON

27
28
29
30

33.2
31.5
31.5
32.4

7h
1h
7h
6h

?
CON
CON
CON

?
@

31.9 7h
30.1 7h
28.6 7h

CON
CON
?

?
DOC 9
DOC 9c
DOC 11
(var.)
DOC 11
(var.)

DOC 28a
DOC 28c
DOC 29a
(var.)
DOC 28c?
DOC 28d

34 Uncertain: 33.0 8h
AD 51838

CON

Justin I
Justinian I
Justinian I
Justinian I

31 Justinian I
32 Justinian I
33 Justinian I

25.1
29.7
30.8
31.6

7h
7h
12h
7h

321

498512
498512
498512

Traces of overstriking on rev.;

punch-mark: o 2
mm, (Metcalf
1969, p. 92, fig. 4)
on reverse, on left
top of M
Punch-mark on 498512
rev. to right of M
498512
Punch-mark
2mm, (Metcalf
1969 p. 92, fig. 4)
on reverse, to right
of M. Half circle,
probably made by
striking a circular
punch at an angle
498512
51827
51827
DOC 11, officina B 51827
DOC 11, officina B 51827

51827
527538/9
527538/9
Obverse with cross 527538/9
in front of bust
527538/9
527538/9
If CON then DOC 527538/9
28c
518538/9?
Possibly 518
538/9 because
post-512
Anastasius folles
are very rare in
Beirut

322

COIN HOARDS 2005

WORKS CITED
Bellinger 1938
Butcher 19978
Butcher 20012

Casey 1996
R. Darawcheh,
R. Sbeinati,
C. Margottini and
S. Paolini 2000
Grierson 1982
Hahn 1973
Harl 1996
Metcalf 1964
Metcalf 1969
Noeske 2000

Perring 19978

Phillips 2004
Steiner n.d

Steiner 2001
Taylor 1977

A.R. Bellinger, Coins from Jerash, 19281934, ANSNNM 81

(New York).
K. Butcher, Coinage in sixth century Beirut: preliminary
observations, Berytus 43, pp. 17380.
K. Butcher, Small change in ancient Beirut. The coin finds from
BEY 006 and 045. Persian, Hellenistic, Roman and Byzantine
Periods. Archaeology of the Beirut Souks 1, Berytus 4546.
P.J. Casey, Justinian, the limitanei, and Arab-Byzantine
relations in the 6th century, JRA 9, pp. 21422.
The 9 July 551 AD Beirut earthquake, eastern Mediterranean
region, Journal of Earthquake Engineering 4, pp. 40314.

P. Grierson, Byzantine Coins (London).

W. Hahn, MIB I: Von Anastasius I. bis Justinianus. (491-565)
(Vienna).
K. Harl, Coinage in the Roman Economy, 300 B.C. to A.D. 700
(Baltimore).
D.M. Metcalf, Some Byzantine and Arab-Byzantine coins from
Palestina Prima, INJ II/34, pp. 3246.
D.M. Metcalf, The Origins of the Anastasian Currency Reforms
(Amsterdam).
Noeske, H.-C., Mnzfunde aus gypten I. Die Mnzfunde des
gyptischen Pilgerzentrums Abu Mina und die Vergleichsfunde
aus den Dioecesen Aegyptus und Oriens vom 4. 8. Jh. N. Chr.
Prolegomena zu einer Geschichte des sptrmischen
Mnzumlaufs in gypten und Syrien. Band I, in: M. R.-Alfldi
and H.-M. von Kaenel, Studien zu Fundmnzen der Antike, Band
12 (Berlin).
D. Perring, Excavations in the Souks of Beirut: an introduction
to the work of the Anglo-Lebanese team and summary report,
Berytus 43, pp. 9-34.
M. Phillips, Currency in seventh century Syria as a historical
source, Byzantine and Modern Greek Studies 28, pp. 3-31.
M. Steiner, Preliminary report of the excavations of BEY 011 by
the team from Leiden University, Holland. Unpublished
excavation report. (Leiden).
M. Steiner, The Hellenistic to Byzantine Souk. Results of the
excavations at BEY 011, ARAM 13, pp. 113-27.
G. Taylor, A hoard of small module coins of Anastasius, CH 3,
p. 87.