Introduction.

Until not very long ago, no coins were assigned to Andronicus I Gidon of Trebizond (1222-35), but in Sear's "Byzantine Coins and their Values" certain rare types, namely the bronze trachies S.2597-99, were assigned to that ruler.

More recently, in 2002 Simon Bendall proposed reassigning certain other rare types to Gidon as well^a. These are all anonymous types showing a standing Christ with the legend "O Chalkites", namely the silver trachy S.2148, the bronze trachies S.2149 & 2150, and the bronze tetarteron S.2153*, all of which were attributed in Sear to an uncertain ruler of Nicea.

More recently again in 2007 Bendall added several more silver and bronze types to the list of possible issues of Gidon^b,c.

Given that Bendall's theories seem have been adopted with enthusiasm (particularly by dealers following the appearance of a considerable number of examples of S.2148 on the market in recent years), I felt it might be worthwhile to review these ideas a little more critically than has been the case so far, and this is done in current article.

(The proposed attributions are also relevant to a number of problems concerning the attribution of certain other coins of this period, some of which are discussed elsewhere on this site).

*  Until recently, apparently only two examples of S.2148 were recorded in museum collections, and examples of the bronze types almost never appear in trade.

Bendall's Theories.

There are basically five groups of coins which Bendall has now assigned to Gidon, which are discussed in turn below.

1).  The bronze trachy S.2599.

S.2599 is the only coin which can be attributed with certainty to Gidon. It shows a military saint standing on the obverse with the legend "A/N/DR/O-NI/KO/C", and the Virgin standing on the reverse with the legends "M-P-QV" and "KOM/N-N/OC-OGI/DO/N" (we remember that Gidon had married into the Comnenus family and liked to call himself by that name).

This is an extremely rare type - the only two examples that I know of are the one shown as Figs 7-7A in Bendall's 2002 article, and another worn  specimen, apparently from different dies, sold on Ebay in 2003 (this last coin is shown below - unfortunately the weight and provenance of this coin are unknown, and it is probably flattened).

Bronze trachy of Andronicus I Gidon (S.2599).
(Probably flattened).

It should be noted that the obverse figure is very possibly intended to represent Gidon himself - since Andronicus was not a military saint, and there is no "O Agioc", the legend was probably intended to be read on both sides. Given this, it is also worth noting that it not clear from the known examples that the ruler has a forked beard like that evident on some of the other types considered below which are attributed to Gidon.

Although we have only two examples to go on it seems fair to say that this is a solid type, single struck on neat flans.

2).  The bronze trachies S.2597-8.

S.2597 and 2598 are very similar to the Latin imitative Type H, and are presumably derived from the same original type, namely the gold hyperpyron S.1983 of Andronicus I Comnenus. In fact S.2598 has formally almost the same design as the Type H, but differs in minor detail - thus on the Type H the ruler wears a diamond pattern loros, whereas on S.2598 he usually wears a plain loros, or sometimes a chlamys. On Type H the ruler's beard is short, but on 2598 it is usually either forked or at least pointed, and finally on Type H the legend basically copies the standard legend of Andronicus Comnenus, namely "Andronikoc Despothc", but while on 2598 the legend is sometimes  "Andronik Decp", it is equally likely to be something like "Andronik ToKom". The related type S.2597 is similar to 2598, but the Virgin is seated on a backless throne, as in the example below.

Bronze trachy S.2597, possibly of Andronicus I Gidon. (3.96g).
(Note forked beard, and legend "...K-D C P").

While certainly scarce types, these two issues are rather commoner than the first type above, and die statistics seem to suggest that they were originally issued in reasonably large numbers, so that they may be rather better known in Asia Minor (the only place where they are known to have been found) than they currently are in the west.

The question now is who issued these types, and where? The first thing to remember is the fact that, like all the types considered here, they are single struck, in contrast to all other regular large module Byzantine trachies, which are usually double struck. This strongly suggests that these types are not issues of a regular Byzantine mint. Secondly, the only find spots that we have for these types are in Asia Minor, as just noted, so presumably this is where they originated. As to the issuer, given that he is named as Andronicus, three possibilities come to mind.

Firstly, these types could perhaps be 12th century Asian issues of Andronicus I Comnenus, since we know that at least some coins (folles) were issued in the name of Alexius I at Trebizond , and rare Asian trachea in the name of Manuel and Isaac Comnenus are also known (Grierson Pl. 58, DOC IV Pl. XXV). The odd legends and somewhat crude style of S.2597-8 could then possibly be explained by their provincial origin, and the relatively heavy weights of these types are certainly more consistent with the 12th century than the 13th. The problem here is that, as Bendall points out, these types differ in their style of fabrication from the Asian issues of Manuel and Isaac mentioned above (which often feature thick clipped polygonal flans evidently cut from sheet), and hence he argues, quite reasonably, that they are not 12th century Comnenan issues.

Secondly, S.2597-9 could conveivably be issues of the crusaders in their territories in north-west Asia-Minor, since we have some indications that other such issues were struck (see Hendy's Latin Type X in "The Post-Hendy Latin Types"). Against the idea is the fact that on all confirmed Latin types the ruler is basically anonymous (if often with a fake name), and hence is depicted with a generic short beard, although perhaps it could be supposed that in Asia the crusaders copied the Comnenan original more closely than their European counterparts did. Another possible problem is the fact that the normal Latin Type H, which we might have expected to have inspired S.2597-8, dates from around 1230 (see "Sequencing the Latin Types"), by which time the crusaders had largely been expelled from Asia.

The third possibility is of course Bendall's assignment to Andronicus Gidon, but apart from some vague Asian provenances (one of which does seem to have been closer to Trebizond than Nicea), there is little specific to confirm this idea, and there is clearly a contrast between the workmanship of of S.2597-8, which are rather crudely made types, struck on ragged flans of widely varying weights, and the neat style of S.2599. Also, neither the legend nor style of these types are consistent with S.2599.

Clearly, we need some more hard evidence if we are to definitely separate the above three alternative assignments for these types - for example, some more confirmed provenances in eastern Asia Minor would presumably rule out the Latin option. But even if we could exclude that particular possibility, and even if we also accept that the these types are not 12th century issues of Trebizond itself, for the moment there is still no positive evidence from which we could conclusively decide that they are 13th century issues of Gidon, rather than 12th century issues of Comnenus from some Asian mint other than Trebizond.

3).  Other bronze  types; and the Latin Type I.

As well as S.2597-8 Bendall has noted some other very rare bronze trachies  with similar designs and legends which he suggests as possible issues of Trebizond (these are listed as types 1-4 in his second 2007 article^c).

Firstly, we have a type (shown as Types 1 & 2 in Bendall's article) which is essentially the inverse of S.2597, i.e, it shows Christ seated on a backless throne on the obverse, sometimes with a star above the throne, and the Virgin crowning a ruler holding a sceptre and akakia on the reverse. The ruler is named sometimes as "Andronik- Decpotic" and sometimes as "Andronik-ToKo". Four of the known examples of this type were acquired in the the Crimea (possibly during the Crimean war), and since it seems the Crimea was under the control of Trebizond in the early 13th century Bendall nominates this type as an issue of Gidon. This is a not unreasonable suggestion, particularly as this type could simply be a variant of S.2597, but as with the latter type we can't rule out the possibility that it is an Asian issue of Andronicus Comnenus (particularly as the star above the throne on one example recalls Manuel I's 4th coinage at Constantinople).

Secondly we have a type with Christ seated on the obverse and a ruler standing on the reverse holding a long cross and an akakia (Bendall type 3). This is another heavy type, and appears to be the same as Jordanov's unlisted Latin Imitative Type II (see "The Post-Hendy Latin Types"), although the latter has a manus dei above right which does not clearly show on Bendall's example (from the BN). The legend is "Androni- Decp".

Here we actually have some useful provenance information - the origin of the BN coin is unknown, but a very similar, although slightly different, coin was found in the Crimea (with the four coins just mentioned); as well, Jordanov's coin came from the a hoard of Latin period types found at Nisovo in the lower Danube valley. Both these find spots are consistent with an Asian source for this type, and even more significantly the Nisovo hoard dates from around 1240, which would suggest that it is indeed an issue of Andronicus Gidon. (Apart from the Nisovo coin this type is apparently unknown in the Balkans, nor is it found in small module form, which strongly suggests it is not in fact a Latin type).

Finally we have the well known but mysterious Latin Imitative Type I (S.2029, Type 4 in Bendall). As discussed elsewhere ("The Latin Type I") this unique coin can hardly be a Latin issue (at least not from Constantinople), and in fact it would seem to be connected with the other types considered here. However, just where it fits in is not clear; on the one hand both its heavy weight, and its legend - the familiar "AND-TOKO", suggest that it belongs with S.2597-8, but its fine style and neat flan do not. Instead, these features might rather link it to S.2599, although it is very noticable that on Type I the ruler has a long forked beard, very reminiscent of Andronicus I Comnenus.

4).  The "Christ Chalkites" types.

The four "Christ Chalkites" types noted in the introduction clearly form an identifiable group of related issues. This is obvious not just from the fact that they share a particular representation of Christ, but more particularly from their distinctive portrayal of the head of Christ, and especially of the hair and beard. Thus the hair is shown with an exaggerated wave, while the beard, which in the later Byzantine period is usually shown as short, and portrayed simply by dots, is in these types shown relatively long, usually pointed, and not infrequently, even forked. Such a treatment of Christ has no exact parallel that I know of, at least on the coins of the period in question. (Examples of S.2148 can be found on the usual archives sites, including CNG).

So we can certainly group the four Chalkites types together - but to what ruler, or even empire, should they be assigned?

In DOC IV, Hendy simply lists these types as "Uncertain", and of nowhere in particular. In Sear, however, Bendall narrowed the field a little by attributing them to Nicaea, although why he did this is unclear - perhaps it was based on the typically Nicaean mix of types, and the general stylistic similarity of the tetarteron S.2153 to similar types of John III Vatatzes. But this isn't much to go on, and in fact, as far as I know, there is no hard evidence (such as a hoard) which specifically links these types to Nicea. (Note that the bronze Chalkites types are much rarer than the "Nicaean" types with which they are commonly listed, but an example of S.2150 from the "Despot" collection appeared as Lot 156 in LHS Numismatics Sale 97).

In fact, this lack of evidence opened the door for Bendall. In his 2002 article he noted that there are two or three known find spots for these types. Firstly, an example of the the silver trachy was apparently found in the 19th century in Chalcis, Euboea, while an S.2149 bronze trachy was found in 1963 in Amastris (Samsun) on the Black Sea coast of Turkey, and a copy of S.2153 was found somewhere in Trebizond, also apparently on the Black Sea coast, around 1960.

Now the Chalcis coin obviously doesn't help Bendall's cause, but, as he points out, silver coins tend to travel more than bronze types. On the basis of the other two coins, then, Bendall suggested that all these types are emissions of Trebizond, and in fact of Andronicus I Gidon.

As with the earlier attributions to Nicaea, this certainly wasn't much to go on, but as noted above, shortly after Bendall's 2002 article a considerable number of examples of the silver trachy S.2148 appeared on the market, reportedly from an Asian hoard (there were apparently about 90 such coins in the find, together with other types, some of which we will consider later). To the suspicious mind of course this fortuitous appearance might suggest the possibility that not all of these offerings are genuine, but given the convincing style of the coins this idea seems very unlikely, so let us discount it for the purposes of the current analysis.

The argument then is that the Chalkites types are of Asian origin, that the weights of the silver coins show that they must postdate c.1220, and that their few known find spots show that they are issues of Trebizond rather than Nicaea.

But perhaps these types could still be Nicaean, issued by a mint in, say, Philadelphia, or maybe north-west Asia Minor. What evidence is there that specifically links them with Gidon? Well, nothing really, apart from arguments based on style and fabric. Thus we notice is that, as with some of the other issues considered here, the Chalkites types, both bronze and silver, are well made, on neat flans, and are never double struck, supporting the idea that they are in fact issues of Trebizond, or at least that they can't be issues of a regular Nicaean mint. We also note that the workmanship of the Chalkites bronzes in particular matches that of S.2599, although the possible lack of a forked beard on the ruler on S.2599 (presuming  it is Gidon himself) could be a problem.

All this supports the assignment of the Chalkites types to Gidon, but it can hardly be said that the case has been proved, and in the end the assignment is really based on an argument by exhaustion - that is, who else could have issued them, given the idea, based partly on provenance, and partly on fabric and style, that these types can hardly be issues of Nicaea.*

Actually, there is one (rather slim) line of evidence linking them to Gidon, namely the Latin Type I - on this named type the obverse, with the distinctive Virgin Soritissa, matches exactly that of the earliest versions of the silver Chalkites trachy S.2148, which might be taken as further evidence, if only indirect, that the Chalkites types are in fact issues of Gidon, rather than Nicaea. On the other hand the similarity of the obverses may not signify any real connection at all, as this particular image of the Virgin is not uncommon in various contexts.

Given that we are talking of style here, we also need to note here the obvious contrast between the neat workmanship of the Chalkites types, both silver and bronze, and that of the trachies S.2597-8 considered earlier, which as we have seen are rather crudely made on ragged flans of varying weights. We also note that on S.2597-8 the the portrayal of Christ is quite conventional, and hence quite different from that on the Chalkites types.

It is therefore clear that even if both series of types are issues of Gidon they can hardly be the product of the same mint at the same time, and in fact, given the disparity in styles, it is difficult to accept that both series can be issues of the same regime. (We note also that S.2597-8 differ in style from the rare but neat silver trachies of Manuel I Comnenus of Trebizond, which evidently date from not long after Gidon).

*  Bendall has also suggested (N. Circ. 2007, p.260-2) that the tetarteron S.1988, normally attributed to Andronicus I Comnenus, is another issue of Gidon. This very rare type is apparently unknown in Greece, but has been found in the Crimea, and hence is very likely an Asian issue. This type is noticable for the lis on the shield of St George, a device also found on some of the silver Chalkites trachies, and so Bendall makes it an issue of Gidon. Actually, it might be better to reverse the argument, because the lis provides a possible (if somewhat tenuous) link between the anonymous Chalkites types and the name Andronicus, and hence might be taken as supporting the idea that the Chalkites types are indeed issues of Gidon. But since there is no separate clear evidence to say that S.1988 is an issue of 13th rather than the 12th century, this line of argument is hardly conclusive.

5).  Other silver types

At about the same time that the new Chalkites trachea surfaced, another unlisted silver trachy appeared on the market, namely Lot 140 in the Gorny and Mosch Sale 155, reportedly from the same hoard as the Chalkites types, and like them, single struck. This new type has Christ seated on the obverse, and a ruler with a long forked beard holding a patriarchal cross with a beardless military saint (St George?) on the reverse. Not surprisingly, this coin has also been suggested as an issue of Andronicus Gidon, mainly on the basis of a legend that aparently includes "..koc..". However, there are problems with this identification since at 4.44 gm this coin is clearly heavier than any of the S.2148 examples reported so far. In fact, given its relatively heavy weight, and the fact that the ruler has the long forked beard of the Nicean emperors, the obvious suspicion is that this coin is a silver issue of Theodore I Lascaris, or even, assuming the legend is correct, a debased electrum issue of Andronicus I Comnenus, a ruler also show on coins with a long forked beard. However, this last option would seem unlikely - this coin certainly seems to be silver rather than electrum, which very likely means, as Bendall argues, that it dates from the early 13th century.

In his first 2007 article Bendall also noted some examples of two more types (or two varieties of one type) of silver trachea, with designs copying that of the electrum hyperpyron S.1984 of Andronicus Comnenus, which are also said to be from the same hoard as the Chalkites types S.2148. These types were in fact previously known, and some examples of them have the legend "ANDPO-TwKO", similar to that of S.2597-8, and Bendall assigns them to Gidon as well. Once again, these types are single struck. The problem here is the same as for the previous new silver type, i.e, they are clearly heavier than the 2148's - this might also suggest that these types are Asian issues of Andronicus Comnenus, but again they are appear to be silver rather than electrum, so the earlier 13th century seems a more likely date for these types, which as before really only leaves Gidon as the only candidate for their issuer.

It is tempting, although certainly not necessary, to assume that these last two silver types belong with the bronze types S.2597-8, and that's what Bendall does, thus dating them all to the 13th century. 

In fact Bendall also wants to date these silver types after the Chalkites types, but surely it must be the other way round, given the lower weights of the silver trachy S.2148*.  In fact, I would be happy to date the Chalkites types after Gidon, although if they are actually issues of Trebizond then they can hardly be later than c.1240, given that the silver trachy was evidently replaced by the flat asper (of similar weight to S.2148) early the reign of Manuel I Comnenus.

*  Bendall wants to date the larger silver types after the Chalkites issues because of their relatively pristine condition. However, assuming that all these types are issues of Gidon at Trebizond, which is by no means certain, then most likely there was a weight reduction of the silver coinage at some stage during his reign. If so, then maybe the recent hoard, if that's what it really is, could have constituted domestic savings put together over some time, so that it included both some older and heavier types along with the newer and lighter Chalkites types. Or maybe the "hoard" actually combined types from more than one source.

But anyway, whatever the sequence of events, it has to be said that the difference in weights between the various silver types raises problems in assigning all of them to Gidon. In fact, given the heavy weights of the silver versions of S.1984, the almost exact similarity between these types and their electrum prototype suggests that they may be 12th century issues after all, i.e, issues of Andronicus Comnenus in Asia.

Conclusion.

Except for the anonymous Chalkites series (and perhaps the Gorny and Mosch silver coin), all the types considered here are named issues of a ruler named Andronicus, and the problem is to determine which Andronicus is meant, the 12th century Byzantine emperor Andronicus I Comnenus, or the 13th ruler Andronicus I Gidon of Trebizond. Judging from the few find spots that are known, almost all of these types seem to be Asian issues; this helps to some degree, but still doesn't settle the matter, as it is quite possible that coins were issued in the name of the Comnenan Andronicus at Trebizond, or  somewhere else in Asia Minor, or even, in some cases, in the Crimea.

Furthermore, the very noticable absence of the name Gidon on almost all of these types must in any case raise prima facie doubts as to their assignment to that ruler, and it is not clear that Bendall's various arguments have overcome those doubts.

In the end, of  the various types considered here, only S.2599 can be considered as a certain issue of Andronicus Gidon. Of the others the most convincing candidates are the rare silver types copying S.1984, and Jordanov's "Latin" bronze trachy, and even then the silver types are doubtful, as we have to rely largely on the argument that they can't be issues of the 12th century because they aren't obviously electrum, a premise that might not be applicable in Asia.

As we have seen, the rest certainly all seem to be Asian issues, and probably not of a regular Nicaean mint, but their assignment to Gidon at Trebizond, or even to the 13th century, has in most cases not been securely established. In particular, the rough types S.2597-8 could still possibly be issues of Andronicus Comnenus at some Asian mint, although probably not Trebizond itself, as could the similar and presumably related "inverse" types from the Crimea (Bendall 2007^c Types 1-2).

The Chalkites types are quite clearly issues of the 13th century, and they may well be issues of Gidon, but there is nothing definite to say that they are; in fact, on the current evidence they could just as well be, as the weights and mix of types have always suggested, the product of a late Nicaean mint, although presumably one east or north of Magnesia.

Overall, Bendall's assignments of most of the types considered here to Gidon, rather than to a Comnenan or Nicaean ruler, while seemingly quite reasonable, rely largely on indirect arguments such as silver content or the style of the flans, rather than on the hard evidence of datable finds, together with the assumption of who else?

It is true Bendall can appeal to some find evidence, but it is very limited, and of rather uncertain value; what is obviously lacking, at least for the moment, are reported finds of these types in and around the city of Trebizond itself. Clearly, what is now needed are some reliable finds for these types, particularly finds datable to the 13th century, although in their absence a survey of museum holdings in north-eastern Turkey (and for that matter, the Crimea) might well go a long way towards clarifying the situation.

Also, some figures for the alloy compositions of the various silver types might also help to clarify the relations between them.

a:   S. Bendall, "An Early Coinage of Trebizond?", N. Circ. 2002, p.113-5.

b:   S. Bendall, "A hoard of coins of Andronicus I Gidon of Trebizond", N. Circ. 2007, p.8-16.

c:   S. Bendall, "Andronicus I Gidon of Trebizond Again", N. Circ. 2007, p.75-80.

Ross Glanfield

September 2009.

10 Aug. '10.  Minor changes made to various arguments.