Cleopatra Selene1, daughter of Cleopatra VII and Antony2, born 403, possibly declared queen of Cyrene at the Donations of Alexandria in autumn 344, placed under the guardianship of Octavia, sister of Augustus, after the annexation of Egypt5, married Juba II king of Mauretania6 between 25 and 207, as his first wife8, by whom she had a son, Ptolemy, king of Mauretania9, through whom she probably had further descendants10, probably a daughter, Unknown (here suggested as Cleopatra)11, and possibly an older son Unknown (??Ptolemy) who died young11.1; she probably died c. 512.
[1] PP VI 14526, PIR2 C 1148. Gr: Kleopatra h Selhnh. The name "Selene" (the Moon) is explained by Plutarch, Antony 36.3, as the counterpart to her twin brother Alexander "Helios" (the Sun). Ý
[3] See discussion under Alexander Helios. Ý
[4] So Dio Cassius 49.41. However, Plutarch, Antony 54.4, has the territory (called "Libya") restored to Cleopatra VII. For the date, see discussion under Cleopatra VII.
The question can perhaps be resolved through coinage. Svoronos 1899 (imaged here) was dated to Antony's notional third consulate, i.e. 31. It was very rare to have a purely textual coin of this type, and Svoronos argued that it must be of Cyrenean origin, since he felt that there was more evidence of this type of coin from there. However, T. Schrapel, Das Reich der Kleopatra 47ff., noting that such coins were also rare in Cyrene, prefers a proposal by M. Grant, From Imperium to Auctoritas 46, that the coin was a camp issue made shortly before Actium. This seems reasonable to me. Octavian was at war with Antony and had withdrawn his consulate, so the coin is in essence a claim by Antony to be the rightful Roman leader.
Schrapel reviews other numismatic evidence from this period in order to narrow down the dates of portrait coins of Antony and Cleopatra VII from Cyrene. He establishes that Roman provincial coinage naming P. Licinius / P. Lepidus, L. Lollius, [P. Canidius] Crassus and L. Pinarius Scarpus demonstrates that normal Roman rule was in place till 34. Therefore the Antony/Cleopatra coinage must date from the period 34-30, showing that Cleopatra VII was the nominal coruler of Cyrene, not her daughter. Ý
[6] Plutarch, Antony 87, Dio Cassius 51.15. Ý
[7] The date is usually given as 20, based on numismatic evidence. According to Dio Cassius 53.26.2 Juba was granted Gaeticulan territory by Augustus in his 9th consulate, i.e. 25, which establishes the date of his first regnal year as 25/4. Tacitus, Annals 4.5, mentions him as still alive in the ninth year of Tiberius = 23 AD, but in the next year mentions Ptolemy's recognition by the Senate (Tacitus, Annals 4.26). Coins are known up to year 48, which matches this chronology perfectly. The only dated coin for Juba before year 23 is dated to his sixth year = 20/19 (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque no 357), and shows Juba on one side and Cleopatra Selene on the other. It is therefore argued that this must be the year of the marriage.
This establishes a terminus ante quem. D. W. Roller, The World of Juba II and Kleopatra Selene 86, notes that this date makes Selene about 20 and Juba nearly 30 at the time of the marriage, and that in 25, when Juba II left Rome to become king, both were already of marriageable age (15 and 25). He therefore proposes that they were married before they left Rome, and suggests that the coin marks the fifth anniversary of the marriage. This is perfectly possible, thuogh it cannot yet be proved. Ý
[8] His second wife was Glaphyra, daughter of Archelaus king of Cappadocia and great-granddaughter of Archelaus, the husband of Berenice IV. The marriage is given by Josephus, Antiquities of the Jews 17.13.4 and Jewish War 2.7.4. She is also named in OGIS 363 = IG II2 3437/8 as "[q]ueen [G]laphyra daughter of kin[g] Archelaus wif[e] of king Jub[a]". The dates are uncertain, but are bounded by the fact that her marriage to Juba occurred between her first and third marriages, which were to two sons of Herod of Judea: Alexander, who Herod executed shortly before his own death in 4 BC, and Archelaus, who was deposed as ethnarch c. 6/7 AD. Josephus also says that Glaphyra married Archelaus after the death of Juba, but cannot be correct, since Juba died in 23 AD. Presumably there was a divorce, but Josephus was unaware of this (it not being important for his purposes).
There is to date no trace of Glaphyra in North African inscriptions. It has been conjectured (e.g. A. Bouché-Leclercq, Histoire des Lagides II 366f.) that Juba spent the years c. A.D. 1-4 in the east, possibly as part of the mission of Gaius Caesar; this would account for the Athenian inscriptions naming him and would give him the opportunity to meet and marry Glaphyra. On this conjecture, it is probable that the marriage ended with his return to Mauretania in about 4 AD. It has been suggested (F. Jacoby, RE 9, 2386) that Josephus' source used an ambiguous word such as "departed" to explain the end of his marriage with Glaphyra, and that Josephus understood this to mean his death. Ý
[9] PIR2 P 1025. Parentage: Suetonius, Caligula 26. He was king of Mauretania 21-40 AD, initially in coregency with his father. He was murdered by Caligula who then annexed Mauretania to the empire, possibly in relation to the conspiracy of Cn. Cornelius Lentulus Gaetulicus. The terminus post quem for his birth, based on the date of his mother's marriage, is 24/19 BC. B. Chanler, Cleopatra's Daughter 356f. n. 188, notes that Tacitus, Annals 4.23, describes him as boyish and reckless at the time of the war with Tacfarinas, in the early A.D. 20s, and so argues that he is unlikely to have been born much earlier than 5 BC. The numismatic evidence begins with coins of Juba's year 30 = 5 AD (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque no 375) which show him with a young child, who is diademed, and in an undated version (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque nos 379-381) is called "king Ptolemy", while slightly later coins of year 36 = 11 AD show him with a second head who is bearded (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque no 383).
On the other hand, Tacitus, Annals 4.26 notes that Ptolemy was instrumental in winning the war against Tacfarinas, which suggests some military experience. D. Fishwick & B. D. Shaw, Historia 25 (1976) 491, suggest that this was gained in the Numidian campaign of years 4-6 AD, an experience which would have linked him to the Roman proconsul Cossus Cornelius Lentulus, whose son, Cn. Cornelius Lentulus Gaetulicus, led the consipracy against Caligula. For this to be true, an earlier date rather than a later one is to be preferred for his birth. However, the linkage to Lentulus is only hypothetical. Without stronger evidence, the development of his portrait on the dated coins makes a birthdate in the years c. 10-5 still seem the best match to me.
D. W. Roller, the World of Juba II and Kleopatra Selene, 256, estimates a birth-date of c. 13-9 BC on the assumption that he would have been born before Selene turned 30. This is a little earlier than my estimate, but not out of line with it. Ý
[10] He probably married a queen [Julia] Urania (PIR2 I 710), known only through a funerary inscription of her freedwoman Julia Bodina at Cherchel (J. Carcopino, Mélanges Grat I 31). Since, from other inscriptions, freedmen and women took on the gentilicium of their former masters, Urania probably had the gentilicium "Julia", which happens to be that of the royal family. Carcopino, noting the Greek name Urania, the apparent absence of a gentilicium of her own, and the absence of any other information about her, considers her to be a concubine of servile origin. Since Juba appears to have been monogamous, he suggests that Urania's husband must have been Ptolemy, and proposes that he revived the ancestral Numidian custom of harem polygamy; the title of "queen" ascribed to her by Julia Bodina was a local courtesy. M. Coltelloni-Trannoy, Le royaume de Maurétanie sous Juba II et Ptolémée 38, simply accepts this argument, though she also notes (p36) that polygamy was forbidden to Roman citizens as an objection to the theory that Juba II married Glaphyra while Selene was still alive, and does not explain why this same objection does not apply to Ptolemy, who was also a Roman citizen. D. W. Roller, The World of Juba II and Kleopatra Selene 252 n. 41, also follows this line, dismissing Urania as probably "a member of the Mauretanian court" who was "queen in more an informal than a legal sense". He suggests that "Urania" was a nickname taken from the muse, given to a "favoured member of the harem".
To my mind, a more plausible reconstruction is given by C. Settipani, Continuité gentilice et continuité familiale dans les familles sénatoriales romaines à l'époque impériale: mythe et realité 438f n. 11. He points out that the name Thea Urania (=Astarte) was held by a wife of Phraates IV of Parthia, and also notes a certain Uranius Antoninus, an Emesan high priest of Astarte and usurper of the mid-third century. He suggests that the name is in origin Emesan, and that Julia Urania was connected to the Emesan royal house, which became one of the leading petty principalities of the Roman East in this period. It may be noted that this dynasty also took on the gentilicium Julius. On this theory, Urania is called "queen" because that was her actual status, and there is no need to invoke a revival of the earlier custom of polygamy to explain her existence. In theory, she could then be a wife to either Juba II, late in his life, or of Ptolemy; of the two, Ptolemy seems more likely.
Ptolemy
was probably the father of Drusilla, wife of Felix, procurator of
Judea under Claudius (PIR2 A
828, Tacitus,
Histories 5.9),
who is described as a "granddaughter of Antony and
Cleopatra". Felix is also said by Suetonius,
Claudius
28,
to have become, in consequence of his elevation, "husband of
three queens". Josephus,
Antiquities
of the Jews 20.7.2,
describes the marriage of Felix to Drusilla, sister of Herod Agrippa
and former wife of Azizus, king of Emesa in Claudius' 13th year (54 AD).
At first sight, since there is no suggestion that Alexander Helios or Ptolemy Philadelphus had any children, one would conclude that Drusilla, granddaughter of Cleopatra, must be the daughter of Cleopatra Selene. She cannot be the Jewish Drusilla, who was maternal granddaughter of a marriage between the nephew of Herod the Great and his daughter, and so cannot be a descendant of Antony and Cleopatra VII. But these passages, particularly in combination, present some problems.
i) One could infer from Suetonius' wording that Felix married three times after becoming governor of Judea, that is, all the marriages happened in the 50s AD or later. This is not necessarily so. Tacitus notes that Felix had married the granddaughter of Antony and Cleopatra immediately after stating that Claudius made him governor, and compares Felix, as grandson-in-law to Antony, to Claudius as Antony's grandson. This would seem to indicate that the marriage occurred shortly before his appointment. Certainly, it seems hard to push back the marriage back past the late 30s, when Felix, a former slave, was most likely freed. Yet Selene was too old to have children after about 5 AD, even if she lived that long. So the marriage would have had to have been to a woman who was at least in her forties if Tacitus is correct, and probably at least a decade older. The marriage to a granddaughter of Cleopatra Selene is much more plausible biologically.
ii) The name "Drusilla" requires explanation. As G. M. Macurdy, Hellenistic Queens 226 correctly notes, it is very unlikely that Cleopatra Selene would have given her daughter such an un-Ptolemaic name. A. Bouché-Leclercq, Histoire des Lagides II 365, suggests she was named after Livia Drusilla, the wife of Augustus. This explanation could also apply to a granddaughter of Selene. However, I think the clue to the name lies in the history of the Jewish Drusilla, whose name is equally unexpected in the Herodian family. Josephus, Antiquities of the Jews 19.9.1, tells us that she was 6 years old when her father, Herod Agrippa, died in 44 AD. Clearly, she was born in the reign of Caligula and was named after that emperor's much-beloved sister. The Herodian family provides other examples of this behaviour: Agrippa I, named after M. Vipsanius Agrippa, born in 11 B.C., very shortly after the latter's death the previous year (Josephus, Antiquities of the Jews 19.8.2) and Drusus, a short-lived son of Agrippa, who was probably born shortly after the death of Agrippa's friend Drusus Caesar, the son of Tiberius, in A.D. 23 (Josephus, Antiquities of the Jews 18.5.4, cf. N. Kokkinos, The Herodian Dynasty, 76).
This suggests that Drusilla was born shortly after the death of a prominent member of the imperial family for whom she was named. The only two Drusillas are Augustus' wife Livia Drusilla and Caligula's sister. But Livia was adopted into the Julian gens by the terms of Augustus' will, and she died as Julia Augusta, not as Drusilla (Tacitus, Annals 1.8). That leaves Caligula's sister, after whom Felix' second wife was also named. The fact that Felix married two queens of the name Drusilla is not proof, as Macurdy thinks, that Tacitus got confused, but rather it is evidence that both women were born at about the same time. While Ptolemy could have been as old as his late fifties in Caligula's reign, he is more likely to have been in his forties. In either case, there is nothing inherently implausible in the idea that he had a daughter at this time of his life. Since Ptolemy was a cousin of Caligula's, it is also very likely that he would name such a daughter Drusilla. Thus, all these factors are consistent with the conjecture that the Drusilla mentioned by Tacitus was born in the late 30s AD.
D. W. Roller, The World of Juba II and Kleopatra Selene 252, who identifies Drusilla with Selene's daughter, suggests she was named after Augustus' stepson Drusus Claudius Nero, who died in 9 (Livy, Periochae 142). He suggests she married Felix in c. A.D. 5/10, though he admits that this implies a very long career for Felix. Moreover, at this time Felix was almost certainly still a slave -- certainly so if his gentilicium was Ti. Claudius rather than M. Antonius -- see below). However, this suggestion does raise the possibility, which I had not previously considered, that she was named after Drusus Caesar, and so was born c. A.D. 23. On this chronology, she would have been of marriageable age in the late 30s, around the time of the death of Felix' mistress Antonia Augusta. He was most likely freed at that time or in the next few years, which removes the objection of slave status.
While either date can be made to work, I am nevertheless still inclined to favour the idea that she was named after Caligula's sister on grounds of the shared gender.
On either chronology, Drusilla must be a great-granddaughter of Antony and Cleopatra, and almost certainly a daughter of Ptolemy rather than of his sister. On both, she was presumably brought into the imperial household on her father's execution and the annexation of his kingdom, and she was married to Felix in her teens, i.e. he was her first husband. Assuming, as seems reasonable, that she was one of the three queens which Suetonius ascribes to Felix, we must still explain how she became a queen. Suetonius only uses the word "regina" to describe a ruling queen or the wife of a king. But at the time of her marriage to Felix Drusilla was only the daughter of the king of a former kingdom. Unless the title was purely honorary, she must have married a king after her marriage to Felix. Since Felix married the Jewish Drusilla in 54 AD, he must first have divorced his Mauretanian wife at the same time. As to the identity of his third queen, we know absolutely nothing.
If Drusilla married Felix shortly before he became procurator, they can have had at most one child, and, given her proposed age, very probably none. If she was married to him a decade or so earlier, however, there is a good chance they had children. Several descendants of Felix are known or suggested, and I list the ones known to me here for completeness:
i) (Antonius) Agrippa (PIR2 A 809), son of Felix and the Jewish Drusilla, perished with his wife in the eruption of Vesuvius in 79 AD (Josephus, Antiquities of the Jews 20.7.2);
ii) Antonia Agrippina (PIR2 A 887), whose name is written in a graffito in an Egyptian royal tomb, may be descended from this marriage;
iii) Antonia Clementiana (PIR2 A 889), probably a daughter of Felix, grandmother of
iv) L. Anneius Domitius Proculus (PIR2 A 620), greatgrandson of an Antonius Felix, probably ours;
v) Antonius Felix Magnus (PIR2 A 829), puer clarissimus, possibly to be identified with M. Antonius Fe..., priest of Augustus and Claudius in 225 AD, and his father
vi) Antonius Fronto Salvianus (PIR2 A 832), vir clarissimus and quaestor.
With the exception of the first, where the genealogy is certain but the nomen is not explicitly given, these proposals all hinge on the accuracy of Tacitus' statement that Felix's gentilicium was Antonius, i.e. that his full name was M. Antonius Felix. This is the generally accepted position, and it is supported by the facts that his brother Pallas was certainly a slave of Claudius' mother Antonia, daughter of Antony and Octavia, in 31, and that M. Antonius Pallas is recorded as a landowner in Egypt in the reign of Tiberius in pLondon 2.195, suggesting that he was freed by Antonia. It is presumed that Felix was also a slave of Antonia and was freed at the same time, or in her will. However, the received texts of Josephus, Antiquities of the Jews 20.7.1 give Felix the gentilicium Claudius (though this is not always reflected in translation). C. Settipani, Continuité gentilice et continuité familiale dans les familles sénatoriales romaines à l'époque impériale: mythe et realité 439 n. 2 points out this it only takes an emendation of one letter to convert the sense from "He sent Claudius Felix" to "Claudius sent Felix". Nevertheless, Felix himself is only known as a freedman of Claudius, and N. Kokkinos, Latomus 49 (1990) 126 has persuasively argued that an inscription found at Bir el-Malik in Israel should be restored as naming Ti. Claudius [Felix] procurator of [Judea]; he also points to evidence that a Ti. Claudius Felix was a treasurer under Claudius (as was Pallas) and in charge of the Roman water supply system.
As to the identity of the royal husband of the Mauretanian Drusilla, we know nothing. But a reasonable guess on the reconstruction proposed here is that he was C. Julius Sohaemus, king of Emesa, who succeeded his brother C. Julius Azizus in the first year of Nero, i.e. 54 AD (Josephus, Antiquities of the Jews 20.8.4), very shortly after latter's wife, the Jewish Drusilla, was married to Felix, and after Felix had divorced the Mauretanian Drusilla. The royal family of Emesa is very imperfectly known. However, it was very probably connected to, if not the same as, the family of the priests of Baal at Emesa, who later married into the Severan imperial family, and from whom two Roman emperors -- Elagabalus and Alexander Severus -- were descended. Emesa -- modern Homs -- is a fully active modern city and has never been systematically explored archaeologically, so we can hope that more information about the Emesan family may yet be found.
Finally,
there is the tradition (Trebellius
Pollio, Historia
Augusta: Tyranni Triginta 27.1,
30.2)
that Zenobia,
queen of Palmyra in the late 3rd century AD (PIR2 S 355),
was descended from Cleopatra
VII, Dido and Semiramis. This has been dismissed as "wild
fantasy" on the part of the author(s) of the Historia Augusta,
but the claim to a descent from Cleopatra
VII was almost certainly made. An imperial declaration sent to
the people of Alexandria describing Alexandria as "my ancestral
city" only fits Vaballathus (PIR2 S 347),
son of Zenobia (P. J. Parsons, CdE 42
(1967) 397), and Kallinikos of Petra, a contemporary of the emperor
Gallienus, dedicated a book on the history of Alexandria to
"Cleopatra", who can only be Zenobia (Suda, KallinikoV;
see A. Stein, Hermes
58 (1923) 448).
The reality of this tradition cannot be proved, but in light of the likelihood of links between the family of Cleopatra Selene and Emesa as discussed above I see nothing farfetched about it. C. Settipani, Continuité gentilice et continuité familiale dans les familles sénatoriales romaines à l'époque impériale: mythe et realité 433ff. has examined the state of this problem in some detail, and the following discussion summarises his results. Such Syrian links would presumably explain the claim to Semiramis, e.g. through the Seleucids and the Achaemenids, while the fact that Dido is also claimed as an ancestor points specifically to a link through Cleopatra Selene, since her husband Juba claimed to be a descendant of the sister of Hannibal (Scol. Lucan, Pharsalia 8.287), and Hannibal's family, the Barcids, claimed to be descended from a younger brother of Dido (Silius Italicus Punica I 71-7). Although the near ancestry of Zenobia herself is not certainly known, she is probably closely related to J. Aurelius Zenobius, governor of Palmyra in 229 AD, whose paternal ancestry is traceable for 6 generations, and includes a Sampsigeramus, which is the name of the founder of the Emesan dynasty, and a C. Julius Bassus, priest of Baal, whose name recalls that of Julius Bassianus, high priest of Baal at Emesa and ancestor of the Severan emperors Caracalla, Alexander Severus and Elagabalus.
After her defeat by Aurelian, Zenobia is said to have been taken to Rome where she married an unnamed senator, by whom she had descendants traceable into at least the fourth century AD. The reality of this tradition seems confirmed by an inscription found in Rome for the gloriously-named L. Septimia Patavinia Balbilla Tyria Nepotilla Odaenathiania (PIR2 S 354), whose first and last names would seem to connect her directly to Zenobia's husband, Septimius Odaenathus, king of Palmyra. Also, the philosopher Libanius mentions amongst his friends in the late 390s a certain Eusebius son of Oedenathus and descendant of king Oedenathus, and in the 9th century Photius noted that around 500 AD the pagan philosopher Damascius dedicated a book to Theodora, daughter of Diogenes, son of Eusebius son of Flavianus and descendant of king Sampsigeramus of Emesa. Settipani notes that these two fragments tie in very well with each other chronologically if we make Flavianus son of Libanius' friend Eusebius, and at least support the idea of strong linkages between Emesa and Palmyra, even if these individuals were descended from Oedenathus through his first marriage rather than through Zenobia. Ý
[11] An Athenian inscription, IG II2 3439, was dedicated by a "daughter of king Juba" whose name is not given. This does not name her mother, but Cleopatra Selene is a strong presumption, since the marriage to Glaphyra was apparently short. M. Coltelloni-Trannoy, Le royaume de Maurétanie sous Juba II et Ptolémée 218f., lists three other inscriptions from Athens naming Juba, including one (OGIS 197 = IG II2 3445) that was dedicated by his son king Ptolemy.
Another reference is IG III1 1309 (not listed by Coltelloni-Trannoy), an epitaph set up in Athens by a king's daughter from Libya in memory of an attendant whose name is unfortunately lost. A. Wilhelm, Mélanges Bidez II 1007, proposes that this princess is Cleopatra VII, visiting Athens with Ptolemy XII during his exile, a suggestion accepted by M. Grant, Cleopatra 15. The description of her as "Libyan" is explained away as a loose Greek description for any part of North Africa, known in this sense since Homer. But it seems to me highly unlikely that a daughter of any king of Egypt would be described this way. It makes much more sense to identify this princess as the daughter of king Juba, who we know from Pausanius 1.17.2 was known in Athens as "Juba the Libyan".
D. W. Roller, Cleopatra 191 n. 34, notes my comment but still regards the assignment of IG III1 1309 to Cleopatra VII as "more probable", without giving a reason.
It is sometimes argued, for example by D. W. Roller, The World of Juba II and Kleopatra Selene 252, that this daughter can be identified with Drusilla, described by Tacitus as a granddaughter of Antony and Cleopatra, but it is hard to make this work chronologically. D. W. Roller, Cleopatra 155, acknowledges that it is problematic and that the alternative reconstruction proposed here is "possible".
Given Selene's insistence on advertising her Ptolemaic connections and the name of her son, "Cleopatra" seems to me a much more likely name than "Drusilla" for this daughter. Ý
[11.1] D. W. Roller, The World of Juba II and Kleopatra Selene 247 notes that the relatively long period between the marriage and the birth of Ptolemy of Mauretania allows for the possibility of extra children who died young. D.E.E. Kleiner & B. A. Buxton, AJA 112, 57 at 83-86, suggest on iconographic and circumstantial grounds that a child prince depicted on the Ara Pacis, usually identified as Gaius Caesar, Augustus' grandson, was instead a son of Juba and Selene, sent to Rome as an honoured hostage. If so, then he would have been born c. 20, about the time that Selene and Juba apear on joint coimnage, which would then have been issued to celebrate the birth of an heir. Since there is no other trace of him, he presumably died shortly after the Ara Pacis was built in 13. While the suggestion is intriguing, and plausible, it is rather speculative.
D. W. Roller, The World of Juba II and Kleopatra Selene 247 suggests that the fact that Ptolemy of Mauretania bore a dynastic name does not mean he was the first born son, on the grounds that neither of Cleopatra VII's first two sons were called Ptolemy. This is of course incorrect -- Caesarion certainly was called Ptolemy -- and besides homonomy was never an objection to choosing a name in the dynasty; cf. Ptolemy VI and Ptolemy VIII, to take the best known example. In any case, if the theory is correct, there is a good chance that this prince died before Ptolemy of Mauretania was born. It seems to me that, if he existed at all, it is most likely that his name was also Ptolemy. Ý
[12] Three dates have been proposed for her death:
i) c. 5 BC (A. Bouché-Leclercq, Histoire des Lagides II 366, estimates 4-3 BC). This is based on the assumption that Juba's marriage to Glaphyra indicates that he was a widower at that time, since Juba was a Roman citizen and therefore required to be monogamous by Roman law. The exact date comes from Epigram 18 by Crinagoras (Anthologia Palatina 7.633), which describes a darkening of the moon on her death. This was thought to refer to a lunar eclipse, in particular the eclipse of 23 March 5 BC.
ii) c. 5/6 AD. B. Chanler, Cleopatra's Daughter 356f. n. 188, argued that the date of 5 BC was impossible because it meant that Ptolemy was at least 30 years old when he succeeded Juba as sole king, at a time when Tacitus, Annals 4.23 describes him as a reckless youth. She also noted that Ptolemy begins to appear on dated coins with his father in 5 AD and supposes that this is because Juba wished to start emphasising dynastic continuity immediately after her death. M. Coltelloni-Trannoy, Le royaume de Maurétanie sous Juba II et Ptolémée 38 makes the same argument, and also notes the disappearance of all imagery related to Selene (Egyptian crocodiles, Isiac symbolism etc) in these dated coins.
Chanler also notes notes a fragmentary inscription, now CIL VIII 9343, a monument to the king and queen of Mauretania. She follows the reconstruction of L. Charrier, Description des Monnaies de la Numidie et de la Maurétanie 95, who reconstructs the inscription as referring to the "happy return" of the king and his victory over the Gaetuli, and therefore dates it to c. AD 6. She argues that in this year there was another lunar eclipse visible in Mauretania (now dated to 7 AD), as the date of her death.
iii) c. 18 AD. A hoard of coins, including coins of Juba II dated between year 36 = 11 AD and year 42 = 17 AD, was discovered at Ksar in Morocco in 1907. The hoard included a number of undated coins in mint condition with the head of Cleopatra Selene as well as that of Juba. K. Regling, ZfN 28 (1910) 9, 11, distinguished the style of two engravers in the coins of Juba II, with class II having a smaller head, rougher treatment of the hair and sharper and harder edges that class I. Class II coins were generally in much better condition than class I coins. Dated coins were associated with Class I up to year 36, and with class II in years 41 and 42. Many of the coins showing Cleopatra Selene, although undated, were of class II. Therefore, Regling argued that either Cleopatra's portrait and symbols continued to be minted long after her death, or Juba must have remarried her after his brief marriage to Glaphyra, and she did not die until after the date of the hoard, i.e. 18 or later. The second solution was accepted by G. H. Macurdy, Hellenistic Queens 227f. and J. E. G. Whitehorne, Cleopatras 201, though G. H. Macurdy, Vassal Queens 55, preferred the first. Be that as it may, Regling's analysis is ignored by M. Coltelloni-Trannoy, Le royaume de Maurétanie sous Juba II et Ptolémée 38 and J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque.
My views on these arguments are as follows:
i) The use of Tacitus, Annals 4.23 to argue that Ptolemy was a youth on his accession carries no weight against the development of his portraits on the dated coins of 5 AD onwards. Tacitus is explicitly referring to his immature policies, which need not reflect a biological immaturity. Hence this argument does not require a death date after c. 5 BC.
ii) Coltelloni-Trannoy's point that the joint series with Ptolemy shows no traces of Egyptian imagery seems to me to be a strong argument against the idea that Cleopatra Selene was also portrayed on coins with Juba during the same period. Also, coins showing Cleopatra Selene are never dated. After the wedding commemorative, the earliest known date of Juba is year 23 = 3 BC (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque no 293). Therefore, if one attaches more significance to the iconography of the coins than to their style, these factors argue for a death date before 3 BC.
iii) Crinagoras 18 is ambiguous evidence, since the passage could well simply be a poetic metaphor for her death playing on the lunar aspect of her name. Assuming however that it does refer to a lunar eclipse, such eclipses visible from Mauretania occurred in 9 and 5 BC and in AD 3, 7, 10, 14 and 21. None of the last three are easy to correlate with theory that the Ksar hoard shows a Cleopatra Selene surviving till shortly after AD 17. Nor is an eclipse date of either AD 3 or 7 easy to correlate with a change in policy towards the succession and towards Egyptian motifs occurring c AD 5. Hence, if the epigram is valid evidence, the circumstantial evidence best fits a date of 5 BC. The change in coinage styles and the adoption of dated coinage most likely simply reflects Juba's return from the east.
iv) Charrier's reconstruction of CIL VIII 9343 is rather adventurous and is not accepted in CIL itself. Charrier reconstructs it as: [Genio Iuba] REGIS / [et Cleopatrae] REGINAE / [....]NAE MAGNAE / [Templ]UM VICTORIAE / [Regressu]S FELIX / [Restituti]O PACI. In CIL the reconstruction is: [...] REGIS / [...] REGINAE / [...Bo]NAE MAGNAE / [...sign]UM VICTORIAE / [...]S FELIX / [...decret]O PAGI. Clearly, the key to the different reconstruction is the reading PACI (Charrier) rather than PAGI (CIL).
v) Unfortunately Regling's article is not illustrated. It is therefore not easy to verify his stylistic analysis. Certain of the features defining his "style II", such as the presence of sharper and harder edges, may simply reflect the fact that these coins were in much better conditon than class I. A number of the Cleopatra Selene coins are also said to be "transitional" in style between style I and II, which does not instil confidence that the criteria are well identified. Also, the possibility of posthumous coinages cannot be excluded: S. Gsell, Histoire ancienne de l'Afrique du Nord VIII 221 notes that coins with the portrait of Juba's ancestor Masinissa were struck long after his death.
vi) Any date later than c. 1 AD raises the problem of Juba's marital relationships with Cleopatra Selene and Glaphyra. Either we must suppose that Juba married Glaphyra bigamously -- against Roman law and Hellenistic custom, but permissible in Numidian tradition -- or that Juba divorced Cleopatra Selene before his conjectured departure to the east. Assuming a divorce then, on Regling's chronology, he must later have remarried her. None of these difficulties arise if we assume that she died before Juba's marriage to Glaphyra. However, possible evidence for the divorce theory is the existence of coins (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque nos 392-395) for Cleopatra Selene alone, which might imply that after a divorce she had her own domain. It has been suggested that these coins represent her regency during Juba's absence in the east, but this seems unlikely in view of a coin of Juba's dated to year 28 = 3 AD (J. Mazard, Corpus Nummorum Numidiae Mauretaniaeque no 294). If there was never a rift between Juba and Selene then these coins remain an unexplained puzzle. The suggestion of D. W. Roller, The World of Juba II and Kleopatra Selene 249 n. 26, that they represent her autonomous power of coinage is probably on the right track; possibly such coins were not preferred in circulation, which would account for their apparently good condition.
On the whole, and mostly because of Juba's marriage to Glaphyra, I am inclined to favour a date in the region of 5 BC for Selene's death. Ý
Update Notes:
11 Feb 2002: Added individual trees
28 Feb 2002: Split into separate entry
14 April 2002: Added discussion of numismatic evidence
on Cyrene betweenn 34 and 30.
18 May 2003: Changed Plutarch Xrefs to the Lacus
Curtius edition
24 Feb 2004: Changed Suetonius Xrefs to the Lacus
Curtius edition; added Xref to online Silius Italicus
7 Nov 2004: Extended discussions of date of marriage,
Urania, Drusilla, date of death based on Roller's comments.
16 Nov 2004: Finally tracked down CIL VIII 9343 and
countered the Charrier/Chanler argument for an AD6 death date based
on it
29 Dec 2004: Changed Historia Augusta Xref for Zenobia
to LacusCurtius edition
20 Jan 2005: Added Xref to Suda entry on Kallinikos
11 Mar 2005: Added Greek transcription
16 Sep 2006: Added link to Packard Humanities DB
18 July 2010: Comments on Roller's
reaction to my assignment of IG III 1309 to a daughter of Juba;
note Kleiner/Buxton proposal for an older son.
9 Dec 2010: Fix broken Perseus & DDbDP links
Website © Chris Bennett, 2001-2011 -- All rights reserved