Ammianus Marcellinus and Osroëne

di R. M. Frakes

The Ancient History Bulletin, 7/4 (1993), pp. 143-147

In Book XIV, the first extant book of the Res Gestae, Ammianus Marcellinus has a famous description of the Eastern Provinces of the Roman Empire (XIV.8.1-14). In the middle of this digression, at XIV.8.7, the Antiochene historian presents a cross-reference regarding the Roman province of Osroëne which has been the source of dispute for scholars. Closer examination of the passage in context will not only indicate a more accurate explanation of this cross-reference but will also yield some new information about the lost books of his history.

At XIV.8.7, Ammianus states:

And first after Osdroene, which we have omitted from this description, as has been said, gradually rises Commagene (now called Euphratensis); it is famous for the great cities of Hierapolis (the ancient Ninus) and Samosata.¹

The reference made by Ammianus to an earlier passage where he stated he omitted Osroëne from his description must refer to something written earlier in the Res Gestae. Most modern commentaries which analyze this passage agree that XIV.8.7 refers to XIV.3.2.² Furthermore, a thorough study of Ammianus’ use of reference verbs also indicates the same result; that is, that XIV.8.7 refers to XIV.3.2.³ However, this standard view has recently been contested.⁴ A closer examination of the context of this second passage, XIV.3.2 will confirm that the standard view will not hold and will lead to some further questions on the nature of the relationship between the extant and lost books of Ammianus’ history.

Chapter three of Book XIV of the Res Gestae describes an unsuccessful plot of the Persians against the Eastern provinces of the Roman Empire. Having discussed the genesis of the plot between the Persian King Sapor II and his general Nohodares, Ammianus states at XIV.3.2:

And because all the districts of Mesopotamia, being exposed to repeated raids, were guarded by frontier units and country garrisons, he [Nohodares] shifted his approach to the left and set up an ambush for the most remote parts of Osdroene,attempting a new and almost unprecedented attack; if he had succeeded, he would have laid waste the entire region like a thunderbolt. What he planned was this.⁵

After this passage, Ammianus describes the Persian general’s plan and how it failed. What is significant here is that Ammianus clearly is not stating what so many modern scholars have indicated he is. Namely, the Antiochene historian is not writing here that he will exclude Osroëne from his account. Rather, he is just mentioning Osroëne to give his audience a geographical reference point for the events he is describing.

Since the identification of XIV.3.2 as Ammianus’ reference for XIV.8.7 will not hold, the question which naturally arises is, where did Ammianus first state that he would omit Osroëne? Nowhere in the extant books does Ammianus state he has discussed, or omitted, Osroëne elsewhere.⁶ There are two explanations for the cross-reference at XIV.8.7. The first could be that Ammianus simply made a mistake and had not mentioned Osroëne anywhere else in his history. While possible, it has been proven that Ammianus is usually quite thorough and dependable in his cross-references.⁷

A second possible explanation is that Ammianus had written that he could omit Osroëne somewhere in the lost books (perhaps even in Book XIII Since we only have Books XIV-XXXI of his work, there no doubt could have been many places in the lost first thirteen books where Ammianus might have had need of mentioning Osroëne. Analysis of style can be of help here. Verbal analysis shows that Ammianus tended to use forms of dico to refer to things which he had described fairly recently.⁸ Since the construction at XIV.8.7 uses the phrase ut dictum est, this suggests that Ammianus had recently stated he would omit Osroëne — perhaps in Book XII or XIII.

Another part of Book XIV helps clarify what Ammianus is doing in his narrative with regard to Osroëne. At XIV.7.21, only shortly before XIV.8.7, Ammianus gives some fruitful information which appears at the end of a fairly long description

of atrocities committed in the Eastern Roman Empire under the Caesar Gallus in 354. At the end of XIV.7.21, Ammianus states:

... all justice left the courts and the legitimate defense of cases was silenced; the butcher and confiscator of property scoured for executions and property to seize throughout the Eastern provinces; these I think now an opportune time to review, except for Mesopotamia, which was described with the Parthian Wars, and Egypt which we will return to at the necessary time.⁹

Ammianus’ description of the Eastern Provinces (XIV.8.1-14) follows. When placed in this context, it seems likely that when the Antiochene historian states at XIV.8.7 that he would omit Osroëne as he had said he would, he was thinking back to what he had just written. That is, Ammianus would not describe Osroëne as he already had described it together with Mesopotamia in the context of his description of the Parthian Wars (which do not appear in the extant books and thus must have been included somewhere in the lost books of his narrative).

This slight confusion on Ammianus’ part is not unreasonable. First, the province of Osroëne is contiguous and south-west of the Roman province of Mesopotamia. This proximity alone could explain Ammianus’ confusion. But even more important is the fact that Osroëne had traditionally been part of Mesopotamia and earlier in Roman history had played a role as a dependant state of the Parthians and as a buffer zone between the Roman and Parthian Empires.¹⁰ Under Trajan, Mesopotamia, along with the district of Osroëne had been annexed by the Roman state. It was abandoned by Hadrian and reconquered by Lucius Verus.¹¹ This district played a role in Roman campaigns against Persia under Septimius Severus, Gordian III, Valerian and Aurelian as well as those of Carus and Galerius. After Galerius’ victorious campaign against the Persian usurper Narses from 297-298, Rome gained further territories of Mesopotamia from Persia. Indeed, it was as a direct result of Galerius’ victory that Diocletian reorganized the Roman Eastern frontier and severed Osroëne from Mesopotamia to make two provinces.¹² Therefore, Ammianus probably remembered stating he would pass over Osroëne as he had said before, because he had included a description of Osroëne in his earlier digression on Mesopotamia (which he mentions at XIV.7.21).

The cross-reference at XIV.8.7 to XIV.7.21 raises some interesting questions about the nature of the digression on Mesopotamia in the lost books. There has been some debate among scholars as to the context of events in which Ammianus may have developed this description.¹³ It has been suggested that this digression could have occurred during Ammianus’ description of Constantius II’s early reign (that is, from 337-353).¹⁴ However, Ammianus’ allusion to Osroëne indicates he must have placed his digression on Mesopotamia sometime before he discussed the reign of Diocletian in his lost books. That is, Ammianus probably made his description of Mesopotamia when Osroëne was still part of it. This analysis would date the Mesopotamian digression to the period before the tetrarchy’s treaty with Persia in 298. Can this digression be dated any more closely? Because of Osroëne’s location, it would seem likely that Ammianus developed this digression in context with a conflict between Rows and Pent. Such large-scale events were probably covered in the lost books of his history. Therefore, it would seem appropriate that he could have written this description during his account of emperors such as Trajan, Hadrian, Lucius Verus, Septimius Severus, Gordon III, Valerian, Gallienus, Aurelian or Probus.

Other cross-references to the lost books of the Res Gestae indicate that he did discuss most of these emperors.¹⁵ It is tempting to try to place the description of Mesopotamia during the period of the Barracks Emperors where it could have aided in describing the events of several campaigns to the Antiochene historian’s audience. Moreover, there are two other reasons for dating the Mesopotamian digression to somewhere between 240-297. First, Ammianus would have had access to witnesses, or at least second-hand witnesses, to some of these events during the formative years of his life.¹⁶ Second, Ammianus’ usage of the cross-reference verb narro is

usually employed to refer to something recently stated in his narrative.¹⁷ This consistent aspect of Ammianus’ style suggests that his description of Mesopotamia (narrarentur) would have appeared in Book XII or XIII. While there is no exact certainty as to when this lost digression appeared, a date in the context of events ranging somewhere between 240-297 would be the most likely possibility.¹⁸

Footnotes

1     XIV.8.7: Et prima post Osdroenam quam (ut dictum est) ab hac descriptione discrevimus, Commagena (nunc Euphratensis) clementer assurgit, Hierapoli (vetere Nino) et Samosata civitatibus amplis illustris.

2     Pieter De Jonge, Sprachlicher und historischer Kommentar zu Ammianus Marcellinus XIV 7-11 (Groningen, 1935) 67; Wolfgang Seyfarth, Ammianus Marcellinus: Römische Geschichte, (Berlin 1968), 265, note 165. E. Gallatier and J. Fontaine, Ammien Marcellin, Tome 1 (Livres XIV-XVI) (Paris, 1968), have only a note on the history of the city of Ninus on this passage.

3     Helena Cichocka, ‘O powtórzeniach w “Res Gestae” Ammiana Marcellina’ [‘De locis quibusdam, qui in Ammiani Marcellini opere q. i. Res gestae iterantur’], Eos 64 (1976), 203-222. In Polish with Latin summary and clear charts of the Antiochene’s use of reference verbs. The chart which refers from XIV.8.7 to XIV.3.2 appears on p. 213.

4     Alanna Emmett, ‘The Digressions in the Lost Books of Ammianus Marcellinus’, in History and Historians in Late Antiquity, edd. Brian Croke and Alanna Emmett (Sydney, 1983), 45-47 suggests that XIV.8.7 refers to XIV.7.21 - a position which this article will support. However, she lists XIV.8.7 as a separate reference to the lost books of the Res Gestae (52, n. 32). This identification undercuts her case. XIV.8.7 is either a reference to XIV.7.21 or a reference to the lost books.

5     XIV.3.2: Et quia Mesopotamia tractus omnes crebro inquietari sueti, praetenturis et stationibus servabantur agrariis, laevorsum flexo itinere, Osdroenae subsiderat extimas partes, novum parumque aliquando temptatum commentum aggressus; quod si impetrasset, fulminis modo cuncta vastarat. Erat autem quod cogitabat huius modi.

6     Ammianus does mention Osroëne directly in two other places in the extant books of the Res Gestae (XXIII.2.7 and XXIV.1.2). These descriptions are both strictly locational and do not shed any light on this problem. He also mentions Edessa, the capital of the province of Osroëne, but these references also do not help with the problem of the reference at XIV.8.7.

7     See Cichocka, 213-218 for chart which testifies to the dependability of Ammianus’ cross references. The one famous exception is XXII. 8. 35 where Ammianus says he will discuss the Halcyon birds at a later point but never does.

8     Cichocka, 213-214.

9     X1V.7.21: ... e iudiciis fas omne discessit et causarum legitima silente defensione, carnifex rapinarum sequester, et obductio capitum, et bonorum ubique multatio versabatur per orientales provincias; quas recensere puto nunc opportunum, absque Mesopotamia, iam digesta cum bella Parthica narrarentur, et Aegypto, quam necessario ahud reieciemus ad tempus.

10     For a brief general overview of Osroëne, see O. Krückmann,‘Osroëne’, RE XVIII.2, col. 1589-1590.

11     W.T. Arnold, Roman Provincial Administration, third edition, 278; on the history of Osroëne in general see A. von Gutschmid, Untersuchungen über die Geschichte des Königreichs Osroëne (1887).

12     C.S. Lightfoot, The Eastern Frontier of The Roman Empire with special reference to the Reign of Constantius II, D.Phil. thesis (Oxford University, 1981), 1-10, describes the history of Rome’s encounters with Persia from the time of Septimius Severus up to the beginning of the fourth century. For further information on the campaigns of 297/98 and the resulting peace treaty see A.H.M. Jones, The Later Roman Empire (Baltimore, 1986), vol. 1, p. 39.

13     Victor Gardthausen, vol. 1, 4; Jean Gimazane, Ammien Marcellin: sa vie et son œuvre (Toulouse, 1889), 409; Hugo Michael, Die verlorenen Bücher des Ammianus Marcellinus (Breslau, 1880), 6; John C. Rolfe, Ammianus Marcellinus, vol. 1, 65; Cichocka, 217; and especially Alanna Emmett, 45-47. For a recent discussion of the issue of the lost books of Ammianus, see John Matthews, The Roman Empire of Ammianus (London, 1989), 27-30.

14     De Jonge, p. 23 on XIV.7.7 argues for the Mesopotamian digression occurring at the time of the escalation of the Persian war following Constantine c. 337 (‘Dieser geographische Exkurs kommt wahrscheinlich in der verschollenen Beschreibung des Perserkrieges nach 337 vor’). See also H.T. Rowell, ‘The first mention of Rome in Ammianus’ extant books and the nature of the History’, Mélanges Carcopino (Paris, 1966), 843-844, who argues for the description taking place during Ammianus’ discussion of Constantius II’s early Persian campaigns before the extant books begin.

15     XXII.15.1 discusses Hadrian and Septimius Severus; XXV 8.5 describes Trajan and Septimius Severus; XXIII.6.24 mentions Verus; and XXIII.5.7 discusses the Gordians.

16     Ammianus’ birth date is not exactly known, but most scholars would agree on a date somewhere in the late 320s or at the latest in the early 330s. See E.A. Thompson, The Historical Work of Ammianus Marcellinus (London, 1947), 1. Such a birthdate would range from 30 to 70 years after these events. Ammianus as a boy could have heard stories of these struggles just as the author of this article (who was born in 1962) can remember his grandfather discussing how he caught typhus in WW1 France as well as his grandmother describing a hold-up in territorial Arizona in the late 1890s which she witnessed as a young girl. Of course, Ammianus himself also visited Osroëne during Julian’s Persian expedition (XXIII.2.7) and he may have gathered some information of geographical and historical nature there as well.

17     Cichocka, 217, shows that XVII.13.1 refers to XVII.12; XXVIII.1.1 refers to XXVII.12; XXVII.1.1 refers to XXVI.6-10; and XXVIII.6.29 refers to XXVIII.6.