Bryn Mawr Classical Review 97.11.11

W. Pohl (ed.), Kingdoms of the Empire. Leiden: Brill, 1997. Pp. x+ 230. $75.00. ISBN 90-0410-845-9.

Reviewed by Robin Seager, Dept. of Classics and Ancient History, University of Liverpool,

Prudently disregarding the ludicrously grandiose blurb, the reader will find here two papers on the nature of treaties in late antiquity (Wirth, Heather) and two papers and an oracular brief note on the question of barbarian settlement (land versus tax income) (Liebeschuetz, Durliat, Wolfram), linked by one that treats both topics (Pohl) and framed by an introduction (Pohl) and a conclusion (Chrysos). Contributions written in German have been translated into English, though Wirth's in particular would have been much easier to understand had it been left in the original; Durliat's appears in the original French.

In his introduction Pohl on the subject of treaties rightly stresses first that the rhetoric is a part of reality, second that the rhetorical reality may have little to do with the balance of power between contracting parties. On barbarian settlement he ingenuously remarks: 'The unbiased reader will possibly find that most of the evidence fits both models (or all three ).' This seems somewhat optimistic. No sane judge would hang the proverbial yaller dawg on the 'evidence' presented in the relevant contributions, much less a theory.

Wirth's professed subject is Rome and its German partners in the fourth century, but it takes him twenty-eight pages to reach it, by way of a rambling history of deditio, the earlier part of which is otiose, the latter so allusive as to be often obscure. He then argues cogently enough that fourth century treaties are to be seen against a background of deditio, admission and integration. But there is little trace of planned control by Rome or a positive policy directed towards integration, though a distinction may be drawn between the fluidity of the situation on the Rhine, where Rome was dealing with small tribal units in constant flux, and the relative stability of the Visigoths on the Danube.

Heather on fourth century treaties is admirably lucid and concise. His careful examination of the contemporary evidence of Ammianus, Libanius, Symmachus and the panegyrics shows that most treaties with peoples on the Rhine and Danube were preceded by deditio, which enabled the claim, as it had ever since the time of the republic, that those concerned had been added to the empire. This process need not necessarily involve actual fighting or subsequent provincialisation. Precise terms inevitably varied from case to case with the realities of power, though rhetoric might obscure differences. Thus Ammianus and Themistius suggest that the agreement with the Goths in 369 was just like that of 332, whereas in fact the parties met on much more equal terms, while Macrianus clearly demanded much more respectful treatment than the kings reduced to submission by Julian. But even here Heather notes the significant point that treaties might be couched in terms familiar and therefore acceptable to the barbarians involved.

Pohl's account of treaties and negotiations with the Lombards contains a useful section on terminology, which points to a conclusion similar to Heather's, that the terminology of treaties between both Rome and barbarians and barbarian and barbarian remains more or less static, whatever the underlying realities of power. He makes unnecessarily heavy weather of Proc. BG 5.3.23 f., where the point is surely simple: true allies do not merely promise help, they deliver on demand. The section on Lombards, Gepids and the empire is essentially no more than a brief historical narrative, but its conclusion is wise: 'As always, the significance and the limits of judicial obligations are tested in conflict.' It is true throughout Roman (and perhaps not only Roman) history that attitudes to treaties are determined by the advantage of the moment and the number of tanks possessed by either side.

Pohl also addresses the question of the accommodation of the Lombards, discussing in detail the two essential pieces of evidence, HL 2.32 and 3.16. His criticisms of the views of Gaupp, Goffart and Durliat are sound enough, demonstrating that all are flawed and in particular involve excessively drastic 'interpretation' of 3.16. But his own argument leads to the conclusion that a difference between the two passages is barely discernible. This would mean in effect that reliqui (sc. nobiles) in 2.32 and populi adgravati in 3.16 are virtually synonymous, which is hard to believe. Nor does his assertion that HL 'did not conceive of the settlement of the Lombards in terms of the division of land' come close to being proven. But it is easy to agree with him that things on the ground were very fluid and much less consensual than the texts suggest, that systematic terminology and stringent theories of international law have no place in the enquiry, and that major changes in the situations described produce only minor variations in the rhetoric applied to them.

Pohl also makes the entertaining remark that the models of Goffart and Durliat may be useful provided that we assume that they do not 'account for most of what really happened', thus assigning them an intellectual and moral standing more or less equal with that of an airline timetable. The 'land or tax income' controversy is indeed a methodological horror-show. Space precludes more than a few brief observations here.

(1) 'If land, why no fuss?' Traditionalists, represented here by Liebeschuetz, reply that (a) there was vacant land in Italy (b) many of Theodoric's men could be accommodated on land originally given to the forces of Odoacer. Therefore the problem was not as great as it might at first appear. To deny this revisionists must either say that Procopius is lying or does not know what he is talking about (Goffart) or, more daringly but no more convincingly, that he too is speaking of tax allowances, not land (Durliat).

(2) Liebeschuetz makes the valid point that the barbarians might well have wanted the security of tenure. One might add that, however sophisticated they or at least their leaders might be in diplomacy and its rhetoric, the rank and file at least might have thought in terms only of land and been satisfied with nothing less.

(3) Cass.Var. 2.16.5 and LBurg. 54 and 55.2, 5 appear to be talking about land, not 'fiscal plates', as is admitted even by those who deny that in fact they are. It is just conceivable that the letter might be a systematic exercise in misrepresentation designed to flatter Liberius by exaggerating the difficulty of the task he had accomplished with such skill. Liebeschuetz appears to believe that this would involve so great a departure from the truth as to be self-stultifying, even in panegyric. That need not be so. The falsification would be no greater than the presentation of purely diplomatic negotiations as total victory in a major war, on which Augustus embarked without turning a hair in the matter of the recovery of the standards from Parthia. But the revisionists still need to come up with a convincing explanation of why Gundobad or his jurists should also have spoken in riddles.

(4) Liebeschuetz's refutation of Durliat's claim that the cities had control of any imperial tax revenue is entirely convincing, and Durliat's use of evidence is elsewhere open to question. In Lib. Or. 39.12 georgia simply means 'source of income' (as LSJ). It has nothing to do with land, taxes levied on it, or peasants, though Libanius' choice of metaphor was no doubt dictated by the fact that in his previous example of Mixidemus' villainy the victims were farmers. Nor can the interpretation of Malchus fr.16 which Durliat offers to make it fit his views be described as 'hors de toute theorie'.

In his conclusion Chrysos is sound on this issue, though he does not indicate his reasons. He also cites Straub's doctrine that 'if we do not understand the sources we should respect them'. As long as that principle is observed, then in the matter of barbarian settlement the burden of proof, pace Durliat, still rests on the revisionists. To invent a model, then tailor the sources to fit it is at best pointless. Only by respecting what the sources say and attempting to answer the questions that arise is progress possible. If the questions prove unanswerable, that is frustrating. But that frustration should not lead us to turn on the sources and rewrite or discard them to suit our preconceptions.