1. 31.12.1–7. All references are to Ammianus unless otherwise indicated.

2. See Burns, T.S., ‘The Battle at Adrianople: A Reconsideration’, Historia 22 (1973) 336–45.Google Scholar

3. On the Psychology of Military Incompetence (London 1976)Google Scholar esp. chapter 22. The ‘Bloody Fool’ explanation does not in fact explain incompetence by otherwise intelligent people.

4. The classic studies are by Adorno, T. et al., The Authoritarian Personality (New York 1950)Google Scholar; Rokeach, M, The Open and Closed Mind (New York 1960)Google Scholar. See too Grossman, J.C., ‘Experimental Manipulation of Authoritarianism and its Effect on Creativity’, Journal of Consulting and Clinical Psychology 36 (1971) 238–44;CrossRefGoogle ScholarPubMedChristie, R. and Jahoda, M. (edd.), The Authoritarian Personality (Glencoe 1954)Google ScholarPubMed; Simonton, D.K., Genius, Creativity and Leadership (Cambridge, Mass. 1984) 54CrossRefGoogle Scholar; Easterbrook, J.A., The Determinants of Free Will (London 1978) 80, 109.Google Scholar

5. Cf. Dixon (n.3 above) 216; Easterbrook (n.4 above) 13, 63; Alfoldy, G., ‘The Crisis of the Third Century as Seen by Contemporaries’, GRBS 15 (1974) 89-111.Google Scholar

6. Adorno (n.4 above) 248–50.

7. Storch, R., ‘The XII Panegyrici Latini and the Perfect Prince’, Acta Classica 15 (1972) 71–76Google Scholar; Seager, R. ‘Some Imperial Virtues in the Latin Prose Panegyrics’, Papers of the Liverpool Latin Seminar 4 (1983) 129–66Google Scholar.

8. Cf. Adorno (n.4 above) 733; authoritarian acceptance of ideology is not based on understanding or belief in its content but on what immediate use can be made of it. Cf. A. Hitler: ‘One can only die for a belief one does not understand.’

9. Cf. McMullen, R., ‘What Difference did Christianity Make?’, Historia 32 (1986) 322-44Google Scholar.

10. Lefcourt, H.M., Locus of Control (2nd ed., London 1982) 42Google Scholar.

11. 22.4.3f. Cf. Frank, R. I., ‘Ammianus on Roman Taxation’, AJP 93 (1972) 69–86Google Scholar.

12. Auerbach, E., Mimesis, trans. Trask, W. (New York 1953) 48fGoogle Scholar.

13. Most people were poor and poor people are more inclined to attribute success to luck, social acceptance, power and status, and less to internal factors like personal initiative, independent judgement, self-control or competence: Easterbrook (n.4 above) 13. Cf. Carney, T.F., The Shape of the Past (Lawrence 1975) 61, 128.Google Scholar

14. As Brown, Peter reminds us, ‘ruling the Roman world meant war, torture and execution’: Society and the Holy in Late Antiquity (London 1982) 87.Google Scholar

15. Thompson, E.A., The Historical Work of Ammianus Marcellinus (Groningen 1947/69) 83Google Scholar; Blockley, R.C., Ammianus Marcellinus: A Study of his Historiography and Political Thought (Brussels 1975) 143Google Scholar, with many references. Cf. 159: ‘Ammianus was certainly a snob.’

16. Rowell, H.T., Ammianus Marcellinus: Soldier Historian of the Late Roman Empire (Cincinnati 1964) 46.Google Scholar

17. Fortune could be earned by merit but is mostly fickle, malicious, variable, and unjust, a classically field-dependent and authoritarian perspective. Cf. Naude, C.P., ‘Fortuna in Ammianus Marcellinus’, Acta Classica 7 (1964) 70–88Google Scholar. On the significance of natural events, cf. the authoritarian belief that ‘it is more than just chance that Japan had an earthquake on Pearl Harbour Day, December 7, 1944’: Adorno (n.4 above) 249.

18. See the emphatic statements at 31.4.9, and also Austin, N.J., Ammianus Marcellinus on Warfare (Brussels 1979) 72Google Scholar; Thompson (n.15 above) 121.

19. Physiognomy is an externally based guide to character that discourages psychological probing. Feelings and motives are attributed, notable uncertainty and the stress of command (16.3.3, 17.3.1 and 13 other references — often enough to make one wonder if there is not a projected fear of responsibility) but one misses the speculation about inner states so prominent in Tacitus. Cf. Blockley (n.15 above) 27ff, 82ff.; Crump, G. A., Ammianus Marcellinus as a Military Historian, Historia Einzelschriften 27 (1975) 59ffGoogle Scholar.; Sabbah, G., La Méthode d’Ammien Marcellin (Paris 1978) 421ff.Google Scholar, 493ff.

20. His two Roman digressions may owe something to Juvenal who, like Suetonius and Procopius (Anecdota), manifest typical authoritarian preoccupation with sexual ‘goings-on’. Ammianus does not betray the same exaggerated concern. This may be due to his concept of the dignity of his work (26.2.11, 28.1.15). He does, however, give emperors a chastity rating (Julian, Constantius, Valentinian — good; Jovian — poor), and abhors the lustfulness and sexual habits of Saracens, Persians and Taifali (14.4.4, 23.4.7, 31.9.5). Compare his attitude to adultery and the Roman elite (28.4.2,9).

21. 31.5.11. Cf. Alonso-Nuñez, J.M., The Ages of Rome (Amsterdam 1983) 19–23Google Scholar; Paschoud, F., Roma Aeterna (Rome 1967) 67ff.Google Scholar

22. Adorno (n.4 above) 248, illustrating authoritarian submission.

23. Dixon (n.3 above) 169; Abrahamsson, B., Military Professionalisation and Political Power (London 1972) 75ffGoogle Scholar. On Ammianus’ interest in gathering military intelligence, see Austin (n.18 above) 22ff., 117ff.

24. Cf. Auerbach (n.12 above) 46: ‘magical and sensory dehumanization’. The dehumanized mob of rioters over whom Leontius presides like a lion tamer (15.7.1–5) reflect the dehumanized subjects in the contemporary obelisk of Theodosius. Cf. Momigliano, A., ‘The Lonely Historian Ammianus Marcellinus’ in Essays in Ancient and Modern Historiography (Oxford 1977) 127–40Google Scholar, at 136.

25. Fifty-two references are given by Blockley (n.15 above) 183, of which seven are favourable or sympathetic: Ursicinus and Valentinian like lions (each 1), Roman auxiliaries like birds (1), victims of cruelty like dumb animals or cattle (4). I am aware that some physiognomic and allegorical thinking may be involved here (cf. Sabbah [n. 19 above] 310) but an element of denigratory aggression is unmistakable. Think of people as animals and they will be seen as such.

26. Cf. Frézouls, E., ‘Les deux politiques de Rome face aux barbares d’après Ammien Marcellin’, Crise el Redressement (Strasbourg 1983) 175–97Google Scholar; Wiedemann, T.E., ‘Between Man and Beasts: Barbarians in Ammianus Marcellinus’ in I. Moxon (ed.), Past Perspectives (Cambridge 1984) 189–202.Google Scholar

27. Blockley (n.15 above) 154. Compare A. Momigliano, ‘Pagan and Christian Historiography in the Fourth Century A.D.’ in Essays (n.24 above) 107–26, who finds his autobiographical passages curiously unrevealing and considers it symbolic that the greatest feat of his military career is the escape from Amida (122).

28. Herodotus (120) scored lowest. I hope to publish and discuss more fully these findings elsewhere. See Winter, D.G., The Power Motive (New York 1973) 249–55Google Scholar and passim.

29. Padgett, V.R. and Jorgenson, D.O., ‘Superstition and Economic Threat: Germany 1918–40’, Personality and Social Psychology Bulletin 8 (1982) 736–41CrossRefGoogle Scholar. People also favour a more authoritarian brand of religion: Sales, S., ‘Economic Threat as a Determinant of Conversion Rates in Authoritarian and Non-authoritarian Churches’, Journal of Personality and Social Psychology 23 (1972) 420–28CrossRefGoogle Scholar.

30. Newbold, R. F., ‘Boundaries and Bodies in Late Antiquity’, Arethusa 12 (1979) 93–114Google Scholar, esp. 102–4; id., ‘Patterns of Movements and Communication in Ammianus and Gregory of Tours’ in Croke, B. and Emmett, A.M. (edd.), History and Historians in Late Antiquity (Sydney 1983) 66–81Google Scholar, esp. 68 (Ammianus’ environment is more unstable than that of Livy, Tacitus or Gregory, but is also the most imposed upon).

31. For discussion of this complex and often contradictory element, the oscillation between providentialism and determinism, see Blockley (n.15 above) 168ff.; Matthews, J.F., ‘Ammianus Marcellinus’, Ancient Writers vol. 2, ed. Luce, TJ. (New York 1982), 1117–38.Google Scholar

32. Auerbach(n.l2 above) 51.

33. 15.5.6, 16.8.7, 21.6.9, 27.6.14, 29.1.8, 29.1.36–39.

34. 15.1.3; Sabbah (n.19 above) 452f.

35. ‘He [Ammianus] has larger sympathies than Livy or Tacitus’: Glover, T.R., Life and Letters in the Fourth Century (Cambridge 1901), 36.Google Scholar

36. Hunt, E.D., ‘Christians and Christianity in Ammianus’, CQ 35 (1985) 186–200.CrossRefGoogle Scholar

37. Cf. 14.9.If., where whatever Ursicinus does is wrong. Cf. Christie and Jahoda (n.4 above) 207; Anderson, R.L., The Rise and Fall of Middle-Class Loyalty to the Roman Empire: A Social Study of Velleius Paterculus and Ammianus Marcellinus (Diss. U. Cal. 1962) 129ff.Google Scholar; Matthews (n.31 above) 1132–35.

38. This bohemianism is, however, much less socially destructive than the immoderation and wilfulness of authoritarian tyranny (27.7.9). Cf. Mackinnon, D.W., ‘The Nature and Nurture of Creative Talent’, American Psychologist 17 (1962) 484–95.CrossRefGoogle Scholar

39. It may be significant that, of all the historians mentioned in the study above, Ammianus most frequently mentions the effect of power on the recipient, i.e., the perspective of the object.

40. Writers tend to be comparatively flexible and to have a high non-authoritarian valuation of the role of the intellect and/or imagination: Barron, F., Creative Person and Creative Process (New York 1967) 71Google Scholar. Cf. Sabbah (n.19 above) 595, on Ammianus’ effort after objectivity, and Paschoud (n.21 above) 109, on Ammianus as the most clear-sighted man of his time.

41. Bowder, D., The Age of Constantine and Julian (London 1978)Google Scholar; McMullen, R., ‘Social Mobility and the Theodosian Code’, JRS 54 (1964) 49–57, at 49.Google Scholar

42. Cf. Auerbach (n.12 above) 58–61.

43. I.e., no credit is claimed for services rendered, an evolution from the do ut des of pagan philanthropy. The emperor may have claimed to be God’s viceregent but could not, for reasons of political survival, disclaim all credit for benefactions. Cf. Newbold, R.F., ‘Personality Structure and Response to Adversity in Early Christian Hagiography’, Numen 31 (1984), 199–215CrossRefGoogle Scholar; id. ‘Power Motivation in Sidonius Apollinaris, Eugippius and Nonnus’, Florilegium 7 (1985) 1–16.Google Scholar

44. Newbold, R.F., ‘Centre, Periphery and Eye in the Late Roman Empire’, Florilegium 3 (1981)72–103.Google Scholar

45. ‘The French Military Collapse in 1940: a Psychohistorical Interpretation’, Journal of Psychohistory 12 (1984–5) 313–37.Google Scholar The list is not complete since some indicators refer specifically to the Third French Republic. Kirkland analysed 614 articles altogether. 112 were either neutral or too close to call. Ominously, the number of autonomous articles fell from 68% in the early period of the study to 38% in the late: the dependent portion rose correspondingly from 32% to 62%. On autonomy and dependency, see further Deci, E.L., The Psychology of Self-Determination (Lexington 1980)Google Scholar; Gosling, R. (ed.), Support, Innovation and Autonomy (London 1973) 6,27.Google Scholar

46. Except for the scythed chariots. Cf. Thompson, E.A., A Roman Reformer and Inventor (Oxford 1952)Google Scholar; Baldwin, B., ‘The De Rebus Bellicis’, Eirene 18 (1978) 23–40Google Scholar; Hassall, M. (ed.), Aspects of the De Rebus Bellicis, BAR, Int. Ser. 63 (1979)Google Scholar; Astin, A.E., ‘Observations in the De Rebus Bellicis’ in Deroux, C. (ed.), Studies in Latin Literature and Roman History (Brussels 1980) 388–439Google Scholar; Wiedemann, T., ‘Petitioning a Fourth Century Emperor: The De Rebus Bellicis’, Florilegium 1 (1979) 140–50Google Scholar; Paschoud (n.21 above) 118ff.

47. Cf. Gordon, CD., ‘Vegetius and his Proposed Reforms of the Army’ in Evans, J.A.S. (ed.), Polis and Imperium (Toronto 1974) 35–55Google Scholar; Goffart, W., ‘The Date and Purpose of Vegetius’ De Re Militari’, Viator 33 (1977) 65–100Google Scholar, whose arguments for a fifth century date are unconvincing; Paschoud (n.21 above) 108ff., who berates Vegetius’ flattery, pessimism and docility; and Silhanek, D.K., Vegetius’ Epitome, Books 1 & 2: A Translation and Commentary (Diss. U. New York 1972)Google Scholar, who, though well-disposed towards Vegetius, admits his ‘Maginot line mentality’.

48. 22.4.6–8. Cf. Dixon (n.3 above) chapter 19, on authoritarian fear of effeminacy. Manpower was the greater problem — Ammianus shows some awareness of this( 19.11.7,31.4.4) — though the two problems are linked in that barbarian contingents were hard to control. See below, n.49.

49. See Cameron, A.D., Claudian: Poetry and Propaganda at the Court of Honorius (Oxford 1970) 158ff.Google Scholar

50. Austin (n. 18 above) 143–51.

51. 31.7.2. Cf. 14.2.7f., 3.2; Austin (n.18 above) 70, 109, 161; Crump (n.19 above) 46, 66.

52. 28.9.4; cf. 14.10.4; Blockley (n. 15 above) 117.

53. Ridley, R., ‘Notes on Julian’s Persian Campaign’, Historia 22 (1973) 317–30Google Scholar; Austin, N.J., ‘Julian at Ctesiphon: A Fresh Look at Ammianus’ Account’, Athenaeum 50 (1972) 301–9.Google Scholar

54. Twelve uses of confidere etc. are neutral or indeterminable. Thirty-one uses of audere etc. out of fifty-four are negative, twelve neutral. For references, see Chiabo, M. (ed.), Index Verborum Ammiani Marcellini (Hildesheim 1983)Google Scholar. On meritorious caution in commanders, cf. 16.2.11, 30.3.2, 31.10.22. At a time when manpower was such a problem, caution was indeed a virtue but more battles are lost by caution than by daring (Dixon [n.3 above], 221) and it is the general who attacks first who usually wins (Simonton [n.4 above], 153).

55. Thompson (n.15 above) 79. Cf. Austin (n.18 above) 85 (‘the military conservatism of the historian’), 17.8.2 (‘careful planning overcomes almost all difficulties’), and Ammianus’ doubts about Valentinian provoking the Quadi with garrisons across the Danube, 29.6.2–16.

56. Julian was autocratic, certainly, but not authoritarian, unless we choose to see him as nothing but a cunning and unscrupulous religious fanatic, as does Bowersock, G., Julian the Apostate (London 1978)Google Scholar. (For telling criticism of Bowersock, see the review by Athanassi-Fowden, P., Journal of Theological Studies, 30 [1979] 331–35)CrossRefGoogle Scholar. Autocratic behaviour has its time and place and is featured by appropriate obedience to authority, and direction of aggression, and an openness, warmth and flexibility that are the marks of successful leaders. Authoritarianism is essentially defensive, repressed and uncritical of authority: Dixon (n.3 above) 256–87. Ammianus disapproved of Julian preferring philosophers’ interpretations of omens rather than the traditional seers’ (23.5.10f., 25.2.7f.), but it may have been a blow for rationality as Julian saw it.

57. Browning, R., The Emperor Julian (London 1975) 224Google Scholar4: ‘He survived a youth that would have turned most men to cynics if not to psychopaths’; Head, C., The Emperor Julian (Boston 1976).Google Scholar