1. The analogy between the system of emperors and that of caliphs is a close one: see G. Dagron, Empereur et prêtre, étude sur le ‘césaropapisme’ byzantin, Paris, Gallimard 1996, pp. 70-3.

2. Dagron, op. cit., p. 72.

3. Ibid., p. 70, cf. 72.

4. J. Béranger, Recherches sur l'aspect idéologique du principat, Basel, Schweizerische Beiträge zur Altertumswissenschaft, 1953, p. 72.

5. An unwritten law excluded from the throne any man from the Greek civilization: out of 100 emperors or usurpers whose origins are known or suspected, not one is from a Greek background. In the fourth and fifth centuries when the western part of the Empire was in fact a German protectorate, another unwritten law prevented the all-powerful generals of German origin from ascending to the throne, so they created puppet emperors to rule in their shadow.

6. Dagron, op. cit., pp. 42-3.

7. Staatsrecht, II (2: 1133).

8. Egon Flaig, Den Kaiser herausfordern, Frankfurt and New York, Campus Verlag, 1992, p. 559.

9. Flaig, op. cit., p. 126.

10. The word had two meanings: the public interest (preventing a barbarian invasion was serving the Republic) and the traditional institutions, Senate, consulate, etc., which were like the spelling of the name Roman, the face of Rome.

11. Tacitus, Histories, II, 72: ‘in Istria there were still inherited clients of the old Crassus family, their country estates and the influence that went with their name’. In 69 an ex-officer of the imperial guard brought with him into Vespasian's party his native town of Fréjus, which was completely devoted to him ‘because he was the town's champion and it hoped he would gain power in the future’ (Histories, III, 43).

12. A. Wallace-Hadrill, ‘Civilis princes: between citizen and king’, in Journal of Roman Studies 72 (1982: 32-48). P. Veyne, Le Pain et le cirque, Paris, Editions du Seuil 1976, p. 718: ‘the imperial system was based on an absurdity: although he was sovereign by subjective right, the emperor was created by his subjects; could they respect their creature unconditionally?’

13. Tocqueville, De la démocratie en Amérique, I, 130.

14. Fables, I, 2 (3), 30 (the frogs complain to Jupiter about their wicked leader): ‘citizens, bear your present misfortune, the god told them, for fear a worse one should befall you’. II, 16, 1: ‘by changing their leader the middling citizens (cives pauperes) only change masters’.

15. Veyne, Le Pain et le cirque, p. 635, quoted by Flaig, op. cit., p. 122, n. 94: ‘zur Entscheidung nicht nur unfähig, sondern auch unwillig’.

16. Seneca, De beneficiis, II, 12: Caligula held out his foot to a senator to be kissed; ‘is that not trampling the Republic underfoot?’

17. Pliny the Younger, Trajan's Panegyricus, LIV, 5; and LXVI, 4.

18. Pliny mentions the Senate's ‘idleness’ under the tyrant Domitian (letter VIII, 14, 8-9); but under the best of leaders, Trajan, he also writes (III, 20, 12) that ‘everything depends on the whim of one man who, in the general interest, has assumed all positions, all tasks; however, by a beneficial mitigation, a few rivulets flowing from that generous source run down to us’. This is very different from the Panegyricus.

19. Tacitus, Annals, I, 73.

20. But, in order to mark himself off from the tyrants, it was also appropriate that he should refuse some of the divine honours his subjects granted him. This was another aspect of the farce of refusing power. Nero, who was an atypical tyrant (he did not have himself made a god), would occasionally refuse divine honours.

21. There is an astounding anecdote in Seneca, De beneficiis, III, 26, or a terribly vulgar one in the first of The Lives of Lucan, 4.

22. Tacitus, Annals, XI, 4; Ammianus Marcellinus, XV, 3, 5 (the senior police officer Mercurius, ‘count of dreams’) and XIX, 12.

23. There is a true-life spy story anecdote in Tacitus, Annals, IV, 69.

24. L. Friedländer, Sittengeschichte Roms, I, pp. 256-8. Use by the police of soldiers in plain clothes who would provoke people into speaking ill of the emperor (Epictetus, IV, 13, 5) and of courtesans (Pliny, Natural History, XXX, 15); some of Vitellius's soldiers crept into Rome to spy on public opinion; everyone kept quiet, everyone was afraid (Tacitus, Histories, I, 85).

25. Martial, X, 48, 21, under Trajan.

26. R. Syme, Tacitus, Oxford, Clarendon Press, 1963, p. 422, n. 6.

27. In 202 the reliefs on the arch of the Severi in the Forum represented the biggest stylistic break in the whole of Roman art; as Rodenwaldt has demonstrated, these reliefs reproduced - or rather were a skilful pastiche of - the paintings in popular style that were exhibited in triumphs to show the people how the war had proceeded. We might imagine that, in order to glorify Napoleon as an emperor close to the people, the reliefs that decorate the Arc de Triomphe at the Etoile in Paris were reproductions or condescending pastiches of images d'Epinal (simple country genre prints); see Ernst Kissinger, Byzantine Art in the Making, Harvard, 1977 (1995), pp. 10-13. In official reliefs the academic style of the century of the Antonines was followed by what Eugenia Strong called a ‘Flemish tapestry style’. In my view this stylistic break has not been sufficiently taken account of and its political significance has been misunderstood. The change was probably an initiative of the artist himself rather than an order from the emperor. Henceforth official bas-reliefs followed a separate path (for instance, on the arch of Constantine in the Roman Forum): they remained faithful to this style, which was designed to be, and considered itself to be, popular, not without a slightly haughty condescension, in contrast to the classical, academic Hellenizing style of the first two centuries that continued to be the style the aristocracy used for the bas-reliefs on sarcophagi.

28. There was no court life or court festivals at the palace. The emperor in his palace was not surrounded by the senators as a king was by his nobles. Far from having a royal style of life, he lived like any other aristocrat: each morning he was greeted by his crowd of clients and he invited senators and equestrians to dine. He had ‘friends’, ‘companions’ or ‘counts’, comites, but did they live at the palace? It is very doubtful; he had his freedmen, but the most important of them lived elsewhere in their splendid mansions (domus).

29. A conversion he admitted he wished for but did not feel he had the right to impose (Eusebius, Life of Constantine, II, 56 and 60).

30. In the new Cambridge Ancient History, XI, The High Empire, Cambridge, Cambridge University Press, 2000, p. 79.

31. Philo of Alexandria, Legatio ad Gaium, 9-11.

32. Fustel de Coulanges, Histoire des institutions politiques de l'ancienne France, I, La Gaule romaine, Paris, Hachette 1900, p. 191.

33. Antiquitates Iudaicae, XIX, 3, 228.

34. Similarly the Byzantine emperors were only isapostoloi, ‘equal to the apostles’. So they were not true apostles.

35. So this love is one of the affects intended to help individuals fit in with their world, allowing them to ‘conquer themselves rather than Fortune’ (as Descartes said) and to judge the grapes too green. Ideologies that are designed to deceive others are unimportant compared to those that are designed to let people make a virtue of necessity. On this fit between reality and what we think of it, see Jon Elster, Le Laboureur et ses enfants: deux essais sur les limites de la rationalité, French translation, Paris, Editions de Minuit 1987; Psychologie politique: Veyne, Zinovie, Tocqueville, Paris, Editions de Minuit 1990; L. Festinger, A Theory of Cognitive Dissonance, Stanford CA, Stanford University Press, 1987, and, for the limits to this theory, J.-P. Poitou, La Dissonance cognitive, Paris, Armand Colin 1974; D. Kahnemann, P. Slovic and A. Tversky, Judgement under Uncertainty, Cambridge, Cambridge University Press, 1982.

36. Jacques Krynen, L'Empire du roi: idées et croyances politiques en France, XIIIe-XVe siècles, Paris, Gallimard 1994, p. 458: ‘The study of love as a political virtue has not yet been carried out’; ‘The feeling of love for our kings seemed natural’, writes Maine de Biran in 1814; ‘this love was a religious feeling like divine love; it was a sort of worship that elevated the soul and, like honour, could command every sort of sacrifice of personal interest, even life’; and he deplores the fact that young men born after 1789 have never experienced this feeling and cannot understand it: they equate it, he writes, with self-interested calculating, part of the career plan (Journal intime, Valette and Monbrun (eds), Paris, Plon 1927, I, p. 78).

37. Quoted by Sainte-Beuve, ‘Relation inédite de la dernière maladie de Louis XV’, in his Portraits littéraires, III.

38. G. Posener, De la divinité du pharaon, Cahiers de la Société Asiatique, XV, 1960.

39. Diatribai, III, 4, 8.

40. This ethological inclination is recognized by the most classical of thinkers, but rationalized as functional because of its natural purpose: discussing the English constitution, Bagehot wrote that the monarch existed in order to make the community comprehensible to the people.

41. In the organization of modern societies there may be an individual, a president or a dictator, at the apex of the hierarchy who is the only one with the power to make the major decisions, such as pressing the button to unleash nuclear weapons (Raymond Aron, Etudes politiques, Paris, Gallimard 1972, p. 191). But it would be another rationalization to explain the ‘mythical’ image of the monarch on the basis of this fact.

42. J.-M. Schaeffer in the journal Communications, no. 72 (2002: 110, n. 6): ‘The human race has a bicephalous evolutionary destiny governed by both the slow pace of genetic evolution (or deviation) and the rapid pace of cultural evolution’.