that had always been the custom; for the council', for the Apodectai, for the Logistai, for all the other offices, the lot must have been deliberately introduced for political purposes. Whether this was done when these offices were themselves created we do not know; we do not even know when they were created; though it cannot well have been before Cleisthenes, nor after 450 B.C.2 It is in its application to all these offices that the essentially democratic nature of the lot consists, much more even than in its later use for the archonship. It is a neglect of this truth which has made Fustel de Coulanges overestimate the continued importance of the religious associations3; and a similar omission to recognise the essential differences between the two has made Grote and others, who have rightly estimated the political use in later times, ignore the truth that it could have had a quite different value when applied only to the archonship. The discussion on the whole matter is instructive because it shows how easy it is to go wrong by "abstraction," and to look on a custom like "election by lot" as though it had an absolute value,

1 Fustel de Coulanges suggests that the Prytanies were elected by lot because their duties were originally religious (p. 390). This is very improbable.

2 The earliest mention of the lot for the council is in an inscription of about 450, which, though it applies to Erythrae, is evidence for the use of the lot at Athens then.

It is still possible that the discovery of some early inscription will give us certain knowledge on the matter.

3 After showing rightly enough that the lot was originally religious and must have existed before the democracy, he goes on to say that it never was democratic. "Il n'était pas d'ailleurs un procédé démocratique," 1. c., n. 1. Even if, as he and Curtius suggest, the institution of the generals by Cleisthenes was a democratic measure, that does not in the least affect the point that the lot as afterwards used was still more democratic.

which would be the same at all times and in all circumstances1.

1 It does not seem possible to discover with any certainty when election by lot was abandoned.

Bergk, N. Rhein. Museum, xix. 605, refers to an inscription published in the Ephemeris Archeologeia, 3793, which contains a list of archons. This is of the time of Mithradates, and in it one man Mýdelos is mentioned as holding the office three years running. This shows that by this time the constitution had been modified, and the archonship was now filled by election from among the richer citizens. The change is generally supposed to have taken place in 146. (Cf. Hertzfeld Griechenland unter den Römern,' 310. C. I. A. iii. 87.) For a minor office we find it existing after this. (Cf. C. I. A. iii. 81. Wachsmuth, die Stadt Athen, i. 651.)

I should however expect to find that the alteration was made before this. Whatever the forms may have been, it is impossible that the reality of the democratic constitution can have been maintained so long as this. It is disappointing that nothing is said about the matter in the accounts of the change of constitution in 322 B.C. Cf. Droysen, Hellenismus, ii. 80-81; Diod. xviii. 65.

CHAPTER III.

Rotation in Office.

THE OFFICIALS.

WE must now turn from the council to a consideration of the other offices to which men were appointed by lot; and in regard to these also, if we would understand the lot, we must consider it in connection with the other parts of the constitution to which it belonged. The lot might be used in other states with other results: at Athens, of which we alone have knowledge, it was only one (though an essential) part of the democratic system. It helped to secure perfect equality among all citizens, a regular rotation in office, and the undisputed authority of the assembly.

The system is really the extension of an oligarchic principle to the whole body of citizens. In a strictly oligarchic state the council is supreme, not the assembly; but the council is chosen only from the members of certain families, and all magisterial appointments and all administrative posts of influence are reserved for members of the oligarchic clique. But in such a state, though a seat in the council is for life, the tenure of all other offices is short, and every member of the ruling order has a right to succeed to office in his turn. By this means

the supremacy of the council over all offices is secured as well as equality among those who are eligible. The lot might have been used to secure this end, just as it was used at Athens; probably indeed it was in operation in some states. However this may be, the principle of rotation in office was not peculiar to Athens, nor was it essentially democratic. Aristotle especially recommends it for his πολιτεία, and he makes it an objection to the state proposed in the Republic of Plato that there the same men will always hold office. He maintains repeatedly that within the select circle of the governing class in any state it is undesirable to make distinctions, and that all "peers" ought to have a share in office'. What was done at Athens was to apply the principle which is found in every oligarchy to a much larger class. There were to be no half-citizens; a man if he was a citizen at all was to have his share in the pleasures and responsibilities of magisterial authority. Here, as in an oligarchy, the Senate consisted of past or future magistrates: but the past and future magistrates included all the citizens. The garland of office came to an Athenian by right of birth, just as did the fasces to a Metellus, or his uniform to a Hohenzollern2.

1 Ar. polit. ii. 2, 1261. Aristotle maintains that, where it is impossible for the best men always to remain in office, whether office be good or bad, it is right that all should take part in it.

Cf. ii. 5, 1264. And again viii. 8, 1308a he specially recommends that in a ToTela the offices should all be made of short duration, so that all qualified persons may be admitted.

Cf. also vii. 2, 1317.

2 At Rome the principle was practically maintained that all

Character

istics of

and Democracy.

There are however important differences between

Oligarchy the working of the principle of rotation in an Oligarchy and a Democracy, since in an oligarchy the officers are few in number, and their duties are in consequence complicated. Accumulation of offices is common; and re-election is practised, if the offices are not held for some considerable period. This is necessary, since the men from whom officials are chosen are few in number. Further, though public responsibility is recognised, and an account can be demanded of the conduct of each official, yet this is generally of but little avail. Account is to be rendered to the same body from whom the magistrates are chosen; the individual is protected by the order to which he belongs, because the interests of all are identical. Each member of the order feels an attack on one of his fellows as a possible curtailment of his own privileges, and the supremacy of the order is endangered when the right of popular criticism is allowed.

At Athens the offices were numerous; all offices were collegiate, and the colleagues were seldom less than ten in number; re-election was almost unknown1; accumulation of office rare; office was

members of the senatorial families should succeed in turn to the

great offices. Had Rome been governed by a pure oligarchy this could have been done by the lot; as it was, this result was practically secured notwithstanding the nominal existence of popular election.

1 As we have seen, re-election to the Bouλn was possible if not common. For other offices however the matter was different. So far as I am aware, there is no definite statement on the matter in any contemporary authority: but