¹ Hansen, Mogens Herman, The Athenian Democracy in the Age of Demosthenes, trans. Crook, J.A. (Oxford: Blackwell, 1991), 310Google Scholar.

² Gasaway, Robert R. and Parrish, Ashley C., ‘Administrative Law in Flux: An Opportunity for Constitutional Reassessment’, George Mason Law Review 24 (2017) 361–416Google Scholar, at 361, emphasis original.

³ Janet McLean, ‘Constitutional Boundaries: Public Office and Public Law’, 2017 MSS on file with the author, 18, citing a case reported variously in R v Mersham (Inhabitants) 7 East 167, 103 Eng Rep 64 (1806), and John Prince Smith Reports (1807).

⁴ Guerin, Benoit, McCrae, Julian, and Shepheard, Marcus, Accountability in modern government: what are the issues? A discussion paper (London: Institute for Government, 2018), 7Google Scholar.

⁵ Op. cit. note 4, 7.

⁶ Landauer, Matthew, ‘The Idiōtēs and the Tyrant: Two Faces of Unaccountability in Democratic Athens’, Political Theory 42 (2014), 139–66CrossRefGoogle Scholar, at 153, citing Aristophanes’ Wasps 587 (in Landauer's translation, modifying Sommerstein's: ‘“And doing all these things we cannot be called to account—a power not given to anyone else”’), and 156, citing Aeschines 3.232-3, which includes the observation (again in Landauer's translation) that ‘in a democratic city the private citizen rules as king through the law and his vote’. See also Hoekstra, Kinch, ‘Athenian democracy and popular tyranny’, in Bourke, Richard and Skinner, Quentin (eds) Popular Sovereignty in Historical Perspective (Cambridge: Cambridge University Press, 2016), 15–51CrossRefGoogle Scholar.

⁷ ‘Archaic elite-led regimes’: Matthew Simonton, W., Classical Greek Oligarchy: A Political History (Princeton and Oxford: Princeton University Press, 2017)Google Scholar, 5 n.17 and passim. Simonton argues more broadly that oligarchy should not be confused with these earlier regimes, but should instead be understood as emerging in the mid fifth century, in reaction to the crystallization of democracy.

⁸ Fröhlich, Pierre, Les cités grecques et le contrôle des magistrats (IVe-Ier siècle avant J.-C.) (Genève: Droz, 2004), 529Google Scholar: ‘Le contrôle des magistrats est un des traits caractéristiques des cités grecques, à la époque classique et à l’époque hellénistiques…Il faut parler des cités grecques, et non des seules démocraties. Car, même si la documentation n'est pas très riche sur ce point, le contrôle des magistrats se retrouve également dans les oligarchies’.

⁹ Willetts, R.F., The Law Code of Gortyn (Berlin: de Gruyter, 1967), 8CrossRefGoogle Scholar.

¹⁰ I. Cret. IV, 181-183 n^o. 79.

¹¹ McLean, ‘Constitutional Boundaries’, 7. Of course, the English system involved property and patronage in ways distinct from that of classical Athens or other Greek poleis.

¹² (2) continues: ‘by the imposition of a fine whose maximum varied, according to the magistrature, from fifty to five hundred drachmas, or else of sending him before the courts for severer punishment’; see Glotz, Gustave, The Greek City and its Institutions, trans. Mallinson, N. (London: Routledge and Kegan Paul, 1929, rpt. 1965), 204Google Scholar.

¹³ Op. cit. note 12,204. It should be noted that other Athenian administrative functions were carried out by public slaves (dēmosioi), who did not occupy an archē but often worked closely to assist those who did. See Ismard, Paulin, Democracy's Slaves: A Political History of Ancient Greece, trans. Todd, Janet Marie (Cambridge, MA and London: Harvard University Press, 2017), 12, 52–53CrossRefGoogle Scholar (this is a translation of Ismard, La démocratie contre les experts: les esclaves publics en Grèce ancienne (Paris: Éditions du Seuil, 2015)).

¹⁴ The full sentence in Greek, citing from the Oxford Classical Text, is ἔστι δὲ πολιτεία πόλεως τάξις τῶν τε ἄλλῶν ἀρχῶν καὶ μάλιστα τῆς κυρίας πάντων (1278b8-10). I discuss its complications in Lane, Melissa, ‘Popular Sovereignty as Control of Officeholders: Aristotle on Greek Democracy’, in Popular Sovereignty in Historical Perspective, edited by Bourke, Richard and Skinner, Quentin (Cambridge: Cambridge University Press, 2016), 52-72, at 65-7Google Scholar.

¹⁵ Aristotle in the Politics describes the constitution of Megara as a dēmokratia led by demagogues that was destroyed because of its ataxia (lack of order) and anarchia (1302b30, 1304b34–40; see also 1300a16–19).

¹⁶ So as one scholar has explained: ‘magistrates [meaning public officials generally, not judges specifically] were referred to as hai archai [plural of the noun] or hoi archontes [plural of the participle of the verb], terms which relate interchangeably to both the office itself and the person who held that office (singular: archē). But hoi archontes does not refer simply to any group of people in public service; rather, it connotes a defined set of responsibilities and expectations as prescribed by law’. See McAuley, Alex, ‘Officials and Office-Holding’, in Beck, Hans (ed.) A Companion to Ancient Greek Government (Chichester, West Sussex: Wiley-Blackwell, 2013), 176–90CrossRefGoogle Scholar, at 179.

¹⁷ I believe that this fact is important, and is overlooked by Jochen Bleicken's overly reductive approach to offices in his magisterial work on Athenian democracy. While the tasks of the officeholders may have been restricted and the authority of the ‘Volk’ in controlling them significant, their very role was to command and indeed to take executive (if not legislative) initiative: contra Bleicken, Jochen, Die athenische Demokratie (Paderborn, Munich, Vienna and Zurich: Ferdinand Schöningh, 1985), 153Google Scholar (‘es konnte nicht einmal Initiative von ihnen ausgehen: Die Beamten waren reine Verwaltungsträger’), and more generally 145-58 and passim. I am grateful to Benjamin Straumann for calling the importance of Bleicken's interpretation on this point to my attention, though I disagree with it as he does not.

¹⁸ As summarily stated by Matthew Landauer, op. cit. note 6, 140: ‘Athenian accountability institutions were remarkable, both in scope and in the intensity of their practice’.

¹⁹ Guerin et al., op. cit. note 4, 7.

²⁰ On all these mechanisms, see inter alia as a standard study, Roberts, Jennifer Tolbert, Accountability in Athenian Government (Madison, WI: University of Wisconsin Press, 1982)Google Scholar.

²¹ For an exhaustive study of dokimasia in and also beyond Athens, see Feyel, Christophe, Dokimasia: la place et le rôle de l'examen préliminaire dans les institutions des cités grecques (Nancy: Association pour la diffusion de la recherche sur l'antiquité, 2009)Google Scholar. Athens also deployed a procedure called dokimasia in a different (but related) context, that of scrutinizing eighteen-year-olds’ qualifications to be entered on the citizen rolls in their respective demes.

²² Fröhlich, op. cit. note 8, 323.

²³ Tolbert Roberts, op. cit. n.20, 6.

²⁴ Tolbert Roberts, op. cit. n.20, 15-17.

²⁵ Pownall, Frances, ‘Public Administration’, in Beck, Hans (ed.) A Companion to Ancient Greek Government (Chichester, West Sussex: Wiley-Blackwell, 2013), 287–301Google Scholar, at 299.

²⁶ Again, the general practice of requiring the rendering of accounts at the end of a magistrate's term in office was widespread, although the word εὔθυνα was infrequently used outside Athens and the expression τὰς εὐθύνας διδόναι was characteristically tied to Athenian procedures as well, where the euthunai took place in two stages.

²⁷ Tolbert Roberts, op. cit. note 8, 18.

²⁸ Hansen, op. cit. note 1, 222.

²⁹ Hansen, op. cit. note 1, 223.

³⁰ Guerin et al., op. cit. note 4, 29 and passim.

³¹ Fröhlich, op. cit. note 8, 57, 58, 60, 64.

³² As translated in The Oresteia, trans Grene and Doniger O'Flaherty; Greek from Teubner (West). Alan H. Sommerstein, in the Cambridge Greek and Latin Classics series, Aeschylus. Eumenides (Cambridge: Cambridge University Press, 1989), ad loc., remarks that this ‘may allude to a procedure of Athenian law’, and points out that in Aeschylus’ Persians (l. 828) ‘Zeus as chastiser of the arrogant is called εὔθυνος βαρύς’.

³³ See Demosthenes, Against Meidias (21.31-34). A dramatic example of Athenian councilors removing their crowns of office before stoning a traitor (Lycidas) to death is provided by Lyc. 1.22 (cf. Hdt. 9.5). I owe these references to Owen Phillips.

³⁴ Thompson, Dennis F., ‘Bureaucracy and Democracy’, in his Restoring Responsibility: Ethics in Government, Business, and Healthcare (Cambridge: Cambridge University Press, 2005), 50–70Google Scholar, at 53.

³⁵ Thompson, ‘Bureaucracy and Democracy’, 56.

³⁶ Guerin et al., op. cit. note 4, 14-15.

³⁷ Guerin et al., op. cit. note 4, 9.

³⁸ McLean, ‘Constitutional Boundaries’, 6-7, notes that there was recourse (on the basis of a warrant) to practices of auditing within the English administration in the late eighteenth and early nineteenth centuries: ‘the process distringas ad computandum…could force the delivery of accounts and be used to determine whether more money was taken than had been accounted for. In theory at least, there could be recourse to an office holder personally for arrears’.

³⁹ https://www.nytimes.com/2016/05/19/world/what-in-the-world/high-wycombe-england-annual-weigh-in.html?_r=1; www.mayorofwycombe.com

⁴⁰ Carawan, Edwin, The Athenian Amnesty and Reconstructing the Law (Oxford: Oxford University Press, 2013), 84CrossRefGoogle Scholar.

⁴¹ I am grateful to Victoria McGeer, whose question about what made their regime problematic made me realize that I needed to address these two standard views directly, as well as to Danielle Allen for help in formulating my preferred view.

⁴² No one accused of having ‘himself performed the act of killing or wounding’ was to be exempted by any of the provisions of amnesty from a homicide trial (Ath. Pol. 39.5-6). See also Andoc. 1.90.

⁴³ Among scholars who make what I take to be this interpretative mistake are Edwin Carawan and Adriaan Lanni. Carawan, Athenian Amnesty, 76, discusses ‘those who held office during the period of civil conflict’ – as including ‘the incumbent Ten who continued in office through the transition, and any officers from the previous regime (the Thirty et al.) who might now present themselves for accounting’, as well as ‘the democratic officers who had governed Peiraeus since the spring’ – saying that all of these were simply being covered by ‘the basic rule that officers must face accountings was nothing new’ (ibid.). But this wrongly or at best incautiously assumes that the Thirty simply had been, or were to be considered to have been, officeholders in the normal sense. Even more starkly, Adriaan Lanni remarks that the requirement of euthuna imposed as part of the amnesty agreement ‘was not an extraordinary transitional justice institution but the standard procedure faced by all outgoing officials under the democracy both before and after the [oligarchical] revolution’, in Law and Order in Ancient Athens (Cambridge: Cambridge University Press, 2016), 181.

⁴⁴ Lysias’ speech Against Eratosthenes is the sole candidate surviving text to be a prosecution at a euthuna of a member of the Thirty, though scholars contest whether it was rather ‘a subsequent prosecution for homicide, or indeed whether the speech was actually delivered’: Lysias, trans. S.C. Todd, The Oratory of Classical Greece, vol. 2 (Austin: University of Texas Press, 2000), ‘Introduction’ to Against Eratosthenes (Lys. 12), 113-16, at 113; he himself dismisses the cogency of the homicide prosecution theory, 113-14. He adds that ‘We have…no evidence for the result of the trial (if, that is, the speech was actually delivered)’ (115). More broadly, three of the Thirty had been killed in the fighting of 403, and most, perhaps all, of the others seem to have availed themselves of the provision in the agreement allowing their faction to withdraw from Athenian citizenship and establish their own self-governing enclave instead in Eleusis. (Here too there was careful provision made as to the future allocation of offices, with the stipulation that ‘No one living at Eleusis should be entitled to hold any office in the city, without cancelling his registration and returning to live in the city’ (AP 39.6).

It must also be acknowledged that Lysias does in ‘Against Eratosthenes’ describe the actions of Eratosthenes ‘once [he had been] established in office’ (εἰς τὴν ἀρχὴν καταστὰς, 12.48), and again refers to the Ten who supplanted the Thirty with the language of office (12.57); the latter is more understandable as an effort was being made to rectify the constitutional nature of rule, while the former may be explained as a willingness to use everyday imprecise language. Lysias elsewhere describes the Thirty as being ‘established’ (κατεστάθησαν), using a form of the same verb (καθίστημι) at 13.35, but adds no noun for office there: despite the fact that Todd translates as ‘took office’ (Lysias, ad loc., 146), the meaning in this context could be simply ‘established in power’.

⁴⁵ We do have five full or fragmentary speeches by Lysias challenging various candidates for office at their dokimasia. Four of them do so on the grounds of their having been among the Three Thousand (Lys. 31; 24; fragment on P. Ryl.), and in one case also having been a member of the cavalry (hippeis) who were especially supportive of them (Lys. 16) – the outcomes of all of which dokimasiai are unknown (Krentz, The Thirty, 117-18). A fifth dokimasia speech was made by Lysias at the dokimasia of Evandrus (Lys. 26), who had been selected for the position of archon (Pythodorus’ position) in 382/1, and whom the orator charged ‘with membership in the boule and service in the cavalry under the Thirty’ – despite which charges we know that ‘Evandrus passed his dokimasia’ (Krentz, The Thirty, 118). On these speeches, see also Loening, Thomas Clark, The Reconciliation Agreement of 403/402 B.C. in Athens (Stuttgart: Franz Steiner Verlag Wiesbaden GMBH, 1987), 103–117Google Scholar.

⁴⁶ Hansen, Mogens Herman, ‘Seven Hundred Archai in Classical Athens’, GRBS 21 (1980) 151–73Google Scholar, makes no mention of any suggrapheus in his list of Athenian offices, though he does list the anagrapheus among those offices not mentioned in the AP (156 with n.12, though he later (at 162) concedes that ‘it cannot be precluded that one or two of the magistrates listed above may have been councillors serving on a committee’ rather than constituting an office proper – and names the anagrapheus as one of the two possible examples of this caveat). Allen, Danielle S., ‘Appendix A: The Number of Magistrates in Athens’, in her The World of Prometheus: The Politics of Punishing in Democratic Athens (Princeton: Princeton University Press, 2000)Google Scholar, disagrees with Hansen on some points, but like him makes no mention of suggrapheis as established magistrates.

⁴⁷ Foucart, Paul-François, ‘Inscription d’Éleusis du Ve siècle’, Bulletin de correspondance hellénique 4 (1880) 225–56CrossRefGoogle Scholar, at 252-3. He also remarks, 252, on the precedents for συγγραφεῖς in the fifth century: ‘A proprement parler, les συγγραφείς n'avaient aucun pouvoir. Ils étaient chargés d'une mission qui consistait à rédiger une ordonnance sur un point désigné’.

⁴⁸ Todd, Stephen, ‘Lysias against Nikomachos: The Fate of the Expert in Athenian Law’, in Foxhall, L. and Lewis, A.D.E. (eds) Greek Law in its Political Setting: Justifications not Justice (Oxford: Clarendon Press, 1996), 101–31Google Scholar, at 111. Todd notes that this remark excludes a passage in Lysias (30.5) about the role of an anagrapheus (a passage which is itself controversial), itself a role whose subjection to euthuna is debated.

⁴⁹ Meiggs, Russell and Lewis, David M. (eds) A Selection of Greek Historical Inscriptions to the End of the Fifth Century B.C., rev. ed. (Oxford: Clarendon Press, 1988)Google Scholar, no. 73, in discussion ad loc. at 220, of two inscriptions. See also Ostwald, Martin, From Popular Sovereignty to the Sovereignty of Law: Law, Society, and Politics in Fifth-Century Athens (Berkeley, Los Angeles, and London: University of California Press, 1986), 416Google Scholar.

⁵⁰ I cannot review here all the debate in the scholarship about evidence claimed for and against the existence of individual suggrapheus or collective suggrapheis between these years, or for the nature of the Four Hundred compared to the Thirty. Suffice it to say that the Four Hundred had come to power in fact in a markedly more irregular assembly meeting in 411 (held at Colonus rather than in Athens), on the basis of a resolution that had initially banned anyone from holding office under the preexisting democratic constitution. So the Four Hundred had acted as suggrapheis in a highly irregular way and at least for the most part without making use of subordinate preexisting offices.

⁵¹ I owe this phrase to Scheppele, Kim Lane, ‘Autocratic Legalism’, The University of Chicago Law Review 85 (2018), 545–584Google Scholar.

⁵² If I am right in hypothesizing that they eschewed holding proper offices, from the absence of evidence to the contrary, this would contrast with the practice of the Four Hundred as described at one point by Thucydides, at least one of whom (Theramenes) was ‘in office’ (ἐν ἀρχαῖς) in their oligarchy (8.89.2) and would become a member of the Thirty as well. It is striking that in this episode, Theramenes and some others fall out with the broader oligarchical group (just as Theramenes would do again with the others of the Thirty, to fatal results, as I remark in my conclusion below). Perhaps this taught the recidivist oligarchs the danger of giving the tools of office to some of their number but not others, making them sensitive to the expectations of accountability that holding office might engender (for good or ill) among the dēmos.

⁵³ See discussion in Krentz, op. cit. note 53, 51-6.

⁵⁴ As does Shear, Julia L., Polis and Revolution: responding to oligarchy in Classical Athens (Cambridge: Cambridge University Press, 2011), 171Google Scholar.

⁵⁵ Krentz, op. cit. note 53, 57 with n. 3, naming Epichares (from Andoc. 1.95), Evandrus (Lys. 26.10), and Teisias (Isoc. 16.43), none of whom figure in the complete list of the names of the Thirty that he gives elsewhere in the work (at 51).

⁵⁶ Here I correct a mistake on this point in earlier unpublished lecture versions of this paper. While the liability of councilors to euthunai is not discussed at any length by Hansen, op. cit. note 1, it is mentioned in passing on 222; see also S.C. Todd, The Shape of Athenian Law (Oxford: Clarendon Press, 1993), 112 n.17.

⁵⁷ To be sure, there are other respects in which contemporary procedures are generally more Athenian than Roman, namely in allowing prosecution during one's term of office, even there may be some forms of immunity from prosecution for certain officeholders during their tenure. Alex McCauley notes that ‘While in the Roman republic magistrates enjoyed immunity from prosecution and sacrosanctity of their persons for their entire term of office, Greek office-holders never had such institutional protection’; modern procedures are also more akin to Greek ones in that, in the latter, according to McCauley, ‘It was not as if a magistrate were untouchable during his term of office; if anything as soon as one assumed authority one automatically became vulnerable to more vectors of accusation and oversight than any average citizen’: McAuley, op. cit. note 16,185,187 respectively.

⁵⁸ In Verrem II. 76. Translation from Cicero's Orations, trans. Charles Duke Yonge (Mineola, NY: Dover Publications, 2018).

⁵⁹ Both short quotations are from McAuley, op. cit. note 16, 177 and 176 respectively.

⁶⁰ McCormick, John P., Machiavellian Democracy (Cambridge: Cambridge University Press, 2011), 183CrossRefGoogle Scholar.

⁶¹ The research in this paper on office and accountability in ancient Greece was presented first in the 2017 Philip Hallie Lecture at Wesleyan University and discussed as part of a manuscript workshop at Princeton University in the same year. It was presented in revised form in the second of the six Carlyle Lectures which I delivered at Oxford University in the Hilary Term of 2018, and then further revised for delivery as the 2018 Royal Institute of Philosophy / Royal Society of Edinburgh Lecture that same year. It was also discussed in 2018 seminars in London, at the Institute for Historical Research and the Civil Service Leadership Academy, and in Cambridge and Oxford. I am most grateful to all who convened and commented on each of those occasions and in subsequent interchanges, and especially to John Haldane for convening the congenial occasion in Edinburgh on behalf of the Royal Institute of Philosophy together with the officers of the Royal Society of Edinburgh, and to Anthony O'Hear as editor of this journal.