Cammack: Deliberation in Ancient Greek Assemblies

A paper by Daniela Cammack, Yale University:

When an ancient Greek dêmos (“people,” “assembly”) deliberated, what did it do? On one view, it engaged in a form of public conversation along the lines theorized by contemporary deliberative democrats; on another, a small number of “active” citizens debated before a much larger, more “passive” audience. On either account, deliberation is represented as an external, speech-centered activity rather than an internal, thought-centered one. The democratic ideal, it is argued, was at least occasional participation in public speech.

This article questions that interpretation. A study of βουλεύομαι, “deliberate,” from Homer to Aristotle reveals three models of deliberation: internal, dialogical, and a partial combination that I shall call “guided,” in which speaking and deliberating were performed by advisers and decision-makers respectively. Assembly deliberation was almost always represented as guided deliberation. The dêmos—which is to say the audience—deliberated (ἐβουλεύετο), while those who spoke before it advised (συνεβούλευσε). Citizens thus did not fall short of a democratic ideal when they did not speak publicly. To the contrary, internal reflection, culminating in a vote, was precisely how the dêmos was expected to exercise its authority. The implications for our conceptualization of ancient Greek democracy are significant.

Full text

7 Responses

  1. This is an important paper for those who seek to model their modern proposals on Athenian democracy, as it seeks to overturn the rich participatory perspective of Habermasian and other deliberative democrats. This is illustrated by the author’s conclusion:

    Although deliberation in ancient Greek assemblies is usually interpreted as group discussion or debate and associated above all with those who spoke publicly, this article has shown that βουλεύομαι, “deliberate,” was historically associated with internal decision-making and was typically ascribed to assembly audiences** rather than to those who addressed them. Assembly deliberation was almost always represented as what I have called “guided deliberation”; those who spoke did not deliberate but advised those who did. Metaphorically at least, in coming forward to advise, speakers cast themselves outside the deliberating dêmos. They posed as counsellors (σύμβουλοι), analogous to the counsellors to kings found elsewhere in the ancient evidence.

    Cammack’s binary perspective — advisors and deliberators — maps well onto the argumentative theory of reasoning (Mercier and Landemore, 2012). According to this hypothesis reasoning serves two distinct survival-related functions: a) convincing people and b) evaluating the arguments of others. The second module pertains to Cammack’s perspective on deliberation and requires listening (open-mindedly), deliberating within and then deciding. The first module, however, pertains to advisors, as speakers “have completed their deliberation in advance” (Cammack, p. 27) and thereby benefit from the confirmatory bias, an important tool for forensic rhetoric (Mercier and Landemore, 2012, pp. 251-4). The derivation of the term “deliberation” from the Latin liber (weight) gives additional credence to Cammack’s perspective — Fishkin arguing that deliberation involves the “weighing” of competing arguments.

    Extrapolating to modern states, a large random sample of the demos would serve the second function as it would deliberate and vote in the same way as other samples (by definition). But the selection of advisors is an altogether different matter and a good case could be made for election.

    ** This effective division of labour in the ecclesia was formalised in the fourth-century legislative courts.


    Mercier, H., & Landemore, H. (2012). Reasoning is for arguing: Understanding the successes and failures of deliberation. Political Psychology, 33(2), 243-258.


  2. Keith,
    Are there new facts or analysis in this piece, or is it mostly the same as her 2012 paper “Deliberation in Classical Athens: Not Talking, But Thinking (and Voting)?”


  3. Terry,

    The original paper has been extensively developed (argumentatively and evidentially) and split into two, the present one destined for Classical Philology and the theoretical/deliberative one for The Journal of Political Philosophy. It’s benefited from criticisms from Seyla Benhabib, Daniel Betti, Paul Cammack, Marianne Constable, Stefan Eich, Amos Espeland, Conor Farrington, Bryan Garsten, Eugene Garver, Robert Goodin, David Grewal, Beth Janairo, Helene Landemore, Melissa Lane, David Langslow, Vanessa Lim, Jane Mansbridge, Josiah Ober, Gary Remer, David Runciman, Daniel Schillinger, George Scialabba, Peter Spiegler, Richard Tuck, Philippe Urfalino, several anonymous referees and audiences at the 2012 meeting of the Midwest Political Science Association, the 2013 meeting of the Northeastern Political Science Association, Harvard, Cambridge, Stanford, Yale, and Berkeley.

    In addition to its scholarly impact it should have a profound effect on how both deliberative democrats and sortition theorists formulate their practical proposals.


  4. Keith,
    This article doesn’t have an impact on me, because I was already convinced of the value of mute weighing of arguments before I read here earlier paper several years ago. However, I disagree that it contradicts the more modern research on the benefits of active deliberation within a minipublic for refining proposals in seeking win-win options, etc. (things like diversity, fact-finding, impartiality, etc.). Regular readers of this Blog already know my view that the mute listening to arguments and secret voting should be the final step of a multi-body sortition process that ALSO includes prior active deliberation by a separate mini-public (and prior to that, the generation of raw proposals by any and all who wish.)


  5. Terry,

    My point is purely regarding the provenance of the deliberative democracy movement. Some authors (including those who ought to know better) argue that they are recreating Athenian democracy and Cammack has shown this to be untrue. Whilst I’m sure your proposal has great normative value it has never been tried, so it remains purely that — an ideal type.


  6. *** A draft of the article may be found in Internet (draft May 9, 2018 on
    *** The article is interesting from a linguistic point of view, especially when the author underlines inaccurate translations.
    *** That the deliberations by assemblies of several thousand and juries of two hundred or more were not dialogical deliberations, that actually did not need linguistic demonstrations.
    *** Daniela Cammack ends the article by « Isocrates, Aeschines and Aristotle all likened the dêmos in democracies of their day to a collective monarch, and they were not wrong. Listening to speeches, considering the issues internally and voting en masse was exactly what deliberation—and thus, more importantly, rule—by an ancient Greek dêmos looked like. » That describes well the final step of an Athenian political process.
    *** But there are two flaws in Daniela Cammack’s study. I will consider here one : she forgets in the political processes the step of informal microdeliberations.
    *** This step is easily forgotten because it is undignified, it is scattered, it is not public, it is not easily studied (even by sociologically-minded political scientists in our societies). In the last French presidential election, I am convinced that it was especially important, because conservative-leaning electors who should vote Fillon could not stand (in families, with friends, in the canteens) the weight of others’ opinion about the « money-hungry » candidate. In ancient Greek Cities, with so important an Agora-life, and with citizens conscious of their political power, the informal microdeliberations step was of paramount importance. Let’s consider a proposal for a change in laws in the Second Athenian democracy. The proposal had to be published before the Monument for the Eponymous Heroes in the Agora. Afterwards it had to be read out in the Assembly. A following Assembly could decide to convene a legislative jury. Only later came the formal decisive deliberation which is the step Cammack is interested about. It is easy to see that the proposal had the time needed to be the subject of many informal microdeliberations. Especially in a simple and static society, where the data of the legislative deliberation did not evolve rapidly.
    *** We must not compare the situation with a contemporary criminal jury, who usually does not know anything of the case before the trial (and in some countries is forbidden to know).
    *** The ancient informal microdeliberations step had good sides ; especially jurors or assemblymen were acquainted with the issue, and not so easily carried by a good orator ; and the debates was less factional, with more mutual exchange and understanding between ordinary citizens. But there was a bad side : these (dialogical) microdeliberations did not follow good deliberation procedures, and isêgoria was not well guaranteed (it is a problem often underlined by Keith Sutherland). Socrates was sentenced by a jury, but very probably he had be previously sentenced by many informal microdeliberations as « the teacher of the bloody Tyrants Critias and Charmides », without in the dialogical microgroup any advocate of the philosopher. It was not easy task to convince the court jurors in such a situation.
    *** In a modern society, the informal microdeliberations will be always less important than in an Agora-centered ancient society ; and the Internet kind of microdeliberations may easily be ultra-factional. Therefore I think that at least an important element of modern ortho-democratic deliberation must be dialogical deliberation by small groups, but with procedures bringing a high level of isêgoria, including for instance injection of orator’s discourses and conversation with allotted samples of the social categories the subject especially impacts on.


  7. Big thanks to Andre, as always, for his thoughtful comments. Cammack’s paper is an exercise in Greek history — it’s relevance to modern proposals is my suggestion. Andre is right about the two principal differences (time and microdeliberations), and I agree that a modern orthodemocracy would need to introduce time gaps in between the formal deliberations. In 4th century Athens there were 7 stages in the legislative process and careful constitutional design would be needed to ensure something like this in modernity (Nadia Urbinati illustrates this in her consideration of Condorcet’s constitutional proposals).

    As for the need for microdeliberation, both the classical and modern examples indicate that this largely served to reinforce prejudice. The Athenian jurors might not have convicted Socrates if they had not already made their minds up before the trial; conservative-leaning French voters were swayed by (media-inspired) gossip regarding Fillon’s financial affairs (I assume by “stand”, he means “withstand”). But I don’t think the answer is dialogic deliberation by small groups, it’s the vital need to ensure representative isegoria. This is an extremely complex and messy task — it’s the longest chapter in my thesis and could arguably merit a full PhD. The protection of the interests of minorities is hard to ensure by democratic procedures (deliberative or otherwise), and would require constitutional safeguards.

    Look forward to hearing about the second “flaw” in Cammack’s study.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: