Mencken: The two kinds of democracy

H.L. Mencken‘s 1927 book Notes on Democracy is an interesting document. On the one hand it is a candid expression of a proud elitist worldview. Mencken spends considerable space explicitly denigrating the average person. In short:

There are men who are naturally intelligent and can learn, and there are men who are naturally stupid and cannot. (p. 17)

Such views cannot be expressed in polite society today, and although it is pretty clear that Mencken is aware that his stridency is politically incorrect, it is also pretty clear that he is expressing ideas that were acceptable, even conventional wisdom, in elite circles of his time.

On the other hand, Mencken devotes much attention to the problems of the electoral process as well (which he identifies with democracy). His anti-democratic attitude allows him to criticize the electoral system in a way that those with commitments either to the existing system or to democracy usually cannot afford. As Mencken damns voters for being stupid and electoral politicians for being scoundrels, Mencken points at several problematic fundamental characteristics of the system, belying his main thrust which focuses on personal characteristics. Here, for example, is the principle of distinction:

Democratic man is stupid, but he is not so stupid that he does not see the government as a group of men devoted to his exploitation that is, as a group external to his own group, and with antagonistic interests. (p. 197)

Mencken’s treatment of “direct democracy” – the standard remedy for the problems of the electoral system – is rather insightful:

A great deal of paper and ink has been wasted discussing the difference between representative government and direct democracy. The theme is a favourite one with university pundits, and also engages and enchants the stall-fed Rousseaus who arise intermittently in the cow States, and occasionally penetrate to Governors’ mansions and the United States Senate. It is generally held that representative government, as practically encountered in the world, is full of defects, some of them amounting to organic disease. Not only does it take the initiative in law-making out of the hands of the plain people and leave them only the function of referees; it also raises certain obvious obstacles to their free exercise of that function. Scattered as they are, and unorganized save in huge, unworkable groups, they are unable, it is argued, to formulate their virtuous desires quickly and clearly, or to bring to the resolution of vexed questions the full potency of their native sagacity. Worse, they find it difficult to enforce their decisions, even when they have decided. Every Liberal knows this sad story, and has shed tears telling it. The remedy he offers almost always consists of a resort to what he calls a purer democracy.

That is to say, he proposes to set up the recall, the initiative and referendum, or something else of the sort,and so convert the representative into a mere clerk or messenger. The final determination of all important public questions, he argues, ought to be in the hands of the voters themselves. They alone can muster enough wisdom for the business, and they alone are without guile. The cure for the evils of democracy is more democracy.


The truth is that the difference between representative democracy and direct democracy is a great deal less marked than political sentimentalists assume. Under both forms the sovereign mob must employ agents to execute its will, and in either case the agents may have ideas of their own, based upon interests of their own, and the means at hand to do and get what they will. Moreover, their very position gives them a power of influencing the electors that is far above that of any ordinary citizen: they become politicians ex officio, and usually end by selling such influence as remains after they have used all they need for their own ends. (pp 80-83)

13 Responses

  1. >Under both forms the sovereign mob must employ agents to execute its will, and in either case the agents may have ideas of their own, based upon interests of their own, and the means at hand to do and get what they will.

    By “agents” I presume Mencken is referring to government officials — possessing only a delegated mandate from the sovereign legislature. Why would that be any different in an allotted democracy (or any other form of government, for that matter). The only instance in which the problem does not occur is under full rotation — impossible in large modern states. It gets even worse if one considers allotted representatives as agents, as they certainly would have “ideas of their own, based on interests of their own, and the means at hand to do and get what they will.”

    So I’m not sure what the relevance of this passage is for those of us interested in sortition — Mencken would clearly have no time for it.


  2. This is my FB status update. I live in Wisconsin btw:

    “the stall-fed Rousseaus who arise intermittently in the cow States” says anti-democrat H.L.Mencken of our LaFollete in his 1927 book.
    Dem are fightin’ words Menckydenk. Lucky you’re already dead.


  3. Humans, exercising judgement. We know how elected representatives behave; why do we expect ‘the mob’ — a randomly selected jury to do any better?


  4. >why do we expect ‘the mob’ — a randomly selected jury to do any better?

    1. Because randomly-selected jurors are less likely to fall victim to partisan and factional forces.

    2. They will be obliged to listen to the arguments first before voting (in the UK most MPs simply turn up for the division).

    3. The principal of large numbers means that randomly-selected jurors are likely to be a lot more typical of average citizens than elected members, and as such will be better able to represent their interests.

    4. The cognitive diversity of an allotted group is more likely to arrive at the optimal outcome than a more homogeneous group, for all the reasons that Aristotle outlined.

    Note though that all these benefits depend on a) ensuring that the judgment of each juror is truly independent and b) restricting the jury to a judgment-only role. It also presupposes balanced information and advocacy.


  5. @conallboyle
    adding to the last comment:
    Moreover, random sampling avoids precisely the “mob rule” that you and Mencken fear. An appropriately sized group, at leisure to deliberate, free of outside pressure, will arrive at better outcomes.

    I disagree with the last comment in that I do not think that ensuring that each “juror” is independent etc…is necessary.

    In my view, the beauty of random sampling, why it’s used in statistics and elsewhere is that you do not need many checks to ensure representativeness. “Restricting” the assembly means no more than the way legislatures are already “restricted” by constitutions in almost every democratic government today.


  6. Ahmed, the Condorcet Jury Theorem requires independence if you are going to avoid groupthink and get the best epistemic outcome. Descriptive representativeness requires restriction to aggregate functions (otherwise the group will no longer be representative). To confirm this all you need to do is to refer to your own example (statistics).

    If you are not going to secure the epistemic benefits and are going to permit actions that destroy statistical representativity then what is the point of sortition?


  7. Keith,

    I am repeating myself, but there are two distinctly different benefits of sortition for decision making bodies that can’t coexist in the same body…You focus on the wisdom of crowds benefits of a descriptively representative body, where debate among members can cause social dynamics (status, elocution, etc.) to foil the wisdom of crowds. You know this, and there is lots of research about this that Yoram seems not to be concerned with. Thus you advocate voting by the allotted body without internal debate (speech-acts), as the Athenian courts did. But there is ANOTHER benefit that your scheme forfeits…and that is the benefits of SHARING latent information that is distributed across the population, where cognitive diversity allows a diverse group to make BETTER decisions than a group of experts could make (Page’s diversity trumps expertise theorem). Again, there is real-world research that shows this. This is one of the advantages of sortition that Yoram seeks to advance, that you, Keith, seem not to appreciate how you would LOSE under your scheme. This is why I favor a system that has at least TWO allotted bodies, one that utilizes each approach — a fully deliberative body that debates and crafts legislation, and a jury that passes judgment on the product of the first body.


  8. Terry,

    I agree with you 100%, and only wish I could put it as well as you do. My concern is how to retain the epistemic benefits of diverse input without sacrificing democratic equality and this would require an additional popular filter between the two groups (the stage that I’ve referred to as a votation referendum in which the options proposed by the first group are put to public vote, prior to deliberative scrutiny by the second group). If we take the Athenian parallel, this triadic process would correspond (roughly) to boule, ecclesia and nomothetai. Of course there are other ways of developing diverse policy proposals (I prefer citizen initiative and e-petitions) but I accept that these are easier to hijack than the first allotted chamber. My principal concern is that there should be a modern equivalent of stage 2, as the eccelesia was always seen as the quintessence of Athenian democracy.

    I can’t tell you how reassuring it is to have at least ONE person (other than Helene Landemore) who understands, and is sympathetic to, my concerns, as I often feel that I’m banging my head against a brick wall. You don’t agree with my argument but at least you understand it and have the good grace to not question my motives in raising it. Helene also disagrees with me but she admits that is because her concerns are epistemic, rather than democratic.

    I do wonder, though, why you still feel the need for sortition as the only principle, rather than some kind of hybrid model that would include direct democracy, sortition and (perhaps) even elections as a way of retaining an element of retroactive accountability. I would be happy with policy generation by a wide variety of sources, including an allotted proposing body, popular initiative and election. The important thing is that the multitude of options would be whittled down at the votation stage. I accept that all forms of referendum are open to manipulation by the “rich-‘n-powerful” but can’t see any other way of retaining the right that we ALL have to say how we wish to be governed (apologies to Ahmed for using the g word).


  9. Keith, I suppose it all depends upon the respect in which people are supposed to be equal before a system counts as democratic. As you know, that’s a point that gets argued incessantly. I’m personally skeptical that one can articulate democratic equality at the level of institutional features, rather as a system of rights more generally. That’s probably very vague, and I haven’t had my coffee this morning yet, but what I mean is that democratic equality requires things like rights of access to the levers of power, rights to petition to have laws changed, etc. That’s very different from the demand that certain decision-making bodies select everyone with equal probability, or that they include all segments of society.


  10. Keith,

    I think that raw proposals should be able to come from anybody, or group. All should have equal right and opportunity to initiate policy proposals (laws). And at the END of the process we want a representative jury to make final decisions. The MIDDLE filtering stage is CRUCIAL (and this is where Yoram gets extremely frustrated with you). I agree with Yoram that an allotted body should perform this winnowing process to protect against elite manipulation. The elite can manipulate an election or votation process much more easily than they can an allotted body (though they will certainly try). We want a body that has a built-in resistance to corruption and power-hoarding. I think sortition is the perfect tool for this…utilizing its anti-corruption properties as much, or more than its descriptive representation qualities for the filtering function.


  11. Peter,

    Agree completely regarding the limited role of sortition in instituting democratic equality. Apart from determining the outcome of a debate by voting it’s hard to see how sortition can contribute much to the other functions that you mention. Everyone should have the right to petition to have laws changed, not just those chosen by lot.


    We could argue until the year dot regarding the merits of lottery-appointed bodies doing everything in politics, but do you seriously think there is any chance that such a system would gain approval by either the existing elite or “the people”? I’m a practical person and would suggest an incremental approach and where better to start than extending the analogy of the trial jury to determining the outcome of legislative “trials”. There is good historical precedent and people can easily understand the possibility of extending the jury system. Why make proposals that have a zero probability of being accepted?


  12. The tactics of social change is important, but it is also important to have a notion about a long-term goal. It may be that the transition to sortition might first occur within a narrow functional area (like zoning or healthcare), where the public might be particularly upset with the status quo, or where the elite might be less defensive. Designing a complete (ideal) system with decision-making authority within a narrow area might be a better approach than, for example giving a merely advisory role to a single allotted body at a national legislative level. On EqualityByLot, it seems to me, we are generally more focused on discussing the ideal model, than discussing tactical considerations. Both tasks are important, but we shouldn’t shortchange the “vision thing” allowing current tactical thinking to shut out big ideas.


  13. Internet forums are open access and if the elites are informed in advance that we propose to put them out of business then we might anticipate some opposition. Much better, IMO, to focus on something we know sortition can do rather than speculate on the possibility of it doing everything. Once that’s established then by all means let’s indulge in speculations as to how best to establish heaven on earth, but in the short-term it’s entirely counterproductive — especially as it leads to the impression that there is far more that divides us than unites us. I’m reminded of the factional divisions within the left during the 1970’s and 1980’s.


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 )

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: