Epistemic Deference and Epistemic Arguments for Conservatism I

A couple of months ago, there was a very interesting debate between Jim Manzi and others on the question of whether social science knowledge was sufficiently well established to warrant radical policy innovations. Manzi argued in a widely blogged article that the methodological limitations of the social sciences – in particular, the impossibility of appropriately measuring the counterfactuals of policy interventions through, for example, randomized experiments – implied that the findings of social science should normally not “outweigh settled practice and even common sense” when we are thinking about changing existing practices or institutions. His critics argued, among other things, that social science can provide evidence that overturns such settled practices, since, after all, the status quo is typically the result of power struggles, not rational discussion; why should we give it more credence than the findings of social science, which emerge from a process that at least approximates rational debate?

We might put the general point thus: Manzi thinks that, where the complexity of social reality makes full scientific knowledge all but impossible, it makes better sense to give “epistemic deference” to settled practice; his critics think that the practices of “social science” as a whole (not necessarily a single random study) often have more “epistemic authority” than settled practice. This sort of division about the locus of epistemic authority also maps into a political division between “conservative” and “non-conservative” positions, with the “conservatives” granting some settled practices or institutions the bulk of epistemic authority, while the “non-conservative” position insists that some other person or institution, usually of more recent vintage, should be given greater epistemic deference than settled practice. Is there any way of adjudicating under what conditions are these positions likely to be right?

The conservative position is typically defended by arguments about the “limits of reason.” The classic statement is in Burke’s Reflections on the Revolution in France:

We are afraid to put men to live and trade each on his own private stock of reason, because we suspect that this stock in each man is small, and that the individuals would do better to avail themselves of the general bank and capital of nations and of ages. Many of our men of speculation, instead of exploding general prejudices, employ their sagacity to discover the latent wisdom which prevails in them. If they find what they seek, and they seldom fail, they think it more wise to continue the prejudice, with the reason involved, than to cast away the coat of prejudice and to leave nothing but the naked reason; because prejudice, with its reason, has a motive to give action to that reason, and an affection which will give it permanence. Prejudice is of ready application in the emergency; it previously engages the mind in a steady course of wisdom and virtue and does not leave the man hesitating in the moment of decision skeptical, puzzled, and unresolved. Prejudice renders a man's virtue his habit, and not a series of unconnected acts. Through just prejudice, his duty becomes a part of his nature.

As Burke puts it, the comparison makes sense; individual reason (or, for that matter, the individual social science study) is highly limited in its epistemic power in comparison to settled social practice. There is typically some reasonable basis for even highly perplexing social practices; and individual reason is likely to be highly misleading in many circumstances. Individually, we suffer from so many cognitive biases and defects that it is a wonder we get up in the morning; and even highly trained experts are often wrong, even in their own fields. (I read somewhere that most peer-reviewed research is eventually overturned). 


But as an argument the comparison is flawed; the relevant comparison should not be that between settled practice and individual reason, but between settled practice and some alternative social practice (e.g., the social practice of science, with its various self-correction mechanisms), or between settled practice and some other collective judgment (e.g., the collective judgment of an assembly or a market). Moreover, the “general bank and capital of ages” is often a “depreciating asset,” as Adrian Vermeule puts it, especially insofar as later people make no contributions to it. From this point of view, Burkean concerns about the limits of individual reason cut much less deeply; the question is about the relative “epistemic power” of different social practices, institutions, or processes. 

Yet I think there is often something to these “epistemic” arguments for conservatism. In later posts, I want to try to investigate under what conditions epistemic arguments for conservatism make sense. (This blog is mostly my scratch research notes, so read at your own risk). As a starting point, I think we might divide these arguments into three kinds: selection arguments (where the epistemic power of settled social practices comes from the operation of a selection filter that is not found in alternative social practices), computational arguments (where the epistemic power of settled social practices comes from the way in which they organize expertise and information), and resilience arguments (where settled social practices are found to be preferable to alternatives because they are adapted to a wide variety of circumstances, even if not optimal for any particular one of them). (Resilience arguments are really a subtype of selection arguments, but I want to think about them separately). More on each of these later.


[update 2:35: edited a few words for clarity].

Footnotes on things I've been reading: Adrian Vermeule's "Law and The Limits of Reason"

Adrian Vermeule, Law and the Limits of Reason (2009).

I read this book for somewhat hazy reasons; perhaps I found the title in a footnote of some paper and it seemed interesting, perhaps I ran across it by doing a search in our library catalog for something else and remembered that someone had once recommended  to me another one of Vermeule’s books. But at any rate, I’m glad I did read it, even though the main topic of the book is something that would normally not interest me very much.

The book is basically an attack on the “constitutional” common law, written primarily for (USA) legal scholars. As I reconstruct it, the basic problem is as follows. 200 and some years ago a few very smart people wrote a constitution for a country. (We’ll leave the name of the country blank, since the problem is pretty general; and I want to move away from any particularities of the American situation). Conditions at the time were very different from conditions today, however; so the problem arises as to whether, and how, to adapt the constitution to current circumstances. Should we change the constitution, and if so, who should change it?

At one extreme we can find the pure originalist position: the constitution should never be changed by anyone. This position is not as crazy as it may look at first glance. Assume that the constitution has survived intact for 200 years (and it has not been ignored during this period). It may then be that the very fact that the constitution has enabled the country to survive for 200 years implies that current constitutional rules are “resilient” – that is, they are broadly adapted to all kinds of circumstances, even if not optimal for any of them. They have been tested by time, with its varied challenges, and come out relatively well. Though different constitutional rules might have produced better outcomes at some points during those 200 years, no other set of rules we could design today would have plausibly produced a better outcome overall. We could perhaps temporarily achieve better outcomes (however defined) by adjusting the rules today, but then we could not be sure that the new rules would be optimal for other, as yet unforeseen circumstances, and we would have to adjust them again and again. And given some assumptions about the likelihood and costs of error, the relative wisdom of the framers versus our current wisdom, and our estimates of the rate of environmental change, it might be that it would never be worth our while to try to change these constitutional rules. We can think of this argument in terms of a trade-off between the benefits of adaptation to current circumstances and the benefits of resilience in changing conditions, given the limits of reason. If our estimates of the benefits of adaptation are low enough, of the resilience of current rules high enough, and of the uncertainty involved in formulating “perfect” rules for current circumstances high enough, perfect conservatism is entirely reasonable. (The earliest version of this argument I know of goes back to Plato’s Statesman, which I discuss extensively here. We also see something like this trade-off in nature: an organism can be too well adapted to its environment to survive drastic changes there. “Generalists” may thrive in changing circumstances, while “specialists” die since they cannot adapt quickly enough to these changes).

But suppose that we do not agree with the pure originalist position. (Few people do). The framers may have been very wise, but they did not foresee everything; and we may have a clearly good idea about how to improve existing constitutional rules (e.g., emancipating the slaves). But the problem still remains, since not every potential change is good: when should constitutional rules be changed, and by whom? The constitutional common lawyers argue that the constitution should (normally) not be changed by legislative assemblies or the “explicit” amendment mechanisms indicated in the constitution itself; instead, it should be changed gradually via the decisions of judges, just as proponents of the “regular” common law argue more generally that law should be changed not via legislative codification but rather through the gradual decisions of judges. Common-law constitutionalism thus tends to be associated with arguments for robust forms of judicial review and a general conservatism on matters of constitutional change, the key assumption being that the limits of reason are less binding on judges than on legislatures or the public, though still quite important.

Vermeule subjects this position to withering (though sometimes repetitive) scrutiny, arguing instead for a position that might be called “Benthamite” or “Thayerian” (as he labels it). On this view, legislatures face fewer epistemic constraints than courts in deciding to adapt constitutional rules to current circumstances, and hence courts should in general defer to them. (Note that this argument does not say much about incentives: it may be that legislatures face fewer epistemic constraints than judges in constitutional matters, but have worse incentives, and it is not clear which constraints would be “binding” in that situation). I am less interested in his ultimate conclusions, however, which are a bit too US-centric (including a plea for including at least one non-lawyer in the Supreme Court for interesting reasons having to do with the virtues of diversity in deliberative groups), than on his general approach to the problem of the limits of reason in politics and law, which is all about trying to figure out how "collective" mechanisms affect those limits.

Vermeule identifies two basic “epistemic” arguments for the constitutional common law position. On one view, the constitutional common law is better than the constitutional law created by legislatures or other explicit mechanisms because it combines the judgments of many judges over time to produce a judgment that is somehow better than the judgments of any individual judge (for reasons having to do with the Condorcet jury theorem, for example, though other mechanisms might be imagined). We can call this the “aggregation” argument. On the second view, which is associated with Hayek and his followers, the constitutional common law is better than the “explicit” constitutional law of legislatures because the process through which it is produced includes a selection “filter” that weeds out bad rules (the litigation and appeals process, for example) in ways that are not possible for explicitly created constitutional rules. We can call this the “selection” argument.

I am not going to discuss in detail all the considerations Vermeule deploys against these arguments. At any rate, it seems to me that the book is written very much like a lawyer’s brief: the constitutional common-law position is attacked from every possible perspective, some of which are less convincing than others. But a couple of points he makes are quite interesting. First, he argues that the “aggregation” arguments tend to suffer from a “nirvana” fallacy, where the benefits of aggregation are assumed to apply only to judge-made law. But in fact laws that are made by legislatures with many people should also benefit from considerations about aggregation (at least assuming equal incentives for information acquisition and sincere voting, which Vermeule discusses in some detail, though not always satisfactorily). And the kind of aggregation over time postulated by common-law constitutionalists may also suffer from informational cascades and epistemic “free-riding”: judges that defer to past decisions at a later time do not add any new information to the law, leaving the first few judges as the real legislators, for example. (“Epistemic free-riding” may also be a problem in legislatures, especially if the legislature is large, since then incentives for information acquisition are smaller. But this is more or less balanced – under some conditions, at least – by the greater epistemic diversity of legislatures, which should improve the quality of their decisions). Second, drawing on some interesting-looking work by Andrei Schleifer, he suggests that there is no plausible “selection” mechanism for the constitutional common law (though he is agnostic on the regular common law) that ensures bad rules get weeded out. Taken together, he is basically arguing that the limits of reason bind judges at least as much, if not more than, as they bind legislators; there are no special advantages to common-law constitutionalism over other forms of constitutionalism, and many apparent disadvantages.

Anyway, this is probably not a work for everyone (even if it is not particularly difficult to read); but I’m rather interested in “epistemic” arguments for conservatism, and this book offered a good overview of how Burkean arguments about the “wisdom of the past” and Hayekian arguments about the wisdom inherent in decentralized systems can be used or misused. 

On the flu, and a tale of two earthquakes

Apologies to my (very few) readers for the lack of blogging. Last week we got the flu - first my daughter, then me, then my wife - and we have been basically out of commission for 10 days (only now getting back to normal). Some of the flu viruses are wily antagonists, cleverly hijacking the body's cellular machinery to replicate themselves prolifically, and it seems to take a lot of the body's considerable resources to get rid of them (with much collateral damage in the form of large amounts of inflammatory cytokines, it has to be said).

In the meantime, there was a large earthquake in Christchurch, about 400km from where we live: 7.1 magnitude in the Richter scale. (It probably was what woke me up at 4:30am on Saturday morning, though I didn't consciously feel anything and in my state at the time it could have been anything). We don't know many people in Christchurch, but thankfully there seem to have been few casualties and no deaths as a direct result of the quake. Lots of damage, yes, but even then few building collapses. Yet compare this with the outcome of Haiti's January earthquake in Léogâne, measuring 7.0 magnitude in the Richter scale (a bit less than the Christchurch one, and also centered close to the surface): the estimates for deaths directly attributable to the quake range from 92,000 to 230,000 people; hundreds of thousands of people remain living in tent camps; the capital city remains shattered. Why the difference?

The time of the day the shaking happened may have made a difference (at least according to the useful summary in the wikipedia page on the Christchurch earthquake, which was up a day after the event). Yet I'm not sure I can see a clear causal story for how this is supposed to work: in Christchurch the earthquake happened when most people were sleeping inside buildings, in Haiti when most people were out and about; you would think the wee hours of the morning, when lots of people are basically unconscious under potentially falling objects, are the most dangerous time for an earthquake, but apparently not. Anybody care to enlighten me?

There is also the fact that most houses in New Zealand are timber framed, which means they bend a bit with the shaking, rather than quickly breaking or pancaking on top of their occupants, whereas most houses in Haiti are not, and that New Zealand has a strong building code that is actually enforced. But ultimately the factors that make the biggest difference seem to be the level of development and the quality of institutions of accountability. From the abstract of a recent (2005) paper by Matthew Kahn (gated):

Using a new data set on annual deaths from disasters in 73 nations from 1980 to 2002, this paper tests several hypotheses concerning natural-disaster mitigation. Though richer nations do not experience fewer natural disasters than poorer nations, richer nations do suffer less death from disaster. Economic development provides implicit insurance against nature's shocks. Democracies and nations with higher-quality institutions suffer less death from natural disaster. Because climate change is expected to increase the frequency of natural disasters such as floods, these results have implications for the incidence of global warming.

Higher levels of economic development and higher quality institutions make it possible to draft and enforce building codes in a relatively uncorrupt way, for example. And good institutions of accountability put the fear of their constituents into politicians, who may become very responsive to their needs in the aftermath of a disaster. Furthermore, in open societies information flows relatively freely and in generally useful ways: consider how Twitter was used to quickly aggregate information - and some misinformation - about the quake soon afterwards, particularly through the hashtag #eqnz; it's almost enough to make me rethink my principled avoidance of the service. Of course, this is itself only possible because the Internet backbone links and the electrical power supplies were not too badly damaged or were quickly repaired, and worked fine to begin with, something that economic development itself makes possible. It's another case of the unfairness of country borders: nature is harsher when you are born in a poor country (or in a non-democratic, not very accountable regime). By the same token, the best insurance against the destructiveness of nature seems to be economic development and democracy.

Using CiteULike for Teaching

I recently rediscovered CiteULike, a free web service for bibliography management. I know I had used it in the past (I still had an account, though I had not logged in for four years) but stopped using it in favor of Zotero and Endnote (because of the latter's integration with Word). But then I stopped using Zotero, partly because it tied me to Firefox (I now use Chrome almost exclusively, and love it), and partly because it was often slow and occasionally crashed (I think it's better now - it even has word processor integration! - but I had one too many bad experiences).

Anyway, there's no software to install with CiteULike and no browser dependencies - it's just a bookmarklet - and you can easily transfer your data to and from Endnote and BibTex. (Incidentally, LaTex is great for Greek-heavy writing on Ancient philosophy with one of the Betacode packages. It's too bad most of the places I submit my work to tend to demand Word files, and that converting a file from LaTex to Word can be such a pain). But it also occurs to me that something like CiteULike would be great for teaching as well.

I typically maintain a large bibliography of recommended works and suggested readings for the essays in my courses, and it's a bit of a pain to add new entries to it and make sure the new entries appear in all the appropriate places and are appropriately organized and formatted (partly this is just the fault of the Blackboard course management system we use at Victoria - which, despite some good points, is still a pretty cumbersome piece of software). So, for example, for my POLS 209 course on Dictatorships and Revolutions I might have some recommended readings on Venezuelan politics (which I suggest the students consult before writing an essay on the relevant topic in the course), or a set of readings on the comparative economic performance of dictatorships vs democracies (for students who are writing on the topic, or might want to learn more). But every time I need to add a new reading, I have to change things in four or five different places, and manually add links and other things. Moreover, though students can add items to the general bibliography (it's maintained in an internal wiki - though the Blackboard wiki tool is terrible) they may add things formatted in odd ways, or in the wrong categories, and they cannot easily download the sources to a bibliography manager.

But with CiteULike I can maintain the entire course bibliography in one place, and then provide links to specific topics using tags (you can even subscribe to the whole thing or to specific tags using RSS, if you want to keep up with the latest additions). Readings on Venezuelan politics? Here! Or I might tag some readings as recommended for the first week of the course, or some particular lecture, and post the link in the course outline and in the appropriate place in Blackboard. And students can of course download the citations to the citation manager of their choice, click on the handy links to read the actual articles (the point of the exercise, after all), or simply copy and paste the plain text of the citations they used to their bibliographies when they finish writing their essays (too much student time is, I think, spent worrying on minor details of bibliography formatting rather than actually reading the sources).

Moreover, students who create an account with CiteULike can easily contribute readings to the common course pool, perhaps with their own review notes (I can always tag them appropriately); they just have to join the class group. (I already incentivize this sort of thing in the course outline - adding to the course bibliography in appropriate ways, or contributing to the course blog/discussion board/wiki, are all things that help students get points in my courses). In fact, that's the part I'm most interested in: I keep trying to foster collaborative research in the classroom, and if students are easily able to contribute sources to the common pool of readings, they can basically help each other in constructive ways. And they can help me as well - I have found some very interesting sources for the course (even whole new literatures) in essays that students have written in the past - sources which sometimes end up in the course bibliography (you really do learn lots from your students).

I am actually considering requiring my honours students (in POLS 401, a course on contemporary political thought) to actually create a CiteULike account and contribute readings to it periodically, perhaps with some review notes. (I tried something like this once with a wiki, but the wiki is too cumbersome for this). To be sure, there's some upfront investment on their part (I probably will have to demonstrate how to use the software in the first few classes, or write a detailed handout with an explanation and pictures) but once they learn, I imagine it could be helpful to them (both in this and other courses).

Has anybody else done similar things? What potential problems am I missing?

Sunday Extremophile Blogging: The ARMAN and Thermoplasmatales archaea

ARMAN archaea, orange circles, with an arcaheon of the thermoplasma order, below them. The small orange particles are viruses. Nobody knows what the needle-like things do. Credit: Luis R. Cronolli/LBNL via ScienceDaily 

The ARMAN archaeon is a mysterious one (mentioned in the previous instalment of Sunday extremophiles). For one thing, it is found only in highly acidic pools of an abandoned mine in Northern California (pH < 1.5 or so; the water there is more like concentrated sulphuric acid. Related organisms are found in other acidic pools, according to Wikipedia), and it has one of the smallest genomes of all free-living organisms (that is, not parasites). Indeed, it is so small that it seems to be at the size limit for metabolic, non-parasitic life (some viruses -no metabolism, so no "life" in a sense- are bigger), with a correspondingly tiny genome (1 million base pairs, compared to hundreds of billions for complex organisms) and few ribosomes. And yet of this small genome (1 million base pairs), about 45% seems to be genes that are not found anywhere else. ARMAN could thus be the "minimal" metabolic organism - with the smallest set of genes necessary for metabolism, though it apparently interacts in some unknown way with other acidophilic archaea, perhaps to get some extra nutrition (the thermoplasmatales, among which are found the most acidophilic organisms known, thriving in pH 0.06 environments). 

I find these weird niches - far removed from human possibility - really fascinating. Are there any limits to the environments in which life can thrive, ultimately? ARMAN lives in an environment extremely hostile to most forms of life - yet even there, it is not alone (in fact, it is possible that life started in similar environments, so perhaps ARMAN's lineage is very old). Endless forms most beautiful indeed.

A Puzzle about Character and Democracy

Elections in modern democracies function as mechanisms of both selection and discipline. With a multiplicity of candidates competing for office, elections should in theory enable voters to select the person who they believe would do the best job of governing them (e.g., the most virtuous candidate), and to discipline incumbent candidates who have failed to do what they want by not re-electing them (assuming re-election is possible), instead electing someone else. Ideally, these two functions of elections would complement each other: voters would select candidates whose characters they trust and whose promises can be determined to be credible, and then credibly discipline them if they do not perform appropriately.

We have some evidence that elections as "disciplinary" institutions really do produce better performance from political leaders, at least in purely economic terms: politicians who are in danger of losing their jobs unless they deliver results that are satisfactory to a large number of people tend to deliver policies that are at the very least not very bad (see, for starters, here and here, though the literature, both theoretical and empirical, is vast, and there are some circumstances under which elections appear to worsen economic performance). Indeed, the literature on the comparative economic performance of democracies and non-democracies typically shows that democracies are at least no worse, and often better, than non-democracies in this respect, contrary to some popular beliefs (as Dani Rodrik notes). I like to show the students in my dictatorships and revolutions class this density plot to illustrate the point:

(Solid line for democratic country-years, dotted line for non-democratic country-years. The data on democracy is taken from the Polity project; we take democratic years for a given country as those where the country scores more than zero in the Polity autocracy+democracy measure, which is only barely democratic, given the coding criteria, but indicates real electoral competition. The data on growth come from the Penn World Tables. The period covered is 1950-2007, with data for most countries with population over 500,000 people. Basically, the plot says that "autocratic" countries with less political competition tended to have a smaller median growth rate but a higher variance in that rate over this period – more miraculous years but also more disastrous years).

Yet politicians as a class are not typically held in high esteem: most people in democracies do not seem to think that politicians are people of good moral character when considered as a group. The complaint about the need for "better" leaders, and the disappointment over actual leaders, seem nearly universal (and go back a long way, to Plato: typically, the praise of the political class always applies to the past, to the ancestors who were better, as in Cicero). I admit, the evidence I have for this proposition is fragmentary, but consider for example this Gallup poll question on how US residents evaluate the "honesty and ethical standards" of different professions:

Though the low ranking of senators and members of congress in this figure is presumably driven in part by bad economic news in late 2009, I do not think it is very atypical. Other polling data also tends to show voters in the US being much more dissatisfied with Congress as a whole than with their own congressman or senator (who tend to be re-elected at an astonishing rate), which suggests that they have a low opinion of the political class as a whole even though they may have a good opinion of their own representative. And the pattern in other countries seems similar: in Italy, politicians are routinely referred to as a "caste" who enjoy immense and undeserved privileges, even as some of them (including, inexplicably, Berlusconi) remain relatively popular; complaints about the quality of the political class in India (where many members of the Lok Sabha have criminal convictions) are extremely common; and distrust of politicians in Venezuela (my native country) seems to be quite general, even if some politicians (like Chavez) are very popular. (I should look up NZ data, but I'm too lazy for that right now). I am sure most people have heard or expressed views about how most politicians are "crooks," and not only the politicians of the "other side" (whatever side one does not support) but also those of one's own side – how they are spineless, too concerned with their electoral fortunes and their image, hypocrites, quarrel too much, etc. When was the last time you heard people praise the overall character of politicians?

Yet this is somewhat puzzling. Voters presumably prefer higher-quality characters among politicians: so why are "virtuous" politicians not being elected, or, if they are being elected, why is this fact not being recognized among the electorate? (Assuming, of course, that the evidence of voter discontent with the political class noted above holds more generally, and reflects a more or less accurate judgment of the character of politicians; it could also be that voters are simply mistaken about the character of politicians as a whole, though that fact would itself demand explanation). Moreover, it's not as if politicians are a low-status group of people; buildings are named after them, resources are expended on ceremonies for them, we address them with honorific titles, and so on, but this makes the disparagement of their character as a class even more puzzling, since their high status presumably biases many judgments of their character upwards rather than downwards, especially for those politicians who are commonly thought to be successful. Politicians are both the occasional subjects of hagiographies and yet are held to have bad moral character when considered as a group.

More precisely stated, the puzzle is that given an implicit preference for high-quality politicians (in terms of character), and a competitive electoral system, voters do not elect more politicians they think actually have high-quality characters, i.e., "virtuous" politicians (for some plausible conception of virtue). What could account for this?

I suspect that there are two basic answers to this question, though they are not necessarily incompatible. The first, and most obvious, blames supply problems. Perhaps there are extremely few virtuous politicians out there, and they do not want to rule (though they occasionally rule; hence the hagiographies). This is the Socratic argument in the Republic; hence philosopher kings must be forced to rule, since they would not naturally want to. Stated differently, perhaps the rewards of political office are more appealing to the non-virtuous than to the virtuous (given some plausible definition of virtue), leading to a pool of candidates heavily weighted towards the non-virtuous (though wouldn't one expect that the virtuous, who would want to help their fellows, would also disproportionately run for office?). Or perhaps the typical political environment of modern democracies disadvantages the virtuous vis à vis the non-virtuous. This is the Machiavellian position, suitably adapted to the 21^st century: if those campaigners capable of using virtuous and vicious tactics (e.g., dishonest lying, duplicity, etc.) are more successful than those who are only capable of using virtuous tactics, then the former will tend to be elected more often than the latter. This answer leads to a different sort of position, however, which attributes the problem less to the actual supply of virtuous politicians than to the epistemic constraints on voters who must distinguish between virtuous and non-virtuous politicians.

From this point of view, there are also a number of possibilities. A Platonic answer is that voters cannot distinguish between virtuous and non-virtuous politicians because they do not know what virtue is; they elect the non-virtuous while believing that they are electing the virtuous, and the non-virtuous do not even have to deceive them about this fact very much. But this answer does not work for us right now, since it implies that voters would be satisfied with the character of politicians they actually elect, but they are not.

Perhaps the problem then is that the epistemic demands of "selection" are higher than those of "discipline." In order to discipline a politician, the voter only needs to know his own situation: if he or she thinks that his life is going well, he can vote for the incumbent, whereas if he does not think so, he can vote against the incumbent (contrast with ancient methods of discipline, which, given term limits, were always quasi-judicial and placed more stringent epistemic demands on citizens). Even if the voter substantially misallocates responsibility for his or her material situation to the political class (perhaps the country's economic performance is substantially affected by factors not under the control of the political class), the incentives created by this sort of behaviour might still induce politicians to, on average and over the long run, provide relatively good policy (or at least operate as a kind of selection filter, so that even if politicians provide policy more or less at random, only relatively successful policies get rewarded). 


By contrast, in order to select a politician properly, the voter needs a lot of information that is not readily available, or that is easily subject to falsification or concealment (by the candidates themselves, among others), especially in large-scale societies where voters cannot know their leaders on a personal basis. (Thus Aristotle's distrust of large political communities, where he thinks selection institutions cannot work well because citizens can never really know each other's characters). 


More specifically, if the pool of candidates includes more non-virtuous than virtuous candidates, but voters cannot easily distinguish between them, they will tend to select the non-virtuous at a higher rate. Since, moreover, when the voter decides whether to punish the incumbent he or she has to also decide whether to take a risk on the challenger's character being bad, the voter might decide to live with leaders with bad characters but who deliver acceptable performance rather than take a chance on untested candidates who may both have bad character and deliver bad performance, knowing that his or her ability to distinguish between good and bad characters is not very good. From this point of view, disciplinary concerns about performance would normally dominate selection concerns about character, and voters would learn to live with rulers whose character they despise (at least in the aggregate), even though they care about both character and performance. (I think there arguments along these lines, though more technically expressed, in Tim Besley's Principled Agents. I really need to read that book and learn some game theory). However, this seems not to account for the phenomenon "congress is bad but my congressman is great" that we sometimes see in the US.

Two other possibilities that seem to me to be less plausible are the following. One is that character is simply too malleable. So voters elect people of good character, but such people quickly become bad upon gaining political power. Voters are thus constantly disappointed: they think they are selecting the best leaders (and in fact, they do) but these people easily become corrupted. I suspect, however, that we should see more incumbent turnover under this hypothesis than we actually see: people would learn that politicians quickly become corrupt, and hence that the best way of dealing with the problem is to turn the out of office quickly, or to impose term limits (as a kind of paternalistic policy, preventing themselves from being tempted to re-elect corrupt people simply because they provide good performance), though perhaps voters evaluate a tradeoff between competence and corruption potential and decide that good performance is worth a certain amount of corruption, as in the previous argument (except that voters in this model are always able to distinguish the corrupt from the virtuous: the difference is that the virtuous turn corrupt at a high rate). 


Finally, I suppose that the puzzle could simply be attributed to partisan biases. On this view, disagreements about policy would be easily translated into attributions of character: only people with bad characters could support the policies the opposition supports! Since even in two party systems around half the population supports positions one does not support, and elects politicians accordingly, this might result in the feeling that most of the political class is worthless. But this theory would seem to imply that politicians would be more despised in multiparty systems than in two-party systems; though I have no data on this question, I suspect this is not the case. Moreover, people often despise politicians even from their own party; they like their policies, but dislike their characters.

Other possibilities? Perhaps the disparagement of the characters of politicians results from human ambivalence about domination? What other ideas are there? (Or perhaps this is a pseudo-puzzle?)

Pet peeves: the neglect of ancient concepts, a continuing series

(Warning: some complaints about the neglect of ancient concepts by classical scholars). 


I just read a nice piece by W. Jeffrey Tatum on "Roman Democracy?"  in the Blackwell Companion to Greek and Roman Political Thought that makes a good case for taking the "democratic" features of the late republican Roman political system seriously, contrary to the common opinion among scholars since the late 19th century that republican Rome was essentially a senatorial oligarchy. 


Tatum sensibly notes that Rome was not a full democracy in the strong sense of the term (pace Fergus Millar), yet he also seems needlessly dismissive of Polybius' and Cicero's judgment that the Roman constitution was "mixed" (pp. 215, 223). Rather, he seems to want to classify Rome as some kind of imperfect democracy, and in the process of casting about for a suitable concept ends up suggesting that Rome was a "delegative democracy" (p. 226). This is a term invented by my old teacher Guillermo O'Donnell to describe political systems like that of Argentina under Menem or Peru under Fujimori, but it is wholly unsuitable for describing the late Roman republic: a "delegative democracy" is a democracy without mechanisms of horizontal accountability, and the Roman republic certainly did not lack that! 


Why not simply say that the best way of describing the Roman political system is through something like the Polybian concept of the "mixed" constitution, or alternatively through the concept of a "hybrid" regime, to use contemporary terminology (though the concept of a hybrid regime today is much less well developed and lacks the normative associations of the idea of the mixed constitution)? The evidence Tatum cites shows quite well that the Roman regime fits the basic ancient criteria for classification as a mixed constitution. It involved a number of distinct centers of power with with complementary but also competing interests (the tribunes, whose power derived from their connection with the popular assemblies and their ability to veto senatorial proposals; the senate, which was the executive committee of the Roman upper classes; and the consuls, whose power derived from their control of military forces in the field) which were not always equally balanced against each other, to be sure, but then again the idea of the mixed constitution did not require that its constituent parts be fully balanced. And it is true that the idea of the mixed constitution had a large normative baggage and implied a certain theory about political stability (which I am currently exploring in this project), but that is no reason to abandon it entirely. 


Aside from concerns about the overwhelming dominance of the senate (which are basically refuted by Tatum), scholars tend to argue that Rome could not be a "mixed" constitution in the Polybian sense because the consuls (the "monarchical" element in the constitution, in the Polybian scheme) were senators and returned to that body, representing senatorial interests all the time. But it seems reasonably clear that in Rome the consuls' power derived not just from the senate, but from their control of armed force and their ability to extract resources from the provinces (as proconsuls after their terms); so Polybius' characterization of the consuls as the "monarchical" center of power (cf. πολιτευμάτων, Histories 6.10.6, which it seems to me could be translated as such) does not seem far-fetched. 


So why the dismissive attitude towards the ancient conceptual apparatus? Rather than throw it out, why not develop it further and see whether it can be adapted to modern conditions? (Another example in the same book: the piece by Forsdyke on the idea of tyranny in classical thought. She seems to suggest that the notion of tyranny in the 5th and 4th centuries was simply a tool in the ideological struggle between the upper and lower classes in the polis, rather than an attempt to actually grasp a real phenomenon, however imperfectly, though the last lines of the piece soften the impression a bit). 

The Class Struggle in Classical Greece

From Aristotle, Politics V.1310a9-11:

at present in some oligarchies they [the oligarchs] swear "And I will be hostile to the people and plan whatever evil I can against them"

Sunday Strange Life Blogging

Not exactly extremophiles tonight. But I found a really neat article about a lovely sea slug, Elysia Chlorotica, which "steals" the chloroplasts of certain algae and derives nutrition from their photosynthesis: in a sense, an animal turns into a plant (picture above, from a neat blog called "Small Things Considered", which is the blog of the American Society for Microbiology; picture source here).

Also via Small Things Considered, a real extremophile, an archaeon that lives in water so hot and acidic it can melt metal rivets (probably worthy of another Sunday extremophile blogging), and the mind-boggling sex lives of mushrooms. The diversity of life never ceases to amaze me.

The most surprising sentence I read today

...while philosophers mandated self-control for kings in regard to both alcohol and sex, the Antigonid king Demetrius the Besieger appropriated the Parthenon itself for the use of his personal harem (ca. 290 BC), and the lead float in the great procession of Ptolemy II in Alexandria in 271/270 was a penis 150ft long with a 20-foot star coming out of its tip.

From a piece by Arthur M. Eckstein on "Hellenistic Monarchy," in the Blackwell Companion to Greek and Roman Political Thought, p. 256. The reference to the harem of Demetrius is from Plutarch (Demetrius, 23); the reference to Ptolemy's float is from Athenaeus (Deipnosophistae, 196a-203b). The feminist criticism basically writes itself.

The Hellenistic kings, I learned, were basically warlords (like most kings most of the time - most small coalition, small selectorate political systems are depressingly similar), though they were also more upfront about it than most (their kingdoms were more or less explicitly premised on usurpation and force; they never appeared in civilian garb; they called their territories "spear-won land").

The piece also makes good sense of the idea of deification (which happened with some frequency among Hellenistic kings and later Roman emperors): in a world were gods are basically powerful beings who can grant or deny your wishes, powerful warlords were not functionally different from gods; they too, could grant or deny your wishes, so why not build temples to them? I suspect deification as ideological control (as in modern totalitarian regimes) was probably not used much in ancient times.

One small point of contention. Eckstein thinks that the classical Greek portrayal of the Achaemenid monarchs of Persia as absolute despots (in Herodotus or Plato) was an ideological "fantasy," but the Hellenistic kings were apparently the real deal. The Persian kings, according to him, were constrained by custom and a powerful aristocracy, while the Hellenistic monarchs were (apparently) not. This may be partly true, but I doubt it settles the matter; kings constrained by small aristocracies can be as absolute as kings not so constrained (and even the Hellenistic monarchs must have had a small winning coalition that had to be kept happy, the so-called "friends" [philoi] that Eckstein mentions - friends here is an actual title, not just an informal relation). The well-attested opulence of the Persian court certainly suggests a very high extractive capacity and a not very constrained monarch from the point of view of their subjects, though perhaps their rule was legitimated in a different way and their freedom of action with respect to their selectorates and winning coalitions was smaller than the freedom of action of the Hellenistic kings.

Education, Selection, Discipline

Consider the problem of virtuous leadership in politics. It would seem to be pretty uncontroversial to say that a government is better the more its leaders are virtuous individuals, and that a regime is better the higher the probability that virtuous leaders will be consistently selected to lead it over time. If our political leaders were just and courageous, fair and wise, that would seem to be a good thing, right? To be sure, there are important disagreements about what counts as virtuous leadership and what counts as a political virtue, and at any rate virtuous leadership is not the only thing that matters in politics, but let us put these questions aside for the moment and consider a different question: assuming a certain conception of virtue, what are the possible institutional solutions to the problem of ensuring the rule of virtue, or a close enough approximation? What possible institutions either increase the chances of virtuous leadership, or reduce its need? 

It seems to me that there are basically three solutions to this problem, which we can call selection institutions, disciplinary institutions, and educational institutions.

“Selection” institutions more or less efficiently select the virtuous as rulers and leaders (via properly structured elections, for example), taking the distribution of character and virtue in society as given, or else efficiently select good policies directly by institutional mechanisms that amplify the good qualities of citizens and dampen their bad qualities (via deliberative or aggregative mechanisms, for example). Theories of “epistemic” democracy from Rousseau and Condorcet onwards tend to emphasize these solutions, but the basic idea is quite old. For example, Plato, Aristotle, and Cicero, none of whom were “democrats” in any important sense of the word, all suggested that though regular people are not individually virtuous, collectively they may yet be good judges of character, especially in small communities where everyone knows everyone else, so that properly structured elections under the right conditions (and there is a lot riding on the qualifications) can result in a relatively good chance of selecting virtuous leadership. The same logic can be applied to the direct selection of policy: if virtuous leadership cannot be ensured, perhaps specific institutions can be created (e.g., referenda or other forms of “direct democracy”) that amplify and aggregate the good qualities of citizens (their private information and local knowledge, for example) so that the kinds of policies that a virtuous leader would implement tend to be selected (I hasten to add that the justification of referenda and other direct democratic institutions is not simply or solely that they tend to result in policies that would have been chosen by ideally virtuous rulers; typically they are justified by the values of participation, autonomy, and the like).

“Disciplinary” institutions more or less efficiently discipline rulers and leaders to act in virtuous ways even if they are not virtuous, taking the outcomes of selection as given. This is the focus of the many theories of political regimes that prescribe separation of power and other such mechanisms; public choice normative theories are an obvious contemporary example, though the basic idea, again, is quite old. So, for example, the theory of the mixed constitution in antiquity is in part a theory of how to induce behaviour that is approximately virtuous by diminishing the incentives towards arrogance and hubris among those who are powerful; it assumes that people who are overly wealthy or overly secure in their power will tend to act in ways that are not beneficial to the community, and hence that other parts of the community need to be given the tools to act as counterweights.

“Educational” institutions, finally, attempt to increase the supply of virtuous individuals in society (who can then be selected into positions of power and responsibility). Many classical writers certainly emphasized such solutions, but one can find examples in contemporary calls for civic education. Educational institutions may aim to directly increase the supply of potential leaders, or may aim to increase the overall level of virtue in the population so that selection institutions will work more efficiently (because more virtuous citizens will be more likely to select leaders properly, perhaps). They may also work directly (e.g., through laws or explicit curricula and in formal settings) or indirectly (as a by-product of other interactions, or as the result of interventions into the broader cultural environment). But the basic point is in both cases to construct institutions, or to shape the cultural environment, in ways that increase the supply of virtue, whether at the citizen level or at the level of candidates for leadership.

It is worth stressing that particular institutions may serve more than one of these purposes. For example, selection institutions such as competitive elections also serve as disciplinary institutions (they are typically selective ex ante and disciplinary ex post). And deliberative mechanisms are typically both selective (of policy) and educational (of participants). But the general analytical distinction between selection, discipline, and education is still, I think, generally useful, as the power of particular effects in any given institutional setting will vary: so elections may, for example, be more efficient at disciplining politicians (i.e., making them behave in ways that approximate virtuous behaviour) under some conditions than at selecting virtuous leaders ex ante.

At any rate, the relationships between these institutions are complex, both within a static context and in the longer run dynamics of a particular system of institutions. From a static perspective, the appropriate balance between selective, disciplinary, and educational institutions (or between the selective, disciplinary, and educational effects of institutions) will depend in great part on a judgment about the relative “efficiency” of institutions for the production and selection of virtuous leaders or the disciplining of non-virtuous leaders. It seems to me, for example, that the greater emphasis on education characteristic of classical thought reflects in part a greater confidence in the effectiveness of the educational technology available to smaller face to face societies; educational institutions for virtue do not seem to scale well in modern, large-scale societies. Similarly, it may be that in modern mass-mediated societies it is increasingly hard to select virtuous leaders ex ante (the noise to signal ratio about their characters is too high, and too susceptible to manipulation or distortion), so that elections need to be reconceived more and more as disciplinary institutions, a thought that most classical thinkers did not seem to consider. For them, elections were primarily selection institutions, and hence did not need to be especially competitive; discipline was produced through the use of the court system or specialized quasi-judicial institutions like the audit (euthuna) and short terms of office, not induced by the regularization of competition common today (note also the prevalence of “term limits” in ancient elections). The parts of the institutional context that appear to be immutable also matter; in an age of monarchy (like the middle ages), where selection institutions (hereditary succession, basically, which is equivalent to more or less random selection) and disciplinary institutions (local noble assemblies, basically) are limited in their effectiveness, it is inevitable that educational solutions would be stressed (hence the many “mirrors of princes” we find, and the relative neglect of discussions of disciplinary and selection institutions).

A potentially more interesting but also more difficult set of questions concern the longer-run dynamics of these systems of institutions. I can imagine, for example, that even though from a static perspective all three solutions are equivalent (so that we could in principle substitute a highly efficient disciplinary system for a highly efficient educational or selection system: consider Kant’s famous comment about the possibility of designing a good constitution for a “race of devils”), from a dynamic perspective things might look quite different. Perhaps an over-reliance on disciplinary institutions would tend to induce a downward shift in the supply of virtue, since all disciplinary procedures can catch people who are actually virtuous, so that potentially virtuous candidates for leadership would be discouraged; or perhaps excessive reliance on selection institutions would increase the risk of really bad outcomes (as on occasion a really bad leader or a really bad policy would slip through). Ancient writers accordingly stressed a relatively balanced mixture of discipline, education, and selection as a solution to the problem of good leadership; but perhaps the appropriate balance will always depend on the actual efficacy of the respective organizational technologies and on material and cultural conditions (so that perhaps we are today right to more or less disregard proposals for extensive civic education as unlikely to lead to significant changes in the virtue of leaders).

What am I missing? Are there other possibilities, other than education, selection, and discipline? 

Endnotes

Things that caught my eye in the last week or so:

• I had not known that in some societies, paternity is thought to be "partible", because they believe that "a fetus is made of accumulated semen" and hence can have multiple biological fathers (the quote is from this book, which sounds interesting, p. 60, via the Less Wrong post linked above, which is also interesting).
• The art of sleeping in seminars. Must remember all the rules next time I sleep in one.
• I found that I had mixed feelings about the possibility of a world without mosquitoes: though apparently their extermination would not be especially damaging (other species would apparently do many of the things they currently do, with the usual caveats that our understanding of their ecological role is limited), I have a quasi-aesthetic negative reflex reaction to the possibility of intentionally exterminating completely another life-form, even one that causes a lot of misery (and even as I know species go extinct from natural causes all the time). I wonder if there is anything out there about the morality of species-cide? I know there was some discussion at some time about the possibility of completely eradicating the smallpox virus, for example, but I don't know if there is serious literature about the morality of the practice (not that we are about to exterminate mosquitoes, mind you - the possibility is purely theoretical). 
• Theatre for audiences of one. I suppose this is closer to conceptual art than to traditional theatre.
• The intelligence of Octopuses. The thing about octopuses is that their "intelligence" is not a by-product of their sociality: they are solitary creatures. Presumably this means that octopi could not develop intelligence that depends on the manipulation of meaningful symbols or language?
• Why the one-eyed do not rule among the blind. An interesting point on the social mediation of insight. 

Polybius and the Dialectic of Forgetting (Or, Theoretical Models in the Classical World)

(Warning: A very long footnote – 2,500 words - about Polybius, human moral psychology, and the use of “models” in the Classical World, written as part of research for this project.)

The extant fragments of book VI of Polybius’ Histories contain a famous (in certain circles) discussion of the “cycle of regimes” (VI.5.10-9.14). The story goes more or less like this.

Human beings start in something like the “state of nature,” without arts or sciences, and in particular without highly developed moral norms, where we herd together like other animals following the strongest or most daring man (the basic primate pattern, we might say today: I can’t help viewing a lot of the stuff I’m reading right now through the lens of Boehm’s book, and this post will be no exception). This is what Polybius calls “monarchy” (μοναρχίαν), where the authority of the leader is limited by his physical strength and daring. This sort of “natural monarchy” then evolves towards kingship (βασιλεία) properly speaking, which is no longer a simple hierarchy sustained by strength and daring but a moral community where the authority of the leader is very much constrained by relatively egalitarian ideas about justice. In such a community, the king does not attempt to distinguish himself from his subjects by their dress or in their food and drink, and he gains the support of others only insofar as he coordinates the enforcement of community norms (VI.6.11-12), even as he may be weak and infirm; tendencies to domination are effectively kept in check.

Polybius explains the development of these ideas about justice as the result of spontaneous reflection about reciprocity (VI.6.5-10): because we expect others to reciprocate our good deeds, and value the good deeds of others, norms codifying those expectations emerge, in turn sustaining the authority of those leaders who coordinate their enforcement. Our “natural” ideas of justice are thus quite egalitarian (cf. the use of “democratic” language to describe these norms in all regimes: VI.8.4 πολιτικῆς ἰσότητος καὶ παρρησίας, VI.9.4 τὴν ἰσηγορίαν καὶ τὴν παρρησίαν), though they will accommodate some hierarchy to the degree that a leader can enforce them. But Polybius argues that this sort of “egalitarian” kingship is not stable; insofar as kingship becomes hereditary (something that is common, Polybius suggests, due to the popular belief in the inheritability of virtuous dispositions) it develops necessarily into tyranny, i.e., a regime where egalitarian community norms no longer constrain the leader. The increase in the security of the position of the king’s heirs, which comes also as a result of other changes (e.g., the emergence of resources such as fortified places that can be monopolized by the leader), increases their temptations to try to dominate others. They begin by trying to distinguish themselves from the others by dress or other signals (VI.7.7), and end by trying to obtain more material resources and reproductive advantages than anyone else in the community. The basic mechanism of change in the Polybian theory of change appears here for the first time: security of position enhances a human tendency to domination (as I was writing this, I came across this recent piece by Ryan Balot that makes a similar point).

But this tendency to domination is counterbalanced by an apparently equally natural tendency to resent domination in the name of the earlier egalitarian community norms. The “best” men – those who are most high-spirited and resentful of domination – will tend to rise up and overthrow the tyrant in the name of these norms, setting themselves up in their place. As speculative political history and anthropology, this is perhaps not self-evident (why would tyranny necessarily give rise to aristocracy rather than democracy or a renewed kingship?), though in general I think Polybius gets the basic features of long-run political development right, even if modern anthropologists would insist on a finer gradation of steps from the basic primate hierarchy to relatively acephalous egalitarian societies to “big man” societies to the kind of morally constrained “chiefdom” that corresponds to Polybius’ notion of kingship and eventually to tyranny. At any rate, the transition from tyranny to aristocracy does fit the history of Rome well enough (and remember, Polybius is writing a history of the growth of Roman power). But as a depiction of a common political psychology, Polybius’ idea is very much on target, insofar as human nature does seem to contain both tendencies towards domination (at least among males) and tendencies to resent domination, and our “default” social norms are mostly egalitarian.

We could thus say that political change, in the Polybian story, is all about the emergence of egalitarian norms, their violation by individuals capable of accumulating resources, and the restoration of such norms by those outside the “winning coalition” who still value such norms. The pattern is repeated at the next step in the cycle. As time and generations pass, the sons of aristocrats become again secure in their position, and again engage in attempts to dominate others in contravention of community norms: this is the beginning of oligarchy. Polybius describes how these new leaders no longer have any experience of the previous egalitarian norms (ἄπειροι δὲ καθόλου πολιτικῆς ἰσότητος καὶ παρρησίας), and have forgotten the misfortunes that led their parents to rise up against tyrants (VI.8.4-5). There is a kind of normative drift: signs of distinction that had been freely given to their parents for their services are interpreted by their sons as things that they deserve naturally, so that the relatively egalitarian norms that had regulated the conduct of their parents no longer regulate their own conduct. What happens instead is that the oligarchs abandon begin to pursue unrestrainedly material and other advantages: they take from others without reciprocating. But the egalitarian norms still remain strong within the rest of the population, and so eventually the people rise up and overthrow the oligarchs, banding behind any leaders who credibly promise to enforce the old moral norms. The people, however, remember with fear both the kings (and their transformation into tyrants) as well as the more recent oligarchs (note the emphasis on memory, which is partly the historian’s domain); they thus decide to manage their affairs by themselves, and democracy is born.

In democracy, as in aristocracy or kingship before, egalitarian norms remain strong as long as some are alive who experienced the previous form of domination; but when the experience is lost (with the grandchildren of the founders of democracy, VI.9.4-5), their influence weakens. People begin to take such norms for granted, and those with resources (the wealthy) begin to attempt to “aim at pre-eminence,” i.e., attempt to dominate others. This competitive struggle among rich individuals sets in motion a process of far-reaching social disintegration, where each wealthy individual corrupts the people by turning their desires in the direction of material accumulation and accustoming them to getting what they want by raiding the wealth of other wealthy individuals. This turns the democracy into a rule of violence (χειροκρατίαν), where multiple demagogic leaders compete to dominate by promising each other’s wealth to the people; the process ends with the masses turning into beasts again (cf. ἀποτεθηριωμένον, VI.9.9), i.e., returning to the state described at the beginning of the cycle, herding like the other animals without real social norms, and eventually getting a new “master,” a new leader who is only limited by his capacity for violence, though under social conditions that are different from those operating at the beginning of the cycle.

In summary, Polybius claims that simple regimes develop according to the following cyclical pattern: pre-social state of nature/natural monarchy->;kingship->;tyranny->;aristocracy->;oligarchy->;democracy->;mob rule/post-social state of nature->;natural despotic monarchy (repeat). This “theory” has sometimes been criticized by modern scholars (see, e.g., von Fritz, p. 84) for its apparently “deterministic” or “rigid” sequence of changes, and indeed Polybius’ presentation is a far cry from the nuanced discussions of political change found in Plato (on which he claims to draw, though Polybius’ discussion is very different from what we find in books VIII and IX of the Republic and in the Statesman) and Aristotle (whose discussions of the problem of political change in book V of the Politics he may not have known, given the possible loss from public view of a lot of Aristotelian writing between the third and the first century ADBC). Moreover, the “theory” of the cycle of regimes sits uneasily with Polybius’ “historical” sensibilities; it is abundantly evident to any observer of historical reality that regimes sometimes change in ways that do not fit a clear pattern (as it was to Aristotle, for example, who criticized – wrongly - Plato for a similar idea). Democracies sometimes turn into oligarchies (as when the Thirty took power in Athens in 404BC), monarchies into democracies, and in general any kind of regime into any other kind, as Aristotle documented in exhaustive detail in book V of the Politics. Writers who knew their Polybius well often tactfully pointed this out after summarizing Polybius’ “theory.” Thus Cicero notes in De Re Publica I.68, after a discussion of regime change that is obviously influenced by Polybius, that regimes tend to change in ways that do not necessarily fit any simple pattern (a remark that is attributed to Scipio, who had been a friend of Polybius in real life), and Machiavelli notes that a community would be unlikely to experience all the stages of change described in the theory before it was taken over by a more stable and better organized state (Discourses I.2 – there is a bit of a puzzle here, for though Machiavelli is clearly describing something like Polybius’ theory, he knew no Greek, and book 6 of Polybius had not been translated at the time). Given that Polybius was not politically or historically naïve, it seems unlikely that these observations would have escaped him.

Yet the criticism is unfair. Polybius explicitly indicates that he is simplifying the “more precise” discussions found in Plato and other philosophers (VI.5.10), which he finds too complicated for pragmatic purposes. The “simplified” theory of the cycle of regimes is not to be taken as an accurate representation of historical reality, but as something like a “model” in the sense in which economists use the term: a distillation of the incentives and other influences affecting the main political actors in a regime, and pragmatically useful as a tool for analyzing political changes in more complex regimes, like the Roman one. (Polybius’ “solution” to the difficulties of the cycle, the “mixed regime,” is also best understood as that sort of model, though that is a subject for another post). These incentives are not simply incentives to behaviour (as in much modern rational choice theorizing) but to character: the model describes how certain characters emerge endogenously from certain regimes, given some assumptions about human nature and about the preservation of historical memory.

What is especially neat about Polybius’ presentation of the “cycle” of regimes is how he ties experience and forgetting with the natural tendencies to both attempt and resent domination in the explanation of political change. Constitutions change for the worse because people forget their experience of earlier attempts of some to dominate others, or rather, because the people who overthrow bad regimes are unable to pass on this experience with sufficient clarity to their children as the children’s position in society becomes increasingly secure. Without real fear of domination (which mostly comes from actual experience of such attempts), egalitarian norms do not survive, a point that applies as much to the newly victorious Rome of Polybius’ time as to other polities. (Incidentally, this seems to indicate that the historian’s role is to remind his audience of these misfortunes, a point argued at some length by Balot with many examples from the rest of the Histories). Thus we have a dialectic between the tendencies to resent domination (which sustain egalitarian norms) and the tendencies to dominate (which corrupt these norms): as one gains the upper hand, it immediately begins to weaken. This dialectic is also a process of corruption: healthy norms are first destroyed among a small elite, then among a larger elite, and finally among the entire people as the historical memory of domination is lost first among the heirs of the monarch, later among the heirs of the larger elite that overthrew the tyrant, and finally among the people themselves through the corrupting effects of the competitive struggle for position among rich individuals. A secure, healthy simple regime is in a sense bad for the education of its leaders, whereas a certain kind of misfortune is a good education, a theme that Polybius emphasizes throughout the Histories.

Ultimately, of course, this “model” of political change is used by Polybius to understand and analyze Roman history: Rome was successful insofar as its leaders instinctively chose courses of action that constantly prevented them from giving full rein to their dominating tendencies (e.g., the “checks and balances” of the Roman mixed constitution, which has, I think, been much misunderstood by modern historians who point out, rightly, that Rome during Polybius’ lifetime was basically an oligarchy; but that is another topic) and will be unsuccessful insofar as it achieves full security (as is evident from the later books of Polybius’ Histories, or at least of what remains of them). The model may be historically implausible (in its assertion that typically tyrannies turn into aristocracies, and oligarchies into democracies, for example) but it is no more implausible than some of the rational choice models of political change in use today (which are useful too, I should note), and it has the added benefit of incorporating a plausible moral psychology. By contrast, Aristotle’s exhaustive description of political change in book V of the Politics, though empirically better informed and starting from a basically plausible principle about how the violation of norms about justice leads to political change (where justice is always understood as a form of equality, though differently in different regimes), seems to miss the forest for the trees. Aristotle spends too much time looking at purely accidental causes of political change (“exogenous shocks” in the contemporary economic jargon, such as foreign conquest or institutional drift), or exploring varieties of a single cause of political change (the attempts to dominate others by high-status individuals), whereas Polybius’ model almost begs to be formalized as a story about the interaction between tendencies to domination and tendencies to resent such domination in the absence of perfect memory about the consequences of domination, and it seems readily applicable to a variety of cases even when it cannot explain them completely. From this point of view, it does not surprise me that Polybius’ simplified story was much more influential, historically, than Aristotle’s complex and empirically informed catalogue of the causes of political change: it identifies a key cause of political change for ill, and suggests ready remedies (basically, keeping the fear of domination alive through “checks and balances,” a remedy that is mentioned by Aristotle as well, but only among many other suggestions, some major, some minor).