Democracy and Political Ignorance – Why Smaller Government is Smarter

Ilya Somin, Stanford University Press, 2013, 280pp

ROBERT HENDERSON is unimpressed by an earnest screed against uninformed politics

Does the ignorance of voters matter in a system of representative democracy? Somin, a professor of law at George Mason University School of Law, thinks it has very serious consequences because it leads voters to make ‘wrong’ decisions. His book is primarily concerned with the effects and implications of voter ignorance on the American political system, but has implications for any political system, democratic or otherwise.

Political ignorance can certainly be found in abundance in both the United States and Britain. Listen to a vox pop or phone-in on a political subject and the ignorance of the general public can be startling when it comes to the detail of  politics, not least because educated respondents are frequently as much at sea with political subjects as the uneducated. Somin cites a large number of prime examples of crass political ignorance amongst Americans. For example, two 2006  polls respectively found that only 42% of Americans could name the three branches of the federal government – the executive (President), legislature (Congress) and judiciary (Supreme Court) – and only 28% could name two or more of the five rights guaranteed by the first amendment (p.19). As for specific policies, a good example is a 2010 survey which showed that 67% of the population did not know that the economy had grown the previous year, despite the economy being judged as one of the most important policy areas by Americans (p.21).

This may be dismaying at first glance, but in practice it is irrelevant how limited is the detailed political knowledge of an electorate. This is because no individual, however diligent, erudite, insightful and intelligent, could be seriously knowledgeable about all but a very small proportion of  the problems and policies arising in a  minimalist state constructed on the Hayek model, let alone the vast ocean of policy areas which are covered in the modern industrial state. That would apply even if political power was devolved. Indeed, in a devolved situation (and Somin is strongly in favour of devolved power) the position could be even worse, because there could be more to know and understand, with multiple jurisdictions to vote for on important issues.

Does this mean that representative democracy should be done away with? Not a bit of it. Even though he is worried about democratic outcomes based on ignorance and sceptical about the chances of improving political knowledge amongst  voters, Somin in the end comes down in favour of it:

Despite political ignorance, democracy retains many advantages over rival systems of government. (p.199)

Indeed it does. Whether electors can make considered decisions on all matters or even the vast majority of issues  is not really the point of representative democratic politics. What matters is the fact that such a political system can best restrain the naturally abusive tendencies of elites and provide by far the best legal mechanisms for the formal and peaceful transition of power, something which makes coups and civil war much less probable.

Voters can meaningfully answer the big political questions. They can oppose mass immigration. They can say whether they want their country to go to war. They can approve or disapprove of their country’s legal system. They can say whether they feel more comfortable with a welfare state or no welfare state. They can make a meaningful choice on whether they wish their country to be part of a supranational bloc, such as the EU. They can decide what punishment should be meted out to criminals. They can say yea or nay to whether essential industries should be in public hands. Electors can also make purely rational decisions (for example, those made simply on arithmetical grounds) on competition for resources; for example, it is perfectly rational to oppose immigration on the grounds that it increases competition for housing, education, jobs and welfare.

The fact that voters’ answers to such questions, if they were ever allowed to vote on them in referenda, would  generally run contrary to the wishes of elites in countries such as the USA and Britain and are routinely thwarted by those elites, tells us that the real reason voters are denied the chance to directly make decisions about policy is not that they are incapable of doing so on many major issues,  but rather that the opinions of voters are opposed to those with power, wealth and influence.

A major problem with Somin’s approach to his subject is that he wants politics to be a science, to have an objective reality like physics. When I was a history and politics undergraduate, I had  to take a compulsory course entitled Modern Political Analysis. This involved flow charts, graphs and formulae which purported to elevate the  study of politics to the level of a science. Politics students were solemnly expected to take seriously, say, a flow chart which started with a box marked electorate, had boxes marked with words such as election and  government, before ending with a box marked democratic outcome. Democracy and Political Ignorance is cut from the same misdirected intellectual cloth, nothing like so crudely but still in a manner which assumes that the democratic process can be reduced to quantifiable data. He even has a few formulae such as this example:

Assume that UV equals utility of voting, CV equals the cost of voting and D equals the expected difference in welfare per person if the voter’s preferred candidate defeats her opponent. Let us further assume that this is a presidential election in a nation with three hundred people, that the voter’s ballot has only a one in one hundred chance of being decisive, and that the voter values the welfare of his fellow citizens an average of a thousand time less than his own…thus we get the following equation D(300 million/1000)/ (100 million) – CV = Uv (p.67)

That is the general error of the book, to imagine that human behaviour can be reduced to a miscellany of objective fact which can be used to determine how people should (or even would of necessity) behave if only they were in full possession of these facts. This matters greatly because the vast majority of political decisions have no objective truth or falsity. The particular mistake Somin makes is to imagine that there is such a thing as perfect information which leads to objectively right answers to political questions.

Somin adds bias to this error, for he makes it clear that he is on the politically correct wing of politics. Everyone who is interested in politics has a bias and that is fair enough if the bias is both acknowledged and opinions are not put forward as objective facts. Somin does not do this. Rather in the way that people claim they are in favour of free expression and then allow censorship of  things they disapprove of, so the author wants a distinction between good and bad values, viz:

This book does not provide a defence of any particular vision of political morality. But unless we adopt the view that all values are equally good – including those of racists and Nazis [note that he does not include Marxists who have been responsible for even more deaths than the Nazis] –  we must admit that good political knowledge might sometimes be put in the service of “bad” values. (p.55)

Tellingly, he also feels that some political knowledge can be damaging:

Why might political knowledge exacerbate the harm caused by an electorate with bad values? Consider an electoral majority that is highly racist and wants to inflict as much harm as possible on  a despised racial minority. If such racist voters become more knowledgeable about the effects of government policies, they might force elected officials to implement policies that increase the minority group’s suffering. (p.54)

That might seem a reasonable position at first glance, but what would constitute ‘racism’? After all, governments of all colours routinely favour one group over another,  whether the group be defined by race, ethnicity or class. Trying to objectively define what was racist behaviour by a government would in practice would be impossible because inevitably judgments would be highly subjective.

Political correctness also colours Somin’s judgements in important ways. For example, he claims  that the  mistreatment of blacks in post-slavery America was in part built on the belief of whites that blacks were prone to excessive criminality and every black man was just waiting to rape white women. But whether post-slavery white America did genuinely fear black criminality is not necessarily the real issue. Human beings will use justifications for likes and dislikes which are not the real reasons for their choices, when they feel either that they simply do not like something without having any clear idea why, or are afraid for legal and social reasons that their motivation for holding a view would be unacceptable or even dangerous for them if expressed. Whites in the old slave-owning states may have used any number of rationalisations for segregation, while their actual motivation was that they did not see blacks as their equals or, more fundamentally, simply as different, as not part of the national American “tribe”. There is also the fundamental difficulty of how any objectively true information could exist in some instances. It is not irrational to have a fear that an enslaved group once set free might wreck physical revenge on the group which had held them enslaved. That being so, it is difficult to see how American whites who believed that could have their fears assuaged by more knowledge. In the nature of things there could be no such knowledge available to decide the question of whether freed slaves and their descendants would be violently criminal.

Interestingly, Somin does not address the fact that it is not just a lack of interest or education which stops people becoming politically knowledgeable, but also lack of innate qualities such as intelligence, intellectual inclination and  extroversion. Perhaps that is because his politics debar him from believing that people will or will not do or be something because that is the way they are born. That would fit into his mindset. IQ is particularly important because the lower the IQ the less ability to handle abstractions or complex data. This is not a trivial matter, because at least ten percent of the population of Western states have IQs of 80 or lower. That is the level which most psychologists working in the field of IQ believe that a person begins to struggle to live an independent life in an advanced modern society.

Somin is much taken with the concepts of rational ignorance and rational irrationality. Rational ignorance is the idea that voters do not devote time to educating themselves about political issues because they make a rational decision that their votes will count for next to nothing. I sincerely doubt whether anyone actually makes a decision to remain ignorant on that basis, although they may use it as an excuse for being politically ignorant.

But even if voters did make a considered decision to remain ignorant it would not self-evidently be a rational decision. To begin with there are many electoral circumstances where a vote is important. That is true where the electorate is small or a seat is marginal. Under the first past the post system used in Britain there are a considerable number of seats where the main party candidates are near enough in their support to make voting a far from redundant business. But even where there is no main party candidate who appeals to an elector, or one of the main party candidates is odds on certain to win, there is still a point in voting. To begin with, if turnout is persistently low it could be used  by those with power to argue for a restricted franchise or even no franchise at all. Then there is the overall vote a party gets. If, for example, a party or presidential candidate gets elected with less of the popular vote than their main opponent their mandate is weakened. If all else fails, a vote for a candidate of a minor party such as UKIP in Britain, the  minor presidential candidate in the USA, or a spoiled ballot sends a public message about the state of elector dissatisfaction with the mainstream parties. Somin is not entirely blind to such objections, but mysteriously and annoyingly they appear to carry little weight with him.

Rational irrationality is the brainchild of the economist Bryan Caplan. The idea is that voters not only have incentives to remain ignorant but also incentives to “engage in highly biased evaluation of  the information they do have” (p.13). The tempting response to this is a sarcastic “Who would have thought it?” Pursuing the idea of rational irrationality, Somin likens those who are seriously committed to supporting political parties to fans of sports teams who support their team blindly,  generally give weight to information which boosts their team and disregard that which does not. The rewards for doing so are emotional. This of course is not irrational behaviour because it is natural for human beings to indulge their “tribal” instincts and defend their position and that of their group. Where rational ignorance and rational irrationality come together, they are to Somin’s mind the most toxic political democratic cocktail, one which could only be overcome or at least ameliorated if those pesky voters would just become “correctly” informed.

What are Somin’s solutions to reduce what he sees as the harm of voter ignorance? It is to lessen the amount which government does (with much of the slack being taken up by private enterprise) and bring as much as possible of politics to the local or regional level, viz:

Despite political ignorance, democracy retains many advantage over rival systems of government. Nonetheless, political ignorance will probably continue to be a serious weakness of democratic government. We are unlikely to eliminate that weakness completely. [Another example of the blindingly obvious] . But we can reduce its dangers by limiting and decentralising the role of government in society (p.199)

There are real  problems with both of these policies. In a large industrialised society, government of necessity has to do a considerable amount, whether that is at the local or national level. There have to be good communications for people, goods and information. A universal school system is unlikely to exist if it is not in large part funded by the taxpayer. Defence and the maintenance of law and order cannot reasonably be left to private initiatives. Foreign policy, especially for a super-power such as the USA, has wide-reaching ramifications for domestic policy and is frequently very complex to master.

As for decentralisation of politics, the more local the decision making the smaller the pool of political talent available. This may well result in poorer decisions being made, especially where the policy is complex. It is also true that if the number of political bodies which can raise and spend taxes increases, the opportunities for corruption  increase, and this generally means more corruption. As for what should be devolved, that would be a real can of worms. For example, it is difficult to imagine Somin thinking that federal action to enforce politically correct behaviour throughout America would be damaging, or that he would readily tolerate a local jurisdiction which, for example, refused to apply equal rights laws.

Overall, Somin is gloomy about the likelihood of political knowledge increasing. He glumly points to the fact that despite rising IQ scores, educational standards and the great ease of access to information because of the internet, there has been little increase in political knowledge during that time (p.199) or of rationality (in his terms).

Perhaps most damaging  for Somin’s desire for greater political knowledge is research (which he cites)  that suggests that the more knowledgeable voters are “…more biased in their evaluation of  new evidence than those with less prior information”( p.80).  If this is true – and it is very plausible because the more data someone has, the greater the material from which  to construct arguments – then the whole idea of a better educated electorate producing superior outcomes falls completely to pieces.

Where does all this leave us? Somin has expended a great deal of effort on creating vast complications for a subject which is essentially simple. The matter can be reduced to its essentials by simply pointing out the following –

  1. That no person, even if they are a full time politician and have the requisite intellect, can master more than a small part of the political range of the modern political state
  2. That even if the scope of politics was massively reduced by, say, half, the problem of inevitable political ignorance would exist because there would still be a great deal left
  3. That for most political issues there is no objectively right answer, and
  4. That much of the population, even if they had a serious interest in politics, would not have the education and intelligence to understand much of what goes on.

There is an important book to be written about voter ignorance within a democracy. Sadly, this is not it. However, the book is worth reading as a first-rate example of the attempts of those working in what are mistakenly called the “social sciences” to pretend that these subjects  are bona fide sciences just like physics and chemistry, and a very revealing look into the modern liberal mind.

ROBERT HENDERSON blogs at www.livinginamadhouse.com



This entry was posted in QR Home and tagged , , , . Bookmark the permalink.

1 Response to Dimocracy

  1. David Ashton says:

    Isn’t there a case for an 0ccupational franchise, so that people can vote for specialists in matters with which they are familiar? Every adult can be included, e.g. married women at home can have their own corporation/chamber/guild. The ILP socialist Bob Edwards who fought against Franco told me he was in favour of it.

Leave a Reply

Your email address will not be published. Required fields are marked *

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