Pinker argues for the stunning success of the Liberal Enlightenment as shown by massive factual evidence (conveyed in 73 charts of statistical data) of human progress over the past 200 years: because of liberalism today more human beings are living longer, healthier, wealthier, freer, safer, more stimulating, and happier lives than human beings have ever lived at any time in history.
Deneen argues for the utter failure of liberalism as shown by cultural critics of the liberal order: because of liberalism today human beings are living lonely, resentful, alienated, boring, meaningless, and generally desperately unhappy lives.
You see what I mean by two different worlds--the world of liberal progress and the world of liberal failure? Actually, Deneen's world is paradoxical in that he argues that liberalism has failed because it has succeeded: the successful triumph of liberalism in the modern world has manifested the destructiveness of the logic of liberal thought in abolishing the healthy practices and institutions of traditional forms of social life in promoting the degrading life of materialist individualism.
To settle this dispute, we need to judge the three levels of argument in this debate. The first level of argument is about how to interpret liberal political theory. The second level is about analyzing the factual evidence of liberalism's practical performance--its success or failure. The third level is about assessing the illiberal alternatives to liberal order.
In this post, I will consider the second level--the factual evidence. In subsequent posts, I will consider the other two levels of the debate.
Pinker relies much more on factual evidence than does Deneen, which shows the differences in their academic positions. As a professor of psychology at Harvard who sees himself as a natural scientist, Pinker believes that the best source of knowledge is the modern scientific method of testing hypotheses by how well they explain the empirical data of human experience of the natural world. He must also study the history of political philosophy in developing his argument for the classical liberal political philosophy that culminated in the Enlightenment, but he sees that philosophic position as making empirical claims about human nature and human history that must be tested by scientific research.
Deneen is a professor of political science at Notre Dame who specializes in the study of the history of political philosophy, which he sees as devoted primarily to the interpretation of classic texts of political philosophy from Plato to the present. But he also recognizes that applying that philosophic knowledge to modern social and political life requires some appeal to empirical evidence for the success or failure of the liberal order and other regimes.
Deneen's book is similar in some ways to Steven Smith's recent book--Modernity and Its Discontents--which I have written about (here). Like Deneen, Smith makes a Straussian argument against liberal modernity, although Smith is more explicit in his appeal to Leo Strauss.
The thesis of Smith's book is "that modernity has created within itself a rhetoric of antimodernity that has taken philosophical, literary, and political forms" in denouncing the bourgeois life as "a kind of low-minded materialism, moral cowardice, and philistinism" (xi). He thinks that he proves this thesis by restating what some of the antibourgeois writers have asserted in their attacks on bourgeois liberalism. But as I have observed, while this proves the existence of an antibourgeois rhetoric, it does not prove the truth of this rhetoric as confirmed by empirical evidence of what life is like in the Bourgeois Era. Smith says that the goals of bourgeois liberalism are no longer credible, because "leading opinion has increasingly lost confidence in these goals" (4). But then he never wonders whether "leading opinion" might be wrong.
So is this a factually accurate description of life in a liberal social order?
There are at least four separate analyses of data for measuring human happiness. The Gallup Organization, the World Values Survey, and the World Happiness Report (found online) present data on countries around the world (see Pinker 2018, 262-89). The General Social Survey has gathered data for the United States.
In Coming Apart: The State of White America, 1960-2010, Charles Murray defends an Aristotelian conception of happiness as lasting and justified satisfaction with life as a whole; and he uses data from the General Social Survey to show that self-reported happiness among Americans is associated with four kinds of virtues--marriage, industriousness, honesty, and religiosity. I have written about Murray's argument here.
Oddly, Deneen discusses Murray's book, in speaking about inequality in America, but without mentioning Murray's analysis of the data on happiness among Americans (Deneen, 134, 149-53). Deneen endorses Aristotle's account in the Nicomachean Ethics of how the moral and intellectual virtues promote human happiness or flourishing (eudaimonia), but he says nothing about how Murray uses empirical evidence of happiness to support this Aristotelian understanding as applied to the American liberal order (Deneen, 35).
Much of the data on happiness is based on self-reports of well-being: people are asked how happy they are, or they are asked to rank their life satisfaction on a scale from 0 to 10. Although we might wonder about the reliability of such self-reporting, it does correlate with other signs of happiness, such as smiling, having a joyful demeanor, and judgments by other people.
The World Happiness Report 2017 (WHR) ranks the happiness of 155 countries based on answers from 3,000 respondents in each of those countries. Respondents were asked to evaluate their lives on a ladder where 0 represents the worst possible life and 10 the best possible. Three-quarters of the differences among countries were accounted for by differences in six key variables. "These six factors are GDP per capita, healthy years of life expectancy, social support (as measured by having someone to count on in times of trouble), trust (as measured by a perceived absence of corruption in government and business), perceived freedom to make life decisions, and generosity (as measured by recent donations). The top ten countries rank highly on all six of these factors" (WHR, 3).
I have written about the Human Freedom Index (HFI) (here), which applies a classical liberal conception of freedom in measuring human freedom--both personal freedom and economic freedom--for 159 countries. The strong correlation between the ranking in the HFI and the ranking in the WHR suggests that people in liberal social orders with the greatest levels of freedom are also generally the happiest people.
Here are some of the rankings of happiness in the WHR with corresponding rankings of freedom in the HFI in parentheses:
1. Norway (7)
2. Denmark (8)
3. Iceland (31)
4. Switzerland (1)
5. Finland (6)
6. Netherlands (9)
7. Canada (11)
8. New Zealand (3)
9. Australia (5)
10. Sweden (13)
14. United States (17)
16. Germany (16)
19. United Kingdom (9)
25. Mexico (73)
26. Singapore (18)
31. France (33)
34. Spain (30)
49. Russia (126)
79. China (130)
82. Venezuela (158)
108. Iran (154)
151. Rwanda (65)
152. Syria (159)
153. Tanzania (99)
154. Burundi (150)
155. Central African Republic (151)
This is empirical evidence against Deneen's critique of liberalism, unless he can show that there is something wrong with this evidence. He has not done that, because he has chosen to remain silent about this evidence.
If Deneen is right about liberalism making people deeply unhappy, one might expect this to be indicated by a high rate of suicide, since this is the most dramatic way in which people express their unhappiness. And, indeed, some critics of liberalism, like Emile Durkheim, have made this claim. But as Pinker points out, the data don't support this. Plotting the data for suicides in England, Switzerland, and the United States from 1860 to 2014 shows that the rate of suicide has declined in all three countries (Pinker 2018, 279). And remember that Switzerland ranks number 1 on the Human Freedom Index. For these three liberal countries, suicide was more common in the past than it is today, which is not what someone like Deneen would predict.
Deneen claims that in a liberal order, people tend to lack "moral self-command." One observable manifestation of moral self-command is when people refrain from attacking and killing other people. Remarkably, Deneen is totally silent about the evidence presented by Pinker (2011) and others (Eisner 2014; Muchembled 2012; Sharpe 2016) that from high rates of violence and homicide in the Middle Ages, there has been a long decline in modern liberal societies, which shows that liberalism actually does promote moral self-command, and that people in pre-modern illiberal societies suffered from a lack of self-control.
One of the best-supported generalizations in the science of criminology is that the propensity to crime--and particularly violent crime--is increased with any loss or weakening of self-control. And so, for example, young men between the ages of 16 and 30 show on average a higher propensity to violent crime because their impulsive personalities incline them to lose self-control, and any factor that increases that propensity in young men increases the rate of violent crime (Wilson and Herrnstein 1985). Therefore, if Deneen were right about liberalism promoting a loss of self-control, we would expect a steady increase of criminal violence in the history of liberal social orders. That we see just the opposite is powerful evidence against Deneen's claim.
Manuel Eisner (2014) is a criminologist who has assembled the History of Homicide Database, which is the most comprehensive collection of quantitative estimates of homicide levels from 1200 to the present. His data show that the average estimates of homicide rates across Europe from 1200 to about 1450 converge at a rate of about 27 per 100,000 inhabitants. This average rate then begins to decline: 20.1 (1500-1549), 12.0 (1600-1649), 5.5 (1700-1749), 3.5 (1800-1825), 2.0 (1900-1924), and 1.0 (2000-2012). So, over a period of 500 years, the peacetime criminal homicide rate in Europe fell by half every century.
Eisner's data also show that most of this decline in homicide rates was due to a fall in lethal male-to-male fighting of about 99%!
Eisner thinks that the best explanation for this dramatic drop in homicidal violence in Europe over the past 500 years is that there was what Norbert Elias called a "civilizing process" (Elias 2000; Linklater and Mennell 2010). European societies went through a change by which average levels of self-control, standards of decency, and disgust for open displays of cruelty tended to increase, which arose from the move away from the Middle Ages to European modernity.
Elias presented various kinds of historical evidence for this civilizing process. For example, he analyzed books on manners to see how these books taught new standards of civilized behavior. He noted, for instance, that in medieval Europe, ordinary people often stabbed each other over insults at the dinner table; and the new books on manners tried to change this.
Eisner presents quantitative data analysis that supports Elias's theory. For example, he shows a strong correlation between increasing book production and decreasing homicide rates. Every 10% increase in book production is associated with a 3.4% reduction in homicide rates, and these two variables share almost 54% of the variance. The countries that had the highest per capita book production by 1750-1799--England, the Netherlands, and Sweden--had also reached the lowest levels of homicide rates. By contrast, countries such as Italy and Spain, with lower literacy and book production, had higher homicide rates. Although correlation does not prove causation, one can infer from this that literacy and the reading of books tends to inculcate habits of self-control.
Eisner also presents quantitative data to support the application of Elias's theory to the fluctuations in violent crime rates around the world from 1950 to 2010. Beginning in the 1960s, there was a steep rise in violent crime across North America and Europe. Then, in the early 1990s, there began a steep drop in violent crime. And by 2010, crimes rates were lower than they ever had been. Some social scientists, such as Fukuyama (1999) and Pinker (2011), have argued that this showed cultural changes in which there was a decline and then a revival of self-control.
To find data on these cultural trends, Eisner has used the Google Books NGAM corpus, which is a database of 8 million digitized books published between 1500 and 2008, with an interface that allows users to track the frequency of any group of words as a percentage of all words in the corpus over a specified period of time. Eisner tracked three groups of words that might express hedonistic preferences--words concerned with "sex," "drugs," and "narcissism." He also tracked four groups of words that relate to self-control--words concerned with "shame," "politeness and good manners," "conscientiousness," and "honesty."
He found that the frequency of the hedonistic words increased during the years that the homicide rate was increasing and decreased during the years that the homicide rate was decreasing. He found the opposite pattern for the frequency of the self-control words, which declined when homicide rates increased, and increased when homicide rates decreased. This can be seen as indicating a culture shift that brought declining self-control in the 1960s and rising self-control in the 1990s, which is correlated with crime rates rising and declining.
As another quantitative data base that might show this culture shift, Eisner has used German opinion poll surveys from 1967 to 2010 asking Germans about what values they thought parents should teach their children. There were three items related to self-control--"politeness and good manners," "doing work diligently and properly," and "being thrifty in money matters."
Eisner found that the endorsement of values of self-control was weak during the years of increasing homicide rates and strong during the years of decreasing homicide rates.
Contrary to Deneen's claim that modern liberalism must destroy the moral virtues of self-control, this historical evidence suggests that modern liberal cultures can and have inculcated habits of self-control that account for the stunning decline in violence over the past few centuries.
In my next posts, I will consider other kinds of evidence relevant to liberalism's influence on social life, education, inequality, and the environment.
Eisner, Manuel. 2014. "From Swords to Words: Does Macro-Level Change in Self-Control Predict Long-Term Variation in Levels of Homicide?" Crime and Justice 43 (2014): 65-134.
Elias, Norbert. 2000. The Civilizing Process. Revised ed. Cambridge, MA: Blackwell.
Fukuyama, Francis. 1999. The Great Disruption: Human Nature and the Reconstitution of Social Order. New York: Free Press.
Linklater, Andrew, and Stephen Mennell. 2010. "Norbert Elias, The Civilizing Process: Sociogenetic and Psychogenetic Investigations--An Overview and Assessment." History and Theory 49 (October): 384-411.
Muchembled, Robert. 2012. A History of Violence: From the End of the Middle Ages to the Present. Cambridge, MA: Polity.
Sharpe, James. 2016. A Fiery and Furious People: A History of Violence in England. London: Random House Books.
Wilson, James Q., and Richard Herrnstein. 1985. Crime and Human Nature. New York: Simon and Schuster.