Robert Putnam and the New American Indifference

Robert Putnam, Professor of Public Policy at Harvard University, has been an academic celebrity since the publication of Bowling Alone : The Collapse and Revival of American Community, based on an article originally published in 1995 [1]. The book supports a bitterly debated argument: that “social capital” is on the decline in America, its decay being responsible for the weakening of political participation and civic engagement, which in turn threatens American democracy.

Without going into the debate which followed this argument, and particularly the critique of the empirical validity of Bowling Alone, we will take up one of the central questions Putnam’s findings raise: if an erosion of social ties in America did indeed occur, how should we explain it? The recent and resounding article by Putnam, published last June [2], seems to come up with an unacceptable answer to this legitimate question: it is racial diversity, he argues, that is responsible for the social attrition in American society.

Consistent with his research methodology, Putnam bases his argument on the findings of a vast survey, the Social Capital Community Benchmark Survey, carried out in 2000 on a sample of about 30 000 individuals (sub-divided into smaller sample groups) residing in 41 different urban communities across the American territory, and chosen for their heterogeneity (size, location, socioeconomic characteristics…). As usual, Putnam defines “social capital” as “networks and the associated norms of reciprocity and trustworthiness” [3] which arise from such networks. Racial diversity is understood as in the American census, but distinguishes only four groups: Hispanics, non-Hispanic Whites, non-Hispanic Blacks, and Asians. The conclusions Putnam draws are overwhelming:

– The greater the racial diversity, the weaker the inter-personal trust

– Individuals tend to trust their neighbors least in communities with the highest diversity rates

– In such communities, not only is inter-racial trust lower than in other communities, but so is intra-racial trust

– Diversity leads to anomie and social isolation

Getting down to specifics, racial diversity would be directly responsible for the lower level of trust in local government, in local elected officials and in the local media, for decreased confidence in the possibility for an individual to make a difference, poorer participation rate in voting, lessened confidence in the possibility of members of the community acting collectively for the common good, a lower rate of financial contribution to charities, a lower number of friends and confidants per individual, and a lower measure of happiness and of perceived quality of life. Two – and only two – phenomena seem to increase along with diversity: the frequency of demonstrations, and the number of hours spent in front of the television.

Hence, while there is no doubt that diversity has economic advantages, its social cost seems immense. As Putnam metaphorically puts it, diversity leads people to “hunker down – that is, to pull in like a turtle”

In his academic triangulation, Putnam equally rejects two competing theses: “contact theory,” and “conflict theory.” According to the former, diversity leads to harmony if individuals can get to know each other. The opposite thesis maintains that close contact with difference exacerbates hostility and leads to discord, if not plain hatred between different groups. However, Putnam claims that those who support one thesis or the other share a common belief in a negative correlation between inter- and intra-group trust: the greater the love of the other, the smaller the love of one’s own; the greater the love of one’s own, the greater the hatred of the other. Yet, while internal social capital (“bonding” or the strengthening of ties within a group) does differ in nature from external social capital (“bridging” or the development of ties between groups), both are, in fact, related. Whites who have more White friends also have more non-White friends.

Essentially, at a time when the USA is facing more racial diversity than ever, Putnam rejects both the idea of a rainbow concord and of a racial civil war. Instead, he asserts that a new civic indifference is gaining momentum in America. Diversity would be leading to distrust. Distrust would be leading to indifference. Indifference would arise from difference.

From globalization to universal indifference?

The first problem with Putnam’s thesis is its scope. The political scientist already attempted, with mixed fortune, to test the findings of Bowling Alone beyond American borders. Similarly, he begins E Pluribus Unum with general considerations on the unstoppable diversity of contemporary societies, setting a rather broad – if not vague – goal for his project implied in the title of the article: the ambition to assess the impact of diversity on social capital “in the XXIst century”.

True, the alleged negative effect of diversity on trust or on the scope of the welfare State and the sustainability of social policies – the famous “diversity-solidarity dilemma” or “dilemma of progressiveness” posed by David Goodhard in 2004 in the British magazine Prospect – is at the heart of one of the most heated debates in the social sciences well beyond the borders of America. In a (previous) macroeconomic version of Putnam’s argument, Alesina, Glaeser and Sacerdote, for instance, upheld the thesis that by weakening the feeling of solidarity, racial diversity reduces the size of the welfare State (which, according to their calculations, explains at least half the gap between Europe and the USA in social policy). As the case of the “Scandinavian model” is said to prove, successful redistribution would suppose homogeneity.

The causal chain in this theory goes from diversity to trust, then on to solidarity, and finally to social policies promoting equality. But don’t these same policies have an impact on trust and on the state of diversity? Without policies promoting equality, how can trust develop and doors to diversity be opened? Besides, the implications of Putnam’s work on public policies are paradoxical. Should an aging Europe, in the name of the welfare state survival, close its borders to an immigration which might indeed save the welfare state?

In any event, if Putnam’s goal is to establish – or even suggest – a worldwide causal relationship between trust and diversity, the methodology outlined in his introduction (fortunately) cannot allow him to maintain such a conclusion. Graph 1, using data similar to that collected by Putnam, shows that if there is, indeed, a causal relationship between both variables, it is in fact inverted, notably thanks to the USA and Canada, two striking illustrations of the fact that a great diversity and a strong feeling of trust can coexist. Nor do historical dynamics support Putnam’s theses: as shown in graphs 2 and 3, whereas a slightly negative correlation between diversity and trust could be observed in 1980 for European countries for which data are available, this correlation seems to fade away as diversity increases and becomes virtually insignificant in 2000.

Graph 1: Diversity and trust in 27 countries of the world in 2000*

In the World Values Survey, trust is measured by the percentage of “most people can be trusted” answers to the question: “Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people?”

Sources: United Nations, Cahuc and Algan (2007)

Note: following Putnam’s recommendations, Sweden was removed from the regression.

Graphs 2, 3 and 4: Diversity and trust in selected European countries in 1980, 1990 and 2000

Graph 2 : 1980

Graph 3 :1990

Graph 4 : 2000

Sources: United Nations, Cahuc and Algan (2007) and Adam (2006)

Note: the trust data for 1980 were in fact collected in 1981; following Putnam’s recommendations, Sweden was removed from the regressions.

Diversity of what? Trusting who?

If some believe that ‘social capital’ is an unidentified scientific object (there are now tens of accepted interpretations of the concept in the literature), what can be said about ‘diversity’? The founding question of Amartya Sen’s critique of Rawls, “Equality of what?” comes to mind: ‘Diversity’ of what?

And what should be meant by ‘diversity’? The multiplicity of races? Racial parity? The relationship between racial majority and minorities? When a population is made up of individuals with 50 different origins, people of such origins representing only 5% of the total population, is there an impact on the feeling of trust among the remaining 95%? When a population is made up of 5 racially homogeneous groups, all different from each another, and a problem of trust develops, could it not in fact be due to the limited diversity in the population? For all we know, limited diversity might also explain the social resentment felt by ex–majority groups which have become the first minority group among other minorities.

Also, what is the precise relationship between diversity and immigration? At various points in his article, Putnam seems to use the notions interchangeably, only to then deny it and, finally, to contend that two types of diversity (Black/White and immigrant/non-immigrant) exist in the USA and that further research should focus on separating one type from the other (that is marking the difference of diversity).

But most importantly, what is the diversity (and its related effects) Putnam is measuring? Is the diversity of individuals reducible to ethnic or racial diversity (note that ethnicity and race are two different concepts in the US context)? Is that particular form of diversity the ‘Diversity of all diversities’, the key to understanding the phenomenon Putnam is studying? Putnam maintains that his findings resist the introduction of other variables such as socioeconomic inequalities. But (and Putnam admits this) the relationship between social capital and economic equality is so strong that it is almost impossible to think that the skyrocketing of socioeconomic inequalities in America since the mid-1980’s had no major effect on the decline of social capital. Once again, correlations need to be disentangled.

Finally, one of Putnam’s most surprising findings, the decline of inter- and intra-group solidarity with higher rates of diversity needs to be examined more closely. Not satisfied with the idea of just preventing the birth of new forms of solidarity, the poison of diversity would also attack existing forms of community. But as the Hispanic population illustrates, diversity itself is diverse in the USA. Seen as monolithic by other racial groups, the Hispanic population is in fact incredibly heterogeneous, and the solidarity networks and relationships of trust (and distrust) which form between and among Porto Ricans, Dominicans, Cubans or Mexicans remain largely unknown today. Diversity within diversity is one of the major challenges for the future of the country, yet Putnam’s focus on the racial question does not allow us to even begin to address it.

Diversity, trust and crime

The decrease in crime which began at the start of the 1990s in America is one of the great enigmas of contemporary social sciences and the source of many controversies due to a common temptation to use what remains a social mystery to further some research agenda or personal inclinations (often both).

Putnam himself notes a causal relationship between social capital and low crime rates. But if we believe his findings and compare them to the evolution of crime in the USA, a paradox appears: how can the decline of social capital coexist with such a striking decline in crime? Moreover, how do the increase in diversity due to the acceleration of immigration in the USA in 1990 and 2000 and the decline of social capital both converge towards the plummeting of crime?

If we overlap questions of urban environment and diversity, assuming that the largest American cities are among the most diverse places in the world (Queens in New York, for instance, being the most diverse county in the USA), the paradox becomes even greater. Indeed, although crime has started to increase again since 2004-2005, it has done so in every category of city except cities of over 1, 000 000 inhabitants (9 cities in the USA), which push crime rates downward (as shown in Table 1 for New York and Los Angeles).

Table 1 : Crime in four American cities

Source : FBI, Crime in the United States 2005

It is thus not absurd to wonder, along with the New York Times and a number of scholars, whether, contrary to Putnam’s analysis, it is the increase in immigration and therefore in diversity that has in fact contributed to diminishing crime, factor and symptom of vitality of social capital.

The poison and the antidote

It is impossible to ignore the political context within which Putnam’s article was published, an America where the rise of the Hispanic community is feeding a growing political malaise.

But Robert Putnam is a conscientious researcher. Shocked at his discovery of empirical proof that diversity had negative impacts on society, he chose to defer publication of his findings for five years in order to ensure that they were robust and to consider the consequences they could have on public debate. Certain particularly irresponsible American scholars have much to learn from this concern for ethics.

It is actually not the first time Putnam has tried to control the impact of his findings on public debate. Indeed, the Saguaro seminar on civic engagement in America held at the John F. Kennedy School of Government at Harvard was prompted by Bowling Alone. The report “Better Together”, which followed from the seminar, first published in December 2000 and then again after 9/11, focused on finding a way to restore trust in the USA based on examples taken from existing communities and grassroots organizations. Putnam had actually anticipated the debate following the publication of E Pluribus Unum. His latest opus, published in 2003, was indeed intended to be a sort of textbook against the findings of his own research. Among the various success stories presented was “A neighborhood in Boston that has been revitalized by a civic association that overcame ethnic differences and now plays an ongoing role in the neighborhood”. But Putnam’s antidote is weaker than his poison.

There is no doubt that Putnam has good intentions: “It would be unfortunate if a politically correct progressivism were to deny the reality of the challenge to social solidarity posed by diversity. It would be equally unfortunate if an ahistorical and ethnocentric conservatism were to deny that addressing that challenge is both feasible and desirable”. But when, at the end of his article, he ‘discovers’ that race, like religion, is a socially constructed concept in America, he is doubly disappointing. Firstly in that he recognizes the fundamental limits of his findings far too late and with a doubtful candor. Secondly, both unconvinced and unconvincing, he invokes a great American mix under the auspices of a Barak Obama, who is more than ever being asked to give proofs of his “Blackness” in the current presidential campaign. The point in contemporary society is not to “melt together” multiple identities, but rather to understand their composition and the conditions under which they can prosper.

E pluribus nullum?

Putnam’s article, overly pessimistic in its analysis and overly optimistic in its conclusions, obscures two contradictory American dynamics. For the first time in its history, the United States is showing more interest in the world than the world is showing in it, and soon, a new and improbable character, the American emigrant, will make its appearance in the world. That said, the Americanization of the United States under the Hispanic pressure is feeding an underground feeling of resentment whose effects go far beyond indifference.

Carried by genuine idealism, Putnam sees himself as predicting the end of American identities, swamped in the waves of the diversity of the world. In reality, he may be announcing the appeasement of the growing intolerance crisis in America through the development of emigration and strategies of segregation.