The usages of the word trust in this chapter are intentionally eclectic and inclusive. My argument does not require the reader to commit to a single meaning or school of thought. It allows that the meanings of trust and the motivations behind it are many and can vary across people and settings. It is inclusive, in the sense that the chapter discusses not only interpersonal trust but also generalized trust (in strangers) and institutional trust (in politics and government). For that purpose, it draws on a third stream of thought, one in which trust is allowed to mean whatever people have in mind when they give scaled answers to social survey questions about trust, even though such questions may leave unstated the precise individual who is trusted or the precise gain that is expected from trust.[302] The most common form of such questions is “In general how much do you trust most people?”; or “In general how much do you trust most people you know personally?”; or “Even if you have had very little or no contact with these institutions ... how much [do] you personally trust [your country’s] Parliament? The police? The civil service?”[303]
In relying on survey respondents to work out for themselves what these questions mean, we make a methodological leap like that made by economists and psychologists who study subjective well-being by asking people how they feel as opposed to specifying the elements of a good life in advance.[304]
What is the value of trust? Trust is one of the mechanisms that can promote mutually beneficial cooperation in society. Trust frees the partners in an enterprise from fear of exploitation and betrayal, whether the enterprise is familial, economic, or political. Trust is important to the quality of cooperation.[305] If a task is unskilled, and effort is easily observed, it may be enough to secure cooperation by coercion or threats, but if quality and diligence matter to performance (as in the accurate passing of complex information), and cannot be easily monitored, brute force may not be productive. There must be more than force and fear.[306] From that perspective, trust is more valuable when the quality of performance is at stake.
When trust is defined by social survey respondents, its outcome validity is found when the resulting measures of trust can be causally associated with evidence on outcomes of cooperation that would be impeded by lack of trust. This is not easy, partly because of the scope for reverse causation (just as trust can enable cooperation, the experience of cooperation can build trust), and partly because both cooperation and trust can be influenced by other factors that vary across social settings. When these factors are disentangled, it turns out that higher levels of generalized trust in society, as measured by social surveys, are associated with higher levels of investment and economic development around the world, and some part of the association is probably causaclass="underline" reduced trust causes investment and development to decline.[307]
What was the state of trust in Soviet society? The evidence is both qualitative and quantitative; both kinds of evidence require careful interpretation.
The qualitative testimonies recorded by those with first-hand experience of living in Soviet society have been marshalled by historians. Surveying the field, Geoffrey Hosking explains the destructiveness of Stalin’s purges by “the wildfire spread of generalized social distrust.” After Stalin, he writes, relationships of trust were to some extent restored within families and work collectives, but “The paroxysms of distrust manifested in Stalin’s Terror... left permanent outlines in the Soviet landscape.”[308] These judgements are consistent with other kinds of qualitative evidence to be reviewed later in this chapter. They do not prove that the Soviet Union was a low-trust society compared to others, but they do suggest plausible ways in which aspects of communist rule might have reduced generalized trust.
Quantitative survey evidence has been used to suggest that low trust is a feature of modern Russia.[309] Can that judgement be extended backward to Soviet times? The measurement of trust in society was still in its infancy at the point when European communism collapsed. The Soviet Union lasted just long enough for the Soviet Russian republic to be included in the second (1990) wave of the World Values Survey. Asked how much they could trust people of their own nationality in general, 38 percent of Russian respondents answered either “a lot” or “somewhat.” This placed Russia last but one out of seventeen countries, far behind the leader (India, with 90 percent) and not much above South Africa (35 percent).[310]
Russia may have been a low-trust society as communist rule collapsed. Is this evidence that low trust was a direct legacy of communist rule? In favor of that interpretation, trust in society changes slowly, and seems to have considerable persistence across generations. If low trust was a feature of Russia in 1990, then it was probably also the case in 1980 or 1970.
But then a more difficult question arises: was low trust under late communism a result of the experience of communism, or was trust in Russia always low? A sceptic could point to China, another country with traumatic experiences of communist repression. China often does comparatively well on trust measures and was the middle country by rank in the 1990 World Values Survey on the question reported above. This does not mean communism had no effect. Perhaps the levels of trust found in China might always have been higher than in other countries because of factors peculiar to Chinese culture or history. But, if so, the converse might be true of Russia. Trust might be low in Russian history, just because of Russia.
More fine-grained evidence of the effects of communist policies is again suggestive, although not conclusive. A well-known study of levels of trust in Africa finds that it was severely damaged by the incidence of the slave trade. This is untangled by identifying how trust measured today across ethnic groups is systematically related to the varying historical exposure of each ethnic group to enslavement.[311] Turning to Soviet history, a study of exposure to forced labor under Stalin finds a similar result: the descendants of those citizens more exposed report lower trust in others today.[312] This result can be compared with another study that starts from Russians’ historical exposure to the prerevolutionary institution of serfdom. That study finds present-day trust unaffected.[313] Putting these together, it has proved easier to identify a persistent detrimental effect on trust arising from communist policies than from coercive institutions in Russia before communism.
When the Soviet leaders resorted to methods of rule that plausibly lowered trust in society, was this a bug or a feature? Did the Soviet leaders bring about a low-trust society by design? For the autocratic ruler, a society with low trust among citizens should have the great advantage that those who might otherwise join with each other to engage in collective opposition are less likely to find each other in the first place or, if they do, more likely to avoid the other as a likely agent provocateur of the regime. The Soviet authorities were always ready to create and spread mistrust, using unfounded rumors, forgeries, and disinformation as a weapon to divide hostile movements and states (as discussed in Chapter 8). It is tempting to conclude that a low-trust society was the intended outcome of Soviet policies and institutions.
302
But Hardin, Trust and Trustworthiness, condemns this approach as a “slippage” (55), and a “vernacular usage” allowing “implications that are wrong” (60); generalized trust is not trust, just “optimistic expectations about being able to build successful relationships with certain, perhaps numerous, others (although surely not with just anyone)” (62).
303
These questions take the form recommended by Organisation for Economic Co-operation and Development, Guidelines on Measuring Trust, 197. In each case the answers should be scaled from о (“Not at all”) to 10 (“Completely”).
304
On the external validity of such measures, see Kaiser and Oswald, “Scientific Value of Numerical Measures”; Diener, Inglehart, and Tay, “Theory and Validity of Life Satisfaction Scales.”
305
Many mechanisms can bring about cooperation, and trust is only one of them. Others discussed below include coercion or threats of violence, and rewards based on monitoring. Cook, Hardin, and Levi, Cooperation without Trust.
307
Alesina and Giuliano, “Culture and Institutions”; Algan and Cahuc, “Trust and Growth.”
308
Hosking, “Trust and Distrust in the USSR” (introducing a special issue of the Slavonic and East European review on the same theme), 1,25.
309
Shlapentokh, “Trust in Public Institutions in Russia”; McKee et al., “Do Citizens of the Former Soviet Union Trust State Institutions”; Kuchenkova, “Interpersonal Trust in Russian Society.”
310
Other trust categories were neither (trust nor distrust), not much, not at all, and don’t know. The number of Russian respondents was 1,961. Inglehart et al., World Values Survey: Round Two, 340.