Trust and sociability: on the limits of confidence and role expectations - Special Invited Issue: Money, Trust, Speculation and Social Justice - Part 1: Trust, Confidence, and Crime

American Journal of Economics and Sociology, The, Oct, 1998 by Adam B. Seligman

Vladimir Ilych Lenin is said to have remarked: "Vertraun ist gut, Kontrol noch besser" - trust is good, but control is much better. In this saying we find what I think is a distinction critical to any preliminary understanding of trust-that is, the distinction between trust and confidence (control in Lenin's terms). Control or confidence is what you have when you know what to expect in a situation; trust is what you need to maintain interaction if you do not.(1)

I

Trust and Confidence

Confidence, and the knowledge necessary to confidence can be based on many different things. It can be predicated on the ability to impose sanctions and the knowledge that one's partner to an interaction also knows that sanctions will be imposed if he or she fails to live up to the terms of an agreement. Note too, that sanctions may be formal or informal; they may be based on an intricate web of kinship obligations or on the verities of contract law. They may be immediate or intergenerational, symbolic or material. In all cases, confidence is based on knowledge that an interaction or exchange is set within a context, a system that, qua system will impose sanctions in the case of an abrogation of agreements (this is true whether agreements are based on principles of the market contract between free agents or on status responsibilities among kin group members). Hence when I say that I "trust" the doctor, I am not quite correct. Rather, I mean that I have confidence in her abilities, in the system that awarded her the degree on the wall (and I may have greater confidence if the degree is from Cornell University and less from a west-coast mail order address), as well as in the epistemological assumptions of American medicine. Of course I may also lack such confidence and take my daughter to Lourdes instead, or I may trust (have faith in) the Lord if, for instance, I am a Christian Scientist. Similarly, when Virginia Held says that she "trusts the plumber to do a non-subversive job of plumbing," she also is not quite correct.(2) For she knows that if he does a "subversive" job she will not only not hire him again, and tell her neighbors not to hire him, but she will also complain to the local better business bureau. She may even refuse to pay him. In short she can impose sanctions formal and informal. She knows this, he knows this, she knows that he knows, he knows that she knows that he knows, and so on. They enter into their interaction and exchange by mutual interest, and this interaction is maintained by mutual confidence in the system within which the exchange takes place. Now if she were to rush off to meet a colleague and had to leave her baby with the plumber until her husband came home, that would be a very different story, one involving both parties in a relationship of trust. I leave this potentiality for the moment, but I return to this aspect of trust at the end of this article.

Confidence, to reiterate, is predicated on knowledge of what will be. And this knowledge may in turn be based on the ability to impose sanctions. It may also be based on what we may term familiarity, what I like to call "stickball." Because John over there played stickball as a boy on East 13th street, he shares with me certain codes of conduct, certain moral evaluations, certain ways of being and acting that bring me to have confidence in him. We are alike, the same, and hence I can predict his actions. Knowledge of what will be, confidence and prediction, are here based not on sanctions but on sameness, on familiarity. Of course, the relevant other may not be "the same" at all; we will often draw certain conclusions from modes of dress or speech, where a person went to school, the neighborhood where the person lives or grew up, the person's religion, and so on, all of which allow us to construct a story, a narrative if you will, of sameness that allows us to have confidence. That we often are incorrect in our assumptions and fool ourselves in this matter is another thing completely. Often, indeed, we combine as many bases of confidence as we can before entering interaction: formal sanctions may be costly and involve too great transaction costs, so we like to know we can impose informal sanctions as well. Indeed, how often does conversation turn to place of origin, academic background, family, or even sports-all icons of familiarity, of ways to demonstrate some underlying sameness, to the other as well as to ourselves. We do this all the time, every day, in situations involving as little thought as choosing whom to sit next to on the bus or as much as what architect to employ in redesigning our house. It is the welt and weave of much of our public life.

Trust is something very different. Trust is needed when there is no basis for confidence, for example, when behavior cannot be predicted or when strangers are part of the interaction. Trust is necessary when the other is unknowable, when behavior cannot be imputed or predicted, because either a) there is no system within which sanctions can be imposed or b) there is no underlying sense of or terms of familiarity or sameness that would allow such prediction.