Defining Trust

“…’tis a tale told by an idiot, full of sound and fury, signifying nothing.”

Shakespeare, MacBeth


Note: This post comes out of ongoing discussions with Barbara Kimmel, CEO of of Trust Across America. She and I share a concern (as do many others) about how imprecision in speaking about trust hampers progress. It’s not an easy topic, but we both believe progress can be made. She’ll be writing about the subject soon as well.


  • “Trust in banking is down.”
  • “I don’t trust what car companies say – I trust someone like myself.”
  • “I trust Amazon, but not Google.”

We make statements like these every day in our casual conversations. We intend them to be – and believe them to be – meaningful. We think we’re saying something by uttering them.

And yet – each of those statements is confused, often to the point of meaninglessness. As Shakespeare would have put it, they are full of “sound and fury” –  while signifying very little.

The mistakes inherent in those statements are not just found in casual conversation. They are ubiquitous in the business and general press, and even, occasionally, in academic writings. The level of discussion about trust is fraught with definitional ambiguity pretty much everywhere.

Yet if we can’t talk meaningfully about trust, then we cannot possibly arrive at useful, justified conclusions for action. How can we create trust-based organizations? Cultures of trust? Increased trust in institutions? Have meaningful discussions about cross-generational trends in trust?

Without common definitions, we are reduced to bemoaning the fate of trust, wringing our hands as bystanders, accomplishing nothing. We need basic definitions.

This post doesn’t pretend to offer a comprehensive definition. But it does humbly attempt to provide three simple distinctions for use in talking about trust. All are obvious when pointed out, but they are not observed in practice.

Let’s call them:

a) the Grammar of trust,

b) the Objects of trust, and

c) the Actions of trust.

The Grammar of Trust: Trust as a Noun, a Verb, and an Adjective.

What does it mean to say, “Trust in banking is down?” Does it mean banks have become less trustworthy? Or does it mean public opinion is turning against banks? Or both?

It makes a difference – that is, if our discussions are to have any policy implications. This is loose language “signifying nothing,” unless we clarify what definition of trust we’re using.

  • To trust someone is to take a risk, to put yourself willingly in harm’s way of another. This is the verb, “to trust.” It’s what the psychologists focus on as a propensity to trust; it’s the entry point of business books like Bob Hurley’s The Decision to Trust.
  • Trustworthiness is an adjective – it’s an attribute we ascribe to others. It falls in the category of virtues. We use ‘trustworthy’ to describe people who we think reflect virtues like credibility, reliability, of high integrity, benevolent, un-self-preoccupied. It’s talked about in books like The Trusted Advisor as the Trust Equation.
  • Trust as a noun is the state of a relationship between two parties. It exists or it doesn’t; if it does, it is described as high or low, thick or thin, broad or deep. Sociologists use this to talk about high- or low-trust societies or cultures. In business, Edelman’s Trust Barometer primarily (when it is clear) focuses on the state of trust.

Violations of grammar.  When we see “Trust in banking is down,” we should immediately ask: which meaning of trust is being used here?

  • If we mean banks have become less trustworthy, this is trust as an adjective. If this is the issue, then what data is being used to define trustworthiness? And should we seek industry-based or regulatory-based solutions to the issue?
  • If we mean that people have become less inclined to trust financial institutions, this is trust as a verb. If this is the problem, is it unique to banks? Or is it part of a general decline in propensity to trust? What kind of social intervention are appropriate – industry associations? Public relations campaigns? Awareness and reach-out initiatives?
  • Or do we strictly intend just to indicate a decline in the state of trust? This is trust as a noun. It is something we can track over time; but It should always beg the question, why? What have been the patterns of trustworthiness, and the patterns of propensity to trust? What is driving the state of trust lower?

If you’re not persuaded that this is a meaningful issue, consider the national (US) debate on violent crime. By most indicators, the incidence of violent crime over the last few decades is down. And yet the fear of crime is up. This is a case where the verb (to fear) is unlinked to the adjective (highly criminalized). If we don’t correctly identify the problem, we will continue to fix a “problem” (violent crime) which is not the primary driver of fear.

Objects of Trust: Personal vs. Institutional

“I don’t trust car companies – I trust someone like myself.”  It may seem obvious that trusting a person is not the same as trusting an institution – Citizens United notwithstanding – but the difference is often blurred.  We’re not confused by, “I trust FedEx to deliver my packages, but not to babysit my daughter,” because baby-sitting requires an individual, not a firm, and we don’t think of FedEx delivery people as being in the baby-sitting business anyway. Trusting people is fundamentally different from trusting organizations.

This may sound obvious, but major trust surveys, e.g. the Edelman Trust Barometer, say things like “trust in someone like me” is trending up vs. “trust in government” or “trust in companies.” This is a category mistake. The two types of trust are qualitatively distinct; they do not belong on the same quantitative scale. The blurring of lines is similar to that of “friends” on Facebook – we use the same word to describe our digital tribes that we use to describe our neighbors and old college buddies. The common language use must be recognized and respected – but it doesn’t mean the meanings are the same.

Former Speaker of the House Tip O’Neill famously said, “All politics is local.” In a similar way, most trust is personal. If FedEx misses two deliveries in a week, my “trust” in them is seriously eroded. Yet if my best friend fails to return two calls, I am perplexed – but my trust in them is barely affected. This is not surprising – it’s not the same trust that we’re talking about.

Trust in particular organizations – companies, congress – is “thin” trust. It’s connected to branding, reliability, reputation – but not to the more powerful personal attributes we associate with trusting individuals. Most people “distrust” congress, at the same time they’re more inclined to “trust” their individual congressperson. This is only surprising if we think the same ‘trust’ is at issue.

Companies that consistently score high on broad measures of trust (see for example, Trust Across America’s Most Trustworthy Companies) are usually, on closer examination, companies that assiduously foster trust-based relationships between individuals – between employees and customers, among employees, with local constituent organizations.

Sloppy use of the object of trust – anthropomorphizing trust when we talk about institutions, for example – should be avoided by writers, and sharply pointed out by readers. The word “trusted” means very different things when applied to Toyota, to my LinkedIn affinity group, and to my next-door neighbor. I may ‘trust’ them all, but we are talking about quite different phenomena.

Actions of Trust: Trust to Do What?

I may trust my dog with my life – but not with my ham sandwich. We all get the difference – and yet we see sentences like, “I trust Amazon – but not Google.” The Amazon/Google difference is probably the same as the life/ham sandwich difference – but we don’t usually hear it the same way.

To see why, just ask what it is that we are trusting Amazon and Google to do? Most likely, the utterer of that sentence means that Amazon delivers fast and reliably, and that Google tracks mountains of information about us. Fast delivery and responsible guardianship of private information are very different things – maybe as different as “life” and “sandwich.”  And yet we act as if we’re making a meaningful statement about corporate trustworthiness when we use the “T” word with both companies in the same sentence. We are not – we are expressing distinct opinions about two very different phenomena.

Whenever you read (or write) something comparing levels of trust – whether it’s between people, or organizations (or across people and organizations), always remember to ask – trust to do what?


There are other definition issues of trust – for example the general propensity to trust strangers vs. the more specific (and variable) trust in particular institutions or individuals. As Eric Uslaner says, “If you punch me in the face, my trust in humanity is un-diminished – but you are and I are finished!”).  But if we just more critical readers (and writers) about the above three distinctions, the discussion of trust would be greatly advanced.