Is trust...

Please explain / justify / argue / query in responses. Boosts appreciated.

NB: I'll answer direct clarifying questions, but (hope to) avoid responding to suggestions other than trying to tease more out of them until this poll closes (in 3 days).

Though I'll throw in a teaser: do any nonhuman entities exhibit trust or trust-adjacent patterns?

Answering the poll (see parent) above: what is ?

There's a neighborhood cat who's been lurking about the property for the past few years. Initially exceedingly skittish, she now flops over and rolls on the ground, or trots toward me, when I call.

Is this "trust"?

If so, how is it that animals have "trust"? (And how do they respond when trust is betrayed?)

Is trust some sort of emotional response? Charles Darwin rather famously explored emotions:


But evidence would suggest that the *behaviours* associated with trust -- extend to animals.

Let's backtrack a bit.

I've been using the general definition of trust _as a belief_ concept -- that is, as a state of knowledge, as "belief extending beyond the limits of immediate evidence". That's contrasted with, say, "blind faith", which is belief in the absence *or opposition* to evidence. Or various forms of "revealed knowledge" (generally religious).

Trust is essential in that ...


... our knowledge is _of necesity_ limited, finite, and incomplete. Trust is a form of informational credit that lets us move from one ground state to another. Not necessarily the only one: we could, say, rely on blind chance or random sortition, and often do. But trust _leverages_ an existing knowledge base.

Etymologies are often instructive (though not, as is often noted, definitive). The ultimate PIE root of "trust" is the PIE "*deru-", "to be firm, solid".


... which forms all or part of: betroth; Dante; dendrite; dendro-; dendrochronology; dour; Druid; drupe; dryad; dura mater; durable; durance; duration; duress; during; durum; endure; hamadryad; indurate; obdurate; perdurable; philodendron; rhododendron; shelter; tar (n.1) "viscous liquid;" tray; tree; trig (adj.) "smart, trim;" trim; troth; trough; trow; truce; true; trust; truth; tryst.*deru-

And again, I'm a huge fan of

Backtracing roots is fascinating.


This case of "trust" having a root in a word that's intensely _physical_ is an interesting characteristic of a large number of words concerning information, knowledge, and truth, by the way, which utterly captivates me. We _inform_, _grasp_, and _understand_. A thing that is correct is _right_, in the handedness sense. I could go on for days....

Back to trust: thinking about my friend the cat, I asked, "is trust an emotion?" And, well, you can Quack that:


And I'm definitely not the first person to ask that question or have the idea.

It of course gets to a larger set of questions:

- What is an emotion?
- What is "trust"?
- Does "trust" have multiple components?

To which the answers are, generally, "yes".

If an emotion is "a mental state associated with the nervous system brought on by chemical changes ... [with] a degree of pleasure or displeasure"...

then if we can show some chemical / endocrinal elements ...


... we've got the fundamentals of an emotional response.

And given the counterpoints of _trust_ and _suspicion_ or _apprehensiveness_ (which the cat also exibits, particularly toward others), there's definitely an attraction / repulsion element at play.

From an evolutionary perspective, some innate wiring for trust and wariness also has clear potential benefits, particularly for social (or in the case of cats, quasi-social) animals. We can never know _for certain_ how others will act.


But we can make _informed judgements_ based on experience and belief.

I'd also asked if non-human entities could experience trust. Animals are one case, and "emotion" fits this. Institutions and organisations are another, and there's a whole world of financial, business, and other constructs of humans + other elements which express of trust. Which means the concept transcends just living things.

@syndikalista suggested "trust is a relation between actors":


That's a fascinating answer (and made me wish I'd included "a relationship" on the poll), though ultimately limiting: I can have trust in a situation or entirely non-volitional system (a stone, a piece of wood, a beam, water, heat), which is a non-actor. Trust as a belief-relation of _or_ between actor(s) could have merits though.

@sydneyfalk suggests trust is a mistake:

And further: trust is accepting that betrayal is a possible outcome:


Two other responses of note, says trust is "a choice"

And @yojimbo: "Trust does not have to be absolute, it can be conditional, and you don't have to extend it. Therefore it isn't a state of knowledge."

Both are interesting, though I disagree.

Most of the time we have _no_ choice but to trust, the alternative being utter decisional paralysis or random action. Knowledge is inherently conditional, incomplete, and probabalistic.


Put another way, trust isn't so much a _form_ of knowledge as an _aspect_ of it. To be absolutely certain of a thing is rare enough to be notable (or pathological). We say "I think X...", often with the meaning of "I believe that X is likely...". The uncertainty of our knowledge is an inseperable characteristic.

This brings up a few other points.

I see "knowledge" as better described as "mental models". Those may be of discrete facts, of interactions or dynamics, or of narratives.


Inherent in the notion of a model is that it is a _representation_ or _symbol_ of a thing, and an _approximation_. "All models are wrong, some models are useful."

When we say we _know_ a thing, what we mean is that we have _some model_ of it, with _some range of utility_. It may be recognition, it may be use or interaction, it may be predictability.

The cat, for example has some cognitive model of me, including elements of place (where I'm likely to be encountered), sound and ...


~@dredmorbius @feonixrift All of the above. It's a state of knowledge giving rise to a behaviour linked to an associated emotion. One learns that trust is safe for a given situation, and then performs trust-doing, and thus experiences trustingness.

@dredmorbius Trust does not have to be absolute, it can be conditional, and you don't have to extend it. Therefore it isn't a state of knowledge.

I can distrust someone, as an emotional or knowledge state, but still act as if I trust them, therefore trust (as far as a single action goes) is a behaviour.

@dredmorbius I think so, that's pretty much the definition of knowledge I seem to be using. It doesn't have to be correct of course.

@dredmorbius i would argue, trust is a relation between actors. It drives from behavior and emotions which it also interacts and lifts up. From that knowledge can emerge, but that’s only a side effect.


I don't see the option for "mistake". Should that be 'other', or...? 😉


I would say that trust is an *expectation* of future actions -a state of *belief*.

I don't know if I'd consider a belief or expectation to be either 'knowledge', 'behaviour' or 'emotion'. I'd say it's something different from either of those. But I'm not a philosopher.

I would say that in my opinion/model, beliefs are FORMED BY knowledge (but not perfectly or immediately), and INFLUENCE both behaviour AND emotion (but again, neither perfectly nor immediately, and differently).


But, one might perhaps define all knowledge to be actually just various beliefs, and if so so, then maybe trust is just a kind of knowledge. It's something we think we know about the world. That's why it's so shocking to us when we feel betrayed - something important that we thought we knew, turns out to have been wrong.

We trust that the ground will not give way under us when we walk or build on it. That trust is a kind of learned knowledge. In an earthquake, that changes.


And in a similar way, we learn through experience, same way as we learn any other knowlege, about social conventions and what we can expect/trust other people to do in certain situations. When they don't do that - either because we moved cultures, or our culture changed, or our 'culture' was an unrepresentative bubble, or we hit an outlier - our learned knowledge turns out to be wrong, and behaviour and emotional changes result from that realisation.


Case study in cultural shock and trust issues with emotional and behavioural disturbances resulting from that: All Of The Internet Today.

@dredmorbius I would separate *trust* (the perception/belief/expectation of good future outcomes) from *commitment* (the choice to take actions that require one's evaluation of trust to be correct).

You can commit without trusting (ie: take a perceived risk), and you can trust without committing (ie: not have the opportunity to do something that you think is safe). You can also commit various levels of resources, it's not aways all/nothing.

Commitment is a behaviour, while trust is a belief.


But in a normal human cybernetic system, there isn't going to be a hard boundary between belief and behaviour. Behaviour will generally follow our beliefs, and our beliefs will generally follow our experiences which will follow our behaviour, and so one is often used as a proxy for the other.

That's why the word 'trust' in English often refers to deliberate choice to make a commitment as much as it does to a passive state of belief or the emotion associated with that belief.

@natecull Is knowledge the same as, different from, or intertwingled with, models?

@dredmorbius I would say all knowledge (or rather, belief - judging whether a belief is 'true knowledge' is an Extremely Hard Problem) is a model, and also a prediction, yes. I don't think we have the capability to have knowledge that is *not* in the form of a model. Everything we know is represented in the form of something else we know.

But a lot of actual human knowledge is probably not necessarily capturable as *formal* models in any particular formal sense.

It's intrinsically not knowledge-based (see e.g. discussion in book by Shoshana Zuboff) but rather a conscious or unconscious state of mind/decision leading to/visible through corresponding behavior

@silmathoron Which of Zuboff's books? The one from this year, IIRC, "Surveillance Capitalism", or an earlier work?

Surveillance Capitalism, chapter 20 or 11 I think, close to the discussion on Skinner and behaviourism

@dredmorbius all the above; you use your knowledge of a situation to influence your behavior and we sometimes take these actions consciously but often just feel them as emotions.

@penguin42 Is there a functional breakdown that might help clarify this?

Or some other dissection/analysis?

@dredmorbius heck I don't know! My PhD is in computers not humans, so for me trust is probably much more vague!

@penguin42 You wouldn't be alone. Numerous papers discussing trust start of noting that the concept is ... frustratingly vague.

@dredmorbius I would argue all three - assuming free will anyway. I would say trust is a state of knowledge based upon the statistical probability of the other sides behaviour and goals. It's a behaviour predicated on that and it's an emotion as a bunch of chemical wiring to evaluate all the above.
More interesting for your nonhumans - when ants build chains to cross gaps and rivers do they show trust or are they slaves or both ?

I'm with @natecull - trust is an expectation of future interactions. But I like the idea that our emotional system is a means of evaluating and communicating such things.

I wouldn't call expectation knowledge, except with great care, because it's only probabilistic, and it knows that.

I do like the separation upthread of performative trust vs felt trust. And the idea that trust might be unacted on.

@EtchedPixels @dredmorbius

@EdS My phrasing of a _state_ of knowledge might point to a range of possible knowledge states / levels / formulations.

@natecull 's comments are quite good, I'll allow.

@natecull @EtchedPixels


my first statement was more of my usual lashing out at the world, my apologies

but I stand by the second. do I "trust" each person I walk by not to literally stab me in the back? in the sense that I generally assume people won't, sure.

but when people have motives, trust gets more complicated. and all people have motives, so trust is relative and complex at best.

(I don't trust many people very much, but that's probably obvious)

@sydneyfalk Erm ... trust me ... I read both as such.

The first was clearly tongue-in-cheek in a ha-ha-only-serious sense, the second carries the main dish: the verso of trust is betrayal. And since trust is an extension, a risk, an forray into the realm of ignorance, and probabalistic, there _will_ be betrayal.

Some of the time.


I assumed, but wanted to make sure I was being clear. There have been times in my life that I absolutely considered trust a mistake, and fortunately those times are in the past. ^_^


I kind of suspected we both had a subscription really, yeah :/ sorry, though.

@sydneyfalk It's engendered caution on my part. Also inspired me to look into the phenomenon itself and the what / how / why of it.


I assume I can't really know, and make my best guess. I don't think I can really grasp trust as a concept.

But that's pretty much true of every aspect of my life, really. XD

@sydneyfalk There's probably a reason that the focus of this little tootstorm is my relationship with a cat ;-)

The affection and devotedness of animals is unlike that of most humans. Near total and unconditional.

Though this particular cat _is_ known to nip when annoyed.


> Though this particular cat _is_ known to nip when annoyed.

Everything has limitations, even unlimited things. 😜

@sydneyfalk My working theory is that cats like to tussel, and a strong sign of trust is actually when they'll do this either with limited damage or appropriate protection.

I've an old glove I'd wear with a former cat who would attack that with relish while I'd pet them with my other (ungloved) hand.

The untrusting cat either doesn't engage (fearing retribution) or attacks in fear/anger when there's no glove. Knowing when you can safely push boundaries is part of the trust relation.

Sign in to participate in the conversation

Everyone is welcome as long as you follow our code of conduct! Thank you. is maintained by Sujitech, LLC.