Acts of Speech and States of Mind

Much of this post was originally drafted a couple years ago, so the personal stories described in here took place then. I’m publishing it now in part because the novella that inspired it—Ted Chiang’s Story of your Lifehas recently been made into a feature-length movie (Arrival). In some contexts, it might make sense to say that this post may contain spoilers for SOYL; in this particular one, that would be hilariously ironic. Even after reading this post, the story will be worth reading.

This post begins, like so many of mine, with a conversation with Jean, the founder of the Upstart Collaboratory, where she and I and others are practicing the extreme sport of human relating. Jean remarked that a conversation she’d had earlier that day had been really good, then noted that she’d already told me that.

I replied, “Well, yes, and it was meaningful to me that you said it again. On the most basic level, it implies that on some level you felt you hadn’t yet conveyed just how good the conversation had been.” Then I shared with her something I’d heard from Andrew Critch, at a CFAR workshop. (Quote is from memory)

If someone says “something” to you, then that doesn’t mean that “something” is true. It also doesn’t necessarily mean that that person believes that “something” is true. Incidentally, it also doesn’t necessarily mean that they think that you don’t already agree with them that “something”. It really just means that, at the moment they said it, it made sense to them to say “something”. To you.

This is something that might be taught in the context of manipulation or deception. It’s more general than that though. Often, people aren’t deliberately deceiving you, yet it’s still worth modelling their acts of speech as indeed, acts of speech. Rather than just parsing their utterances directly. The fact that someone said something communicates meaning beyond merely the something that they said.

Acts of Speech

We now turn to Story of my Life. The heptapods, an alien species that the narrator (a linguist) is interacting with, have a nonlinear, nonsequential view of time. This means that they actually have a broad awareness of past and future and present, all at once. They still progress through time, but they know what’s going to happen before it does.

If I could have described this to someone who didn’t already know, she might ask, if the heptapods already knew everything that they would ever say or hear, what was the point of their using language at all? A reasonable question. But language wasn’t only for communication: it was also a form of action. According to speech act theory, statements like “You’re under arrest,” “I christen this vessel,” or “I promise” were all performative: a speaker could perform the action only by uttering the words. For such acts, knowing what would be said didn’t change anything. Everyone at a wedding anticipated the words “I now pronounce you husband and wife,” but until the minister actually said them, the ceremony didn’t count. With performative language, saying equaled doing.

— Stories of your Life and Others, p138

During my conversation with Jean, we were joined by Amanda, another member of the community at the time. She had plans to go for a walk with Jean. Once she finished her miso soup, it was time for the walk. The practice of becoming comfortable saying “no” to requests has been present in our ecosystem for the last few days, so I put out, experimentally, the possibility of me joining them.

Jean deferred the request to Amanda, as I expected. Amanda considered it, then said lightly, “Yeah, I don’t think that makes sense, tonight. I have some patterns around that, almost with you as my little brother trying to steal mom’s attention. Not that I relate to Jean as mom, or you as brother, in general. But it shows up in this particular system, here.” I nodded, and she added, “but I’d be interested sometime in the future maybe, the three of us going for a walk together.”

I remarked, “that’s basically what I expected you’d say. And, I thought it would be cool to actually have that exchange, to have that practice of saying no, of hearing no, and of retaining a feeling of connection.” To Jean, I said, “And so in many ways this is like the thing with the aliens.” As we enacted the culture we’re trying to build, we also built trust in same-sidedness and being able to consciously navigate together.


At the Upstart Collaboratory, we’re never just having conversations. We’re also training ourselves to think differently. We’re a theatre troupe that isn’t just performing but also practising, which involves having the skill and mutual trust so that at any moment any of us can pull out the director’s chair and say “cut” and we go meta and start talking about the way in which we were just talking. This is a key part of being able to help each other level up in this way.

Meta-Communication director chair

The ability to shift into meta-communication is an important faculty to have, because it allows people to break out of double binds. A double-bind is the kind of damned-if-you-do, damned-if-you-don’t structure talked about in the famous novel Catch-22. If you know what to look for, double binds are readily found everywhere in interpersonal power dynamics, persisting because they’re traps (in a systems sense). Anthony Wilden talks about this extensively in Men and Women, War and Peace: The Strategist’s Companion.

Around page 111, he outlines eight “conditions necessary to a pathological double bind”, which include the obvious ones, about self-contradictory injunctions, and then notes that another key condition is

“a third contextual injunction preventing the victim from stepping outside the frame of the situation, an injunction preventing any attempt at comment, metacommunicative action, withdrawal, or escape. The victim thus cannot not respond to the double bind, but at the same time no appropriate response – within the context imposed – is possible.”

Of course, not all kinds of “going meta” are actually meaningfully stepping out of the context imposed. But part of what can make that more possible is having a shared larger context to step into. If you and your conversational partner (or business partner, or romantic partner) have the shared frame of double binds being a thing, and really wanting to be able to get on the same side about things, then it can be easier to make reference to that, because there’s already some common ground.

This is part of why I write blog posts like this: to try to articulate common frames for people to use with me and with each other.

And the particular shared frame that this blog post is trying to offer is:

Pay more attention to not just the meaning of what someone says, but the meaning of them having said it.

This applies to yourself as well, although as the following example shows, that can sometimes be quite uncomfortable.

Use of Language

This bit is adapted from a response to my friend Brienne’s facebook thread about using conscious manipulation of language to produce shifts in thinking.

In being hyper-aware of this stuff myself, I’ve also started noticing it in other people. Working full-time on a project to relate more effectively, I experience a lot of feedback on language usage.

One interesting thing I’ve noticed is a pattern where one person suggests a language/phrasing/framing change to someone else, and the first person kind of defends their original phrasing. Not as desirable per se, just as acceptable. What I think is noteworthy is given that what is being expressed is often very subtle, we need to clarify my language in order to really understand what I was trying to get at.

I’m having trouble thinking of an exact wording that has occurred, but it’s the structure that’s important so I’ll just abstract it:

  • Me: “So it’s like, we’re Xing to Y the Z”
  • Other: “I feel it’s more like, we find that our Zs are Ying Xing?”
  • Me, defensively: “Oh yeah, that’s what I meant by the thing I said.”

Note that the phrasing that I would have originally used in this story would have been one that would run counter to conscious language shifts (and thinking shifts) that both of us are trying to make. (See common-knowledge self-commitments for more on this practice.)

So one possibility is that I was just being careless with my language here. Another is that I was being careless with my thinking. But this means that if I want to get high-quality feedback on my thinking, I need to be really clear with my languaging. And I need to accept that if my brain is unaware enough to speak something in a frame I don’t want, then it’s very plausible that it’s going to be unaware enough to think that thought. And therefore the defense of “that’s what I meant” isn’t enough, because the fact that I said something else indicates that I couldn’t have been fully clear in meaning the thing I meant to mean.

Which isn’t a problem! It’s just a thing I need to notice if I’m going to be able to learn more. Behavior is patterned, so presumably this isn’t just an error in my phrasing but rather a bug in the system that is generating what I say. And what I say is valuable evidence about how I’m thinking. The defensiveness shows up because if I’m not feeling on-the-same-side with someone, it can naturally feel unsafe for them to know what’s going on in my head.

The states of mind that make an act more or less probable

Relatively early on in Harry Potter and the Methods of Rationality, Eliezer Yudkowsky’s version of Quirinius Quirrell points out to an agitated Harry who has just failed to conceal something he wanted to be a secret,

“The import of an act lies not in what that act resembles on the surface, Mr. Potter, but in the states of mind which make that act more or less probable.”

This is related to the earlier quote from Critch. In HPMOR, Quirrell is primarily talking about antagonistic, deliberate actions: one person trying to convince or deceive another. But it goes beyond this. High-quality communication in general is ultimately a matter of using what someone says (and how) to infer their state of mind.

This means that what we’re doing in each moment of communication is, fundamentally, revealing our state of mind to each other, and interpreting each others’ state of mind. Of course, how transparent this is can vary a lot; in many circumstances, people tend to maintain common ignorance about this, and also to maximally mask what they reveal.

But as I’ve gestured at in this post, in a context intended for learning, there’s a lot to be gained from being open about this. By consciously co-directing our attention to what we’re each noticing and learning about each others’ (and our own) states of mind from the communication we’re doing, we can develop increased attunement and higher-quality information flow and co-thinking.

Over time, this allows us to become less subject to the experience of communicating, which lets us get our old patterns of interaction out of the way so we can communicate and learn from each other more effectively.

Question for further reflection: you don’t quite experience the world itself, but your brain’s construction & interpretation of it, based on your assumptions etc. How might, therefore, what you experience someone else saying be not just direct evidence about their state of mind but also about yours?

If you found this thought-provoking, I invite you to subscribe:    
About Malcolm

Constantly consciously expanding the boundaries of thoughtspace and actionspace. Creator of Complice, a system for achieving your important goals.

Personal Website

Have your say!

Have your say!


Name *

Email *