[Intuitive self-models] 4. Trance

post by Steven Byrnes (steve2152) · 2024-10-08T13:30:41.446Z · LW · GW · 7 comments

Contents

  4.1 Post summary / Table of contents
  4.2 Background context: The subtle art of changing an intuitive model
    4.2.1 Back to bistable perception
    4.2.2 Shifting intuitive models is like herding cats: hard but possible
  4.3 Background notation: S(⋯) and D(⋯)
  4.4 What is a “trance”?
    4.4.1 Two properties related to “trance”: a 2×2 square
      4.4.1.1 Property 1 but not Property 2: Flow states
      4.4.1.2 Property 2 but not Property 1: “Lucid trance” and related
      4.4.1.3 Both Property 1 and Property 2: Deep trance, sometimes involving amnesia
    4.4.2 What does “irreconcilably different” mean in Property 2?
  4.5 Explaining various tricks that help start and maintain trance
    4.5.1 Trick 1: Hold in mind beliefs that are incompatible with your mental model of the homunculus
    4.5.2 Trick 2: Avoid any action A that would normally be conceptualized as being caused by the homunculus
    4.5.3 Trick 3: Seek out perceptual illusions where the most salient intuitive explanation of what’s happening is that a different agent is the direct cause of motor or attentional actions
    4.5.4 Trick 4: Find a “possessor” who you see as high status (a leader that you want to follow)
      4.5.4.1 Trick 4A: Indirect version—the high-status person is not the actual “possessor”
  4.6 Explaining properties of trance
    4.6.1 Trance Property 1 causes “losing track of time”
    4.6.2 “Deep” trance with both properties 1 & 2 can be associated with amnesia
    4.6.3 Possessions are often child-like
      4.6.3.1 Learning to talk from scratch
      4.6.3.2 Other examples beyond speech
  4.7 Conclusion
None
7 comments

4.1 Post summary / Table of contents

Part of the Intuitive Self-Models series [? · GW].

“Trance” is an umbrella term for various states of consciousness in which “you lose yourself”, somehow. The first kind that I learned about was hypnotic trance, as depicted in the media:

With examples like that, I quite naturally assumed that hypnotism was fictional.

Other types of trance, particularly “spirit possession” in traditional cultures (e.g. Haitian Vodou), and New Age “channeling”, initially struck me as equally fictional—especially the wild claim that people would “wake up” from their hypnotic or other trance with no memory of what just happened. But when I looked into it a bit more, I found myself believing that these are indeed real experiences, even if I couldn’t explain them.

(…Not veridical experiences! Obviously the New Age “channelers” are not literally “receiving information from paranormal sources”! But real experiences. See §1.3.1 [LW · GW]–1.3.2 [LW · GW] for the difference.)

Meanwhile, the term “trance” is also applied to the more down-to-earth experience of “losing oneself” in drugs, music, slot machines, or (of course) writing good code. Those seemed less weird to me. But I didn’t see how they had anything to do with the more exotic amnesia-inducing types of trance above.

Then finally I read the “Masks and Trance” chapter of Keith Johnstone’s Impro: Improvisation and the Theatre (1979). Not only did he suggest themes and patterns connecting the various types of trance, but he also drew on his many years of personal experience figuring out how to induce trance states in students taking his acting classes. His own trance-induction technique centered around having the students wear masks, look at themselves in the mirror, and thus somehow “become” the mask. He elaborated on that core technique with various other supporting tricks, which I’ll reverse-engineer below. Anyway, that book chapter gave me lots of nuts-and-bolts details to chew on, enough that everything about trance started to click into place.

Thanks to everything we’ve learned in the previous three posts of this series under our belts, I claim that we’re in a great position to understand all the different types of trance. Exactly what are they? What do they have in common? Why do they happen? What makes them tick?

The rest of the post is organized as follows:

• Section 4.2–4.3 give some background, particularly on how we should think about deliberately changing intuitive self-models. Trance induction entails deliberately switching out of the Conventional Intuitive Self-Model (§3.2 [LW · GW]), just as learning physics entails deliberately switching out of the naïve intuitive physics model. In each case, it’s not enough to intellectually want to shift to a different intuitive model. Actually making that happen can be very tricky! There’s a whole art to it.
• Section 4.4 answers the burning question: what is a trance? I think there are actually two related properties that define different types of trance, leading to a 2×2 square: “everyday life”, “flow states”, “lucid trance”, and “deep trance”.
• Section 4.5 goes over lots of different tricks that are used to induce trance, explaining why they work, in terms of intuitive self-models.
• Section 4.6 then goes over some properties of trance states, including “losing track of time”, amnesia, and the frequent phenomenon of “possessed” people acting like impulsive children.

(Sources: Just about everything that I know about trance phenomena in the real world, I learned from that one book chapter by Johnstone. I find him a trustworthy source in general, but you should accordingly take these claims with a grain of salt, and please share in the comments if I’m missing anything important!)

4.2 Background context: The subtle art of changing an intuitive model

4.2.1 Back to bistable perception

Back in §1.2 [LW · GW], I talked about the Spinning Dancer video, which is bistably compatible with either an intuitive model where she spins clockwise, or an intuitive model where she spins counterclockwise.

…But then I mentioned that for some reason, it’s very hard for me to get her to seem to spin clockwise. Every time I look at her, she seems to be spinning counterclockwise.

Suppose I really want her to be spinning clockwise in my mind. What might I do? There’s a comment thread where we were exchanging tips and tricks [LW(p) · GW(p)]. Put some other spinning thing nearby in your field of view? Rotate the screen then flip it back? Use your peripheral vision? You get the idea.

Using one of those tricks, I managed to make the dancer spin clockwise for a few seconds! Until I blinked and lost it. But with a bit more time and effort, I’m sure I could get the Spinning Dancer to flip at will.

4.2.2 Shifting intuitive models is like herding cats: hard but possible

What do we learn from my dancer example?

On the one hand, shifting intuitive models is surprisingly hard! You can’t necessarily just want to have a particular intuitive model, and voluntarily make that happen. It’s more like herding cats—using careful control of attention, actions, and sensory input environment, in order to exert weak and indirect influence on a complex low-level process.

And I should hardly be complaining about how hard it is for me to learn to control my intuitive model of the Spinning Dancer … when there are meditators who spend thousands of hours trying to achieve certain shifts in their intuitive models of their own minds.

But on the other hand, intuitive models can be manipulated with the right techniques. It’s a skill issue.

Of course, people don’t need to understand why intuitive-model-shifting techniques work, in order to use them. They can be discovered by trial-and-error, and then passed around via culture. But still, I want to know why the techniques work. Entering a trance is an intuitive self-model shift, and there are techniques that tend to make it happen. How do they work? Explaining that will be a major theme of this post. But first I need to say what a trance is! One more bit of background first, though:

4.3 Background notation: S(⋯) and D(⋯)

As a reminder from §2.2.3 [LW · GW], “S(apple)” is short for “the concept of apple in a self-reflective frame”, i.e. the self-reflective intuitive model where we’re envisioning the apple as the occupant of conscious awareness. S(apple) is how we think about the possibility of the apple being in awareness, and thus it’s different from the apple actually being in awareness—in other words, S(apple) ≠ apple. Instead, S(apple) involves a frame (in the sense of “frame semantics” in linguistics or “frame languages” in GOFAI) where the “awareness” concept and the “apple” concept are connected into an abstract container / containee relationship.

Also as a reminder from §2.6 [LW · GW], for any action A (either attention-control or motor-control), there’s a common rapid sequence of two consecutive thoughts [S(A) ; A], where more specifically S(A) has positive valence. This sequence is conceptualized as the “intentional” execution of action A, as “an exercise of free will”. This two-step sequence is itself encapsulated as an intuitive model, which I call “D(A)”, short for “deciding to do A”.

And then §3.5 [LW · GW] added the other half of how we conceptualize this sequence. We think of S(A) as having positive valence because the homunculus wants A to happen right now; and then we think of A actually happening because the homunculus did it.

In this post, it’s not always the homunculus that does it. So I’ll need a new ingredient in the notation: “S(X←Z)” and “D(X←Z)”, where Z is the agent that is causing these things to happen via its vitalistic force (§3.3 [LW · GW]). So in the previous posts, S(X) and D(X) are always short for S(X←homunculus) and D(X←homunculus) respectively. But in this post, we’ll see examples where Z is the hypnotist, or spirit, or mask, or character, etc. You could read D(A←Z) as “Z deciding to do action A”.

4.4 What is a “trance”?

Finally, the section we’ve been waiting for!

I don’t think there’s a single standard definition of “trance”, but I think I came up with a nice way to classify the space of trance-related phenomena.

4.4.1 Two properties related to “trance”: a 2×2 square

The word “trance” seems to be used in inconsistent ways. I propose the following definitions:

• TRANCE PROPERTY 1: There are NOT any S(A←homunculus) thoughts, at all, for an extended period.
• TRANCE PROPERTY 2: There are (attention-control and/or motor-control) actions A happening, of a type which under normal circumstances would be conceptualized as D(A←homunculus) (i.e., the rapid sequence [S(A←homunculus) ; A]), but they are instead conceptualized in an irreconcilably different way.

These two properties are obviously not totally independent—for example, when trance property 1 holds, that kinda “opens up space” for an alternative conceptualization (property 2). But the two properties can also come apart. In fact, they fill out all four quadrants of a 2×2 square as follows:

The top-left (“everyday life”) is self-explanatory; let’s talk about the other three:

4.4.1.1 Property 1 but not Property 2: Flow states

Think of getting lost in an engrossing film, or book, or videogame, or slot machine, or dance, or engineering trade space analysis. Or running for your life from an angry bear.

These all have something in common—you “lose yourself” in it, and don’t realize how much time has passed. I’ll explain why later (§4.6).

4.4.1.2 Property 2 but not Property 1: “Lucid trance” and related

In these cases, the homunculus is sometimes conceptualized as taking actions (either motor control, attention control, or both), but some non-homunculus entity is also conceptualized as taking actions.

My main example is “lucid trance”. There, the homunculus is conceptualized as taking attention-control actions from time to time (popping in to “observe” what’s happening), but takes few if any motor actions.

Here’s an example where Impro talks about lucid trance:

Many actors report ‘split’ states of consciousness, or amnesias; they speak of their body acting automatically, or as being inhabited by the character they are playing.

Fanny Kemble:^[1] ‘The curious part of acting, to me, is the sort of double process which the mind carries on at once, the combined operation of one’s faculties, so to speak, in diametrically opposite directions; for instance, in that very last scene of Mrs Beverley, while I was half dead with crying in the midst of real grief, created by an entirely unreal cause, I perceived that my tears were falling like rain all over my silk dress, and spoiling it; and I calculated and measured most accurately the space that my father would require to fall in, and moved myself and my train accordingly in the midst of the anguish I was to feign, and absolutely did endure.’ (William Archer, Masks and Faces, 1988.)

In Kemble’s description there, you can see how her normal homunculus is taking sporadic attention and motor actions to ensure proper stage blocking, while meanwhile her in-character homunculus is outputting most of her actions and dialog.

A different example is a “tulpa”, which Wikipedia describes (in its modern secular form) as “a type of willed imaginary friend which practitioners consider to be sentient and relatively independent”. Here, there’s more of an even split, where the homunculus is conceptualized as taking some attention-control and motor-control actions, but some other entity is also conceptualized as taking some attention-control actions. To be clear, I think tulpas are not normally described as a form of lucid trance. But both of my tulpamancer friends report that their tulpa can execute motor-control actions too, and if that’s not a lucid trance, at least it seems to have an awful lot in common with a lucid trance.^[2]

4.4.1.3 Both Property 1 and Property 2: Deep trance, sometimes involving amnesia

This category includes the most stereotypical “real” trance states, including deep hypnotism, intense possession ceremonies in traditional cultures, New Age “channeling”, and so on. Their most startling property is that they often (or maybe always?) seem to come along with amnesia, for reasons I’ll get to in §4.6.2 below.

4.4.2 What does “irreconcilably different” mean in Property 2?

Here’s another illustrative discussion from Impro, in the context of classes where Johnstone tries to get his students to feel “possessed” by a mask.

The problem is not one of getting the students to experience the ‘presence’ of another personality — almost everyone gets a strong kick from their reflection [wearing a mask] — the difficulty lies in stopping the student from making the change ‘himself’. There’s no reason for the student to start ‘thinking’ when he already ‘knows’ intuitively exactly what sort of creature he is.

To explain what I think he’s getting at: In the brain’s intuitive models, concepts can be stretched to some extent, but eventually break and need to be tossed out and replaced by a different, incompatible concept.

For example, compare (A) a stereotypical baseball, (B) a squashed and malformed (but still recognizable) baseball, (C) a carrot. It’s not just that these get progressively less baseball-ish. Instead, the intuitive “baseball” concept flips off entirely at some point between (B) and (C): (C) is just simply not a baseball at all, in your intuitive conception. If I mumble the word “baseball”, it constitutes evidence against the possibility that I’m talking about (C), whereas it would be evidence for the possibility that I’m talking about (B) or (A).

Back to trance. In the quote above, I claim that Johnstone is suggesting that, when untrained students put on the mask, they often construct a mental model that maintains the homunculus as active, but somehow stretches / modifies / patches it to better fit the current situation. And that’s disappointing to Johnstone—it’s not what he’s going for in the class. Instead, he works hard to find techniques that will lead to a conceptualization where the cause of the actions does not involve the homunculus concept at all, but rather an entirely different intuitive concept.

And that’s what I mean when I say “irreconcilably different” in Property 2.

4.5 Explaining various tricks that help start and maintain trance

4.5.1 Trick 1: Hold in mind beliefs that are incompatible with your mental model of the homunculus

The S(A←homunculus) thought has various properties / associations. One of them is a spatial location—typically the head, as discussed in §3.7 [LW · GW]. Another is the strong association between the homunculus and your concept of the body which the homunculus manipulates, feels, and occupies (cf. “body schema”).

So if you can hold incompatible properties / associations as very active in your head, it prevents the S(A←homunculus) model from arising. It’s analogous to how you can prevent yourself from thinking about a tune by holding a different tune in your head, or how you can prevent yourself from thinking about anything by paying sufficiently rapt attention to your breath, or how strong feelings of anxiety about failure make it very difficult to even imagine the possibility of success.

Suppressing  S(A←homunculus) thoughts directly helps Trance Property 1 and indirectly helps Trance Property 2.

Examples / discussion in Impro:

(We already saw the first part of this quote in §3.7.1 [LW · GW].)

The placing of the personality in a particular part of the body is cultural. Most Europeans place themselves in the head, because they have been taught that they are the brain. In reality of course the brain can’t feel the concave of the skull, and if we believed with Lucretius that the brain was an organ for cooling the blood, we would place ourselves somewhere else. The Greeks and Romans were in the chest, the Japanese a hand’s breadth below the navel, Witla Indians in the whole body, and even outside it. We only imagine ourselves as ‘somewhere’.

Meditation teachers in the East have asked their students to practise placing the mind in different parts of the body, or in the Universe, as a means of inducing trance.… Michael Chekhov, a distinguished acting teacher…suggested that students should practise moving the mind around as an aid to character work. He suggested that they should invent ‘imaginary bodies’ and operate them from ‘imaginary centres’. He write:

‘You are going to imagine that in the same space you occupy with your own, real body there exists another body—the imaginary body of your character . . . you clothe yourself, as it were, with this body; you put it on like a garment. What will be the result of this “masquerade”? After a while (or perhaps in a flash!) you will begin to feel and think of yourself as another person . . .

‘Your whole being, psychologically and physically, will be changed—I would not hesitate to say even possessed—by the character . . . your reasoning mind, however skilful it may be, is apt to leave you cold and passive, whereas the imaginary body has the power to appeal directly to your will and feelings.’ …

…

Someone wears a boiler suit stuffed with balloons to make him ‘huge’. He still looks ‘himself’. I say, ‘Move and imagine that the costume is your body surface’, and suddenly he becomes a ‘fat man’.

Pretending that the costume is the actual body surface has a powerful transforming effect on most people. We all of us have a ‘body image’ which may not be at all the same as our actual body. Some people imagine themselves as a blob with bits sticking out, and others have a finely articulated body image. Sometimes a person who has slimmed will still have, visibly, a ‘fat’ body image.

Likewise, insofar as the homunculus concept is strongly associated with a certain facial appearance, seeing the wrong face in a mirror is incompatible with that concept. Hence the idea of wearing a mask and looking in the mirror:

I then say: ‘Relax. Don’t think of anything. When I show you the mirror, make your mouth fit the Mask and hold it so that the mouth and the Mask make one face.…’

(This mask technique also uses Tricks 2, 3, and 4A below.)

4.5.2 Trick 2: Avoid any action A that would normally be conceptualized as being caused by the homunculus

(These include “deliberate” motor actions and “deliberate” attention control. Spontaneous / impulsive actions can be OK, see §3.5.2 [LW · GW] in the previous post and §4.6.3 below.)

How does this work? When such an A arises, if we self-reflect on what’s happening, we by default explain it as D(A←homunculus), which is counterproductive for any type of trance.

Examples from Impro:

If you lie down and make your body relax, going through it from feet to head, and loosening any points of tension that you find, then you easily float away into fantasy. The substance and shape of your body seem to change. You feel as if the air is breathing you, rather than you breathing the air, and the rhythm is slow and smooth like a great tide. It’s very easy to lose yourself, but if you feel the presence of a hostile person in the room you break this trance, seizing hold of the musculature, and becoming ‘yourself’ once more.

Meditators use stillness as a means of inducing trance. So do present-day hypnotists. The subject doesn’t have to be told to be still, he knows intuitively not to assert control of his body by picking his nose of rapping his feet. …

Many ways of entering trance involve interfering with verbalisation. Repetitive singing or chanting are effective, or holding the mind on to single words; such techniques are often thought of as ‘Oriental’, but they’re universal.

…

‘Don’t have any words in your head’, I say. … I may say ‘When you look in the mirror let the Mask make a sound, and keep the sound going all through the scene.’ This is a meditation technique very effective in blocking verbalisation (like Tibetan monks chanting ‘Oooooommmmm’). …

…

Trance states are likely whenever you abandon control of the musculature. Many people can get an incredible ‘high’ from being moved about while they remain relaxed. Pass them round a circle, lift them, and (especially) roll them about on a soft surface. For some people it’s very liberating, but the movers have to be skilled.

Here’s something vaguely related in Kaj Sotala’s post about Buddhist no-self [LW · GW]:

The opposite strategy is commonly associated with what are so-called nondual techniques. Instead of training an analytical, attention-controlled part of the mind to examine the sense of self, the nondual route is to nudge the mind into a state where those analytical parts of the brain become less active. As those parts also produce the sense of ‘the observer’ in the first place, attenuating their activity can offer a glimpse into a state of consciousness where that sensation is lacking. Some versions of this approach seem to be tapping into some of the same machinery which causes people to experience a state of flow, as flow states also seem to involve a downregulation in both analytical thought and the sense of self.

4.5.3 Trick 3: Seek out perceptual illusions where the most salient intuitive explanation of what’s happening is that a different agent is the direct cause of motor or attentional actions

How does this work? It’s self-explanatory. Recall, Trance Property 2 entails getting lots of D(A←Z) thoughts into your head, where Z is NOT the homunculus. These kinds of perceptual illusions are a great way to increase the intuitive salience and plausibility of those kinds of models.

Some examples from Impro:

I once asked a girl to close her eyes while I put a coin under one of three cups. Secretly I put a coin under each cup. When I asked her to guess which cup the coin was under, she was, of course, correct. After she’d made a correct choice about six times, she was convinced I was somehow controlling her thoughts, and moved into a rather disassociated state, so I explained, and she ‘snapped out of it’. I would suggest this as a possible means of inducing hypnosis. …

Again we see that the subject is made to feel that his body is out of control, and becomes subject to a high-status person. Some hypnotists sit you down, ask you to stare upwards into their eyes and suggest that your eyelids are wanting to close—which works because looking upwards is tiring, and because staring up into a high-status person’s eyes makes you feel inferior. Another method involves getting you to hold your arm out sideways while suggesting that it’s getting heavier. If you think the hypnotist is responsible for the heaviness rather than gravity, then you are likely to accept his control. Hypnotists don’t, as sometimes claimed, ask you to put your hands together and then tell you that you can’t part them, but they do ask you to link them in such a way that it’s awkward to part them. If you believe the hypnotist responsible for such awkwardness, then you may abandon the attempt to separate them. If you squeeze your index fingers hard, and then wait, you’ll feel it starting to swell — I imagine this is an illusion caused by the weakening of the muscles of the compressing hand. This too can be a way of inducing trance so long as the subject doesn’t realise that the ‘swelling’ would be experienced anyway, even without the hypnotist’s suggestion.

One more example, I think:

Crowds are trance-inducing because the anonymity imposed by the crowd absolves you of the need to maintain your identity.

I think what’s happening in this last one is that there’s a salient intuitive model where your body is part of “the crowd”, and “the crowd” is the force controlling your actions.

4.5.4 Trick 4: Find a “possessor” who you see as high status (a leader that you want to follow)

Background: Seeing someone as high status basically means that, if they want me to do X, then I feel motivated to do X as a direct consequence of that.

As discussed in my post Social status part 2/2: everything else [LW · GW], there are two main categories of reasons that I might see someone as high status—(1) maybe I like / admire them [LW · GW]; or (2) maybe I fear them. These correspond to the “prestige” / “dominance” split of Dual Strategies Theory.

However, for the purpose of this discussion, it doesn’t matter why you’re sincerely motivated to do what the leader wants, it just matters that you have that motivation.^[3]

How does this work? Suppose there’s a hypnotist who I see as a leader that I sincerely want to follow, for whatever reason. If the hypnotist says “stand”, then I immediately want to stand, by definition.

So far, this isn’t a trance; I’m just describing a common social dynamic. Specifically, if I’m not in a hypnotic trance, the sequence of thoughts in the above might look like a three-step process:

[S(stand←hypnotist) ; S(stand←homunculus) ; stand]

i.e., in my intuitive model, first, the hypnotist exercises his free will with the intention of me standing; second, I (my homunculus) exercise my own free will with the intention of standing; and third, I actually stand. In this conceptualization, it’s my own free will / vitalistic force / wanting (§3.3.4 [LW · GW]) that causes me to stand. So this is not a trance.

However, my relation to the hypnotist opens up the possibility of a different sequence of thoughts, where the middle step is omitted:

[S(stand←hypnotist) ; stand]

And this sequence is represented by the intuitive model I called “D(stand←hypnotist)”, i.e. the hypnotist expressing his free will through my body, and making me stand.

I’m motivated to stand for the same reason as before—i.e., because I see the hypnotist as a high-status leader whom I’m motivated to follow—but the action is conceptualized differently.

Examples from Impro:

Here’s the basic dynamic:

When a hypnotist takes over the function normally exercised by the personality, there’s no need to leave the trance. Mask teachers, priests in possession cults, and hypnotists all play high status in voice and movement. A high-status person whom you accept as dominant can easily propel you into unusual states of being.

This next quote makes it clear that the S(A←homunculus) thoughts tend to be negative-valence / aversive, whereas the S(A←hypnotist) thoughts tend to be positive-valence, which helps facilitate the latter getting dropped out. Why do they have those valences? I think those both come from the same strong motivation to do what the hypnotist wants: if your homunculus makes a decision, then there’s a risk that you’ll do something the hypnotist doesn’t like, whereas if you’re a medium for the hypnotist’s (inferred) will, then there’s much less risk of that. Here’s the quote:

Hilgard writes: ‘I asked a young woman subject who was practising appearing awake while hypnotised to examine some interesting objects in a box on a table at the far end of the room and to comment to me on them as if she were not hypnotised. She was quite reluctant to make this effort, eventually starting to do it with a final plea: “Do you really want me to do this? I’ll do it if you say so.” ’

Another subject of Hilgard said: ‘Once I was going to swallow, but I decided it wasn’t worth the effort. At one point I was trying to decide if my legs were crossed, but I couldn’t tell, and I didn’t quite have the initiative to find out.’ Another subject said: ‘I panic in an open-ended situation where I am not given specific directions. I like very definite suggestions from the hypnotist.’ Hilgard comments: ‘Thus the planning function, while not entirely lost, is turned over very largely to the hypnotist, willingly and comfortably, with some annoyance being shown when the subject is asked to take responsibility for what he has to do.’ (Ernest R. Hilgard, The Experience of Hypnosis, Harcourt Brace, 1968.)

(Note that the intuition that “I can do something but only with great effort” is associated with negative valence—see discussion here [LW · GW].)

4.5.4.1 Trick 4A: Indirect version—the high-status person is not the actual “possessor”

This is a funny variant where the high-status person doesn’t act as the possessor themselves, but rather directs that something else should be the possessor. I guess that works too. My main example here is the mask-possession technique described at length in Impro.

When the students are “possessed” by the masks, the masks themselves don’t have high status (from the students’ perspective). But the students do know that high-status Johnstone wants them to be possessed by the mask. So we can still get sequences like:

[S(dance ← mask persona) ; dance]

where the mask persona “decides” to dance. Again, there’s positive valence associated with S(dance ← mask persona), which then causes “dance” to happen; and the source of that positive valence is students’ motivation to do what they think Johnstone wants them to do, because they look up to him and trust him. Or at least, that’s some nonzero part of it. Presumably they also want to be better actors, to have a new fun experience, etc.

A related factor is that, if they trust that high-status Johnstone is in charge and keeping them safe, then they can feel more motivated to get deeper into the trance—they don’t need to remain lucid for their own safety.

Examples from Impro:

Johnstone first talks about his own status as part of how he induces trance in his mask classes:

Once the students are ready I change my status, and play ‘high’. I don’t bounce around and wave my arms like I would for a comedy class. I become stiller, ‘serious’ and more ‘adult’. The change in me products a change of feeling in the students which I exploit by assuring them that the Masks are not dangerous, that whatever happens I can handle it, and that all that matters is that they must take off the Mask when I ask them to. …

Then a later section entitled “Dangers” has some more illuminating status-related discussion: 

I did once have a Mask hold up a chair as if it was going to attack me. I walked towards it, said ‘Take the Mask off’, and held the chair while the actor took off the Mask. My confidence stemmed from the fact that there was no reason why the actor should attack me. He relied on my authority to be in a trance in the first place. …

Masks may cause physical harm when the teacher is believed to be in control, but in fact has been distracted. The Mask may be depending on the teacher to say ‘Take the Mask off.’ When the instruction doesn’t come, as a rule the Mask turns itself off, but it might, I suppose, make an error, and hit harder than it ‘intends’. We have the paradox that Masks are safest when the teacher is absent, since the actors then operate their own controls. …

As for actors refusing to remove the Mask, it’s never happened to me in the way people mean, although I imagine it could happen. There are reports of people in clinical hypnosis who have ‘stayed asleep’ (though not for long!) but we have to ask what people would gain from such behaviour. If someone refused to come out of a trance during a public hypnosis show, then he’d be put into a dressing-room to sleep it off, and would miss all the fun. In clinical hypnosis, the only purpose of such an action would lie in the opportunity to embarrass and confuse the hypnotist. If the hypnotist remained calm, then there’d be no pay-off. In case of any trouble with people refusing to remove the Mask, all you’d have to do would be to say ‘OK, fine, good,’ and keep your status. Then the refusal would be pointless. Always remember that unless the subject is crazy, or freaked out on drugs, then his trance has a purpose, and it exists because of the support of the teacher and the rest of the class. Go close to the Mask, put your arm around its shoulders. Your physical proximity to an entranced person usually switches Masks off.

…the teacher’s job is to keep the student safe, and to protect him so that he can regress. This is the opposite of the Freudian view that people regress in search of greater security. In acting class, students only regress when they feel protected by a high-status teacher. …

4.6 Explaining properties of trance

Now that we know more about what trance is, and how it’s induced, we can tackle the question of what happens in a trance, and why.

4.6.1 Trance Property 1 causes “losing track of time”

Recall that Trance Property 1 is the absence of S(⋯←homunculus) thoughts, and includes flow states, engrossing movies, and deep trance states. A property of these is that we often “lose track of time”. I think this has a couple connotations, which I’ll go over.

Connotation 1: “While the thing is happening, we’re not thinking about how long we’ve been doing it.”

Imagine I’m engrossed in a movie. If the prince is fighting the dragon in the river, then I’m thinking about the prince and the dragon and the river. I’m not thinking about how long the movie has been going on.

By contrast, when a self-modeling S(⋯←homunculus) thought pops up, e.g. “I’m watching a movie right now”, one of its salient associations is how watching the movie fits into the narrative of my day and my life, which might bring to mind the question of how long I’ve been watching.

Connotation 2: After the thing is over, we might look back on what just happened and say “I can’t believe how much time passed—it felt like a few minutes but it’s been hours!”

I think the explanation is a combination of auto-associative memory and the availability heuristic. When I want to know how long X took, the default procedure is: (1) try to summon memories that were captured in the midst of X, (2) see how many memories come up, and how easily. (It’s not a reliable procedure, but it’s so easy that people evidently do it anyway. And if they need more reliable data than that, they check the clock!)

Now, there’s a problem with using auto-associative memory for this purpose: if I was engrossed in a movie, and there was a scene where the prince fights the dragon in the river, I can only recall that scene by thinking about “prince”, or “dragon”, or “river”, etc. But I can’t pull up any of those, because they’re not on my mind! It’s a chicken-and-egg problem! So those memories are rather inaccessible.

On the other hand, if I was not engrossed in the movie, but rather periodically thinking self-modeling S(⋯)-related thoughts during the movie (“I am ¾ of the way through the movie now…”), then I can think “myself watching the movie”, and then any or all of those archived S(⋯) thoughts will auto-associatively pop right into my mind. The self-reflective narrative of my day serves as a kind of “hook” for auto-associative recall.

As an exception which proves the rule, I think there are instances where you’re in a flow state for a long time, but there’s an readily-accessible “hook” to pull up lots of different snapshots from within that interval. For example, suppose you were writing for four hours in a flow state. Maybe afterwards, you would say “Where did the time go! Those hours felt like minutes!” But then maybe you would scroll up on your screen, seeing paragraph after paragraph of what you just wrote, and then maybe you’d say “…Huh, well, that was a really epic work session, I sure did do a lot of stuff!”. And now the hours feel like hours again.

To be clear, none of this is amnesia—I can still remember what I did during my flow state at work, and I can still remember the movie plot. Being engrossed makes it harder to summon memories via a life-narrative-level “hook” (“I watched a movie last night, what was happening halfway through it?”) but the memories are still there if you find an object-level “hook” with which to retrieve them (“Seeing that river reminds me of the battle scene in the movie last night…”).

So let’s move on to the deeper form of amnesia:

4.6.2 “Deep” trance with both properties 1 & 2 can be associated with amnesia

At the top, I suggested a distinction between lucid trance states with Property 2 but not Property 1, versus amnesiac trance states which have both properties.

The lucid-versus-amnesiac distinction comes from Johnstone. For example:

If we compare Mask work with ‘possession cults’, then we can see many similarities. It’s true that the possessed person is often supposed to remember nothing that happens during the trance—but this is also observed sometimes in Mask work, even though it’s not demanded. And two types of possession are often described: an amnesiac and a lucid state.

Or again:

It’s reported that voodoo trancers remember nothing about their possessions, but Jane Belo, writing of trance in Indonesia, describes two types of possession: one in which a ‘power is present that is different from his “I”, and makes two simultaneous integrations, and that in which there is a temporary but total change of the personality in which the person is “transformed” into another being or object.’

As far as I can tell from studies of trance amnesia (including posthypnotic amnesia), trance amnesia is the exact same phenomenon as the amnesia across Dissociative Identity Disorder (DID) “alters”. It’s an interesting phenomenon. It’s readily distinguishable from true amnesia, particularly because (1) it only impacts explicit (autobiographical) memories, not other kinds of memory; (2) subjects do worse than chance on certain tests of recall. But the phenomenon is also readily distinguishable from malingering (a.k.a. “just trolling the experimenters”). Anyway, I’ll discuss DID amnesia in the next post [LW · GW], and my opinion is that trance amnesia is the same thing.

The short version is: If I have a tune in my head, then I’m very unlikely to simultaneously recall a memory of a different tune. Likewise, if I’m angry right now, then I’m less likely to recall past memories where I felt happy and forgiving, and vice-versa.

In an analogous way, if I’m in a conventional state of consciousness, where the homunculus is active and salient, then that fact conflicts with the summoning of memories where an irreconcilably-different (§4.4.2) entity is conceptualized as the cause of mental events.

Those kinds of associational conflicts normally have a nonzero but limited impact on memory. However, in the context of trance, it morphs into a far more powerful and effective force.

Again, more in the next post [LW · GW]; we thank you for your patience.

4.6.3 Possessions are often child-like

4.6.3.1 Learning to talk from scratch

Johnstone says that, in mask possessions, the students-as-masks usually need to re-learn to talk from scratch—and he discusses how to give them “speech lessons” in great detail!

But there’s a revealing caveat on that:

Possessed people don’t seem to need speech lessons (which Masks do, as described later), but there are many descriptions of inarticulate sounds preceding speech. And sometimes a deeply possessed Mask will speak from the first moment.

I think what’s happening is:

• If you have a sufficiently strong, vivid, internally-consistent, and stable mental model D(A ← god / mask / etc), then that model can slot in as the “explanation” of the action A = speech act, “explaining it away”, so that the competing intuitive model D(A←homunculus) does not get summoned.
• On the other hand, if D(A ← god / mask / etc) is fragile, shaky, and low-confidence, rather than vivid and strong, then it’s more likely to lose in a head-to-head competition with the far-more-established (high prior probability) D(A←homunculus) thought.

Johnstone’s students are much more towards the “fragile and shaky” end. They’re new to the whole mask-possession business. They come from a culture with no trance traditions. And they’re trying to jump right into a trance, rather than first having their conventional identity softened by an hours-long ceremony of drums, drugs, and dancing. So the students’ initial trance states are quite fragile, and speaking will tend to break it.

Taking that one step further, since the students know that speech acts will break their trance, they’ll preemptively avoid speaking. After all, they want to be in a trance, all things considered. Otherwise, they wouldn’t be in one in the first place (see §4.5.4 above).

4.6.3.2 Other examples beyond speech

We can generalize from this to say that, over life experience, we develop a rich and complex set of behaviors that constitute “my best self” and are always done intentionally (cf. §2.5.1 [LW · GW]). These behaviors are all threats to the stability of a trance (especially a shaky novice trance), and hence novices who are motivated to maintain a trance will “flinch away” from those behaviors.

And if those behaviors are all suppressed, then what remains? Acting on impulse, like a child.

More examples from Impro:

Ghede, God of death, and of sexuality, is consumed by raging hungers, but note the paradox that the supernatural creature who we would expect to be ‘super-adult’ is very childlike — exactly as the Masks are. Ghede, in Deren’s^[4] description, sounds exactly like a Mask. …

For an introductory Mask class I will set up a table with a variety of props on it. They’ll be on a table because the act of bending down may turn a new Mask off. I avoid any props that would present ‘difficulties’. An umbrella might encourage a Mask to think how to open it. An alarm clock might suggest winding it up. Anything that would require a Mask to have a mental age of more than two and a half I would remove. The objects on the table are the sort that would interest young children. …

A new Mask is like a baby that knows nothing about the world. Everything looks astounding to it, and it has little access to its wearer’s skills. Very often a Mask will have to learn how to sit, or bend down, or how to hold things. It’s as if you build up another personality from scratch; it’s as if a part of the mind gets separated, and then develops on its own. There are exceptions, but in most cases the very best Masks start off knowing the least. They don’t know how to take the lids off jars; they don’t understand the idea of wrapping things (given a present they just admire the paper). When objects fall to the floor it’s as if they’ve ceased to exist.

4.7 Conclusion

I think this story hangs together well, and accordingly I now see trance states as natural and expected, as opposed to bizarre and implausible, and fun^[5] rather than scary.

In particular, I think the excerpts in §4.5.4 make it especially clear that the motivation to enter a trance, and the motivation to remain in a trance, and the motivation to do things in a trance, are all conventional motivations that can be understood in conventional terms. There’s no paranormal mind-control happening here! Hold that thought—we’ll be returning to that same theme in Posts 5, 6, 7, and especially 8.

As mentioned at the top, I haven’t exhaustively researched trance, so please leave comments below!

In the next post [LW · GW], we move on to Dissociative Identity Disorder, which will turn out to have some overlap with the ideas I introduced above.

Thanks Thane Ruthenis, Linda Linsefors, Justis Mills, and Johannes Mayer for critical comments on earlier drafts.

1. ^{^}

   Fanny Kemble is a famous 19th-century actress.

2. ^{^}

   This paragraph is pretty much all I plan to say about tulpas in this series—the series is long enough as is.

3. ^{^}

   That said, I would guess that prestige is much more common and effective than dominance, in terms of trance-induction. I say that because I think liking / admiring [LW · GW] tends to leads to more sincere and enthusiastic motivation to “follow” [LW · GW], whereas fear tends to lead to more grudging obedience.

4. ^{^}

   Maya Deren was an avant-garde filmmaker who who filmed, wrote about, and participated in Haitian Vodou rituals in the 1940s–50s.  

5. ^{^}

   I mean, fun for some people. It’s not really my scene.

7 comments

Comments sorted by top scores.