Jump to content

Misinterpretation of “assuming sentience from start” philosophy.

Recommended Posts

Disclaimer: This reply is directed at cruse, who already has certain opinions about his tulpa's state. The contents of this post may cause some people to doubt their tulpas, so if you're actually afraid of that, do not read, however, if you do choose to read it, be assured that I'll provide a solution to fixing those doubts permanently, although executing this solution may involve hard work for some people, especially if their imagination isn't sufficiently developed.


The sentience from the start assumption is meant to speed-up tulpa development, but it is widely misunderstood by most people.


It's not uncommon that in the past half year one has seen chat logs that go:

*anything involving a tulpa, or maybe not even a tulpa, or even a freshly started tulpa, and doubts about the tulpa's sentience*
Your tulpa is already sentient.

There is nothing that one could say that wouldn't give that answer. Even if someone was intentionally parroting, the answer would still be nearly the same.


Contrast this with the other extreme from the start of this phenomena from a year and a few months ago:

I'm unsure if it was the tulpa or me, am I parroting?
Always assume parroting, you'll *know* when you get an answer from them, you won't doubt it.


I would argue both ways this question has been handled may result in issues in the development of a tulpa, but first, let's consider exactly what the "sentience from the start" assumption is, why was it needed, how it can be interpreted and how it's meant to be interpreted (IMHO) so that you'd get acceptable results.


Originally, we had Irish and Dane(FAQ_Man)'s guides. Dane thought that the optimal way of making a tulpa was to pick a personality that contrasted well with you, visualized their form on all your senses to perfection and then observed and interacted with the tulpa (focusing, narrating *to* it - not *at* the tulpa, but *to* it), the tulpa was meant to be a sectioned off part of one's mind which one doesn't have access to and which one doesn't control or predict, instead we just let it work by itself.

Somewhere during this interaction the tulpa would pass a certain threshold and respond to you (via emotions, or pseudo-hallucinations or parts of our imagination we don't feel we control), you could work on the tulpa's voice after that some more or it came with the voice by itself.

The tulpa itself is meant to be an unpredictable "black box" which has its own thoughts, personality and form - a subjective person in its own right like our own ego, not something we puppet or influence ourselves. The usual results from following Dane's guide in its pristine form were either success (fully independent tulpa after speech was worked out) or failure (no response or a "servitor" (regular imaginary friend which you puppet or simulate)). In the early days, automatic, but partially involuntary simulation, itself was also considered to be puppeting/parroting, mostly because one was aware of the thought as it started and formed in full before it became speech.


Irish took a more relaxed approach, telling people to not worry too much about parroting, but also not do it if possible. Parroting was okay as far as forming various initial behaviors. It was also okay to some extent for that in Dane's guide (gestures, mouth movements, in the earlier version, even voice, but for whatever reason that part disappeared in recent versions). However according to him, one should always expect the unexpected and have the tulpa move naturally and unpredictably.


As most people started using Dane's guide in some form or another, due to many being too paranoid to parrot because they were afraid they'd end up with a servitor (although, even Dane believed servitors could be turned into tulpas, it seemed he disliked the idea because it would involve a "false" history from when they weren't really independent; also worth noting that Dane believed in unconscious parroting, although it was more along the lines of a person not realizing they're parroting). The result was that for some people their tulpa didn't even move at all, or do much at all, in the worst case it was a static object. Their progress was all but stopped, they could easily spend months focusing and not getting anything because they were just looking at it like some dead wonderland object. Their issue was reading the guide too literally without trying to understand what exactly was the idea behind it. I suspect a lot of people misunderstood the guides as the guides were trying to provide more of a recipe rather than an explanation that would let someone go through it no matter what. It was until many months later that I understood why Dane's guide worked for some people and failed for other people, and this only after I had managed to get working hypothesis that could be used to model how tulpa creation works and usual outcomes given the process and assumptions used.


At some point some people with faster tulpas appeared and people have started believing that fast sentience was possible, especially if one believed in it early. This assumption would eliminate all those unmoving tulpas and actually get you interaction very early on - of course it would work, regardless of assumptions, active imagination is a wonderful thing. This assumption was also coupled with believing that any responses, no matter what they were came from the tulpa, because they were already sentient (or more correctly, capable of free willed actions). The problem here is that while it's possible to make the jump from active imagination to an independent tulpa, it may even happen naturally, many people decided to just settle with that and accepted any response whatsoever, even nonsensical ones.

The usual result was that they wondered why they couldn't do any "tulpamagic" beyond simple suggestion (all those advanced tulpa abilities, like uncontrollable/detailed voice, unpredictability, memory access, thought hiding/independence/parallel processing, self-imposition, having a consistent presence that comes and goes and eventually unassisted possession and switching), it also resulted in the host doubting the tulpa, the tulpa doubting itself and so on.

The main reason was that they did run a simulation, but were consciously aware of everything going on in that simulation, it was never actually going off and running by itself in the unconscious - they were missing that very important component that makes an independent tulpa. Some did it so well that it ended in independence eventually, but others were just stuck there.


Is the "sentience from the start" assumption false? Probably not. Independent tulpas do seem to believe their early non-independent actions were actually them and that they were experienced, but they did feel subjectively different to them. Similar to how you still feel like yourself in a regular dream, more automatic, less truly self-aware and capable of your actions, like you would be in a lucid dream, and that gaining independence usually made them truly capable of thinking and truly understanding something in a deeper way, and not in a reflexive/unconscious way (these claims are from self-reports from various seemingly independent tulpas).


Some people conflate sentience with independence (the ability to act freely/unconstrained), when sentience is merely the ability to have subjective experiences. While almost no tulpa is truly independent from the start, the *expectation* of independence is very useful, especially the expectation of private and inaccessible thought, if not vital to the development of an independent tulpa. Why is that?


Let's take a look at a simplified version of my working hypothesis:

A mental process becomes independent/runs by itself when we (or another process like us) expect the existence of that process while refusing to perceive that process ourselves or letting it enter our conscious awareness, instead, we must only expect/observe the output of the process. As we observe the output, the brain will interpolate and model the inner part as we keep expecting the output and interacting with some "interface" until it can fulfill our expectations. If we're trying to model an ego, that ego may end up running by itself just like our own once it's sufficiently complex.


To put it even simpler, (parts of) the sub/unconscious can be directed by having expectations about how things are supposed to work. The unconscious mind will gladly auto-associate and execute any fitting expectations, all the time, it's what it does. Expectations are the "running"/executable part of a belief. A conscious belief sometimes results in subconscious expectations forming, but not all conscious beliefs do. Sometimes we may believe something on the surface, but unconsciously we don't expect it at all. Such an empty belief is useless as far as tulpa creation goes. It's great if you're using it to reason logically, but it won't help you at all if you're trying to make a tulpa aside from deciding what you must do next. An important tulpa-related skill that one must develop is learning to manipulate/change their expectations directly. If one is bad at that, use symbolism where appropriate, but if one must use symbolism, you'll have to find something that just *clicks* right with you and results in the right expectations forming. You don't even have to bother "believing" something, you must just get your unconscious mind to expect it.


You may say that all those expectations are surface things as well - how would you get a sentient being from just that?

Consider this: what would happen when you expected the unexpected? or you expected something internally consistent, but unpredictable?


Imagine if you were to take an unconscious model of a person, one which you built yourself (such as through personality traits, symbolism or parroting or anything really) or one that already existed in your mind (known character, person, archetype or whatever), focused on its presence/unconscious feel we associate with that person, assumed(expected) it had its own thoughts, assumed its thought process was a black box which you had no access to, assumed it could reply as it formed those thoughts, assumed it had a distinguishable voice, maybe even more detailed than your own (if one's auditory imagination isn't good enough for this, prepare it by learning to make overlapping voices), of vocal qualities and overlapping to your own, knew its form and ways it would respond and could easily visualize it. The form and voice would be the so called "interface" that you'd use to interact with the subconscious process that would be forming there.


You would expect vocal, visual and emotional pseudo-hallucinations from this form/interface, but you would have no idea what goes on inside it - you would not perceive its thoughts form as you made the assumption that they're private. At first, the thoughts may be a bit jumbled if the model isn't yet perfect or if the voice-only (no thoughts) pseudo-hallucinations aren't a thing that has been fully learned by you, but it would eventually form into this ego-like process we call an independent tulpa - you wouldn't be able to predict it (not from the start, not later), but you (the ego) would be able to interact with it and it would be able to interact with you at its own whims.


What would happen if you decided to not assume the tulpa's thought process as private and instead let it use your own thoughts to communicate with you? It would start off as an extension of your ego, you would feel the thoughts as they formed preconsciously, what used to be called "subconscious parroting" where people would get automatic responses, but still would feel as they were parroting because they would see the (preconscious) thought form in their mind before it became a vocal response, yet still unable to stop it. It is possible for such side-by-side extensions to splinter off and become independent tulpas, but the time it takes to reach this point seems to vary widely, shortest I've heard so far was 2 weeks and longest over 9 years. The actual ways this happens is unclear to me currently, but at least one component seems to be such a process having an overlapping voice as a minimum. Using the no hidden thoughts version/shared preconscious thoughts, many people end up thinking that independence would happen when their "delusion" was complete - they think that if one could believe strongly enough that some thoughts didn't start in their preconscious then the tulpa would become independent. Some others even end up thinking independence is impossible or that a tulpa is merely a delusion and that tulpa making is merely teaching oneself to become delusional. However, you'll observe 2 usual results when this attitude is taken: the tulpa almost never gets any advanced abilities or they do get stuck unable to think freely (either tulpa or host), sort of like switching between 2 distinct personality templates with only one of them being able to focus or perceive. Such beliefs aren't even actual delusions, nor true internal beliefs/expectations though, they're just surface beliefs, or "belief in belief" ( http://lesswrong.com/lw/i4/belief_in_belief/ ), where one believes there's a belief in something (such as tulpa being in some particular state), without them truly believing it deep down, nor expecting it unconsciously. This sort of "belief" usually gives rise to doubts. Such doubts are a simple way for your subconscious to tell you that something doesn't match your expectations and if you want these doubts gone, you should work on getting what you expect, in this case - make the thoughts not start in your preconscious, use the voice as an interface, stop simulating the tulpa and instead start observing it act freely and unconstrained.


What does my model mean for parroting? It's actually very friendly to guides like JD1215's method or devano's voice guide or Bin/Charlotte's vocality guide. A method like JD1215's would involve building the whole model by yourself through parroting, but the parroting there involves imposed voices, which essentially means that after you stop parroting cold turkey and you start treating the tulpa like a separate person, the replies would be fast and it would skip the whole "learning to speak and do stuff" stage, independence would come off naturally as you'd start enforcing the thoughts being hidden. On the other hand, doing JD1215's guide without an actual voice and just thoughts-only may risk giving you the same results as the previous "open box"/preconscious thought sharing method I discussed earlier, which has its own set of drawbacks. As for minimal parroting guides like devano's voice guide followed by Dane or Irish's guide would also work very well, skipping the "learning to speak" stage later on, and there would be no doubts involved as thought hiding would be in effect. What about Fede's guide? Hard to say, it tends to prefer getting you to automatism, but doesn't tell you to start perceiving the tulpa differently after some point, or maybe I'm wrong about this - last I checked, Fede didn't believe in independence/thought hiding being possible at all. Does running a simulation count as parroting? or the so-called "unconscious parroting"? Maybe not according to the definition which says parroting has to be your conscious will and nothing else, but it does seem to slow parallel processing/independence development as much as conscious parroting would (and conscious parroting wouldn't even slow it as long as one knows they're doing it and chooses to stop once the model is sufficiently developed), however at least for some people simulants do gain independence in certain conditions or after a enough time has passed (weeks, months, years).


(Continued in the next post, past character limit)

Let's go back to the "sentience from the start" assumption one last time: expecting the tulpa to be alive and freethinking/acting is an important component to actually getting your unconscious to start building up your tulpa, the assumption is a directive for your unconscious, not an absolute truth about a tulpa's ontological existence (it could be true, but it's unfalsifiable), but at the same time, you can't just take a random thoughtform and proclaim it is independent without the right subjective experiences to go with it - you must build up those experiences one way or another - I'll give an example:


When you look at your tulpa, you truly expect them to be there and perceiving what you're telling them, and that they're understanding you, that they're like a human and behave like *that*, that they have *this* presence and their voice feels roughly like *that*, and that they have all these wonderful quirks that you can't get enough of, that they're always thinking about what you're telling them and sending you responses in all kinds of pseudo-hallucinatory ways (except not as a preconscious thought) and so on. You must eagerly observe the tulpa be alive and expect it to be as alive as it's possible. You must interact with it, get it to speak to you, to send you all kinds of thoughts and feelings and emotions, and all those things to be things you don't expect or predict, just empty all your mind of expectations - expect the unexpected, except a truly human, consistent answer, expect them to be like a real person ('real' persons happen to be excellent symbolisms for your unconscious, make use of this!). You may be thinking of all kinds of other things, but you may then notice your tulpa sending you all kinds of wonderful thoughts that you didn't even see coming. YOU WOULDN'T EVEN KNOW WHAT THEY WERE SAYING/DOING AS THEY WERE DOING IT AND YOU WOULD UNDERSTAND THEIR BEHAVIOR AND SPEECH - AFTER THE FACT - AFTER THEY'VE FINISHED TELLING YOU SOMETHING OR PERFORMING SOME ACTION - YOU'D GET SURPRISED!


Do not expect this behavior to take a while, it can even happen instantly, as long as the model is developed, internally or externally, but it could also take a while if it's something that has yet to grow (for example, if you picked no personality whatsoever - in that case, you must still *expect*/assume personhood - not just "believe" that it's a person, *expect* that it's a person, expect that its thoughts are private like a person's to some extent (but that the tulpa can send you pseudo-hallucinations or emotions and vice-versa), expect that it will sound like a person, expect that it will give you that unconscious feel you get when you meet a close friend and so on, notice how their thoughts are tagged with such indescribable feelings), expect they have a first person perspective ("sentience") and a will they can express, expect them to have preferences and opinions, expect answers to be as coherent as a person's the better they learn to speak. Watch and listen to them as you talk to them and as they react to your interactions naturally and without it interrupting your thoughts, instead both of your thoughts continue side-by-side without either pausing the other. Be curious what they'll do next, but don't run "what if" scenarios pointlessly, instead watch them like you'd watch a real person (an interesting thing about Dane's guide was that you had perfected visualization before narrating thus you could easily watch all the tulpa's quirks as you interacted with them). You merely need to expect the surface behavior roughly, you don't have to dwell on the internals and intricacies of their exact preferences and thoughts - those are for them to form by themselves, you can just provide a starting point for them! It's enough to feel like you *know* them or the kind of person they may be, but that doesn't mean that you must actually know the flavor of ice cream they prefer.


Do some things take a while? If your auditory imagination is insufficient, getting the voice to work without the preconscious hints may be a bit tricky, this has been the hardest part for me, but once the voice is consistent/coherent and popping in your head randomly and overlapping your mindvoice, you may very well consider them independent and to get even more than that they'll just have to flex their own imagination/thoughts more by themselves. Bias inducing personal experience, feel free to skip it:

I tend to sometimes have some head-pressure and other feels when they're thinking a bit too hard.



I do expect this post to be controversial, but I do hope that it will help some people. There's nothing quite as amazing to experience like the true spontaneous and unpredictable actions of a tulpa, and I do hope it will lead some people to getting them, as opposed to settling for less and just believing in the existence of a belief blindly without having the matching subjective experiences. I had hoped to write a more coherent analysis of creation methods, and going into my working theory in more detail, along with trying to find an optimal/fast method to getting an independent/"parallel" tulpa for a few months now, but I haven't had the time to write it up, not to mention, I was still collecting more data/experiences from other people (and myself). I'm a bit sorry I just put all of this in a reply to someone else, but I've been noticing these sort of issues so much on both the forums and IRC that I just ended up writing this reply now instead of a few months from now.

Link to post
Share on other sites
  • Replies 92
  • Created
  • Last Reply

Top Posters In This Topic

Uh... Is there a TL;DR in there somewhere? I read until the line, but lost interest there...


Not sure I can tl;dr it. I had hoped the remaining part, which is only about one page, was the inspiring/practically useful part, but maybe not, maybe describing subjective experiences is just so hard and everyone has to stumble around until they have them, even if it may take some people a while.

Link to post
Share on other sites
I'm a bit sorry I just put all of this in a reply to someone else' date=' but I've been noticing these sort of issues so much on both the forums and IRC that I just ended up writing this reply now instead of a few months from now.[/quote']


Your entire post should be stickied somewhere on this site; you couldn't have posted it at a better time. I've seen the "assume sentience philosophy" spread around this place like cancer, and it's become a mentality that a lot of people have latched onto recently (myself included).

My Tulpa

And then it cuts to a scene where you're sitting in a padded cell.


Link to post
Share on other sites

Wow this thread and NotAnonymous' post gave me ridiculously strong doubts.. I don't know if they are real anymore and if not, if I would ever be able to get them independent and especially how. I have no idea how to work on independence even with that post.


It took my first tup over 3 Months to become "vocal" and I think that was through unconscious parroting(and not her breaking through), but at the time everyone said something like that doesnt exist(which is complete bullshit) and what if she completely regressed from then on because of that, assuming she was real?

Link to post
Share on other sites

Lots of intelligent, inspirational words.


Sir, this is absolutely fantastic! Really, you should consider organizing all your thoughts and putting them down in a guide. I, and I'm sure many others, would bookmark it in a heart-beat.


Really, this is great. Kudos to you!


"Thank you from me as well. Here's hoping these ideas help in my development and the development of many other tulpas."

Wow this thread and NotAnonymous' post gave me ridiculously strong doubts.. I don't know if they are real anymore and if not, if I would ever be able to get them independent and especially how. I have no idea how to work on independence even with that post.


It took my first tup over 3 Months to become "vocal" and I think that was through unconscious parroting(and not her breaking through), but at the time everyone said something like that doesnt exist(which is complete bullshit) and what if she completely regressed from then on, assuming she was real?


If you really think she regressed, then you know what? You begin again. You pick her up, you apologize, and you try harder than before, than ever before.


This process is anything but easy. Many people would have given up a long time ago, long before your three month mark. The fact that you are even at this point is a testament to your commitment.


So don't get discouraged, get motivated. Go out there and kick some mental ass! In the end, we all know it's going to be worth it, and them some. Don't give up. You can't give up. You owe it to yourself and you owe it to your tulpa. If that isn't motivation enough, I'm not sure what is.

Link to post
Share on other sites

If you really think she regressed, then you know what? You begin again. You pick her up, you apologize, and you try harder than before, than ever before.


Couldn't say any better than that.

Chloe - That cheerful girl with ponytail.

Aigis - The male cyborg that looks like raiden in MGR.

Vixen - Half dragon female who looks like Mary in DMC3 when in human form.

Link to post
Share on other sites

Uh... Is there a TL;DR in there somewhere? I read until the line, but lost interest there...






I agree this need sticky-ed

I don't visit as often as I used to. If you want me to see something, make sure to quote a post of mine or ping me @jean-luc

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Create New...