Dreams, Astral Projection and Tulpas

Four things get bundled together under the word astral, and only one of them is a person sitting in bed intending to have a lucid dream. This page separates them: lucid dreaming, out of body experiences, remote viewing and tulpa practice. They have different evidence bases, different reliability, different risks and different methods, and treating them as a single skillset is the source of most of the confusion in this corner of the field. The arrangement below explains the structure first, then each of the four in turn, and finishes with what to do about astral larvae, which is a historical idea that has been absorbed almost entirely into modern occult practice without much examination of whether it belongs there.

The four things, and why they are not one thing

Lucid dreaming is a documented phenomenon with a measurable population base and reliable induction techniques. It is the most scientifically tractable item in this entire domain and it should be treated as the baseline against which the others are assessed.

Out of body experiences are reported by a small percentage of the population, and they tend to arrive near sleep onset, or during illness, or during meditation, rather than at the will of the practitioner. Remote viewing is disputed in its results and uncontroversial in its training value. Tulpa practice is a Tibetan Buddhist method with a specific purpose that is not the one it is usually borrowed for.

• Lucid dreaming: trainable, reliable, and the place to start.
• Out of body: mostly involuntary, and a different question entirely.
• Remote viewing: results disputed, description skill real.
• Tulpa: a deity-yoga method being used for something it was not built for.

What the sleep architecture does for you

REM periods lengthen through the night, and the longest one runs in the final third of the sleep period. That single fact accounts for most of the difference between people who lucid dream frequently and people who have never managed it.

Waking without moving is the other half. A lucid dream is usually lost within seconds, and the loss is caused by the small motor act of changing position. The combination, morning waking plus stillness, is a genuine technique rather than a folk one, and it is testable by anyone willing to run it for a month.

The reality test that actually works

Text on the hands is the standard check, and within a dream it often appears as text that is approximately right and becomes less right the longer you look. That is useful but it is also easy to talk yourself past, and I have talked myself past it more than once.

The check that holds up better is behavioural rather than symbolic. Try to push a finger through the other hand, or run a light switch, or look at a clock twice and check the hands moved. In a dream these mostly fail or behave wrongly. In a waking moment you do not need to be told, and the difference is the actual signal you are looking for.

Out of body, and why it is mostly not at will

Reported out of body experiences cluster near sleep onset, during illness, and in meditation, and they are not usually deliberately produced by the person having them. That is the plain finding and it is inconvenient for the project-based practice built around them.

A sizeable minority of the population report one at least once, which is more than is often claimed and much less than the frameworks built on the phenomenon imply. The sensible position is to treat it as something that happens to some people rather than as a skill to be developed, and to be considerably more interested in the circumstances of occurrence than in the interpretation.

Remote viewing, and the description skill

The work done at research institutes is disputed in its results, which is the honest summary. Prospective trials have not replicated reliably and the controversy is not going to resolve in anybody's favour soon.

The training does reliably improve visual description ability, and this is a genuinely transferable skill. Learning to describe a scene rather than impress with an interpretation of it is worth doing on its own terms, and it is the one part of this practice that survives the failure of the rest.

What tulpas are actually for

The practice comes from Tibetan Buddhist traditions, where the method is used for visualisation and deity yoga. That is a different project from the one it is borrowed for in modern occultism, and the borrowing changes what happens.

A tulpa in the original context is a disciplined visualisation held to a very high standard and eventually dissolved. A tulpa in the popular version is a companion, a servant, or an entity. The visualisation itself is identical in both cases; the intention and the endpoint are not, and the practice is easier to get wrong when the endpoint is a relationship.

Astral larvae, and the honest account

Astral larvae are thought forms, left over from thoughts that were held intensely and then dropped, and the traditional remedy is a banishing. Chaos magic adds a sharper claim: that a larvae is a pattern fragment and that eliminating it means noticing it and letting it go rather than fighting it.

The mechanism is worth taking seriously even if the entity is not. What gets described as a persistent intrusive thought is well documented in ordinary psychology as an unwanted automatic thought, and the response that works for that is acceptance rather than suppression. The occult version and the clinical version have arrived at the same place by different routes, and neither needs the other to be true.

Where the practice starts eating its own substrate

Lucid dream practice is unusually good at producing a distorted relationship with sleep, because the methods that work involve either waking earlier or sleeping later. The result is a practice that improves the skill while degrading the state the skill depends on.

If the practice starts costing you sleep, the correct response is to reduce it rather than to push through, and the tradition's own advice is a short fixed routine rather than a climbing one. Sleep is the substrate, and optimising against it is not a technique.

One month, sequenced the way it actually needs to be

Week one, learn the reality test and use it on ordinary dreams, so that the check is familiar before it matters. Week two, morning waking with stillness on five mornings. Week three, keep a log that records attempts, successes and the specific content, since the content is the part people forget.

Week four, review. Most people get somewhere in that month and a substantial minority get nowhere, and the log is what tells the two groups apart. What you should not do is escalate the method in week five because it did not work, since the usual cause of a null month is an unfamiliarity that needed more time rather than more effort.

1. Week one: the reality test, practised on ordinary dreams.
2. Week two: morning waking, stillness, five mornings.
3. Week three: log attempts, successes and content separately.
4. Week four: review the log before changing anything.

A table of the four, and their evidence

Lucid dreaming has a measurable population base, reliable induction techniques, and a long history of laboratory work. Out of body experiences have a small self-selected reporting base, no reliable voluntary production method, and a strong association with illness and sleep onset. Remote viewing has disputed results and uncontroversial training effects.

Tulpa practice is different again, because it is a documented contemplative method with its own purpose rather than a disputed experiment. Putting four things with four different evidence bases into a single category and calling it astral is the error that the rest of this page is trying to undo.

What the log has to contain

Attempts, successes and content, kept as three separate counts. Collapsing them into one number is the standard mistake, because a month of five attempts and two lucid dreams reads very differently from two attempts and two lucid dreams, and only one of those is remarkable.

Content is the field most often dropped, and it is the most useful over a long run. People reuse dream imagery constantly without noticing, and a content log is the only way that reuse becomes visible rather than becoming a private mythology.

• Attempts, counted, even the ones where you fell back asleep.
• Successes, counted only when a reality test confirmed them.
• Content, written down rather than recalled later.

What to do when a month produces nothing

Nothing, for one month. A null month in a practice with this many variables is the normal state, and the usual response is to escalate, which converts a statistical fact into a self-inflicted problem.

Check the two things that account for most nulls before changing anything. Was the reality test being used, or was success being judged on how vivid the dream felt? And was the practice actually fitted into a real morning routine, or attached to an intention that was never tested? Almost every stuck month is one of those two.

The interference question nobody asks

Lucid dream practice competes directly with ordinary sleep, and the two are not separable in the way most practice write-ups assume. Waking earlier reduces REM pressure later in the night, which means the very practice that produces the skill also erodes the condition that produces it.

This is worth planning for rather than discovering. If the practice is going to last more than a couple of months, it needs a structure that does not depend on losing sleep, and the only one that does not is a very short routine run on ordinary waking mornings. Length and escalation are the two things to resist.

Where a chaos magician would put all of this

The framework people usually reach for is a thought form, described as a pattern fragment rather than a creature, and the instruction is to notice it and let it go rather than to fight it. That is a workable model and it maps onto something ordinary and well documented.

The unwanted automatic thought, which is a recognised feature of ordinary cognition and has a substantial literature on what actually reduces it, is not fixed by suppression and is reliably worsened by it. The occult version of the remedy and the clinical version of the remedy arrived at the same place, and neither one requires the other to be true.

A defensible order to do all of this in

Lucid dreaming first, because it is trainable, because the evidence is best, and because the reality test it requires is the most transferable skill in the group. Remote viewing second, for the description training rather than for the results. The others are not a sequence at all, and treating them as stages of progress is the part of the popular framing that most reliably costs people time.

Frequently Asked Questions

Why do I lucid dream never but other people seem to?

The reliable variables are the morning window, because REM periods lengthen through the night, and stillness at waking, because the smallest movement ends the state. People who dream lucidly are usually long-sleepers or early risers by accident of routine rather than by technique. Most of the rest of the variation is how often they practised and how long they stuck with it.

Can astral projection be learned?

For most people, not reliably, and the evidence does not support treating it as a developed skill. Reported experiences cluster near sleep onset, during illness and in meditation rather than at will. That is worth knowing before you invest in a framework built on the assumption that it is trainable.

What actually happens when I eliminate an astral larva?

Usually nothing observable, which is the honest expected result. The useful part is the practice rather than the outcome: noticing a persistent intrusive thought and letting it go, instead of fighting it, is what changes, and the same technique is described in clinical psychology without any of the entity material.

Is it safe to practise lucid dreaming regularly?

Yes, with one real caveat. The methods that work involve waking earlier or sleeping later, so the practice can quietly damage the sleep it depends on. Reduce it if that happens rather than pushing through, and treat sleep as the substrate rather than as a variable to optimise against.

Does a paid dream journal actually help?

A notebook does the same job and the method is free. A purchase is worth considering if you want a timer, because holding receptive attention on a fixed duration is one of the few things where the app is genuinely better than guessing, and if you want your log searchable later. Neither is a reason to change the practice itself.