I don't know how inspiring "no causation without reification" is as a rallying cry. Probably not very. But as a pithy description of an important insight, I think it's quite nice.

There's lots of reason to care about causation. In fact, it's hard to talk about much of anything without causation showing up. We need some way to talk about the fact that stuff happens, then other stuff happens, and that other stuff happens only because the first stuff happened, or so we hope to claim. The idea that one event causes another seems fairly fundamental to how humans perceive the world.

Formally, we often model causation as the action of one thing implying another, and we might formalize this with mathematical notation like  to mean some event or thing  causes some other event or thing  to happen.

Hopefully that all sounds boring and obvious to you, and has you wondering when I'm going to get to the point.

My point is that the entirety of what we think we mean when we say  is ontological, i.e it's in the map. Not once ever has causality existed in the territory.

Okay, I know, time to smash the downvote button, but hear me out.

Try to imagine what the world is like if you're not modeling it. Are you picturing atoms? Particles? Wave functions? Strings?

Wrong!

Those are all models we impose on the world to make sense of it. Useful ones, usually, but still models.

The territory, noting that the notion that such a thing exists is itself another model, is a kind of soup of stuff that's all mixed up and irreducible. You either have everything or nothing, for all time or no time. If you want anything else, you have to draw a map, make distinctions, and make choices.

Those distinctions are hopefully correlated with how the territory works, because that makes them useful for things like predicting what will happen (and we often call this correspondence "truth"), but they are not the territory itself. The territory just is; always has been, always will be.

This means that there's no aspect of the territory that is causality. There's no , there's no , there's no , there's just "is".

But if it's not in the territory, why is causality so useful that it's largely invisible to us, and why is it so hard for us to wrap our minds around what the world would be like if we didn't perceive its presence?

Well, there clearly is some feature of the territory we're carving out, making into a thing ("reifying"), and calling causality. But, we often forget, almost the instant we've done this, that we were the ones who did the carving, and then imagine that our perception of the world is how it really is.

This is important because it can lead us to making mistakes by thinking we understand something we don't. We fail to notice how confused we are about the aspect of reality we call causality because it's so close to our faces we don't realize it's a lens through which we are looking all the time.

And yet, everything still adds up to normality. That causation feels so natural to us is a sign that its describing something normal. So why worry about this, beyond perhaps philosophical interest?

Because sometimes it does matter. If you're trying to build an AI, and you only get one shot at avoiding existential catastrophe, and if you decide it all hinges on something about how the AI reasons about cause and effect, you better make damn sure you know what you mean here since there's no guarantee the AI will respect your notions of what cause and effect mean and not do something unexpected because you failed to realize the AI could just do something else you didn't expect it could do.

I sometimes see people get sloppy when thinking about causality and forgetting that they are the ones perceiving causality and instead pretending causality is just a natural feature of the world that exists in the territory, totally independent of our perception of it. To be sure, the territory is naturally doing something that we identify as causality, except the territory cares not one bit what we think that is, and so we cannot take for granted that what we think it means for one thing to cause another is as meaningful as we think it is.

This is a hard insight to feel in your bones, and it's tempting to gloss over it because we can tell ourselves that we can ignore uncertainty about causality just this one time because so often everything works out okay when we do. Resist that temptation. The world might just depend on it.

New to LessWrong?

New Comment
38 comments, sorted by Click to highlight new comments since: Today at 9:14 AM

Based on the original post, and OP comments, it sounds like the proposal here is:

Thesis: Not once ever has causality existed in the territory.

Support: There's nothing I can say that does not involve drawing a map.

Which, to me, looks like textbook "begging the question" - OP is picking one specific thing that can be said: "A causes B" and saying "look: causality can't be in the territory because it's just an idea in the map, just like everything else."

But that provides no insight or additional information. There's nothing special about "causality" not being in the territory, any more than saying "apples" aren't in the territory or "atoms aren't in the territory" or "justice isn't in the territory" or "paradoxes aren't in the territory" 

Once you place "the territory" beyond the reach of mere words to describe, it ceases to be possible to talk about what attributes it does or does not have. You can't usefully say "the territory doesn't have causality" when you have already said "I can't talk about the territory"

Even within the meta-map that places the territory outside the reach of mere words, it remains useful to talk about the presence or absence of causality at the limit of what words can reach. Is there time at the quantum interaction level; does it behave in the same way; are there iff relationships between superstrings - I'm certainly not qualified to opine on the answers, but the questions are of at least theoretical interest. (and theory with no practicality can still be useful - imaginary numbers were just a plaything of the mind for hundreds of years, until suddenly they became the math of a key industry)

Once you place "the territory" beyond the reach of mere words to describe, it ceases to be possible to talk about what attributes it does or does not have. You can't usefully say "the territory doesn't have causality" when you have already said "I can't talk about the territory"

But since causality is a thing in the map, of course this can be said usefully. It only doesn't make sense if you mix up the referrer and the referent.

Why is this not a contradiction: "Snorphblats can't be described with words"

"Snorphblats don't have a liver"

Or, how does the above differ from the statements in my previous comment? Please use a similar analogy, if necessary.

Just that. If Snorphblats can't be described with words, and yet you literally just used words to describe Snorphblats, what you must be doing is some action that is situated within the map. This action is itself in the territory (where else would it be?), but it need not be true that Snorphblats exist in the territory independent of any statements about them such that they can be meaningfully described.

This is a small part of a big topic, explored in detail in this classic LW sequence.

At a meta-level, you are hardening my position, not moving me towards your position. While I have stated my objections and questions in, now, five different ways, you have repeated the same point five times. And without either addressing, or even attempting to rephrase my position to show you understand it. To me, this implies that you don't fully understand your thesis.

And no, I'm not necessarily describing Snorphblats. If statement 1 is true, then statement two is just mouth noises without truth value. If statement 2 is true, then statement 1 is false. And, of course, they could both be false.

I keep saying the same thing because it's the same issue every time, and as I've already said there's no good faith way for me to respond to some of your more specific points because I disagree with the premise on which they are based. I think you are failing to understand the point being made, but it's in a subtle way that seems to be beyond my ability to convey to you in these comments.

That I don't "fully understand my thesis" is actually closer to hitting near the heart of the thing than much else, but not in the sense of my not understanding some particular thing, but rather in that it's a specific example of the general epistemological point at issue.

That I'm not being more explicit and actively deconstructing your responses in detail is a choice. Look, I could switch and do that. I've done it in the past. I'm not going to do it here, because experience tells me it has very poor return on investment for effort: it's no more likely to lead my interlocutor to understand, but it requires a great deal more effort on my part. I simply choose to engage at the level I do because I think it's my best tradeoff along the efficiency frontier of "doing my best". That it's not working for you is fine; you can give up anytime you like. Now is not always the right time to hear a thing and understand it in a particular way.

Hume made this point in An Enquiry Concerning Human Understanding. :) 

Edit: added a link. 

My point is that the entirety of what we think we mean when we say  is ontological, i.e it's in the map. Not once ever has causality existed in the territory.

and 

This means that there's no aspect of the territory that is causality. There's no , there's no , there's no , there's just "is".

I'm glad I'm not the only one who finds this self-evident. The world indeed just is. You can think of it in timeless terms, or in evolving terms, but it doesn't change anything. You are a part of the world and so you just are, as well. There is no causality in you except for a physical process in your brain doing what feels like reification.

I thought that the argument you made was quite assertive. I don't think it's nearly as obvious as you claim that causality is in the map, even though it is a possibility that is often underappreciated.

Whenever I read or think about causation, I wish language allowed to make the distinction between the two types:

  •          (A is sufficient to cause B): Too many bullet holes cause death.
  •    (A is necessary to cause B): Lack of vitamin C causes scurvy.

Language does, because it allows you to define necessary but insufficuent causes, unnecessary but sufficient causes, and so on. Language contains a lot of subtle distinctions that aren't made clearer -- quite the contrary -- by folding them all into the graphical symbol of an arrow.

It sounds like you're begging the question here.

Unless you are positing strict determinism, or full timeless physics, or literal "there is no territory," a causing b just means a is either a necessary, sufficient, or both condition for b. Define a and b at whatever level you want to call the "territory."

My point here is epistemological, not physical or metaphysical. I need not posit anything be true about the territory, only that I can say nothing about it without imposing some map, model, categories, etc.. The moment we define something at any level we've left the territory for the map.

Or not. There's a metaphor where concepts and categories are something you see instead of the territory, ie. the veil of perception, but there's also a metaphor where concepts and categories are a lens through which one sees the territory. Making an arbitrary choice of metaphor doesnt prove anything.

So for this we do have some pretty good evidence, namely the process of observation and perception appears to be a physical one, i.e. we have no evidence that anything can be perceived directly just as it is, and in fact we'd expect the world to look different if we could, such as there being a lack of arguments about the boundaries of categories.

So for this we do have some pretty good evidence,

For what? For which of the three or four views you have been putting forward?

Science gives us good evidence against naive realism, but theproblem is that there is more than one alternative to naive realism. So, further arguments are needed to support a specific alternative to naive realism.

I think what you're seeing here is that there's multiple aspects to what I'm saying, which look like different views but really aren't.

Feel free to explain how they come together.

So, what is the difference between that episiotomy (all communication describes a map) and mine (some communication describes the territory)? Like: on what kinds of questions would they lead to different answers? Is one of them faster, or less likely to lead to errors? Is there some other distinction in effect that I just can't conceive?

Neither, the descriptions you give are both partial, and both are true and not in conflict. I think the only issue perhaps is thinking that "all communication describes a map" is the sum total of what I'm saying here.

My constructions are, by definition, in conflict (if you assume map != territory, which is a question of definitions. Let me know if you are using those words differently). How does your thesis differ from "all communication describes a map"?

Where is the conflict? It can be both true that, for example, a cup is a thing reified in the map and there is some aspect of the territory that cup predicts, at least based on how it affects a map.

Of course you want a map that describes the territory, but that does not mean the same thing as ontology = metaphysics.

Can you give an example of describing the territory that does not include reification? I would accept "the indented torus collection of covalent bonded atoms at approximately [1' x 2 ' x 2 ' with my brain as the origin point] emits electromagnetic radiation primarily in the 650-680 nm range" (ie "That cup is red"), but I suspect that you would not.

I can't, because that's impossible. There's nothing I can say that does not involve drawing a map.

So, what is the difference between that episiotomy (anything you can say draws a map) and mine (some communication describes the territory)? Like: on what kinds of questions would they lead to different answers? Is one of them faster, or less likely to lead to errors? Is there some other distinction in effect that I just can't conceive?

This question remains confused so I can't answer it in good faith, because I've already rejected its premise above.

Um, you literally just said two posts up that you can't say anything that does not involve drawing a map. That you are defining "the territory" to be out of reach of mere words.

I'm asserting that "the territory" should be defined to be at the limit of what words can describe, but still within bounds. I'm literally saying "take your meta-map of how maps & territory work, and move one inferential step towards the position "there is no map, it's all territory.""

Then, go back and review my previous questions.

"There is no causation in the territory" is a claim about the territory. It cannot be asserted if the territory is unknowable.

You need to distinguish between a number of subtly different claims:

  1. The territory is knowable, and known not to contain certain things (which are nonetheless useful for mapping purposes . (Eg. Nobody has 2.4 children).

  2. The territory is only knowable , "via" human concepts (Kant)

  3. The territory is unknowable, or equivalently correspondence between map and territory is impossible.

Etc.

We need not be so detailed. To make a claim about the territory is to be in the map, and thus the point is already proved.

To be in the map is to be in the territory.

My point is that the entirety of what we think we mean when we say A⟹B is ontological, i.e it’s in the map. Not once ever has causality existed in the territory.

Is there an argument to support that?

Try to imagine what the world is like if you’re not modeling it. Are you picturing atoms? Particles? Wave functions? Strings?

Wrong!

Those are all models we impose on the world to make sense of it. Useful ones, usually, but still models

And is there an argument to support that?

I can see how the mere usefulness of a map level category isn't sufficient reason to believe that the feature exists in the territory. But, equally, the usefulness of a map level feature is no reason to believe that there is not something corresponding to it in the territory. Nor is the fact that it is a map level symbol. Maps are not necessarily false, nor necessarily true.

What’s an example of a misconception someone might have due to having a mistaken understanding of causality, as you describe here?

Generally, supposing the existence of particular things prior to the experience of them. The key insight is to see that the existence of "things" is not identical to the existence of reality out of which things are carved.

Take literally anything and it's your example: a cup, an atom, experience, causation, dancing, etc.

You can find none of these things in the territory itself, only in your understanding of it (and yet something is there in the territory for you to create a useful understanding of it, but it only becomes a thing by virtue of some perception of it).

You can find the raw material of the things in the territory, but you can't find the template that you are using to categorise them..so says the cookie cutter theory .you are supplying the cutter, not the dough .

Formally, we often model causation as the action of one thing implying another, and we might formalize this with mathematical notation like to mean some event or thing causes some other event or thing to happen.

Causation is not so easy to model. I have a job that requires a degree. This implies that I have a degree. But my having this job did not cause me to have a degree (to clarify, perhaps the expectation of gaining such a job caused me to get the degree, but not the job itself). From ET Jaynes, after a similar example:

This example shows also that the major premise, ‘if A then B’ expresses B only as a logical consequence of A; and not necessarily a causal physical consequence, which could be effective only at a later time. The rain at 10 am is not the physical cause of the clouds at 9:45 am. Nevertheless, the proper logical connection is not in the uncertain causal direction (clouds ⇒ rain), but rather (rain ⇒ clouds), which is certain, although noncausal. We emphasize at the outset that we are concerned here with logical connections, because some discussions and applications of inference have fallen into serious error through failure to see the distinction between logical implication and physical causation. The distinction is analyzed in some depth by Simon and Rescher (1966), who note that all attempts to interpret implication as expressing physical causation founder on the lack of contraposition expressed by the second syllogism (1.2). That is, if we tried to interpret the major premise as ‘A is the physical cause of B’, then we would hardly be able to accept that ‘not-B is the physical cause of not-A’. In Chapter 3 we shall see that attempts to interpret plausible inferences in terms of physical causation fare no better.

Judea Pearl is supposedly working on defining causation, though I know little of it. I think he talks abut "backwards causation" and the examples of it that I've heard (if I recall correctly) sound like they confuse the job with the expectation of the job. Maybe causation is an incoherent idea.

This means that there's no aspect of the territory that is causality. There's no A, there's no B, there's no ⟹, there's just "is".

Here's how I think of it: I have sensory data coming in (what I see and hear and so on), and every word I associate with that data is an abstraction that seems to match with a useful pattern within the data (e.g. "I'm looking at a table"). So I think we kind of agree with the "There's no A or B" (the table is not fundamentally part of the universe, it's my model of some collection of actual universe "stuff"[1]), though I would phrase it in a way in which people might think we disagree. I think it still makes sense to talk about A or B (tables are real), it's just that statements like "A is true" become a lot more slippery (but not in a "Everybody has their own truth, man" way). And this is the case, as you say, even at the level of atoms, though our sensory data is intermediated with other high-tech tools (e.g. electron microscopes or whatever).

Where I think we disagree is "there's no ⟹". Maybe there isn't. But the universe apparently follows some rules. The laws that physicists found may be implications of these rules, but they might be the rules themselves. For the sake of analogy, the "code" that the universe runs on might contain "matter/energy cannot be created or destroyed", and I think it's fair to consider this to be part of the universe (though whether we can establish that a rule is actually in the code is another matter). The rules might also contain something about causation.

Anyway, great post.


  1. Strictly speaking, it's not even that, it's a model of my experiences, which is a filtered and somewhat distorted version of the actual universe. ↩︎

You miss the mark here. You've confused the map for the territory, though to be fair it's easy to do because how do you think about how the world works without modeling it?

You'd have to point me to one of my sentences that you disagree with, since I don't think I've made that mistake.

Perhaps you think the only true fact about the universe is the whole universe itself, so in that case, talking about the "rules in the source code of the universe" wouldn't make sense. I'm having to guess here, since you haven't stated your disagreement. But if that is the case, you'd be assuming something about the nature of the universe. I only said that the universe might innately contain rules.

Where I think we disagree is "there's no ⟹". Maybe there isn't. But the universe apparently follows some rules. The laws that physicists found may be implications of these rules, but they might be the rules themselves. For the sake of analogy, the "code" that the universe runs on might contain "matter/energy cannot be created or destroyed", and I think it's fair to consider this to be part of the universe (though whether we can establish that a rule is actually in the code is another matter). The rules might also contain something about causation.

To posit that there are some rules or not is to have jumped from talking about the epistemological issue of not being able to address the territory except through the lens of the map to already supposing a map that carves up the world into thing that are rules (or not rules). You then go on to suppose these rules might contain something about causation, but now you've traveled miles down the road of accepting the framing of the map and jump towards a metaphysical framing of the territory as one that contains rules.

This is confusing map and territory, because beyond confusion it's both true that causation doesn't inherently exist in the territory (because nothing does) and that you can pick and choose useful structure out of the territory to say something about what you think it is, but only from within the framing of the map you've created.