Scientist by training, coder by previous session,philosopher by inclination, musician against public demand.
Team Piepgrass: “Worried that typical commenters at LW care way less than I expected about good epistemic practice. Hoping I’m wrong.”
TAG
Karma: 1,296
What is a sentence anyway… is there something special about a period, as opposed to other punctuation marks? Many are available: the colon is a possibility; also its half-brother; and the comma,of course...also the ellipsis—even the mighty m-dash!
The idea that grammar is just inflection is misleading: languages that are mostly isolating can have complex ordering rules,like the the notorious adjective ordering of English.
As for french …Moi, je ne me défile pas.
1st person. Sing.
1st person. Sing, again.
Negative.
1st person. Sing, reflexive.
Verb!!!
Negative,again.
I have heard of versions of many-worlds that are supposed to be testable
The’re are versions that are falsified, for all practical purposes, because they fail to.predict broadly classical observations—sharp valued real numbers, without pesky complex numbers or superpositions. I mean mainly the original Everett theory of 1957. There have been various attempts to patch the problems—preferred basis, Decoherence , anthropics, etc, -- so there are various non falsified theories.
The one that I’m most familiar with (“classic many-worlds”?) is much more of a pure interpretation, though: in that version, there is no collapse and the apparent collapse is a matter of perspective. A component of the wavefunction that I perceive as me sees the electron in the spin-down state, but in the big superposition, there’s another component like me but seeing the spin-up state. I can’t communicate with the other me (or “mes,” plural) because we’re just components of a big vector—we don’t interact.
Merely saying that everything is a component of a big vector doesn’t show that observers dont go into superposition with themselves, because the same description applies to anything which is in superposition..it’s a very broad claim.
What you call classic MWI is what I the have-your-cake-and-eat-it … assuming nothing except that collapse doesn’t occur, you conclude that observers make classical observations for not particular reason...you doing even nominate Decoherence or preferred basis as the mechanism that gets rid of the unwanted stuff.
On the other hand, classic decoherence posits that the wavefunction really does collapse, just not to 100% pure states. Although there’s technically a superposition of electrons and a superposition of mes, it’s heavily dominated by one component. Thus, the two interpretations, classic many-worlds and classic decoherence, are different interpretations.
OK. I would call that single world decoherence. Many worlders appeal to Decoherence as well.
So classic decoherence is more falsifiable than classic many-worlds.
If classic MW means Everetts RSI, it’s already false.
I’ve already told you why Im not going to believe chatGpt. Judge for yourself: https://www.researchgate.net/profile/Bruno-Marchal-3.
Why? I was there, it wasn’t.
Bruno Marchal was talking about this stuff in the nineties.
alignment is structurally impossible under competitive pressur
Alignment contrasts with control, as a means to AI safety.
Alignment roughly means the AI has goals, or values similar to human ones (which are assumed, without much evidence to be similar across humans), so that it will do what we want , because it’s what it wants.
Control means that it doesn’t matter what the AI wants, if it wants anything.
In short, there is plenty of competitive pressure towards control , because no wants an AI they can’t control. Control is part of capability.
MWI is more than one theory.
There is an approach to MWI based on coherent superpositions, and a version based on decoherence. These are (for all practical purposes) incompatible opposites, but are treated as interchangeable in Yudkowsky’s writings. Decoherent branches are large, stable, non interacting and irreversible...everything that would be intuitively expected of a “world”. But there is no empirical evidence for them (in the plural) , nor are they obviously supported by the core mathematics of quantum mechanics, the Schrödinger equation.Coherent superpositions are small scale , down to single particles, observer dependent, reversible, and continue to interact (strictly speaking , interfere) after “splitting”. the last point is particularly problematical. because if large scale coherent superposition exist , that would create naked eye evidence at macrocsopic scale:, e.g. ghostly traces of a world where the Nazis won.
We have evidence of small scale coherent superposition, since a number of observed quantum.effects depend on it, and we have evidence of decoherence, since complex superposition are difficult to maintain. What we don’t have evidence of is decoherence into multiple branches. From the theoretical perspective, decoherence is a complex , entropy like process which occurs when a complex system interacts with its environment. Decoherence isn’t simple. But without decoherence, MW doesn’t match observation. So there is no theory of MW that is both simple and empirically adequate, contra Yudkowsky and Deutsch.
Decoherence says that regions of large complex superpositions stop interfering with each other
It says that the “off diagonal” terms vanish, but that would tend to.generate a single predominant outcome (except, perhaps, where the environment is highly symmetrical).
Physicalism doesn’t solve the hard problem, because there is no reason a physical process should feel like anything from the inside.
Computationalism doesn’t solve the hard problem, because there is no reason running an algorithm should feel like anything from the inside.
Formalism doesn’t solve the hard problem, because there is no reason an undecideable proposition should feel like anything from the inside.
Of course, you are not trying to explain qualia as such, you are giving an illusionist style account. But I still don’t see how you are predicting belief in qualia.
And among these fictions, none is more persistent than the one we call qualia.
What’s useful about them? If you are going to predict (the belief in) qualia, on the basis of usefulness , you need to state the usefulness. It’s useful to know there is a sabretooth tiger bearing down in you , but why is an appearance more useful than a belief ..and what’s the use of a belief-in-appearance?
This suggests an unsettling, unprovable truth: the brain does not synthesize qualia in any objective sense but merely commits to the belief in their existence as a regulatory necessity.
What necessity?
ETA:
self-referential, self-regulating system that is formally incomplete (as all sufficiently complex systems are) will generate internally undecidable propositions. These propositions — like “I am in pain” or “I see red” — are not verifiable within the system, but are functionally indispensable for coherent behavior.
I still see no reason why an undecideable proposition should appear like a quale or a belief in qualia.
That failure gets reified as feeling.
Why?
I understand that you invoke the “Phenomenological Objection,” as I also, of course, “feel” qualia. But under EN, that feeling is not a counterargument — it’s the very evidence that you are part of the system being modeled.
Phenomenal conservatism , the idea that if something seems to exist ,you should (defeasibly) assume it does exist,.is the basis for belief in qualia. And it can be defeated by a counterargument, but the counter argument needs to be valid as an argument. Saying X’s are actually Y’s for no particular reason is not valid.
This argument is based on a conflation. It assumes that there’s one single thing, “morality”, and this one thing produces not only answers to “what should you do”, but also, what should we condemn, what should we punish, what should one feel guilty about, and other similar questions; and that, moreover, the answers to these questions are identical (or opposites, as appropriate; here the first one would be the opposite of the others).
Yes: consequentialism, deontology, etc, are are different aspects of morality, and even relate to different things...the permissible versus the desirable, etc. Yet they can be reconciled:-
One of the areas of debate about ethics is where the locus of ethical concern is .. whether it lies in persons (an approach known as virtue ethics), rules (deontology) or the consequences of actions (consequentialism). (There are also other axes, such as objectivity versus subjectivity and cognitivism versus non cognitivism). Consider a case where someone dies in an industrial accident , although all rules were followed: if you think the plant manager should be exonerated because he folowed the rules, you are siding with deontology, whereas if you think he should be punished because a death occurred under his supervision, you are siding with consequentialism.
Many people encounter deontology in the form of “ten commandments” style religious law. From a rational perspective, this kind of deontology is unsatisfactory: for one thing, there are multiple competing systems, and it is not clear why any system should be followed, and it is difficult to adapt traditional deontology to new circumstancs. Likewise, virtue ethics suffers from disagreements about what is virtuous.. for instance, the stength-and-independence cluster of virtues versus the kindness-and-cooperation cluster.
To those who are looking for a rational basis for ethics (which includes most of theose seeking to find a motivating basis for ethics), consequentialism is more attractive. A basis in the preferences and values people actually have, which are cognitevely accessible. Being based on the preferences and values people actually have, goes a considerable way to finding motivation to behave ethically (although there is a considerable wrinkle in blancing “my “preferences against “yours”).. And it is possible to adapt to changing circumstances in terms of the results we would wish to get out of them. These are the advantages of consequentialism.
On the other hand, ethics as it is exists in human societies doesn’t have a strong rational basis. Psyhcological studies show that people’s thinking about ethics is instead intuitive, and often centered on rules and virtues which are taken for granted. However, the arguments for consequentialism so far given don’t add up to arguments for pure consequentialism. The more sophisticated defences of rational ethics can include aspects of deontology and virtue ethics.
The disadvantages of (most forms of) consequentialism include the fact that consequences are impossible to calculate exactly, in general. Secondly, different individuals, approximating conseqential decision making differently, would lead to lack of coordination. For instance, . Thirdly, it is unreasonable to punish people for consequences of intentional actions whose outcomes they could not forsee. Fourthly, it is unreasonable to punish people for what unintentional actions.
Rules that are commonly agreed, and which lead to (approximately) desireable outcomes, in the consequentialist sense solve all these problems. Firstly, it is possible to memorise a set of rules. Secondly, if everyone follows the same rules, it is possible to co-ordinate. Thirdly, it is reasonable to punish someone for failing to follow a rule they knew about and knew they should be followed. Fourthly, new rules can be formulated in response to to changing circumstances, since it is possible to choose rules that lead to desirable expected outcomes.
(“Right” and “wrong”, that is praisweorthiness and blameability are concepts that belong to deontology. A good outcome in the consequentialist sense, one that is a generally desired, is a different concept from deontological right. Ethics does not have a single subject mtter ..it is about goodness in the sense of desireable ends and goodness in the sense of right behaviour and goodness in the sense of virtue.)
The advantages of consequentialism can still be retained by basing rules on expected consequences. That is very much a compromise, though. A finite and cognitively manageble set of rules can only approximate the case-by-case calculations of an ideal ethical reasoner. But ideal ethical reasoners don’t exist. (But some people reason better than others, even though everyone is obliged to follow the same set of rules...)
The explicit construction of rules is apparent in modern, tenchologically advanced societies, since such societies face challenges to adapt socially to their technological innovations. Nonetheless, the rules of a more traditional society can be retrospectively seen as gradual adaptations, existing in order to bring about desirable consequences. And inadmuch as ethical rules exist to fulfil a purpose, bringing about desirable consequences, they can be seen as doing so better or worse.
People need to actually be able to act on morality, which is where virtue (in one sense) comes in. Virtue can mean moral fibre, an inner capacity to do what is not in your selfish interest, or , alternatively moral standing or status that rewards people for being moral. Virtue correlates more with reward, deontology more with punishment.
But there’s something else, which is a very finite legible learning algorithm that can automatically find all those things
Is there? I see a.lot of talk about brain algorithms here, but I have never seen one stated...made “legible”.
—the object-level stuff and the thinking strategies at all levels. The genome builds such an algorithm into the human brain
Does it? Rationalists like to applaud such claims, but I have never seen the proof.
And it seems to work!
Does it? Even If we could answer every question we have ever posed, we could still have fundamental limitations. If you did have a fundamental cognitive deficit, that prevents you from.understanding some specific X how would you know? You need to be able to conceive X before conceiving that you don’t understand X. It would be like the visual blind spot...which you cannot see!
And then I’m guessing your response would be something like: there isn’t just one optimal “legible learning algorithm” as distinct from the stuff that it’s supposed to be learning. And if so, sure
So why bring it up?
there isn’t just one optimal “legible learning algorithm”
Optimality—doing things efficiency—isn’t the issue, the issue is not being able to do certain things at all.
I think this is very related to the idea in Bayesian rationality that priors don’t really matter once you make enough observations.
The idea is wrong. Hypotheses matter , because if you haven’t formulated the right hypothesis , no amount of data will confirm it. Only worrying about weighting of priors is playing in easy mode, because it assumes the hypothesis space is covered. Fundamental cognitive limitations could manifest as the inability to form certain hypotheses. How many hypotheses can a chimp form? You could show a chimp all the evidence in the world, and it’s not going to hypothesize general relativity.
Rationalists always want to reply that Solomonoff inductors avoid the problem on the basis that SIs consider “every” “hypothesis”… but they don’t , several times over. It’s not just that they are uncomputable, it’s also that it’s not know that every hypothesis can be expressed as a programme. The ability to range over a complete space does not equate to the ability to range over Everything.
Here’s an example: If you’ve seen a pattern “A then B then C” recur 10 times in a row, you will start unconsciously expecting AB to be followed by C. But “should” you expect AB to be followed by C after seeing ABC only 2 times? Or what if you’ve seen the pattern ABC recur 72 times in a row, but then saw AB(not C) twice? What “should” a learning algorithm expect in those cases? You can imagine a continuous family of learning algorithms, that operate on the same underlying principles.
A set of underlying principles is a limitation. SIs are limited to computability and the prediction of a sequence of observations. You’re writing as that something like prediction of the next observation is the only problem of interest , but we don’t know that Everything fits that pattern. The fact that Bayes and Solomomoff work that way is of no help, as shown above.
But within this range, I acknowledge that it’s true that some of them will be able to learn different object-level areas of math a bit faster or slower, in a complicated way, for example.
But you haven’t shown that efficiency differences are the only problem. The nonexistence of fundamental no-go areas certainly doesn’t follow from the existence of.efficiency differences.
, it can still figure things out with superhuman speed and competence across the board
The definition of superintelligence means that “across the board” is the range of things humans do, so if there is something humans can’t do at all,an ASI is not definitionally required to be able to do it.
By the same token, nobody ever found the truly optimal hyperparameters for AlphaZero, if those even exist, but AlphaZero was still radically superhuman
The existence of superhuman performance in some areas doesn’t prove adequate performance in all areas, so it is basically irrelevant to the original question, the existence of fundamental limitations in humans.
OP discusses maths from a realist perspective. If you approach it as a human construction, the problem about maths is considerably weakened...but the wider problem remains, because we don’t know that maths is Everything.
this is conflating the reason for why one knows/believes P versus the reason for why P,
Of course, that only makes sense assuming realism.
You are understating your own case, because there is a difference between mere infinity and All Kinds of Everything. An infinite collection of one kind of thing can be relatively tractable.
Again, those are theories of consciousness, not definitions of consciousness.
I would agree that people who use consciousness to denote the computational process vs. the fundamental aspect generally have different theories of consciousness, but they’re also using the term to denote two different things.
But that doesn’t imply that they disagree about (all of) the meaning of the term “qualia”..since denotation (extension, reference)doesn’t exhaust meaning. The other thing is connotation, AKA intension, AKA sense.
https://en.m.wikipedia.org/wiki/Sense_and_reference
Everyone can understand that the qualia are ,minimally, things like the-way-a-tomato-seems-to-you, so that’s agreement on sense , and the disagreement on whether the referent is “physical property”, “nonphysical property” , “information processing”, etc, arises from different theoretical stances.
(I think this is bc consciousness notably different from other phenomena—e.g., fiber decreasing risk of heart disease—where the phenomenon is relatively uncontroversial and only the theory about how the phenomenon is explained is up for debate.
That’s an odd use of “phenomenon”...the physical nature of a heart attack is uncontroversial, and the controversy is about the physical cause. Whereas with qualia, they are phenomenal properly speaking..they are appearences...and yet lack a prima facie interpretation in physical (or information theoretic) terms. Since qualia do present themselves immediately as phenomenal, then outright denial …feigning anaesthesia or zombiehood.. is a particular poor response to the problem. And the problem is different to “how does one physical event cause another one that is subsequent in time”...it’s more like “how or whether qualia, phenomenal consciousness supervenes synchronously on brain states”. .
With consciousness, there are a bunch of “problems” about which people debate whether they’re even real problems at all (e.g., binding problem, hard problem). Those kinds of disagreements are likely causally upstream of inconsistent terminology.)
If you don’t like the terminology, you can invent better terminology. Throughout this exchange , you have been talking in terms of “consciousness” , and I have been replying in terms of “qualia”, because “qualia” is a term that was invented to hone in on the problem, on the aspects of consciousness where it isn’t obviously just information processing. (I’m personally OK with using information theoretic explanations, such as global workplace theory, to address Easy Problem issues , such as Access Consciousness).
Theres a lot to be said for addressing terminological.issues, but it’s not an easy win for camp #1.
If it was that easy to understand, we wouldn’t be here arguing about it.
Definitions are not theories
Even if there is agreement about the meaning of the word, there can also be disagreement about the correct theory of qualia. Definitions always precede theories—we could define “Sun” for thousands of years before we understood its nature as a fusion reactor. Shared definitions are a prerequisite of disagreement , rather than just talking past each other.
The problem of defining qualia—itself, the first stage in specifying the problem—can be much easier than the problem of coming up with a satisfactory theoretical account, a solution. It’s a term that was created by an English speaking philosopher less than a hundred years ago, so it really doesn’t present the semantic challenges of some philosophical jargon.
(The resistance to qualia can also be motivated by unwillingness to give up commitments—bias, bluntly—not just semantic confusion)
My claim is that arguments about qualia are (partially) caused by people actually having different cognitive mechanisms that produce different intuitions about how experience works.
Semantic confusions and ideological rigidity already abound, so there is no need to propose differing cognitive mechanisms.
Theories about how qualia work don’t have to be based on direct intuition. Chalmers arguments are complex, and stretch over 100s of pages.
Well, I’m glad you’ve settled the nature of qualia. There’s a discussion downthread, between TAG and Signer, which contains several thousand words of philosophical discussion of qualia.
Again, the definition is one thing and the “nature”...the correct ontological theory...is another. The definition is explained by Wikipedia, the correct theory , the ultimate explanation is not.
Seriously, I definitely have sensations.
“Sensation” is ambiguous between a functional capacity—stopping at a red light—and a felt quality—what red looks like. The felt quality is definitely over and above the function, but that’s probably not your concern.
I just think some people experience an extra thing on top of sensations, which they think is an indissoluble part of cognition, and which causes them to find some things intuitive that I find incomprehensible.
It’s true that some people have concluded nonphysical theories from qualia… but it doesn’t follow that they must be directly perceiving or intuiting any kind of nonphysicalism in qualia themselves. Because it’s not true that every conclusion has to be arrived at immediately, without any theoretical, argumentative background. Chalmers’ arguments don’t work that way and are in fact quite complex.
Physics is a complex subject that needs to be learnt. To know what is physical is therefore not a matter of direct intuition...so to know that qualia are not physical is also not a matter of direct intuition.
There’s no existing, successful, physical or computational theory of qualia. The people who think qualia aren’t physical, aren’t necessarily basing that on some kind of direct perception, and don’t necessarily know less physics than the people who do.
Consciousness itself is overloaded (go figure!) since it can refer to both “a high-level computational process” and “an ontologically fundamental property of the universe”.
Again, those are theories of consciousness, not definitions of consciousness.
Qualia can be a synonym for consciousness (if you are in Camp #2) or mean something like “this incredibly annoying and ill-defined concept that confused people insist on talking about” (if you’re in Camp #1). I recommend only using this term if you’re talking to a Camp #2 audience.
There are many more than two possibilities. You can take consciousness seriously, in Chalmer’s sense, and accept that there is a Hard Problem, without denying that there are other, easier aspects to consciousness.. Chalmers accepts that there are easy problem as well.
And the definition problem becomes much easier if you remember that definitions aren’t theories.
Free will in the general context means that you are in complete control of the decisions you make, that is farthest from the truth. Sure you can hack your body and brain …
Why “complete” control? You can disprove anything , in a fake sort of way, by setting then bar high—if you define memory as Total Recall, it turns out no-one has a memory.
Who’s this “you” who’s separate from both brain and body? Shouldn’t you be asking how the machine works? A machine doesn’t have to be deterministic , and can be self-modifying.
When Robert Sapolsky says there is no free will, he means that if we know your current body state perfectly, we can predict with 100% accuracy what you will do in the next moment given an input.
We can’t, in general. Theres no perfect predictability in the human sciences.
I specifically mentioned wife instead of a generic friends
Then you you are picking a special case to make a general point.
If we sufficiently understand how the brain and body works we should be able to predict.
Why? Determinism isn’t a fact. We don’t have evidence of physical determinism, so we can’t make a bottom up argument, and we dont have perfect predictability in psychology, either.
Why would anyone ever care if a god could predict their actions, when no such god exists, and humans can only make bad guesses?I
Predictability implies determinism, determinism implies no (libertarian) free will.
Possibly we are just in one of the mathematical universes that happens to have an arrow of time—the arrow seems to arise from fairly simple assumptions, mainly an initial condition and coarse graining
You are misunderstanding the objection. It’s not just an arrow of time in the sense of order, such as increasing entropy, it’s the passingness of the time. An arrow can exist statically, but that’s not how we experience time. We don’t experience it as a simultaneous group of moments that happen to be ordered , we experience one moment at a time. A row of houses is ordered but not one-at-a-time, like a succession of movie frames.
The valence of pleasure and pain is not just a sign change, they serve vastly different psychological functions and evolved for distinct evolutionary reasons.
And the associate qualia? What’s the mathematical theory of qualia? Is it bottom-up …we have some mathematical descriptions of qualia, and it’s only a matter of time before we have the rest...or top-down...everything is mathematical, so qualia must be...?
(Extensively reviesed and edited).
Reductionism
Reductionism is not a positive belief, but rather, a disbelief that the higher levels of simplified multilevel models are out there in the territory.
Things like airplane wings actually are, at least as approximations. I don’t see why you are.approvingly quoting this: it conflates reduction and elimination.
But the way physics really works, as far as we can tell, is that there is only the most basic level—the elementary particle fields and fundamental forces.
If that’s a scientific claim ,it needs to be treated as falsifiable, not as dogma.
You can’t handle the raw truth, but reality can handle it without the slightest simplification. (I wish I knew where Reality got its computing power.)”
It’s not black and white. A simplified model isn’t entirely out there, but it’s partly out there. There’s still a difference between an aeroplane wing and horse feathers.
Vitalistic Force
Vitalistic force (§3.3) is an intuitive concept that we apply to animals, people, cartoon characters, and machines that “seem alive” (as opposed to seeming “inanimate”).
It amounts to a sense that something has intrinsic important unpredictability in its behavior
The intuitive model says that the decisions are caused by the homunculus, and the homunculus is infused with vitalistic force and hence unpredictable. And not just unpredictable as a state of our limited modeling ability, but unpredictable as an intrinsic property of the thing itself—analogous to how it’s very different for something to be “transparent” versus “of unknown color”, or how “a shirt that is red” is very different from “a shirt that appears red in the current lighting conditions
Unpredictability is the absence of a property: predictability. Vitalistic force sounds like the presence of one. It’s difficult to see why a negative property would equate to a positive one. We don’t have to regard an unpredictable entity as quasi-alive. We don’t regard gambling machines in casinos as quasi alive. Our ancestors used to regard the weather as quasi alive, but we don’t—so it’s not all that compulsive. We also don’t have to regard living things as unpredictable—an ox ploughing a furrow is pretty predictable. Unpredictability and vitalism aren’t the same concept, and aren’t very rigidly linked, psychologically.
It doesn’t veridically (§1.3.2) correspond to anything in the real world (§3.3.3).
Except..
Granted, one can argue that observer-independent intrinsic unpredictability does in fact exist “in the territory”. For example, there’s a meaningful distinction between “true” quantum randomness versus pseudorandomness. However, that property in the “territory” has so little correlation with “vitalistic force” in the map, that we should really think of them as two unrelated things.
So let’s say that two different things: unpredictableness , non-pseudo randomness could exist in the territory, and could found a real, non-supernatural version of free will. Vitality could exist in the territory too—reductionism only requires that it is not fundamental, not that it is not real at all. It could be as real as an airplane wing. Reduction is not elimination.
However, that property in the “territory” has so little correlation with “vitalistic force” in the map, that we should really think of them as two unrelated things
So what is the definition of vitalistic force that’s a) different from intrinsic surprisingness b) incapable of existing in the territory even as an approximation?
Homunculi
The strong version of the homunculus , the one-stop-shop that explains everything about consciousness, identity, and free will, is probably false...but bits and pieces of it could still be rescued.
Function: it’s possible that there are control systems even if they don’t have a specific physical location.
Location: Its quite possible for higher brain areas to be a homunculus (or homunculi) lite, in the sense that , they exert executive control, or are where sensory data are correlated. Rejecting ghostly homunculi because they are ghostly doesn’t entail rejecting physical homunculi The sensory and mirror homunculi.
Vitalism: It’s possible for intrinsic surprisingness to exist in the territory, because intrinsic surprisingness is the same thing as indeterminism.
There’s also a further level of confusion about whether your idea of homunculus is observer or observed.
Are “we” are observing “ourselves” as a vitalistic homunculus , observing the rest of ourselves? If the latter, which is the real self, the the observer or the homunculus?
As discussed in Post 1, the cortex’s predictive learning algorithm systematically builds generative models that can predict what’s about to happen
No one has discovered a brain algorithm, so far.
Free Will
the suite of intuitions related to free will has spread its tentacles into every corner of how we think and talk about motivation, desires, akrasia, willpower, self, and more
And now we come to the part of the argument where an objective unbiased assessment of free will. concludes that the concept (or rather concepts) are so utterly broken and wrong that any vestige has to be “rooted out”.
Now, I expect that most people reading this are scoffing right now that they long ago moved past their childhood state of confusion about free will. Isn’t this “Physicalism 101” stuff?
It’s the case that a lot of people think that the age old problem of free will is solved at a stroke by “physics, lol”… but there are also sophisticated naturalistic defences.
There are two dimensions to the problem: the what-we-mean-by-free-will dimension, and the what-reality-offers-us dimension. The question of free will partially depends on how free will is defined, so accepting a basically scientific approach does not avoid the “semantic” issues of how free will, determinism , and so on, are best conceptualised.
I don’t know what people mean by “free will” and I don’t think they usually do either.
Professional philosophers are quite capable of stating their definitions, and you at capable of looking them up.)
Mr. Yudkowsky has no novel insight to offer into how the territory works, nor any novel insight into the correct semantics of free will. He has not solved either sub problem, let alone both. He has proposed a mechanism (not novel) about how the feeling of free will could be a predictable illusion, but that falls short of proving that it is..he basically relies on having an audience who are already strongly biased against free will.
To dismiss fee will, just on the basis of Physicalism, not even deterministic physics, is to tacitly define it as supernatural. Does everyone define it that way? No,there are compatibilists and naturalistic libertarians.
Compatibilism is a naturalistic theory of free will, and libertarianism can be.
(https://insidepoliticalscience.com/libertarian-free-will-vs-compatibilism/)
To provide a mechanism by which the feeling of free will could be an illusion , which he had done, , does not show that it actually is an illusion, because of the usual use laws of modal logic—he needs to show that his model is the only possibility, not just a possibility. (These problems were pointed out long ago, of course).
It is possible, in the right kind universe to have libertarian free will backed by an entirely physical mechanism, since physics be indeterministic … and to have a veridical perception of it. The existence of another possibility, where the sense of free will is illusory, doesn’t negate the veridical possibility. “Yes,but physicalism ” doesn’t either.
You don’t observe your brain processes so you don’t observe them as deterministic or indeterministic .. An assumption of determinism has been smuggled in by a choice of language, the use of the word “algorithm”. But, contrary to what many believe, algorithms can be indeterministic.
If someone demonstrated that brains run on an indeterministic algorithm, that fulfils the various criteria for libertarian free will, would you still deny that humans have any kind of free will?
Didn’t Eliezer Yudkowsky describe free will as “about as easy as a philosophical problem in reductionism can get, while still appearing ‘impossible’ to at least some philosophers”?
Questions can seem easy if you don’t understand their complexities.
Yudkowsky posted his solution to the question of free will along time ago, and the problems were pointed out almost immediately. And ignored for over a decade.
More precisely: If there are deterministic upstream explanations of what the homunculus is doing and why, e.g. via algorithmic or other mechanisms happening under the hood, then that feels like a complete undermining of one’s free will and agency (§3.3.6)
Why? How can you demonstrate that without a definition of free will Obviously , that would have no impact given the compatibilist definition of free will, for instance?
I have had a lot of discussions on the subject , and I have noticed that many laypeople believe in dualism, or a ghost -in-the-machine theory. In that case, I suppose lead that the machine is do it could be devastating. But..I said laypeople. Professional philosophers generally don’t define FW that way, and don’t think that dualism and free will are the same thing.
Typical definitions are:-
-
The ability or discretion to choose; free choice.
-
The power of making choices that are neither determined by natural causality nor predestined by fate or divine will.
-
A person’s natural inclination; unforced choice.
And if there are probabilistic upstream explanations of what the homunculus is doing and why, e.g. the homunculus wants to eat when hungry, then that correspondingly feels like a partial undermining of free will and agency, in proportion to how confident those predictions are.
That’s hardly an undermining of libertarian free will at all..LFW only requires that you could have done otherwise..not that you could have done anything at all, or that you could defy statistical laws.
The way intuitive models work (I claim) is that there are concepts, and associations / implications / connotations of those concepts. There’s a core intuitive concept “carrot”, and it has implications about shape, color, taste, botanical origin, etc. And if you specify the shape, color, etc. of a thing, and they’re somewhat different from most normal carrots, then people will feel like there’s a question “but now is it really a carrot?” that goes beyond the complete list of its actual properties.
There’s way of thinking about free will and selfhood that is just a list of naturalistically respectable properties , and nothing beyond. Libertarianism doesn’t require imperceptible essences, on the naturalistic view, it could just be the operation of a ghost-free machine.I
According to science, the human brain/body is a complex mechanism made up of organs and tissues which are themselves made of cells which are themselves made of proteins, and so on.
Science does not tell you that you are a ghost in a deterministic machine, trapped inside it and unable to control its operation. Or that you are an immaterial soul trapped inside an indetrministic machine. Science tells you that you are, for better or worse, the machine itself.
Although I have used the term “machine”, I do not intend to imply that a, machine is necessarily deterministic. It is not known whether physics is deterministic, so “you are a deterministic machine” does not follow from “you are entirely physical”. The correct conclusion is “you are no more undetermined than physics allows you to be”.
So the scientific question of free will becomes the question of how the machine behaves, whether it has the combination of unpredictability, self direction, self modification and so on, that might characterise free will… depending on how you define free will.
There is a whole science of self-controlling machines: cybernetics. Airplane autopilots and , more recently, self driving cars are examples. Self control, without indeterminism is not sufficient for libertarian free will, but indeterminism without self control is not either
All of those things can be ascertained by looking at a person (or an animal or a machine) from the outside. They don’t require a subjective inner self… unless you define free will that way. If you define free will as dependent on a ghostly inner self, then you are not going to have a scientific model of free will.
Consciousness
As a typical example, Loch Kelly at one point mentions “the boundless ground of the infinite, invisible life source”. OK, I grant that it feels to him like there’s an infinite, invisible life source. But in the real world, there isn’t. I’m picking on Loch Kelly, but his descriptions of PNSE are much less mystical than most of them. ”
I grant that it feels to you like you have certain knowledge of the universe’s true ontology, but at best what you actually have a set of scientific models—mental constructs, maps—that make good predictions. I am not saying I have certain knowledge that the mystical ontology is certainly correct, I am saying we are both behind Kantian veils. Prediction underdermines ontology. So long as boundless life source somehow behaves just like matter, under the right circumstances, physics can’t disprove it—just as physicalism requires matter to behave like consciousness, somehow, under the right circumstances
The old Yudkowsky post “How An Algorithm Feels From Inside” is a great discussion of this point.
As has been pointed out many times, there is no known reason for an algorithm to feel like anything from the inside
-
This Cartesian dualism in various disguises is at the heart of most “paradoxes” of consciousness. P-zombies are beings materially identical to humans but lacking this special res cogitans sauce, and their conceivability requires accepting substance dualism.
Only their physical possibility requires some kind of nonphysicality. Physically impossible things can be conceivable if you don’t know why they are physically impossible, if you can’t see the contradiction between their existence and the laws of physics. The conceivability of zombies is therefore evidence for phenomenal consciousness not having been explained, at least. Which it hasn’t anyway: zombies are in no way necessary to state the HP.
The famous “hard problem of consciousness” asks how a “rich inner life” (i.e., res cogitans) can arise from mere “physical processing” and claims that no study of the physical could ever give a satisfying answer.
A rich inner life is something you have whatever your metaphysics. It doesn’t go.away when you stop believing in it. It’s the phenomenon to be explained. Res Cogitans, or some other dualistic metaphysics, is among an number of ways explaining it...not something needed to pose the problem.
The HP only claims that the problem of phenomenal consciousness is harder-er than other aspects of consciousness. Further arguments by Chalmers tend towards the lack of a physical solution, but you are telescoping them all into the same issue.
We have also solved the mystery of “the dress”:
But not the Hard Problem: the HP is about having any qualia at all, not about ambiguous or anomalous qualia. There would be an HP if everyone just saw the same.uniform shade of red all the time.
As with life, consciousness can be broken into multiple components and aspects that can be explained, predicted, and controlled. If we can do all three we can claim a true understanding of each
If. But we in fact lag in understanding the phenomenal aspect, compared to the others. In that sense, there is a defacto hard-er problem.
The important point here is that “redness” is a property of your brain’s best model for predicting the states of certain neurons. Redness is not “objective” in the sense of being “in the object”.
No, that’s not important. The HP starts with the subjectivity of qualia, it doesn’t stop with it.
Subjectivity isn’t just the trivial issue of being had by a subject, it is the serious issue of incommunicability, or ineffability.
Philosophers of consciousness have committed the same sins as “philosophers of life” before them: they have mistaken their own confusion for a fundamental mystery, and, as with élan vital, they smuggled in foreign substances to cover the gaps. This is René Descartes’ res cogitans, a mental substance that is separate from the material.
No, you can state and justify the HP without assuming dualism.
Are you truly exercising free will or merely following the laws of physics?
Or both?
And how is the topic of free will related to consciousness anyway?
There is no “spooky free will”
There could be non spooky free will...that is more than a mere feeling. Inasmuch as Seth has skipped that issue—whether there is a physically plausible, naturalistic free will—he hasn’t solved free will.
There are ways in which you could have both, because there are multiple definitions of free will, as well as open questions about physics. Apart from compatibilist free will, which is obviously compatible with physics, including deterministic physics, naturalistic libertarian free will is possible in an indeterministic universe. NLFW is just an objectively determinable property of a system, a man-machine. Free will doesn’t have to be explained away, and isn’t direct require an assumption of dualism.
But selfhood is itself just a bundle of perceptions, separable from each other and from experiences like pain or pleasure.
The subjective e, sense -of-self is,.pretty much by definition. Whether there are any further objective facts, that would answer questions about destructive teleportation and the like, is another question. As with free will, explaining the subjective aspect doesn’t explain away the objective.aspect.
First, computationalism doesn’t automatically imply that, without other assumptions, and indeed there are situations where you can’t clone data perfectly,
Thats a rather small nit. The vast majority of computationalists are talking about classical computation.
Indeed, I was basically trying to say that computationalism is so general that it cannot predict any result that doesn’t follow from pure logic/tautologies,
That’s not much of a boast: pure logic can’t solve metaphysical problems about consciousness, time, space, identity, and so on. That’s why they are still problems. There’s a simple logical theory of identity, but it doesn’t answer the metaphysical problems, what I have called the synchronic and diachronic problems.
Secondly, one could semi-reasonably argue that the inability to clone physical states is an artifact of our technological immaturity, and that in the far-future, it will be way easier to clone physical states to a level of fidelity that is way closer to the level of copyability of computer programs.
Physicalism doesn’t answer the problems. You need some extra information about how similar or different physical things are in order to answer questions about whether they are the same or different individuals. At least, if you want to avoid the implications of raw physicalism—along the lines of “if one atom changes, you’re a different person”. An abstraction would be useful—but it needs to be the right one.
Third, I gave a somewhat more specific theory of identity in my linked answer, and it’s compatible with both computationalism and physicalism as presented, I just prefer the computationalist account for the general case and the physicaliskt answer for specialized questions.
You seem to be saying that consciousness is nothing but having a self model, and whatever the self believes about itself is the last word...that there are no inconvenient objective facts that could trump a self assessment (“No you’re not the original Duncan Idaho, you’re ghola number 476. You think you’re the one and only Duncan because you’re brain state is a clone of the original Duncan’s”). That makes things rather easy. But the rationalist approach to the problem of identity generally relies on bullet biting about whatever solution is appealing—if computationalism is is correct, you can be cloned, and the you really are on two places at once.
My main non-trivial claim here is that the sense of a phenomenal experience/awareness fundamentally comes down to the fact that the brain needs to control the body, and vice-versa, so you need a self-model of yourself, which becomes a big part of why we say we have consciousness, because we are referring to our self models when we do that.
Well, how? If you could predict qualia from self control, you’d have a solution—not a dissolution—to the HP.
Another reason why the hard problem seems hard is that way too many philosophers are disinclined to gather any data on the phenomenon of interest at all, because they don’t have backgrounds in neuroscience, and instead want to purely define consciousness without reference to any empirical reality.
Granting that “empirical” means “outer empirical” …. not including introspection.
I don’t think there is much evidence for the “purely”. Chalmers doesn’t disbelieve in the easy problem aspects of conscious.
We’re talking about “physical processes”
We are talking about functionalism—it’s in the title. I am contrasting physical processes with abstract functions.
In ordinary parlance, the function of a physical thing is itself a physical effect...toasters toast, kettles boil, planes fly.
In the philosophy of mind, a function is an abstraction, more like the mathematical sense of a function. In maths, a function takes some inputs and or produces some outputs. Well known examples are familiar arithmetic operations like addition, multiplication , squaring, and so on. But the inputs and outputs are not concrete physical realities. In computation,the inputs and outputs of a functional unit, such as a NAND gate, always have some concrete value, some specific voltage, but not always the same one. Indeed, general Turing complete computers don’t even have to be electrical—they can be implemented in clockwork, hydraulics, photonics, etc.
This is the basis for the idea that a compute programme can be the same as a mind, despite being made of different matter—it implements the same.abstract functions. The abstraction of the abstract, philosopy-of-mind concept of a function is part of its usefulness.
Searle is famous critic of computationalism, and his substitute for it is a biological essentialism in which the generation of consciousness is a brain function—in the concrete sense of function.It’s true that something whose concrete function is to generate consciousness will generate consciousness..but it’s vacuously, trivially true.
The point is that the functions which this physical process is implementing are what’s required for consciousness not the actual physical properties themselves.
If you mean that abstract, computational functions are known to be sufficient to give rise to all.asoexs of consciousness including qualia, that is what I am contesting.
I think I’m more optimistic than you that a moderately accurate functional isomorph of the brain could be built which preserves consciousness (largely due to the reasons I mentioned in my previous comment around robustness.
I’m less optimistic because of my.arguments.
But putting this aside for a second, would you agree that if all the relevant functions could be implemented in silicon then a functional isomorph would be conscious?
No, not necessarily. That , in the “not necessary” form—is what I’ve been arguing all along. I also don’t think that consciousnes had a single meaning , or that there is a agreement about what it means, or that it is a simple binary.
The controversial point is whether consciousness in the hard problem sense—phenomenal consciousness, qualia—will be reproduced with reproduction of function. It’s not controversial that easy problem consciousness—capacities and behaviour—will be reproduced by functional reproduction. I don t know which you believe, because you are only talking about consciousness not otherwise specified.
If you do mean that a functional duplicate will necessarily have phenomenal consciousness, and you are arguing the point, not just holding it as an opinion, you have a heavy burden:-
You need to show some theory of how computation generates conscious experience. Or you need to show why the concrete physical implementation couldn’t possibly make a difference.
@rife
Yes, I’m specifically focused on the behaviour of an honest self-report
Well,. you’re not rejecting phenomenal consciousness wholesale.
fine-grained information becomes irrelevant implementation details. If the neuron still fires, or doesn’t, smaller noise doesn’t matter. The only reason I point this out is specifically as it applies to the behaviour of a self-report (which we will circle back to in a moment). If it doesn’t materially effect the output so powerfully that it alters that final outcome, then it is not responsible for outward behaviour.
But outward behaviour is not what I am talking about. The question is whether functional duplication preserves (full) consciousness. And, as I have said, physicalism is not just about fine grained details. There’s also the basic fact of running on the metal
I’m saying that we have ruled out that a functional duplicate could lack conscious experience because: we have established conscious experience as part of the causal chain
“In humans”. Even if it’s always the case that qualia are causal in humans, it doesn’t follow that reports of qualia in any entity whatsoever are caused by qualia. Yudkowsky’s argument is no help here, because he doesn’t require reports of consciousness to be *directly” caused by consciousness—a computational zombies reports would be caused , not by it’s own consciousness, but by the programming and data created by humans.
to be able to feel something and then output a description through voice or typing that is based on that feeling. If conscious experience was part of that causal chain, and the causal chain consists purely of neuron firings, then conscious experience is contained in that functionality.
Neural firings are specific physical behaviour, not abstract function. Computationalism is about abstract function
Then the phenomenon could be stem from punctuation habits, as @bfinn says. Did you notice that my original comment doesn’t contain a sentence, by your standards?